diff options
| author | Maarten Lankhorst <maarten.lankhorst@linux.intel.com> | 2020-06-29 12:15:51 +0200 |
|---|---|---|
| committer | Maarten Lankhorst <maarten.lankhorst@linux.intel.com> | 2020-06-29 12:16:26 +0200 |
| commit | 60e9eabf41fa916d2ef68c5bf929197975917578 (patch) | |
| tree | 39ce456390ed34d2624aed1260203f43fff94d38 /drivers/gpu/drm/amd | |
| parent | 84e543bc9d1dc550132ba25b72df28d40cc44333 (diff) | |
| parent | 0a19b068acc47d05212f03e494381926dc0381e2 (diff) | |
| download | lwn-60e9eabf41fa916d2ef68c5bf929197975917578.tar.gz lwn-60e9eabf41fa916d2ef68c5bf929197975917578.zip | |
Backmerge remote-tracking branch 'drm/drm-next' into drm-misc-next
Some conflicts with ttm_bo->offset removal, but drm-misc-next needs updating to v5.8.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Diffstat (limited to 'drivers/gpu/drm/amd')
363 files changed, 13643 insertions, 6803 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile index c2bbcdd9c875..210d57a4afc8 100644 --- a/drivers/gpu/drm/amd/amdgpu/Makefile +++ b/drivers/gpu/drm/amd/amdgpu/Makefile @@ -55,7 +55,7 @@ amdgpu-y += amdgpu_device.o amdgpu_kms.o \ amdgpu_vf_error.o amdgpu_sched.o amdgpu_debugfs.o amdgpu_ids.o \ amdgpu_gmc.o amdgpu_mmhub.o amdgpu_xgmi.o amdgpu_csa.o amdgpu_ras.o amdgpu_vm_cpu.o \ amdgpu_vm_sdma.o amdgpu_discovery.o amdgpu_ras_eeprom.o amdgpu_nbio.o \ - amdgpu_umc.o smu_v11_0_i2c.o + amdgpu_umc.o smu_v11_0_i2c.o amdgpu_fru_eeprom.o amdgpu-$(CONFIG_PERF_EVENTS) += amdgpu_pmu.o diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 2992a49ad4a5..cd913986863e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -28,6 +28,18 @@ #ifndef __AMDGPU_H__ #define __AMDGPU_H__ +#ifdef pr_fmt +#undef pr_fmt +#endif + +#define pr_fmt(fmt) "amdgpu: " fmt + +#ifdef dev_fmt +#undef dev_fmt +#endif + +#define dev_fmt(fmt) "amdgpu: " fmt + #include "amdgpu_ctx.h" #include <linux/atomic.h> @@ -161,6 +173,7 @@ extern int amdgpu_gpu_recovery; extern int amdgpu_emu_mode; extern uint amdgpu_smu_memory_pool_size; extern uint amdgpu_dc_feature_mask; +extern uint amdgpu_dc_debug_mask; extern uint amdgpu_dm_abm_level; extern struct amdgpu_mgpu_info mgpu_info; extern int amdgpu_ras_enable; @@ -177,6 +190,8 @@ extern int sched_policy; static const int sched_policy = KFD_SCHED_POLICY_HWS; #endif +extern int amdgpu_tmz; + #ifdef CONFIG_DRM_AMDGPU_SI extern int amdgpu_si_support; #endif @@ -190,8 +205,6 @@ extern int amdgpu_cik_support; #define AMDGPU_WAIT_IDLE_TIMEOUT_IN_MS 3000 #define AMDGPU_MAX_USEC_TIMEOUT 100000 /* 100 ms */ #define AMDGPU_FENCE_JIFFIES_TIMEOUT (HZ / 2) -/* AMDGPU_IB_POOL_SIZE must be a power of 2 */ -#define AMDGPU_IB_POOL_SIZE 16 #define AMDGPU_DEBUGFS_MAX_COMPONENTS 32 #define AMDGPUFB_CONN_LIMIT 4 #define AMDGPU_BIOS_NUM_SCRATCH 16 @@ -439,7 +452,9 @@ struct amdgpu_fpriv { int amdgpu_file_to_fpriv(struct file *filp, struct amdgpu_fpriv **fpriv); int amdgpu_ib_get(struct amdgpu_device *adev, struct amdgpu_vm *vm, - unsigned size, struct amdgpu_ib *ib); + unsigned size, + enum amdgpu_ib_pool_type pool, + struct amdgpu_ib *ib); void amdgpu_ib_free(struct amdgpu_device *adev, struct amdgpu_ib *ib, struct dma_fence *f); int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, @@ -512,7 +527,7 @@ static inline void amdgpu_set_ib_value(struct amdgpu_cs_parser *p, /* * Writeback */ -#define AMDGPU_MAX_WB 128 /* Reserve at most 128 WB slots for amdgpu-owned rings. */ +#define AMDGPU_MAX_WB 256 /* Reserve at most 256 WB slots for amdgpu-owned rings. */ struct amdgpu_wb { struct amdgpu_bo *wb_obj; @@ -724,6 +739,7 @@ struct amdgpu_device { uint32_t rev_id; uint32_t external_rev_id; unsigned long flags; + unsigned long apu_flags; int usec_timeout; const struct amdgpu_asic_funcs *asic_funcs; bool shutdown; @@ -751,7 +767,6 @@ struct amdgpu_device { uint8_t *bios; uint32_t bios_size; struct amdgpu_bo *stolen_vga_memory; - struct amdgpu_bo *discovery_memory; uint32_t bios_scratch_reg_offset; uint32_t bios_scratch[AMDGPU_BIOS_NUM_SCRATCH]; @@ -843,7 +858,8 @@ struct amdgpu_device { unsigned num_rings; struct amdgpu_ring *rings[AMDGPU_MAX_RINGS]; bool ib_pool_ready; - struct amdgpu_sa_manager ring_tmp_bo; + struct amdgpu_sa_manager ib_pools[AMDGPU_IB_POOL_MAX]; + struct amdgpu_sched gpu_sched[AMDGPU_HW_IP_NUM][AMDGPU_RING_PRIO_MAX]; /* interrupts */ struct amdgpu_irq irq; @@ -903,7 +919,9 @@ struct amdgpu_device { struct amdgpu_display_manager dm; /* discovery */ - uint8_t *discovery; + uint8_t *discovery_bin; + uint32_t discovery_tmr_size; + struct amdgpu_bo *discovery_memory; /* mes */ bool enable_mes; @@ -923,7 +941,7 @@ struct amdgpu_device { atomic64_t gart_pin_size; /* soc15 register offset based on ip, instance and segment */ - uint32_t *reg_offset[MAX_HWIP][HWIP_MAX_INSTANCE]; + uint32_t *reg_offset[MAX_HWIP][HWIP_MAX_INSTANCE]; /* delayed work_func for deferring clockgating during resume */ struct delayed_work delayed_init_work; @@ -935,9 +953,6 @@ struct amdgpu_device { /* link all shadow bo */ struct list_head shadow_list; struct mutex shadow_list_lock; - /* keep an lru list of rings by HW IP */ - struct list_head ring_lru_list; - spinlock_t ring_lru_list_lock; /* record hw reset is performed */ bool has_hw_reset; @@ -945,9 +960,8 @@ struct amdgpu_device { /* s3/s4 mask */ bool in_suspend; + bool in_hibernate; - /* record last mm index being written through WREG32*/ - unsigned long last_mm_index; bool in_gpu_reset; enum pp_mp1_state mp1_state; struct mutex lock_reset; @@ -966,14 +980,19 @@ struct amdgpu_device { uint64_t unique_id; uint64_t df_perfmon_config_assign_mask[AMDGPU_MAX_DF_PERFMONS]; - /* device pstate */ - int pstate; /* enable runtime pm on the device */ bool runpm; bool in_runpm; bool pm_sysfs_en; bool ucode_sysfs_en; + + /* Chip product information */ + char product_number[16]; + char product_name[32]; + char serial[16]; + + struct amdgpu_autodump autodump; }; static inline struct amdgpu_device *amdgpu_ttm_adev(struct ttm_bo_device *bdev) @@ -990,10 +1009,10 @@ int amdgpu_gpu_wait_for_idle(struct amdgpu_device *adev); void amdgpu_device_vram_access(struct amdgpu_device *adev, loff_t pos, uint32_t *buf, size_t size, bool write); -uint32_t amdgpu_mm_rreg(struct amdgpu_device *adev, uint32_t reg, +uint32_t amdgpu_device_rreg(struct amdgpu_device *adev, uint32_t reg, + uint32_t acc_flags); +void amdgpu_device_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v, uint32_t acc_flags); -void amdgpu_mm_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v, - uint32_t acc_flags); void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev, uint32_t reg, uint32_t v, uint32_t acc_flags); void amdgpu_mm_wreg8(struct amdgpu_device *adev, uint32_t offset, uint8_t value); @@ -1010,25 +1029,20 @@ int emu_soc_asic_init(struct amdgpu_device *adev); /* * Registers read & write functions. */ - -#define AMDGPU_REGS_IDX (1<<0) #define AMDGPU_REGS_NO_KIQ (1<<1) -#define AMDGPU_REGS_KIQ (1<<2) -#define RREG32_NO_KIQ(reg) amdgpu_mm_rreg(adev, (reg), AMDGPU_REGS_NO_KIQ) -#define WREG32_NO_KIQ(reg, v) amdgpu_mm_wreg(adev, (reg), (v), AMDGPU_REGS_NO_KIQ) +#define RREG32_NO_KIQ(reg) amdgpu_device_rreg(adev, (reg), AMDGPU_REGS_NO_KIQ) +#define WREG32_NO_KIQ(reg, v) amdgpu_device_wreg(adev, (reg), (v), AMDGPU_REGS_NO_KIQ) -#define RREG32_KIQ(reg) amdgpu_mm_rreg(adev, (reg), AMDGPU_REGS_KIQ) -#define WREG32_KIQ(reg, v) amdgpu_mm_wreg(adev, (reg), (v), AMDGPU_REGS_KIQ) +#define RREG32_KIQ(reg) amdgpu_kiq_rreg(adev, (reg)) +#define WREG32_KIQ(reg, v) amdgpu_kiq_wreg(adev, (reg), (v)) #define RREG8(reg) amdgpu_mm_rreg8(adev, (reg)) #define WREG8(reg, v) amdgpu_mm_wreg8(adev, (reg), (v)) -#define RREG32(reg) amdgpu_mm_rreg(adev, (reg), 0) -#define RREG32_IDX(reg) amdgpu_mm_rreg(adev, (reg), AMDGPU_REGS_IDX) -#define DREG32(reg) printk(KERN_INFO "REGISTER: " #reg " : 0x%08X\n", amdgpu_mm_rreg(adev, (reg), 0)) -#define WREG32(reg, v) amdgpu_mm_wreg(adev, (reg), (v), 0) -#define WREG32_IDX(reg, v) amdgpu_mm_wreg(adev, (reg), (v), AMDGPU_REGS_IDX) +#define RREG32(reg) amdgpu_device_rreg(adev, (reg), 0) +#define DREG32(reg) printk(KERN_INFO "REGISTER: " #reg " : 0x%08X\n", amdgpu_device_rreg(adev, (reg), 0)) +#define WREG32(reg, v) amdgpu_device_wreg(adev, (reg), (v), 0) #define REG_SET(FIELD, v) (((v) << FIELD##_SHIFT) & FIELD##_MASK) #define REG_GET(FIELD, v) (((v) << FIELD##_SHIFT) & FIELD##_MASK) #define RREG32_PCIE(reg) adev->pcie_rreg(adev, (reg)) @@ -1065,7 +1079,7 @@ int emu_soc_asic_init(struct amdgpu_device *adev); tmp_ |= ((val) & ~(mask)); \ WREG32_PLL(reg, tmp_); \ } while (0) -#define DREG32_SYS(sqf, adev, reg) seq_printf((sqf), #reg " : 0x%08X\n", amdgpu_mm_rreg((adev), (reg), false)) +#define DREG32_SYS(sqf, adev, reg) seq_printf((sqf), #reg " : 0x%08X\n", amdgpu_device_rreg((adev), (reg), false)) #define RREG32_IO(reg) amdgpu_io_rreg(adev, (reg)) #define WREG32_IO(reg, v) amdgpu_io_wreg(adev, (reg), (v)) @@ -1248,5 +1262,9 @@ _name##_show(struct device *dev, \ \ static struct device_attribute pmu_attr_##_name = __ATTR_RO(_name) -#endif +static inline bool amdgpu_is_tmz(struct amdgpu_device *adev) +{ + return adev->gmc.tmz_enabled; +} +#endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c index 1e41367ef74e..956cbbda4793 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c @@ -444,7 +444,6 @@ static int amdgpu_atif_handler(struct amdgpu_device *adev, DRM_DEBUG_DRIVER("ATIF: %d pending SBIOS requests\n", count); - /* todo: add DC handling */ if ((req.pending & ATIF_PANEL_BRIGHTNESS_CHANGE_REQUEST) && !amdgpu_device_has_dc_support(adev)) { struct amdgpu_encoder *enc = atif->encoder_for_bl; @@ -463,6 +462,27 @@ static int amdgpu_atif_handler(struct amdgpu_device *adev, #endif } } +#if defined(CONFIG_DRM_AMD_DC) +#if defined(CONFIG_BACKLIGHT_CLASS_DEVICE) || defined(CONFIG_BACKLIGHT_CLASS_DEVICE_MODULE) + if ((req.pending & ATIF_PANEL_BRIGHTNESS_CHANGE_REQUEST) && + amdgpu_device_has_dc_support(adev)) { + struct amdgpu_display_manager *dm = &adev->dm; + struct backlight_device *bd = dm->backlight_dev; + + if (bd) { + DRM_DEBUG_DRIVER("Changing brightness to %d\n", + req.backlight_level); + + /* + * XXX backlight_device_set_brightness() is + * hardwired to post BACKLIGHT_UPDATE_SYSFS. + * It probably should accept 'reason' parameter. + */ + backlight_device_set_brightness(bd, req.backlight_level); + } + } +#endif +#endif if (req.pending & ATIF_DGPU_DISPLAY_EVENT) { if (adev->flags & AMD_IS_PX) { pm_runtime_get_sync(adev->ddev->dev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index abfbe89e805e..ad59ac4423b8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -564,6 +564,13 @@ uint32_t amdgpu_amdkfd_get_num_gws(struct kgd_dev *kgd) return adev->gds.gws_size; } +uint32_t amdgpu_amdkfd_get_asic_rev_id(struct kgd_dev *kgd) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)kgd; + + return adev->rev_id; +} + int amdgpu_amdkfd_submit_ib(struct kgd_dev *kgd, enum kgd_engine_type engine, uint32_t vmid, uint64_t gpu_addr, uint32_t *ib_cmd, uint32_t ib_len) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index 13feb313e9b3..ffe149aafc39 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -27,6 +27,7 @@ #include <linux/types.h> #include <linux/mm.h> +#include <linux/kthread.h> #include <linux/workqueue.h> #include <kgd_kfd_interface.h> #include <drm/ttm/ttm_execbuf_util.h> @@ -65,6 +66,7 @@ struct kgd_mem { struct amdgpu_sync sync; bool aql_queue; + bool is_imported; }; /* KFD Memory Eviction */ @@ -148,6 +150,9 @@ int amdgpu_amdkfd_post_reset(struct amdgpu_device *adev); void amdgpu_amdkfd_gpu_reset(struct kgd_dev *kgd); +int amdgpu_queue_mask_bit_to_set_resource_bit(struct amdgpu_device *adev, + int queue_bit); + /* Shared API */ int amdgpu_amdkfd_alloc_gtt_mem(struct kgd_dev *kgd, size_t size, void **mem_obj, uint64_t *gpu_addr, @@ -175,13 +180,14 @@ uint64_t amdgpu_amdkfd_get_hive_id(struct kgd_dev *kgd); uint64_t amdgpu_amdkfd_get_unique_id(struct kgd_dev *kgd); uint64_t amdgpu_amdkfd_get_mmio_remap_phys_addr(struct kgd_dev *kgd); uint32_t amdgpu_amdkfd_get_num_gws(struct kgd_dev *kgd); +uint32_t amdgpu_amdkfd_get_asic_rev_id(struct kgd_dev *kgd); uint8_t amdgpu_amdkfd_get_xgmi_hops_count(struct kgd_dev *dst, struct kgd_dev *src); /* Read user wptr from a specified user address space with page fault * disabled. The memory must be pinned and mapped to the hardware when * this is called in hqd_load functions, so it should never fault in * the first place. This resolves a circular lock dependency involving - * four locks, including the DQM lock and mmap_sem. + * four locks, including the DQM lock and mmap_lock. */ #define read_user_wptr(mmptr, wptr, dst) \ ({ \ @@ -190,10 +196,10 @@ uint8_t amdgpu_amdkfd_get_xgmi_hops_count(struct kgd_dev *dst, struct kgd_dev *s pagefault_disable(); \ if ((mmptr) == current->mm) { \ valid = !get_user((dst), (wptr)); \ - } else if (current->mm == NULL) { \ - use_mm(mmptr); \ + } else if (current->flags & PF_KTHREAD) { \ + kthread_use_mm(mmptr); \ valid = !get_user((dst), (wptr)); \ - unuse_mm(mmptr); \ + kthread_unuse_mm(mmptr); \ } \ pagefault_enable(); \ } \ @@ -218,7 +224,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( void *vm, struct kgd_mem **mem, uint64_t *offset, uint32_t flags); int amdgpu_amdkfd_gpuvm_free_memory_of_gpu( - struct kgd_dev *kgd, struct kgd_mem *mem); + struct kgd_dev *kgd, struct kgd_mem *mem, uint64_t *size); int amdgpu_amdkfd_gpuvm_map_memory_to_gpu( struct kgd_dev *kgd, struct kgd_mem *mem, void *vm); int amdgpu_amdkfd_gpuvm_unmap_memory_from_gpu( diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c index 6529caca88fe..35d4a5ab0228 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c @@ -22,7 +22,6 @@ #include <linux/module.h> #include <linux/fdtable.h> #include <linux/uaccess.h> -#include <linux/mmu_context.h> #include <linux/firmware.h> #include "amdgpu.h" #include "amdgpu_amdkfd.h" diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c index 4ec6d0c03201..bf927f432506 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c @@ -19,7 +19,6 @@ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. */ -#include <linux/mmu_context.h> #include "amdgpu.h" #include "amdgpu_amdkfd.h" #include "gc/gc_10_1_0_offset.h" @@ -543,6 +542,9 @@ static int kgd_hqd_destroy(struct kgd_dev *kgd, void *mqd, uint32_t temp; struct v10_compute_mqd *m = get_mqd(mqd); + if (adev->in_gpu_reset) + return -EIO; + #if 0 unsigned long flags; int retry; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c index 0b7e78748540..744366c7ee85 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c @@ -20,8 +20,6 @@ * OTHER DEALINGS IN THE SOFTWARE. */ -#include <linux/mmu_context.h> - #include "amdgpu.h" #include "amdgpu_amdkfd.h" #include "cikd.h" @@ -237,7 +235,7 @@ static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id, CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_EN, 1); WREG32(mmCP_HQD_PQ_DOORBELL_CONTROL, data); - /* read_user_ptr may take the mm->mmap_sem. + /* read_user_ptr may take the mm->mmap_lock. * release srbm_mutex to avoid circular dependency between * srbm_mutex->mm_sem->reservation_ww_class_mutex->srbm_mutex. */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c index ccd635b812b5..feab4cc6e836 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c @@ -20,8 +20,6 @@ * OTHER DEALINGS IN THE SOFTWARE. */ -#include <linux/mmu_context.h> - #include "amdgpu.h" #include "amdgpu_amdkfd.h" #include "gfx_v8_0.h" @@ -224,7 +222,7 @@ static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id, CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_EN, 1); WREG32(mmCP_HQD_PQ_DOORBELL_CONTROL, data); - /* read_user_ptr may take the mm->mmap_sem. + /* read_user_ptr may take the mm->mmap_lock. * release srbm_mutex to avoid circular dependency between * srbm_mutex->mm_sem->reservation_ww_class_mutex->srbm_mutex. */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c index df841c2ac5e7..c7fd0c47b254 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c @@ -19,8 +19,6 @@ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. */ -#include <linux/mmu_context.h> - #include "amdgpu.h" #include "amdgpu_amdkfd.h" #include "gc/gc_9_0_offset.h" diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index 9dff792c9290..9015c7b76d60 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -362,13 +362,13 @@ static int vm_validate_pt_pd_bos(struct amdgpu_vm *vm) ret = amdgpu_vm_validate_pt_bos(adev, vm, amdgpu_amdkfd_validate, ¶m); if (ret) { - pr_err("amdgpu: failed to validate PT BOs\n"); + pr_err("failed to validate PT BOs\n"); return ret; } ret = amdgpu_amdkfd_validate(¶m, pd); if (ret) { - pr_err("amdgpu: failed to validate PD\n"); + pr_err("failed to validate PD\n"); return ret; } @@ -377,7 +377,7 @@ static int vm_validate_pt_pd_bos(struct amdgpu_vm *vm) if (vm->use_cpu_for_update) { ret = amdgpu_bo_kmap(pd, NULL); if (ret) { - pr_err("amdgpu: failed to kmap PD, ret=%d\n", ret); + pr_err("failed to kmap PD, ret=%d\n", ret); return ret; } } @@ -660,15 +660,15 @@ static int reserve_bo_and_vm(struct kgd_mem *mem, ret = ttm_eu_reserve_buffers(&ctx->ticket, &ctx->list, false, &ctx->duplicates); - if (!ret) - ctx->reserved = true; - else { - pr_err("Failed to reserve buffers in ttm\n"); + if (ret) { + pr_err("Failed to reserve buffers in ttm.\n"); kfree(ctx->vm_pd); ctx->vm_pd = NULL; + return ret; } - return ret; + ctx->reserved = true; + return 0; } /** @@ -733,17 +733,15 @@ static int reserve_bo_and_cond_vms(struct kgd_mem *mem, ret = ttm_eu_reserve_buffers(&ctx->ticket, &ctx->list, false, &ctx->duplicates); - if (!ret) - ctx->reserved = true; - else - pr_err("Failed to reserve buffers in ttm.\n"); - if (ret) { + pr_err("Failed to reserve buffers in ttm.\n"); kfree(ctx->vm_pd); ctx->vm_pd = NULL; + return ret; } - return ret; + ctx->reserved = true; + return 0; } /** @@ -1279,31 +1277,30 @@ err: } int amdgpu_amdkfd_gpuvm_free_memory_of_gpu( - struct kgd_dev *kgd, struct kgd_mem *mem) + struct kgd_dev *kgd, struct kgd_mem *mem, uint64_t *size) { struct amdkfd_process_info *process_info = mem->process_info; unsigned long bo_size = mem->bo->tbo.mem.size; struct kfd_bo_va_list *entry, *tmp; struct bo_vm_reservation_context ctx; struct ttm_validate_buffer *bo_list_entry; + unsigned int mapped_to_gpu_memory; int ret; + bool is_imported = 0; mutex_lock(&mem->lock); - - if (mem->mapped_to_gpu_memory > 0) { - pr_debug("BO VA 0x%llx size 0x%lx is still mapped.\n", - mem->va, bo_size); - mutex_unlock(&mem->lock); - return -EBUSY; - } - + mapped_to_gpu_memory = mem->mapped_to_gpu_memory; + is_imported = mem->is_imported; mutex_unlock(&mem->lock); /* lock is not needed after this, since mem is unused and will * be freed anyway */ - /* No more MMU notifiers */ - amdgpu_mn_unregister(mem->bo); + if (mapped_to_gpu_memory > 0) { + pr_debug("BO VA 0x%llx size 0x%lx is still mapped.\n", + mem->va, bo_size); + return -EBUSY; + } /* Make sure restore workers don't access the BO any more */ bo_list_entry = &mem->validate_list; @@ -1311,6 +1308,9 @@ int amdgpu_amdkfd_gpuvm_free_memory_of_gpu( list_del(&bo_list_entry->head); mutex_unlock(&process_info->lock); + /* No more MMU notifiers */ + amdgpu_mn_unregister(mem->bo); + ret = reserve_bo_and_cond_vms(mem, NULL, BO_VM_ALL, &ctx); if (unlikely(ret)) return ret; @@ -1342,8 +1342,19 @@ int amdgpu_amdkfd_gpuvm_free_memory_of_gpu( kfree(mem->bo->tbo.sg); } + /* Update the size of the BO being freed if it was allocated from + * VRAM and is not imported. + */ + if (size) { + if ((mem->bo->preferred_domains == AMDGPU_GEM_DOMAIN_VRAM) && + (!is_imported)) + *size = bo_size; + else + *size = 0; + } + /* Free the BO*/ - amdgpu_bo_unref(&mem->bo); + drm_gem_object_put(&mem->bo->tbo.base); mutex_destroy(&mem->lock); kfree(mem); @@ -1382,9 +1393,9 @@ int amdgpu_amdkfd_gpuvm_map_memory_to_gpu( * concurrently and the queues are actually stopped */ if (amdgpu_ttm_tt_get_usermm(bo->tbo.ttm)) { - down_write(¤t->mm->mmap_sem); + mmap_write_lock(current->mm); is_invalid_userptr = atomic_read(&mem->invalid); - up_write(¤t->mm->mmap_sem); + mmap_write_unlock(current->mm); } mutex_lock(&mem->lock); @@ -1688,7 +1699,8 @@ int amdgpu_amdkfd_gpuvm_import_dmabuf(struct kgd_dev *kgd, | KFD_IOC_ALLOC_MEM_FLAGS_WRITABLE | KFD_IOC_ALLOC_MEM_FLAGS_EXECUTABLE; - (*mem)->bo = amdgpu_bo_ref(bo); + drm_gem_object_get(&bo->tbo.base); + (*mem)->bo = bo; (*mem)->va = va; (*mem)->domain = (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM) ? AMDGPU_GEM_DOMAIN_VRAM : AMDGPU_GEM_DOMAIN_GTT; @@ -1696,6 +1708,7 @@ int amdgpu_amdkfd_gpuvm_import_dmabuf(struct kgd_dev *kgd, (*mem)->process_info = avm->process_info; add_kgd_mem_to_kfd_bo_list(*mem, avm->process_info, false); amdgpu_sync_create(&(*mem)->sync); + (*mem)->is_imported = true; return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c index d1495e1c9289..d9b35df33806 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c @@ -40,7 +40,7 @@ static int amdgpu_benchmark_do_move(struct amdgpu_device *adev, unsigned size, for (i = 0; i < n; i++) { struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring; r = amdgpu_copy_buffer(ring, saddr, daddr, size, NULL, &fence, - false, false); + false, false, false); if (r) goto exit_do_move; r = dma_fence_wait(fence, false); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c index 031b094607bd..78ac6dbe70d8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c @@ -60,8 +60,6 @@ static uint32_t amdgpu_cgs_read_ind_register(struct cgs_device *cgs_device, { CGS_FUNC_ADEV; switch (space) { - case CGS_IND_REG__MMIO: - return RREG32_IDX(index); case CGS_IND_REG__PCIE: return RREG32_PCIE(index); case CGS_IND_REG__SMC: @@ -77,6 +75,8 @@ static uint32_t amdgpu_cgs_read_ind_register(struct cgs_device *cgs_device, case CGS_IND_REG__AUDIO_ENDPT: DRM_ERROR("audio endpt register access not implemented.\n"); return 0; + default: + BUG(); } WARN(1, "Invalid indirect register space"); return 0; @@ -88,8 +88,6 @@ static void amdgpu_cgs_write_ind_register(struct cgs_device *cgs_device, { CGS_FUNC_ADEV; switch (space) { - case CGS_IND_REG__MMIO: - return WREG32_IDX(index, value); case CGS_IND_REG__PCIE: return WREG32_PCIE(index, value); case CGS_IND_REG__SMC: @@ -105,6 +103,8 @@ static void amdgpu_cgs_write_ind_register(struct cgs_device *cgs_device, case CGS_IND_REG__AUDIO_ENDPT: DRM_ERROR("audio endpt register access not implemented.\n"); return; + default: + BUG(); } WARN(1, "Invalid indirect register space"); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 76ea245f2106..a25fb59c127c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -924,7 +924,8 @@ static int amdgpu_cs_ib_fill(struct amdgpu_device *adev, ring = to_amdgpu_ring(entity->rq->sched); r = amdgpu_ib_get(adev, vm, ring->funcs->parse_cs ? - chunk_ib->ib_bytes : 0, ib); + chunk_ib->ib_bytes : 0, + AMDGPU_IB_POOL_DELAYED, ib); if (r) { DRM_ERROR("Failed to get ib !\n"); return r; @@ -1207,7 +1208,6 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p, { struct amdgpu_fpriv *fpriv = p->filp->driver_priv; struct drm_sched_entity *entity = p->entity; - enum drm_sched_priority priority; struct amdgpu_bo_list_entry *e; struct amdgpu_job *job; uint64_t seq; @@ -1257,7 +1257,6 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p, trace_amdgpu_cs_ioctl(job); amdgpu_vm_bo_trace_cs(&fpriv->vm, &p->ticket); - priority = job->base.s_priority; drm_sched_entity_push_job(&job->base, entity); amdgpu_vm_move_to_lru_tail(p->adev, &fpriv->vm); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index 6ed36a2c5f73..8842c55d4490 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -26,6 +26,7 @@ #include "amdgpu.h" #include "amdgpu_sched.h" #include "amdgpu_ras.h" +#include <linux/nospec.h> #define to_amdgpu_ctx_entity(e) \ container_of((e), struct amdgpu_ctx_entity, entity) @@ -72,13 +73,30 @@ static enum gfx_pipe_priority amdgpu_ctx_sched_prio_to_compute_prio(enum drm_sch } } -static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, const u32 hw_ip, const u32 ring) +static unsigned int amdgpu_ctx_prio_sched_to_hw(struct amdgpu_device *adev, + enum drm_sched_priority prio, + u32 hw_ip) +{ + unsigned int hw_prio; + + hw_prio = (hw_ip == AMDGPU_HW_IP_COMPUTE) ? + amdgpu_ctx_sched_prio_to_compute_prio(prio) : + AMDGPU_RING_PRIO_DEFAULT; + hw_ip = array_index_nospec(hw_ip, AMDGPU_HW_IP_NUM); + if (adev->gpu_sched[hw_ip][hw_prio].num_scheds == 0) + hw_prio = AMDGPU_RING_PRIO_DEFAULT; + + return hw_prio; +} + +static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, u32 hw_ip, + const u32 ring) { struct amdgpu_device *adev = ctx->adev; struct amdgpu_ctx_entity *entity; struct drm_gpu_scheduler **scheds = NULL, *sched = NULL; unsigned num_scheds = 0; - enum gfx_pipe_priority hw_prio; + unsigned int hw_prio; enum drm_sched_priority priority; int r; @@ -90,52 +108,16 @@ static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, const u32 hw_ip, const entity->sequence = 1; priority = (ctx->override_priority == DRM_SCHED_PRIORITY_UNSET) ? ctx->init_priority : ctx->override_priority; - switch (hw_ip) { - case AMDGPU_HW_IP_GFX: - sched = &adev->gfx.gfx_ring[0].sched; - scheds = &sched; - num_scheds = 1; - break; - case AMDGPU_HW_IP_COMPUTE: - hw_prio = amdgpu_ctx_sched_prio_to_compute_prio(priority); - scheds = adev->gfx.compute_prio_sched[hw_prio]; - num_scheds = adev->gfx.num_compute_sched[hw_prio]; - break; - case AMDGPU_HW_IP_DMA: - scheds = adev->sdma.sdma_sched; - num_scheds = adev->sdma.num_sdma_sched; - break; - case AMDGPU_HW_IP_UVD: - sched = &adev->uvd.inst[0].ring.sched; - scheds = &sched; - num_scheds = 1; - break; - case AMDGPU_HW_IP_VCE: - sched = &adev->vce.ring[0].sched; - scheds = &sched; - num_scheds = 1; - break; - case AMDGPU_HW_IP_UVD_ENC: - sched = &adev->uvd.inst[0].ring_enc[0].sched; - scheds = &sched; - num_scheds = 1; - break; - case AMDGPU_HW_IP_VCN_DEC: - sched = drm_sched_pick_best(adev->vcn.vcn_dec_sched, - adev->vcn.num_vcn_dec_sched); - scheds = &sched; - num_scheds = 1; - break; - case AMDGPU_HW_IP_VCN_ENC: - sched = drm_sched_pick_best(adev->vcn.vcn_enc_sched, - adev->vcn.num_vcn_enc_sched); + hw_prio = amdgpu_ctx_prio_sched_to_hw(adev, priority, hw_ip); + + hw_ip = array_index_nospec(hw_ip, AMDGPU_HW_IP_NUM); + scheds = adev->gpu_sched[hw_ip][hw_prio].sched; + num_scheds = adev->gpu_sched[hw_ip][hw_prio].num_scheds; + + if (hw_ip == AMDGPU_HW_IP_VCN_ENC || hw_ip == AMDGPU_HW_IP_VCN_DEC) { + sched = drm_sched_pick_best(scheds, num_scheds); scheds = &sched; num_scheds = 1; - break; - case AMDGPU_HW_IP_VCN_JPEG: - scheds = adev->jpeg.jpeg_sched; - num_scheds = adev->jpeg.num_jpeg_sched; - break; } r = drm_sched_entity_init(&entity->entity, priority, scheds, num_scheds, @@ -178,7 +160,6 @@ static int amdgpu_ctx_init(struct amdgpu_device *adev, ctx->override_priority = DRM_SCHED_PRIORITY_UNSET; return 0; - } static void amdgpu_ctx_fini_entity(struct amdgpu_ctx_entity *entity) @@ -525,7 +506,7 @@ static void amdgpu_ctx_set_entity_priority(struct amdgpu_ctx *ctx, enum drm_sched_priority priority) { struct amdgpu_device *adev = ctx->adev; - enum gfx_pipe_priority hw_prio; + unsigned int hw_prio; struct drm_gpu_scheduler **scheds = NULL; unsigned num_scheds; @@ -534,9 +515,11 @@ static void amdgpu_ctx_set_entity_priority(struct amdgpu_ctx *ctx, /* set hw priority */ if (hw_ip == AMDGPU_HW_IP_COMPUTE) { - hw_prio = amdgpu_ctx_sched_prio_to_compute_prio(priority); - scheds = adev->gfx.compute_prio_sched[hw_prio]; - num_scheds = adev->gfx.num_compute_sched[hw_prio]; + hw_prio = amdgpu_ctx_prio_sched_to_hw(adev, priority, + AMDGPU_HW_IP_COMPUTE); + hw_prio = array_index_nospec(hw_prio, AMDGPU_RING_PRIO_MAX); + scheds = adev->gpu_sched[hw_ip][hw_prio].sched; + num_scheds = adev->gpu_sched[hw_ip][hw_prio].num_scheds; drm_sched_entity_modify_sched(&aentity->entity, scheds, num_scheds); } @@ -665,78 +648,3 @@ void amdgpu_ctx_mgr_fini(struct amdgpu_ctx_mgr *mgr) idr_destroy(&mgr->ctx_handles); mutex_destroy(&mgr->lock); } - - -static void amdgpu_ctx_init_compute_sched(struct amdgpu_device *adev) -{ - int num_compute_sched_normal = 0; - int num_compute_sched_high = AMDGPU_MAX_COMPUTE_RINGS - 1; - int i; - - /* use one drm sched array, gfx.compute_sched to store both high and - * normal priority drm compute schedulers */ - for (i = 0; i < adev->gfx.num_compute_rings; i++) { - if (!adev->gfx.compute_ring[i].has_high_prio) - adev->gfx.compute_sched[num_compute_sched_normal++] = - &adev->gfx.compute_ring[i].sched; - else - adev->gfx.compute_sched[num_compute_sched_high--] = - &adev->gfx.compute_ring[i].sched; - } - - /* compute ring only has two priority for now */ - i = AMDGPU_GFX_PIPE_PRIO_NORMAL; - adev->gfx.compute_prio_sched[i] = &adev->gfx.compute_sched[0]; - adev->gfx.num_compute_sched[i] = num_compute_sched_normal; - - i = AMDGPU_GFX_PIPE_PRIO_HIGH; - if (num_compute_sched_high == (AMDGPU_MAX_COMPUTE_RINGS - 1)) { - /* When compute has no high priority rings then use */ - /* normal priority sched array */ - adev->gfx.compute_prio_sched[i] = &adev->gfx.compute_sched[0]; - adev->gfx.num_compute_sched[i] = num_compute_sched_normal; - } else { - adev->gfx.compute_prio_sched[i] = - &adev->gfx.compute_sched[num_compute_sched_high - 1]; - adev->gfx.num_compute_sched[i] = - adev->gfx.num_compute_rings - num_compute_sched_normal; - } -} - -void amdgpu_ctx_init_sched(struct amdgpu_device *adev) -{ - int i, j; - - amdgpu_ctx_init_compute_sched(adev); - for (i = 0; i < adev->gfx.num_gfx_rings; i++) { - adev->gfx.gfx_sched[i] = &adev->gfx.gfx_ring[i].sched; - adev->gfx.num_gfx_sched++; - } - - for (i = 0; i < adev->sdma.num_instances; i++) { - adev->sdma.sdma_sched[i] = &adev->sdma.instance[i].ring.sched; - adev->sdma.num_sdma_sched++; - } - - for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { - if (adev->vcn.harvest_config & (1 << i)) - continue; - adev->vcn.vcn_dec_sched[adev->vcn.num_vcn_dec_sched++] = - &adev->vcn.inst[i].ring_dec.sched; - } - - for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { - if (adev->vcn.harvest_config & (1 << i)) - continue; - for (j = 0; j < adev->vcn.num_enc_rings; ++j) - adev->vcn.vcn_enc_sched[adev->vcn.num_vcn_enc_sched++] = - &adev->vcn.inst[i].ring_enc[j].sched; - } - - for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { - if (adev->jpeg.harvest_config & (1 << i)) - continue; - adev->jpeg.jpeg_sched[adev->jpeg.num_jpeg_sched++] = - &adev->jpeg.inst[i].ring_dec.sched; - } -} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h index de490f183af2..f54e10314661 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h @@ -88,7 +88,4 @@ void amdgpu_ctx_mgr_entity_fini(struct amdgpu_ctx_mgr *mgr); long amdgpu_ctx_mgr_entity_flush(struct amdgpu_ctx_mgr *mgr, long timeout); void amdgpu_ctx_mgr_fini(struct amdgpu_ctx_mgr *mgr); -void amdgpu_ctx_init_sched(struct amdgpu_device *adev); - - #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c index c0f9a651dc06..d33cb344be69 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c @@ -27,7 +27,7 @@ #include <linux/pci.h> #include <linux/uaccess.h> #include <linux/pm_runtime.h> - +#include <linux/poll.h> #include <drm/drm_debugfs.h> #include "amdgpu.h" @@ -74,8 +74,82 @@ int amdgpu_debugfs_add_files(struct amdgpu_device *adev, return 0; } +int amdgpu_debugfs_wait_dump(struct amdgpu_device *adev) +{ +#if defined(CONFIG_DEBUG_FS) + unsigned long timeout = 600 * HZ; + int ret; + + wake_up_interruptible(&adev->autodump.gpu_hang); + + ret = wait_for_completion_interruptible_timeout(&adev->autodump.dumping, timeout); + if (ret == 0) { + pr_err("autodump: timeout, move on to gpu recovery\n"); + return -ETIMEDOUT; + } +#endif + return 0; +} + #if defined(CONFIG_DEBUG_FS) +static int amdgpu_debugfs_autodump_open(struct inode *inode, struct file *file) +{ + struct amdgpu_device *adev = inode->i_private; + int ret; + + file->private_data = adev; + + mutex_lock(&adev->lock_reset); + if (adev->autodump.dumping.done) { + reinit_completion(&adev->autodump.dumping); + ret = 0; + } else { + ret = -EBUSY; + } + mutex_unlock(&adev->lock_reset); + + return ret; +} + +static int amdgpu_debugfs_autodump_release(struct inode *inode, struct file *file) +{ + struct amdgpu_device *adev = file->private_data; + + complete_all(&adev->autodump.dumping); + return 0; +} + +static unsigned int amdgpu_debugfs_autodump_poll(struct file *file, struct poll_table_struct *poll_table) +{ + struct amdgpu_device *adev = file->private_data; + + poll_wait(file, &adev->autodump.gpu_hang, poll_table); + + if (adev->in_gpu_reset) + return POLLIN | POLLRDNORM | POLLWRNORM; + + return 0; +} + +static const struct file_operations autodump_debug_fops = { + .owner = THIS_MODULE, + .open = amdgpu_debugfs_autodump_open, + .poll = amdgpu_debugfs_autodump_poll, + .release = amdgpu_debugfs_autodump_release, +}; + +static void amdgpu_debugfs_autodump_init(struct amdgpu_device *adev) +{ + init_completion(&adev->autodump.dumping); + complete_all(&adev->autodump.dumping); + init_waitqueue_head(&adev->autodump.gpu_hang); + + debugfs_create_file("amdgpu_autodump", 0600, + adev->ddev->primary->debugfs_root, + adev, &autodump_debug_fops); +} + /** * amdgpu_debugfs_process_reg_op - Handle MMIO register reads/writes * @@ -152,11 +226,16 @@ static int amdgpu_debugfs_process_reg_op(bool read, struct file *f, if (r < 0) return r; + r = amdgpu_virt_enable_access_debugfs(adev); + if (r < 0) + return r; + if (use_bank) { if ((sh_bank != 0xFFFFFFFF && sh_bank >= adev->gfx.config.max_sh_per_se) || (se_bank != 0xFFFFFFFF && se_bank >= adev->gfx.config.max_shader_engines)) { pm_runtime_mark_last_busy(adev->ddev->dev); pm_runtime_put_autosuspend(adev->ddev->dev); + amdgpu_virt_disable_access_debugfs(adev); return -EINVAL; } mutex_lock(&adev->grbm_idx_mutex); @@ -207,6 +286,7 @@ end: pm_runtime_mark_last_busy(adev->ddev->dev); pm_runtime_put_autosuspend(adev->ddev->dev); + amdgpu_virt_disable_access_debugfs(adev); return result; } @@ -255,6 +335,10 @@ static ssize_t amdgpu_debugfs_regs_pcie_read(struct file *f, char __user *buf, if (r < 0) return r; + r = amdgpu_virt_enable_access_debugfs(adev); + if (r < 0) + return r; + while (size) { uint32_t value; @@ -263,6 +347,7 @@ static ssize_t amdgpu_debugfs_regs_pcie_read(struct file *f, char __user *buf, if (r) { pm_runtime_mark_last_busy(adev->ddev->dev); pm_runtime_put_autosuspend(adev->ddev->dev); + amdgpu_virt_disable_access_debugfs(adev); return r; } @@ -275,6 +360,7 @@ static ssize_t amdgpu_debugfs_regs_pcie_read(struct file *f, char __user *buf, pm_runtime_mark_last_busy(adev->ddev->dev); pm_runtime_put_autosuspend(adev->ddev->dev); + amdgpu_virt_disable_access_debugfs(adev); return result; } @@ -304,6 +390,10 @@ static ssize_t amdgpu_debugfs_regs_pcie_write(struct file *f, const char __user if (r < 0) return r; + r = amdgpu_virt_enable_access_debugfs(adev); + if (r < 0) + return r; + while (size) { uint32_t value; @@ -311,6 +401,7 @@ static ssize_t amdgpu_debugfs_regs_pcie_write(struct file *f, const char __user if (r) { pm_runtime_mark_last_busy(adev->ddev->dev); pm_runtime_put_autosuspend(adev->ddev->dev); + amdgpu_virt_disable_access_debugfs(adev); return r; } @@ -325,6 +416,7 @@ static ssize_t amdgpu_debugfs_regs_pcie_write(struct file *f, const char __user pm_runtime_mark_last_busy(adev->ddev->dev); pm_runtime_put_autosuspend(adev->ddev->dev); + amdgpu_virt_disable_access_debugfs(adev); return result; } @@ -354,6 +446,10 @@ static ssize_t amdgpu_debugfs_regs_didt_read(struct file *f, char __user *buf, if (r < 0) return r; + r = amdgpu_virt_enable_access_debugfs(adev); + if (r < 0) + return r; + while (size) { uint32_t value; @@ -362,6 +458,7 @@ static ssize_t amdgpu_debugfs_regs_didt_read(struct file *f, char __user *buf, if (r) { pm_runtime_mark_last_busy(adev->ddev->dev); pm_runtime_put_autosuspend(adev->ddev->dev); + amdgpu_virt_disable_access_debugfs(adev); return r; } @@ -374,6 +471,7 @@ static ssize_t amdgpu_debugfs_regs_didt_read(struct file *f, char __user *buf, pm_runtime_mark_last_busy(adev->ddev->dev); pm_runtime_put_autosuspend(adev->ddev->dev); + amdgpu_virt_disable_access_debugfs(adev); return result; } @@ -403,6 +501,10 @@ static ssize_t amdgpu_debugfs_regs_didt_write(struct file *f, const char __user if (r < 0) return r; + r = amdgpu_virt_enable_access_debugfs(adev); + if (r < 0) + return r; + while (size) { uint32_t value; @@ -410,6 +512,7 @@ static ssize_t amdgpu_debugfs_regs_didt_write(struct file *f, const char __user if (r) { pm_runtime_mark_last_busy(adev->ddev->dev); pm_runtime_put_autosuspend(adev->ddev->dev); + amdgpu_virt_disable_access_debugfs(adev); return r; } @@ -424,6 +527,7 @@ static ssize_t amdgpu_debugfs_regs_didt_write(struct file *f, const char __user pm_runtime_mark_last_busy(adev->ddev->dev); pm_runtime_put_autosuspend(adev->ddev->dev); + amdgpu_virt_disable_access_debugfs(adev); return result; } @@ -453,6 +557,10 @@ static ssize_t amdgpu_debugfs_regs_smc_read(struct file *f, char __user *buf, if (r < 0) return r; + r = amdgpu_virt_enable_access_debugfs(adev); + if (r < 0) + return r; + while (size) { uint32_t value; @@ -461,6 +569,7 @@ static ssize_t amdgpu_debugfs_regs_smc_read(struct file *f, char __user *buf, if (r) { pm_runtime_mark_last_busy(adev->ddev->dev); pm_runtime_put_autosuspend(adev->ddev->dev); + amdgpu_virt_disable_access_debugfs(adev); return r; } @@ -473,6 +582,7 @@ static ssize_t amdgpu_debugfs_regs_smc_read(struct file *f, char __user *buf, pm_runtime_mark_last_busy(adev->ddev->dev); pm_runtime_put_autosuspend(adev->ddev->dev); + amdgpu_virt_disable_access_debugfs(adev); return result; } @@ -502,6 +612,10 @@ static ssize_t amdgpu_debugfs_regs_smc_write(struct file *f, const char __user * if (r < 0) return r; + r = amdgpu_virt_enable_access_debugfs(adev); + if (r < 0) + return r; + while (size) { uint32_t value; @@ -509,6 +623,7 @@ static ssize_t amdgpu_debugfs_regs_smc_write(struct file *f, const char __user * if (r) { pm_runtime_mark_last_busy(adev->ddev->dev); pm_runtime_put_autosuspend(adev->ddev->dev); + amdgpu_virt_disable_access_debugfs(adev); return r; } @@ -523,6 +638,7 @@ static ssize_t amdgpu_debugfs_regs_smc_write(struct file *f, const char __user * pm_runtime_mark_last_busy(adev->ddev->dev); pm_runtime_put_autosuspend(adev->ddev->dev); + amdgpu_virt_disable_access_debugfs(adev); return result; } @@ -651,16 +767,24 @@ static ssize_t amdgpu_debugfs_sensor_read(struct file *f, char __user *buf, if (r < 0) return r; + r = amdgpu_virt_enable_access_debugfs(adev); + if (r < 0) + return r; + r = amdgpu_dpm_read_sensor(adev, idx, &values[0], &valuesize); pm_runtime_mark_last_busy(adev->ddev->dev); pm_runtime_put_autosuspend(adev->ddev->dev); - if (r) + if (r) { + amdgpu_virt_disable_access_debugfs(adev); return r; + } - if (size > valuesize) + if (size > valuesize) { + amdgpu_virt_disable_access_debugfs(adev); return -EINVAL; + } outsize = 0; x = 0; @@ -673,6 +797,7 @@ static ssize_t amdgpu_debugfs_sensor_read(struct file *f, char __user *buf, } } + amdgpu_virt_disable_access_debugfs(adev); return !r ? outsize : r; } @@ -720,6 +845,10 @@ static ssize_t amdgpu_debugfs_wave_read(struct file *f, char __user *buf, if (r < 0) return r; + r = amdgpu_virt_enable_access_debugfs(adev); + if (r < 0) + return r; + /* switch to the specific se/sh/cu */ mutex_lock(&adev->grbm_idx_mutex); amdgpu_gfx_select_se_sh(adev, se, sh, cu); @@ -734,16 +863,20 @@ static ssize_t amdgpu_debugfs_wave_read(struct file *f, char __user *buf, pm_runtime_mark_last_busy(adev->ddev->dev); pm_runtime_put_autosuspend(adev->ddev->dev); - if (!x) + if (!x) { + amdgpu_virt_disable_access_debugfs(adev); return -EINVAL; + } while (size && (offset < x * 4)) { uint32_t value; value = data[offset >> 2]; r = put_user(value, (uint32_t *)buf); - if (r) + if (r) { + amdgpu_virt_disable_access_debugfs(adev); return r; + } result += 4; buf += 4; @@ -751,6 +884,7 @@ static ssize_t amdgpu_debugfs_wave_read(struct file *f, char __user *buf, size -= 4; } + amdgpu_virt_disable_access_debugfs(adev); return result; } @@ -805,6 +939,10 @@ static ssize_t amdgpu_debugfs_gpr_read(struct file *f, char __user *buf, if (r < 0) return r; + r = amdgpu_virt_enable_access_debugfs(adev); + if (r < 0) + return r; + /* switch to the specific se/sh/cu */ mutex_lock(&adev->grbm_idx_mutex); amdgpu_gfx_select_se_sh(adev, se, sh, cu); @@ -840,6 +978,7 @@ static ssize_t amdgpu_debugfs_gpr_read(struct file *f, char __user *buf, err: kfree(data); + amdgpu_virt_disable_access_debugfs(adev); return result; } @@ -1369,6 +1508,8 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev) amdgpu_ras_debugfs_create_all(adev); + amdgpu_debugfs_autodump_init(adev); + return amdgpu_debugfs_add_files(adev, amdgpu_debugfs_list, ARRAY_SIZE(amdgpu_debugfs_list)); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h index de12d1101526..2803884d338d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h @@ -31,6 +31,11 @@ struct amdgpu_debugfs { unsigned num_files; }; +struct amdgpu_autodump { + struct completion dumping; + struct wait_queue_head gpu_hang; +}; + int amdgpu_debugfs_regs_init(struct amdgpu_device *adev); int amdgpu_debugfs_init(struct amdgpu_device *adev); void amdgpu_debugfs_fini(struct amdgpu_device *adev); @@ -40,3 +45,4 @@ int amdgpu_debugfs_add_files(struct amdgpu_device *adev, int amdgpu_debugfs_fence_init(struct amdgpu_device *adev); int amdgpu_debugfs_firmware_init(struct amdgpu_device *adev); int amdgpu_debugfs_gem_init(struct amdgpu_device *adev); +int amdgpu_debugfs_wait_dump(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 559dc24ef436..a027a8f7b281 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -64,9 +64,11 @@ #include "amdgpu_xgmi.h" #include "amdgpu_ras.h" #include "amdgpu_pmu.h" +#include "amdgpu_fru_eeprom.h" #include <linux/suspend.h> #include <drm/task_barrier.h> +#include <linux/pm_runtime.h> MODULE_FIRMWARE("amdgpu/vega10_gpu_info.bin"); MODULE_FIRMWARE("amdgpu/vega12_gpu_info.bin"); @@ -138,6 +140,72 @@ static DEVICE_ATTR(pcie_replay_count, S_IRUGO, static void amdgpu_device_get_pcie_info(struct amdgpu_device *adev); /** + * DOC: product_name + * + * The amdgpu driver provides a sysfs API for reporting the product name + * for the device + * The file serial_number is used for this and returns the product name + * as returned from the FRU. + * NOTE: This is only available for certain server cards + */ + +static ssize_t amdgpu_device_get_product_name(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = ddev->dev_private; + + return snprintf(buf, PAGE_SIZE, "%s\n", adev->product_name); +} + +static DEVICE_ATTR(product_name, S_IRUGO, + amdgpu_device_get_product_name, NULL); + +/** + * DOC: product_number + * + * The amdgpu driver provides a sysfs API for reporting the part number + * for the device + * The file serial_number is used for this and returns the part number + * as returned from the FRU. + * NOTE: This is only available for certain server cards + */ + +static ssize_t amdgpu_device_get_product_number(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = ddev->dev_private; + + return snprintf(buf, PAGE_SIZE, "%s\n", adev->product_number); +} + +static DEVICE_ATTR(product_number, S_IRUGO, + amdgpu_device_get_product_number, NULL); + +/** + * DOC: serial_number + * + * The amdgpu driver provides a sysfs API for reporting the serial number + * for the device + * The file serial_number is used for this and returns the serial number + * as returned from the FRU. + * NOTE: This is only available for certain server cards + */ + +static ssize_t amdgpu_device_get_serial_number(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = ddev->dev_private; + + return snprintf(buf, PAGE_SIZE, "%s\n", adev->serial); +} + +static DEVICE_ATTR(serial_number, S_IRUGO, + amdgpu_device_get_serial_number, NULL); + +/** * amdgpu_device_supports_boco - Is the device a dGPU with HG/PX power control * * @dev: drm_device pointer @@ -231,10 +299,10 @@ void amdgpu_device_vram_access(struct amdgpu_device *adev, loff_t pos, } /* - * MMIO register access helper functions. + * device register access helper functions. */ /** - * amdgpu_mm_rreg - read a memory mapped IO register + * amdgpu_device_rreg - read a register * * @adev: amdgpu_device pointer * @reg: dword aligned register offset @@ -242,25 +310,19 @@ void amdgpu_device_vram_access(struct amdgpu_device *adev, loff_t pos, * * Returns the 32 bit value from the offset specified. */ -uint32_t amdgpu_mm_rreg(struct amdgpu_device *adev, uint32_t reg, - uint32_t acc_flags) +uint32_t amdgpu_device_rreg(struct amdgpu_device *adev, uint32_t reg, + uint32_t acc_flags) { uint32_t ret; - if ((acc_flags & AMDGPU_REGS_KIQ) || (!(acc_flags & AMDGPU_REGS_NO_KIQ) && amdgpu_sriov_runtime(adev))) + if (!(acc_flags & AMDGPU_REGS_NO_KIQ) && amdgpu_sriov_runtime(adev)) return amdgpu_kiq_rreg(adev, reg); - if ((reg * 4) < adev->rmmio_size && !(acc_flags & AMDGPU_REGS_IDX)) + if ((reg * 4) < adev->rmmio_size) ret = readl(((void __iomem *)adev->rmmio) + (reg * 4)); - else { - unsigned long flags; - - spin_lock_irqsave(&adev->mmio_idx_lock, flags); - writel((reg * 4), ((void __iomem *)adev->rmmio) + (mmMM_INDEX * 4)); - ret = readl(((void __iomem *)adev->rmmio) + (mmMM_DATA * 4)); - spin_unlock_irqrestore(&adev->mmio_idx_lock, flags); - } - trace_amdgpu_mm_rreg(adev->pdev->device, reg, ret); + else + ret = adev->pcie_rreg(adev, (reg * 4)); + trace_amdgpu_device_rreg(adev->pdev->device, reg, ret); return ret; } @@ -306,28 +368,19 @@ void amdgpu_mm_wreg8(struct amdgpu_device *adev, uint32_t offset, uint8_t value) BUG(); } -void static inline amdgpu_mm_wreg_mmio(struct amdgpu_device *adev, uint32_t reg, uint32_t v, uint32_t acc_flags) +void static inline amdgpu_device_wreg_no_kiq(struct amdgpu_device *adev, uint32_t reg, + uint32_t v, uint32_t acc_flags) { - trace_amdgpu_mm_wreg(adev->pdev->device, reg, v); + trace_amdgpu_device_wreg(adev->pdev->device, reg, v); - if ((reg * 4) < adev->rmmio_size && !(acc_flags & AMDGPU_REGS_IDX)) + if ((reg * 4) < adev->rmmio_size) writel(v, ((void __iomem *)adev->rmmio) + (reg * 4)); - else { - unsigned long flags; - - spin_lock_irqsave(&adev->mmio_idx_lock, flags); - writel((reg * 4), ((void __iomem *)adev->rmmio) + (mmMM_INDEX * 4)); - writel(v, ((void __iomem *)adev->rmmio) + (mmMM_DATA * 4)); - spin_unlock_irqrestore(&adev->mmio_idx_lock, flags); - } - - if (adev->asic_type >= CHIP_VEGA10 && reg == 1 && adev->last_mm_index == 0x5702C) { - udelay(500); - } + else + adev->pcie_wreg(adev, (reg * 4), v); } /** - * amdgpu_mm_wreg - write to a memory mapped IO register + * amdgpu_device_wreg - write to a register * * @adev: amdgpu_device pointer * @reg: dword aligned register offset @@ -336,17 +389,13 @@ void static inline amdgpu_mm_wreg_mmio(struct amdgpu_device *adev, uint32_t reg, * * Writes the value specified to the offset specified. */ -void amdgpu_mm_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v, - uint32_t acc_flags) +void amdgpu_device_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v, + uint32_t acc_flags) { - if (adev->asic_type >= CHIP_VEGA10 && reg == 0) { - adev->last_mm_index = v; - } - - if ((acc_flags & AMDGPU_REGS_KIQ) || (!(acc_flags & AMDGPU_REGS_NO_KIQ) && amdgpu_sriov_runtime(adev))) + if (!(acc_flags & AMDGPU_REGS_NO_KIQ) && amdgpu_sriov_runtime(adev)) return amdgpu_kiq_wreg(adev, reg, v); - amdgpu_mm_wreg_mmio(adev, reg, v, acc_flags); + amdgpu_device_wreg_no_kiq(adev, reg, v, acc_flags); } /* @@ -365,7 +414,7 @@ void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev, uint32_t reg, uint32_t return adev->gfx.rlc.funcs->rlcg_wreg(adev, reg, v); } - amdgpu_mm_wreg_mmio(adev, reg, v, acc_flags); + amdgpu_device_wreg_no_kiq(adev, reg, v, acc_flags); } /** @@ -397,20 +446,12 @@ u32 amdgpu_io_rreg(struct amdgpu_device *adev, u32 reg) */ void amdgpu_io_wreg(struct amdgpu_device *adev, u32 reg, u32 v) { - if (adev->asic_type >= CHIP_VEGA10 && reg == 0) { - adev->last_mm_index = v; - } - if ((reg * 4) < adev->rio_mem_size) iowrite32(v, adev->rio_mem + (reg * 4)); else { iowrite32((reg * 4), adev->rio_mem + (mmMM_INDEX * 4)); iowrite32(v, adev->rio_mem + (mmMM_DATA * 4)); } - - if (adev->asic_type >= CHIP_VEGA10 && reg == 1 && adev->last_mm_index == 0x5702C) { - udelay(500); - } } /** @@ -1126,6 +1167,8 @@ static int amdgpu_device_check_arguments(struct amdgpu_device *adev) adev->firmware.load_type = amdgpu_ucode_get_load_type(adev, amdgpu_fw_load_type); + amdgpu_gmc_tmz_set(adev); + return 0; } @@ -1147,7 +1190,7 @@ static void amdgpu_switcheroo_set_state(struct pci_dev *pdev, enum vga_switchero return; if (state == VGA_SWITCHEROO_ON) { - pr_info("amdgpu: switched on\n"); + pr_info("switched on\n"); /* don't suspend or resume card normally */ dev->switch_power_state = DRM_SWITCH_POWER_CHANGING; @@ -1161,7 +1204,7 @@ static void amdgpu_switcheroo_set_state(struct pci_dev *pdev, enum vga_switchero dev->switch_power_state = DRM_SWITCH_POWER_ON; drm_kms_helper_poll_enable(dev); } else { - pr_info("amdgpu: switched off\n"); + pr_info("switched off\n"); drm_kms_helper_poll_disable(dev); dev->switch_power_state = DRM_SWITCH_POWER_CHANGING; amdgpu_device_suspend(dev, true); @@ -1524,9 +1567,9 @@ static int amdgpu_device_parse_gpu_info_fw(struct amdgpu_device *adev) chip_name = "vega12"; break; case CHIP_RAVEN: - if (adev->rev_id >= 8) + if (adev->apu_flags & AMD_APU_IS_RAVEN2) chip_name = "raven2"; - else if (adev->pdev->device == 0x15d8) + else if (adev->apu_flags & AMD_APU_IS_PICASSO) chip_name = "picasso"; else chip_name = "raven"; @@ -1574,8 +1617,10 @@ static int amdgpu_device_parse_gpu_info_fw(struct amdgpu_device *adev) (const struct gpu_info_firmware_v1_0 *)(adev->firmware.gpu_info_fw->data + le32_to_cpu(hdr->header.ucode_array_offset_bytes)); - if (amdgpu_discovery && adev->asic_type >= CHIP_NAVI10) + if (amdgpu_discovery && adev->asic_type >= CHIP_NAVI10) { + amdgpu_discovery_get_gfx_info(adev); goto parse_soc_bounding_box; + } adev->gfx.config.max_shader_engines = le32_to_cpu(gpu_info_fw->gc_num_se); adev->gfx.config.max_cu_per_sh = le32_to_cpu(gpu_info_fw->gc_num_cu_per_sh); @@ -1721,19 +1766,31 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev) return -EINVAL; } - r = amdgpu_device_parse_gpu_info_fw(adev); - if (r) - return r; - - if (amdgpu_discovery && adev->asic_type >= CHIP_NAVI10) - amdgpu_discovery_get_gfx_info(adev); - amdgpu_amdkfd_device_probe(adev); if (amdgpu_sriov_vf(adev)) { + /* handle vbios stuff prior full access mode for new handshake */ + if (adev->virt.req_init_data_ver == 1) { + if (!amdgpu_get_bios(adev)) { + DRM_ERROR("failed to get vbios\n"); + return -EINVAL; + } + + r = amdgpu_atombios_init(adev); + if (r) { + dev_err(adev->dev, "amdgpu_atombios_init failed\n"); + amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_ATOMBIOS_INIT_FAIL, 0, 0); + return r; + } + } + } + + /* we need to send REQ_GPU here for legacy handshaker otherwise the vbios + * will not be prepared by host for this VF */ + if (amdgpu_sriov_vf(adev) && adev->virt.req_init_data_ver < 1) { r = amdgpu_virt_request_full_gpu(adev, true); if (r) - return -EAGAIN; + return r; } adev->pm.pp_feature = amdgpu_pp_feature_mask; @@ -1763,6 +1820,14 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev) } /* get the vbios after the asic_funcs are set up */ if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_COMMON) { + r = amdgpu_device_parse_gpu_info_fw(adev); + if (r) + return r; + + /* skip vbios handling for new handshake */ + if (amdgpu_sriov_vf(adev) && adev->virt.req_init_data_ver == 1) + continue; + /* Read BIOS */ if (!amdgpu_get_bios(adev)) return -EINVAL; @@ -1889,6 +1954,12 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) if (r) return r; + if (amdgpu_sriov_vf(adev) && adev->virt.req_init_data_ver > 0) { + r = amdgpu_virt_request_full_gpu(adev, true); + if (r) + return -EAGAIN; + } + for (i = 0; i < adev->num_ip_blocks; i++) { if (!adev->ip_blocks[i].status.valid) continue; @@ -1975,6 +2046,8 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) amdgpu_xgmi_add_device(adev); amdgpu_amdkfd_device_init(adev); + amdgpu_fru_get_product_info(adev); + init_failed: if (amdgpu_sriov_vf(adev)) amdgpu_virt_release_full_gpu(adev, true); @@ -2008,8 +2081,24 @@ static void amdgpu_device_fill_reset_magic(struct amdgpu_device *adev) */ static bool amdgpu_device_check_vram_lost(struct amdgpu_device *adev) { - return !!memcmp(adev->gart.ptr, adev->reset_magic, - AMDGPU_RESET_MAGIC_NUM); + if (memcmp(adev->gart.ptr, adev->reset_magic, + AMDGPU_RESET_MAGIC_NUM)) + return true; + + if (!adev->in_gpu_reset) + return false; + + /* + * For all ASICs with baco/mode1 reset, the VRAM is + * always assumed to be lost. + */ + switch (amdgpu_asic_reset_method(adev)) { + case AMD_RESET_METHOD_BACO: + case AMD_RESET_METHOD_MODE1: + return true; + default: + return false; + } } /** @@ -2155,6 +2244,8 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev) adev->ip_blocks[i].status.late_initialized = true; } + amdgpu_ras_set_error_query_ready(adev, true); + amdgpu_device_set_cg_state(adev, AMD_CG_STATE_GATE); amdgpu_device_set_pg_state(adev, AMD_PG_STATE_GATE); @@ -2187,7 +2278,8 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev) if (gpu_instance->adev->flags & AMD_IS_APU) continue; - r = amdgpu_xgmi_set_pstate(gpu_instance->adev, 0); + r = amdgpu_xgmi_set_pstate(gpu_instance->adev, + AMDGPU_XGMI_PSTATE_MIN); if (r) { DRM_ERROR("pstate setting failed (%d).\n", r); break; @@ -2340,6 +2432,8 @@ static int amdgpu_device_ip_suspend_phase1(struct amdgpu_device *adev) { int i, r; + amdgpu_device_set_pg_state(adev, AMD_PG_STATE_UNGATE); + amdgpu_device_set_cg_state(adev, AMD_CG_STATE_UNGATE); for (i = adev->num_ip_blocks - 1; i >= 0; i--) { if (!adev->ip_blocks[i].status.valid) @@ -2767,12 +2861,12 @@ static int amdgpu_device_get_job_timeout_settings(struct amdgpu_device *adev) * By default timeout for non compute jobs is 10000. * And there is no timeout enforced on compute jobs. * In SR-IOV or passthrough mode, timeout for compute - * jobs are 10000 by default. + * jobs are 60000 by default. */ adev->gfx_timeout = msecs_to_jiffies(10000); adev->sdma_timeout = adev->video_timeout = adev->gfx_timeout; if (amdgpu_sriov_vf(adev) || amdgpu_passthrough(adev)) - adev->compute_timeout = adev->gfx_timeout; + adev->compute_timeout = msecs_to_jiffies(60000); else adev->compute_timeout = MAX_SCHEDULE_TIMEOUT; @@ -2823,6 +2917,14 @@ static int amdgpu_device_get_job_timeout_settings(struct amdgpu_device *adev) return ret; } +static const struct attribute *amdgpu_dev_attributes[] = { + &dev_attr_product_name.attr, + &dev_attr_product_number.attr, + &dev_attr_serial_number.attr, + &dev_attr_pcie_replay_count.attr, + NULL +}; + /** * amdgpu_device_init - initialize the driver * @@ -2924,9 +3026,6 @@ int amdgpu_device_init(struct amdgpu_device *adev, INIT_LIST_HEAD(&adev->shadow_list); mutex_init(&adev->shadow_list_lock); - INIT_LIST_HEAD(&adev->ring_lru_list); - spin_lock_init(&adev->ring_lru_list_lock); - INIT_DELAYED_WORK(&adev->delayed_init_work, amdgpu_device_delayed_init_work_handler); INIT_DELAYED_WORK(&adev->gfx.gfx_off_delay_work, @@ -2935,7 +3034,7 @@ int amdgpu_device_init(struct amdgpu_device *adev, INIT_WORK(&adev->xgmi_reset_work, amdgpu_device_xgmi_reset_func); adev->gfx.gfx_off_req_count = 1; - adev->pm.ac_power = power_supply_is_system_supplied() > 0 ? true : false; + adev->pm.ac_power = power_supply_is_system_supplied() > 0; /* Registers mapping */ /* TODO: block userspace mapping of io register */ @@ -2984,18 +3083,8 @@ int amdgpu_device_init(struct amdgpu_device *adev, if (amdgpu_mes && adev->asic_type >= CHIP_NAVI10) adev->enable_mes = true; - if (amdgpu_discovery && adev->asic_type >= CHIP_NAVI10) { - r = amdgpu_discovery_init(adev); - if (r) { - dev_err(adev->dev, "amdgpu_discovery_init failed\n"); - return r; - } - } - - /* early init functions */ - r = amdgpu_device_ip_early_init(adev); - if (r) - return r; + /* detect hw virtualization here */ + amdgpu_detect_virtualization(adev); r = amdgpu_device_get_job_timeout_settings(adev); if (r) { @@ -3003,6 +3092,11 @@ int amdgpu_device_init(struct amdgpu_device *adev, return r; } + /* early init functions */ + r = amdgpu_device_ip_early_init(adev); + if (r) + return r; + /* doorbell bar mapping and doorbell index init*/ amdgpu_device_doorbell_init(adev); @@ -3109,14 +3203,13 @@ fence_driver_init: goto failed; } - DRM_DEBUG("SE %d, SH per SE %d, CU per SH %d, active_cu_number %d\n", + dev_info(adev->dev, + "SE %d, SH per SE %d, CU per SH %d, active_cu_number %d\n", adev->gfx.config.max_shader_engines, adev->gfx.config.max_sh_per_se, adev->gfx.config.max_cu_per_sh, adev->gfx.cu_info.number); - amdgpu_ctx_init_sched(adev); - adev->accel_working = true; amdgpu_vm_check_compute_bug(adev); @@ -3181,9 +3274,9 @@ fence_driver_init: queue_delayed_work(system_wq, &adev->delayed_init_work, msecs_to_jiffies(AMDGPU_RESUME_MS)); - r = device_create_file(adev->dev, &dev_attr_pcie_replay_count); + r = sysfs_create_files(&adev->dev->kobj, amdgpu_dev_attributes); if (r) { - dev_err(adev->dev, "Could not create pcie_replay_count"); + dev_err(adev->dev, "Could not create amdgpu device attr\n"); return r; } @@ -3266,9 +3359,10 @@ void amdgpu_device_fini(struct amdgpu_device *adev) adev->rmmio = NULL; amdgpu_device_doorbell_fini(adev); - device_remove_file(adev->dev, &dev_attr_pcie_replay_count); if (adev->ucode_sysfs_en) amdgpu_ucode_sysfs_fini(adev); + + sysfs_remove_files(&adev->dev->kobj, amdgpu_dev_attributes); if (IS_ENABLED(CONFIG_PERF_EVENTS)) amdgpu_pmu_fini(adev); if (amdgpu_discovery && adev->asic_type >= CHIP_NAVI10) @@ -3354,15 +3448,12 @@ int amdgpu_device_suspend(struct drm_device *dev, bool fbcon) } } - amdgpu_device_set_pg_state(adev, AMD_PG_STATE_UNGATE); - amdgpu_device_set_cg_state(adev, AMD_CG_STATE_UNGATE); - - amdgpu_amdkfd_suspend(adev, !fbcon); - amdgpu_ras_suspend(adev); r = amdgpu_device_ip_suspend_phase1(adev); + amdgpu_amdkfd_suspend(adev, !fbcon); + /* evict vram memory */ amdgpu_bo_evict_vram(adev); @@ -3739,6 +3830,8 @@ static int amdgpu_device_reset_sriov(struct amdgpu_device *adev, if (r) return r; + amdgpu_amdkfd_pre_reset(adev); + /* Resume IP prior to SMC */ r = amdgpu_device_ip_reinit_early_sriov(adev); if (r) @@ -3833,6 +3926,8 @@ static int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev, int i, r = 0; bool need_full_reset = *need_full_reset_arg; + amdgpu_debugfs_wait_dump(adev); + /* block all schedulers and reset given job's ring */ for (i = 0; i < AMDGPU_MAX_RINGS; ++i) { struct amdgpu_ring *ring = adev->rings[i]; @@ -4037,6 +4132,64 @@ static void amdgpu_device_unlock_adev(struct amdgpu_device *adev) mutex_unlock(&adev->lock_reset); } +static void amdgpu_device_resume_display_audio(struct amdgpu_device *adev) +{ + struct pci_dev *p = NULL; + + p = pci_get_domain_bus_and_slot(pci_domain_nr(adev->pdev->bus), + adev->pdev->bus->number, 1); + if (p) { + pm_runtime_enable(&(p->dev)); + pm_runtime_resume(&(p->dev)); + } +} + +static int amdgpu_device_suspend_display_audio(struct amdgpu_device *adev) +{ + enum amd_reset_method reset_method; + struct pci_dev *p = NULL; + u64 expires; + + /* + * For now, only BACO and mode1 reset are confirmed + * to suffer the audio issue without proper suspended. + */ + reset_method = amdgpu_asic_reset_method(adev); + if ((reset_method != AMD_RESET_METHOD_BACO) && + (reset_method != AMD_RESET_METHOD_MODE1)) + return -EINVAL; + + p = pci_get_domain_bus_and_slot(pci_domain_nr(adev->pdev->bus), + adev->pdev->bus->number, 1); + if (!p) + return -ENODEV; + + expires = pm_runtime_autosuspend_expiration(&(p->dev)); + if (!expires) + /* + * If we cannot get the audio device autosuspend delay, + * a fixed 4S interval will be used. Considering 3S is + * the audio controller default autosuspend delay setting. + * 4S used here is guaranteed to cover that. + */ + expires = ktime_get_mono_fast_ns() + NSEC_PER_SEC * 4ULL; + + while (!pm_runtime_status_suspended(&(p->dev))) { + if (!pm_runtime_suspend(&(p->dev))) + break; + + if (expires < ktime_get_mono_fast_ns()) { + dev_warn(adev->dev, "failed to suspend display audio\n"); + /* TODO: abort the succeeding gpu reset? */ + return -ETIMEDOUT; + } + } + + pm_runtime_disable(&(p->dev)); + + return 0; +} + /** * amdgpu_device_gpu_recover - reset the asic and recover scheduler * @@ -4052,7 +4205,8 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, struct amdgpu_job *job) { struct list_head device_list, *device_list_handle = NULL; - bool need_full_reset, job_signaled; + bool need_full_reset = false; + bool job_signaled = false; struct amdgpu_hive_info *hive = NULL; struct amdgpu_device *tmp_adev = NULL; int i, r = 0; @@ -4060,6 +4214,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, bool use_baco = (amdgpu_asic_reset_method(adev) == AMD_RESET_METHOD_BACO) ? true : false; + bool audio_suspended = false; /* * Flush RAM to disk so that after reboot @@ -4073,16 +4228,9 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, emergency_restart(); } - need_full_reset = job_signaled = false; - INIT_LIST_HEAD(&device_list); - dev_info(adev->dev, "GPU %s begin!\n", (in_ras_intr && !use_baco) ? "jobs stop":"reset"); - cancel_delayed_work_sync(&adev->delayed_init_work); - - hive = amdgpu_get_xgmi_hive(adev, false); - /* * Here we trylock to avoid chain of resets executing from * either trigger by jobs on different adevs in XGMI hive or jobs on @@ -4090,39 +4238,25 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, * We always reset all schedulers for device and all devices for XGMI * hive so that should take care of them too. */ - + hive = amdgpu_get_xgmi_hive(adev, true); if (hive && !mutex_trylock(&hive->reset_lock)) { DRM_INFO("Bailing on TDR for s_job:%llx, hive: %llx as another already in progress", job ? job->base.id : -1, hive->hive_id); + mutex_unlock(&hive->hive_lock); return 0; } - /* Start with adev pre asic reset first for soft reset check.*/ - if (!amdgpu_device_lock_adev(adev, !hive)) { - DRM_INFO("Bailing on TDR for s_job:%llx, as another already in progress", - job ? job->base.id : -1); - return 0; - } - - /* Block kfd: SRIOV would do it separately */ - if (!amdgpu_sriov_vf(adev)) - amdgpu_amdkfd_pre_reset(adev); - - /* Build list of devices to reset */ - if (adev->gmc.xgmi.num_physical_nodes > 1) { - if (!hive) { - /*unlock kfd: SRIOV would do it separately */ - if (!amdgpu_sriov_vf(adev)) - amdgpu_amdkfd_post_reset(adev); - amdgpu_device_unlock_adev(adev); + /* + * Build list of devices to reset. + * In case we are in XGMI hive mode, resort the device list + * to put adev in the 1st position. + */ + INIT_LIST_HEAD(&device_list); + if (adev->gmc.xgmi.num_physical_nodes > 1) { + if (!hive) return -ENODEV; - } - - /* - * In case we are in XGMI hive mode device reset is done for all the - * nodes in the hive to retrain all XGMI links and hence the reset - * sequence is executed in loop on all nodes. - */ + if (!list_is_first(&adev->gmc.xgmi.head, &hive->device_list)) + list_rotate_to_front(&adev->gmc.xgmi.head, &hive->device_list); device_list_handle = &hive->device_list; } else { list_add_tail(&adev->gmc.xgmi.head, &device_list); @@ -4131,19 +4265,40 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, /* block all schedulers and reset given job's ring */ list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) { - if (tmp_adev != adev) { - amdgpu_device_lock_adev(tmp_adev, false); - if (!amdgpu_sriov_vf(tmp_adev)) - amdgpu_amdkfd_pre_reset(tmp_adev); + if (!amdgpu_device_lock_adev(tmp_adev, !hive)) { + DRM_INFO("Bailing on TDR for s_job:%llx, as another already in progress", + job ? job->base.id : -1); + mutex_unlock(&hive->hive_lock); + return 0; } /* + * Try to put the audio codec into suspend state + * before gpu reset started. + * + * Due to the power domain of the graphics device + * is shared with AZ power domain. Without this, + * we may change the audio hardware from behind + * the audio driver's back. That will trigger + * some audio codec errors. + */ + if (!amdgpu_device_suspend_display_audio(tmp_adev)) + audio_suspended = true; + + amdgpu_ras_set_error_query_ready(tmp_adev, false); + + cancel_delayed_work_sync(&tmp_adev->delayed_init_work); + + if (!amdgpu_sriov_vf(tmp_adev)) + amdgpu_amdkfd_pre_reset(tmp_adev); + + /* * Mark these ASICs to be reseted as untracked first * And add them back after reset completed */ amdgpu_unregister_gpu_instance(tmp_adev); - amdgpu_fbdev_set_suspend(adev, 1); + amdgpu_fbdev_set_suspend(tmp_adev, 1); /* disable ras on ALL IPs */ if (!(in_ras_intr && !use_baco) && @@ -4163,7 +4318,6 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, } } - if (in_ras_intr && !use_baco) goto skip_sched_resume; @@ -4174,30 +4328,14 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, * job->base holds a reference to parent fence */ if (job && job->base.s_fence->parent && - dma_fence_is_signaled(job->base.s_fence->parent)) + dma_fence_is_signaled(job->base.s_fence->parent)) { job_signaled = true; - - if (job_signaled) { dev_info(adev->dev, "Guilty job already signaled, skipping HW reset"); goto skip_hw_reset; } - - /* Guilty job will be freed after this*/ - r = amdgpu_device_pre_asic_reset(adev, job, &need_full_reset); - if (r) { - /*TODO Should we stop ?*/ - DRM_ERROR("GPU pre asic reset failed with err, %d for drm dev, %s ", - r, adev->ddev->unique); - adev->asic_reset_res = r; - } - retry: /* Rest of adevs pre asic reset from XGMI hive. */ list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) { - - if (tmp_adev == adev) - continue; - r = amdgpu_device_pre_asic_reset(tmp_adev, NULL, &need_full_reset); @@ -4259,11 +4397,15 @@ skip_sched_resume: /*unlock kfd: SRIOV would do it separately */ if (!(in_ras_intr && !use_baco) && !amdgpu_sriov_vf(tmp_adev)) amdgpu_amdkfd_post_reset(tmp_adev); + if (audio_suspended) + amdgpu_device_resume_display_audio(tmp_adev); amdgpu_device_unlock_adev(tmp_adev); } - if (hive) + if (hive) { mutex_unlock(&hive->reset_lock); + mutex_unlock(&hive->hive_lock); + } if (r) dev_info(adev->dev, "GPU reset end with ret = %d\n", r); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_df.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_df.h index 057f6ea645d7..61a26c15c8dd 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_df.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_df.h @@ -52,9 +52,6 @@ struct amdgpu_df_funcs { uint64_t (*get_fica)(struct amdgpu_device *adev, uint32_t ficaa_val); void (*set_fica)(struct amdgpu_device *adev, uint32_t ficaa_val, uint32_t ficadl_val, uint32_t ficadh_val); - uint64_t (*get_dram_base_addr)(struct amdgpu_device *adev, - uint32_t df_inst); - uint32_t (*get_df_inst_id)(struct amdgpu_device *adev); }; struct amdgpu_df { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index 27d8ae19a7a4..b5d6274952a5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -23,9 +23,7 @@ #include "amdgpu.h" #include "amdgpu_discovery.h" -#include "soc15_common.h" #include "soc15_hw_ip.h" -#include "nbio/nbio_2_3_offset.h" #include "discovery.h" #define mmRCC_CONFIG_MEMSIZE 0xde3 @@ -135,9 +133,10 @@ static int hw_id_map[MAX_HWIP] = { static int amdgpu_discovery_read_binary(struct amdgpu_device *adev, uint8_t *binary) { uint64_t vram_size = (uint64_t)RREG32(mmRCC_CONFIG_MEMSIZE) << 20; - uint64_t pos = vram_size - DISCOVERY_TMR_SIZE; + uint64_t pos = vram_size - adev->discovery_tmr_size; - amdgpu_device_vram_access(adev, pos, (uint32_t *)binary, DISCOVERY_TMR_SIZE, false); + amdgpu_device_vram_access(adev, pos, (uint32_t *)binary, + adev->discovery_tmr_size, false); return 0; } @@ -158,7 +157,7 @@ static inline bool amdgpu_discovery_verify_checksum(uint8_t *data, uint32_t size return !!(amdgpu_discovery_calculate_checksum(data, size) == expected); } -int amdgpu_discovery_init(struct amdgpu_device *adev) +static int amdgpu_discovery_init(struct amdgpu_device *adev) { struct table_info *info; struct binary_header *bhdr; @@ -169,17 +168,18 @@ int amdgpu_discovery_init(struct amdgpu_device *adev) uint16_t checksum; int r; - adev->discovery = kzalloc(DISCOVERY_TMR_SIZE, GFP_KERNEL); - if (!adev->discovery) + adev->discovery_tmr_size = DISCOVERY_TMR_SIZE; + adev->discovery_bin = kzalloc(adev->discovery_tmr_size, GFP_KERNEL); + if (!adev->discovery_bin) return -ENOMEM; - r = amdgpu_discovery_read_binary(adev, adev->discovery); + r = amdgpu_discovery_read_binary(adev, adev->discovery_bin); if (r) { DRM_ERROR("failed to read ip discovery binary\n"); goto out; } - bhdr = (struct binary_header *)adev->discovery; + bhdr = (struct binary_header *)adev->discovery_bin; if (le32_to_cpu(bhdr->binary_signature) != BINARY_SIGNATURE) { DRM_ERROR("invalid ip discovery binary signature\n"); @@ -192,7 +192,7 @@ int amdgpu_discovery_init(struct amdgpu_device *adev) size = bhdr->binary_size - offset; checksum = bhdr->binary_checksum; - if (!amdgpu_discovery_verify_checksum(adev->discovery + offset, + if (!amdgpu_discovery_verify_checksum(adev->discovery_bin + offset, size, checksum)) { DRM_ERROR("invalid ip discovery binary checksum\n"); r = -EINVAL; @@ -202,7 +202,7 @@ int amdgpu_discovery_init(struct amdgpu_device *adev) info = &bhdr->table_list[IP_DISCOVERY]; offset = le16_to_cpu(info->offset); checksum = le16_to_cpu(info->checksum); - ihdr = (struct ip_discovery_header *)(adev->discovery + offset); + ihdr = (struct ip_discovery_header *)(adev->discovery_bin + offset); if (le32_to_cpu(ihdr->signature) != DISCOVERY_TABLE_SIGNATURE) { DRM_ERROR("invalid ip discovery data table signature\n"); @@ -210,7 +210,7 @@ int amdgpu_discovery_init(struct amdgpu_device *adev) goto out; } - if (!amdgpu_discovery_verify_checksum(adev->discovery + offset, + if (!amdgpu_discovery_verify_checksum(adev->discovery_bin + offset, ihdr->size, checksum)) { DRM_ERROR("invalid ip discovery data table checksum\n"); r = -EINVAL; @@ -220,9 +220,9 @@ int amdgpu_discovery_init(struct amdgpu_device *adev) info = &bhdr->table_list[GC]; offset = le16_to_cpu(info->offset); checksum = le16_to_cpu(info->checksum); - ghdr = (struct gpu_info_header *)(adev->discovery + offset); + ghdr = (struct gpu_info_header *)(adev->discovery_bin + offset); - if (!amdgpu_discovery_verify_checksum(adev->discovery + offset, + if (!amdgpu_discovery_verify_checksum(adev->discovery_bin + offset, ghdr->size, checksum)) { DRM_ERROR("invalid gc data table checksum\n"); r = -EINVAL; @@ -232,16 +232,16 @@ int amdgpu_discovery_init(struct amdgpu_device *adev) return 0; out: - kfree(adev->discovery); - adev->discovery = NULL; + kfree(adev->discovery_bin); + adev->discovery_bin = NULL; return r; } void amdgpu_discovery_fini(struct amdgpu_device *adev) { - kfree(adev->discovery); - adev->discovery = NULL; + kfree(adev->discovery_bin); + adev->discovery_bin = NULL; } int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev) @@ -257,14 +257,16 @@ int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev) uint8_t num_base_address; int hw_ip; int i, j, k; + int r; - if (!adev->discovery) { - DRM_ERROR("ip discovery uninitialized\n"); - return -EINVAL; + r = amdgpu_discovery_init(adev); + if (r) { + DRM_ERROR("amdgpu_discovery_init failed\n"); + return r; } - bhdr = (struct binary_header *)adev->discovery; - ihdr = (struct ip_discovery_header *)(adev->discovery + + bhdr = (struct binary_header *)adev->discovery_bin; + ihdr = (struct ip_discovery_header *)(adev->discovery_bin + le16_to_cpu(bhdr->table_list[IP_DISCOVERY].offset)); num_dies = le16_to_cpu(ihdr->num_dies); @@ -272,7 +274,7 @@ int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev) for (i = 0; i < num_dies; i++) { die_offset = le16_to_cpu(ihdr->die_info[i].die_offset); - dhdr = (struct die_header *)(adev->discovery + die_offset); + dhdr = (struct die_header *)(adev->discovery_bin + die_offset); num_ips = le16_to_cpu(dhdr->num_ips); ip_offset = die_offset + sizeof(*dhdr); @@ -286,7 +288,7 @@ int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev) le16_to_cpu(dhdr->die_id), num_ips); for (j = 0; j < num_ips; j++) { - ip = (struct ip *)(adev->discovery + ip_offset); + ip = (struct ip *)(adev->discovery_bin + ip_offset); num_base_address = ip->num_base_address; DRM_DEBUG("%s(%d) #%d v%d.%d.%d:\n", @@ -335,24 +337,24 @@ int amdgpu_discovery_get_ip_version(struct amdgpu_device *adev, int hw_id, uint16_t num_ips; int i, j; - if (!adev->discovery) { + if (!adev->discovery_bin) { DRM_ERROR("ip discovery uninitialized\n"); return -EINVAL; } - bhdr = (struct binary_header *)adev->discovery; - ihdr = (struct ip_discovery_header *)(adev->discovery + + bhdr = (struct binary_header *)adev->discovery_bin; + ihdr = (struct ip_discovery_header *)(adev->discovery_bin + le16_to_cpu(bhdr->table_list[IP_DISCOVERY].offset)); num_dies = le16_to_cpu(ihdr->num_dies); for (i = 0; i < num_dies; i++) { die_offset = le16_to_cpu(ihdr->die_info[i].die_offset); - dhdr = (struct die_header *)(adev->discovery + die_offset); + dhdr = (struct die_header *)(adev->discovery_bin + die_offset); num_ips = le16_to_cpu(dhdr->num_ips); ip_offset = die_offset + sizeof(*dhdr); for (j = 0; j < num_ips; j++) { - ip = (struct ip *)(adev->discovery + ip_offset); + ip = (struct ip *)(adev->discovery_bin + ip_offset); if (le16_to_cpu(ip->hw_id) == hw_id) { if (major) @@ -375,13 +377,13 @@ int amdgpu_discovery_get_gfx_info(struct amdgpu_device *adev) struct binary_header *bhdr; struct gc_info_v1_0 *gc_info; - if (!adev->discovery) { + if (!adev->discovery_bin) { DRM_ERROR("ip discovery uninitialized\n"); return -EINVAL; } - bhdr = (struct binary_header *)adev->discovery; - gc_info = (struct gc_info_v1_0 *)(adev->discovery + + bhdr = (struct binary_header *)adev->discovery_bin; + gc_info = (struct gc_info_v1_0 *)(adev->discovery_bin + le16_to_cpu(bhdr->table_list[GC].offset)); adev->gfx.config.max_shader_engines = le32_to_cpu(gc_info->gc_num_se); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h index ba78e15d9b05..d50d597c45ed 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h @@ -26,7 +26,6 @@ #define DISCOVERY_TMR_SIZE (64 << 10) -int amdgpu_discovery_init(struct amdgpu_device *adev); void amdgpu_discovery_fini(struct amdgpu_device *adev); int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev); int amdgpu_discovery_get_ip_version(struct amdgpu_device *adev, int hw_id, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c index de9b62e2800a..c56438a6c9f2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c @@ -523,7 +523,8 @@ uint32_t amdgpu_display_supported_domains(struct amdgpu_device *adev, break; case CHIP_RAVEN: /* enable S/G on PCO and RV2 */ - if (adev->rev_id >= 0x8 || adev->pdev->device == 0x15d8) + if ((adev->apu_flags & AMD_APU_IS_RAVEN2) || + (adev->apu_flags & AMD_APU_IS_PICASSO)) domain |= AMDGPU_GEM_DOMAIN_GTT; break; default: diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c index 0e8018c9aa8e..d2a105e3bf7c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c @@ -1188,3 +1188,13 @@ int amdgpu_dpm_set_df_cstate(struct amdgpu_device *adev, return ret; } + +int amdgpu_dpm_allow_xgmi_power_down(struct amdgpu_device *adev, bool en) +{ + struct smu_context *smu = &adev->smu; + + if (is_support_sw_smu(adev)) + return smu_allow_xgmi_power_down(smu, en); + + return 0; +}
\ No newline at end of file diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h index 936d85aa0fbc..6a8aae70a0e6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h @@ -450,6 +450,7 @@ struct amdgpu_pm { /* Used for I2C access to various EEPROMs on relevant ASICs */ struct i2c_adapter smu_i2c; + struct list_head pm_attr_list; }; #define R600_SSTU_DFLT 0 @@ -538,4 +539,6 @@ int amdgpu_dpm_baco_enter(struct amdgpu_device *adev); int amdgpu_dpm_set_df_cstate(struct amdgpu_device *adev, uint32_t cstate); +int amdgpu_dpm_allow_xgmi_power_down(struct amdgpu_device *adev, bool en); + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 8ea86ffdea0d..126e74758a34 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -85,9 +85,11 @@ * - 3.34.0 - Non-DC can flip correctly between buffers with different pitches * - 3.35.0 - Add drm_amdgpu_info_device::tcc_disabled_mask * - 3.36.0 - Allow reading more status registers on si/cik + * - 3.37.0 - L2 is invalidated before SDMA IBs, needed for correctness + * - 3.38.0 - Add AMDGPU_IB_FLAG_EMIT_MEM_SYNC */ #define KMS_DRIVER_MAJOR 3 -#define KMS_DRIVER_MINOR 36 +#define KMS_DRIVER_MINOR 38 #define KMS_DRIVER_PATCHLEVEL 0 int amdgpu_vram_limit = 0; @@ -138,12 +140,14 @@ int amdgpu_emu_mode = 0; uint amdgpu_smu_memory_pool_size = 0; /* FBC (bit 0) disabled by default*/ uint amdgpu_dc_feature_mask = 0; +uint amdgpu_dc_debug_mask = 0; int amdgpu_async_gfx_ring = 1; int amdgpu_mcbp = 0; int amdgpu_discovery = -1; int amdgpu_mes = 0; int amdgpu_noretry; int amdgpu_force_asic_type = -1; +int amdgpu_tmz = 0; struct amdgpu_mgpu_info mgpu_info = { .mutex = __MUTEX_INITIALIZER(mgpu_info.mutex), @@ -687,13 +691,12 @@ MODULE_PARM_DESC(halt_if_hws_hang, "Halt if HWS hang is detected (0 = off (defau /** * DOC: hws_gws_support(bool) - * Whether HWS support gws barriers. Default value: false (not supported) - * This will be replaced with a MEC firmware version check once firmware - * is ready + * Assume that HWS supports GWS barriers regardless of what firmware version + * check says. Default value: false (rely on MEC2 firmware version check). */ bool hws_gws_support; module_param(hws_gws_support, bool, 0444); -MODULE_PARM_DESC(hws_gws_support, "MEC FW support gws barriers (false = not supported (Default), true = supported)"); +MODULE_PARM_DESC(hws_gws_support, "Assume MEC2 FW supports GWS barriers (false = rely on FW version check (Default), true = force supported)"); /** * DOC: queue_preemption_timeout_ms (int) @@ -713,6 +716,13 @@ MODULE_PARM_DESC(dcfeaturemask, "all stable DC features enabled (default))"); module_param_named(dcfeaturemask, amdgpu_dc_feature_mask, uint, 0444); /** + * DOC: dcdebugmask (uint) + * Override display features enabled. See enum DC_DEBUG_MASK in drivers/gpu/drm/amd/include/amd_shared.h. + */ +MODULE_PARM_DESC(dcdebugmask, "all debug options disabled (default))"); +module_param_named(dcdebugmask, amdgpu_dc_debug_mask, uint, 0444); + +/** * DOC: abmlevel (uint) * Override the default ABM (Adaptive Backlight Management) level used for DC * enabled hardware. Requires DMCU to be supported and loaded. @@ -728,6 +738,16 @@ uint amdgpu_dm_abm_level = 0; MODULE_PARM_DESC(abmlevel, "ABM level (0 = off (default), 1-4 = backlight reduction level) "); module_param_named(abmlevel, amdgpu_dm_abm_level, uint, 0444); +/** + * DOC: tmz (int) + * Trusted Memory Zone (TMZ) is a method to protect data being written + * to or read from memory. + * + * The default value: 0 (off). TODO: change to auto till it is completed. + */ +MODULE_PARM_DESC(tmz, "Enable TMZ feature (-1 = auto, 0 = off (default), 1 = on)"); +module_param_named(tmz, amdgpu_tmz, int, 0444); + static const struct pci_device_id pciidlist[] = { #ifdef CONFIG_DRM_AMDGPU_SI {0x1002, 0x6780, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TAHITI}, @@ -1163,14 +1183,6 @@ static int amdgpu_pmops_resume(struct device *dev) { struct drm_device *drm_dev = dev_get_drvdata(dev); - /* GPU comes up enabled by the bios on resume */ - if (amdgpu_device_supports_boco(drm_dev) || - amdgpu_device_supports_baco(drm_dev)) { - pm_runtime_disable(dev); - pm_runtime_set_active(dev); - pm_runtime_enable(dev); - } - return amdgpu_device_resume(drm_dev, true); } @@ -1180,7 +1192,9 @@ static int amdgpu_pmops_freeze(struct device *dev) struct amdgpu_device *adev = drm_dev->dev_private; int r; + adev->in_hibernate = true; r = amdgpu_device_suspend(drm_dev, true); + adev->in_hibernate = false; if (r) return r; return amdgpu_asic_reset(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c index ca5dde4ec47d..db731f573f98 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c @@ -133,8 +133,7 @@ static int amdgpufb_create_pinned_object(struct amdgpu_fbdev *rfbdev, u32 cpp; u64 flags = AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED | AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS | - AMDGPU_GEM_CREATE_VRAM_CLEARED | - AMDGPU_GEM_CREATE_CPU_GTT_USWC; + AMDGPU_GEM_CREATE_VRAM_CLEARED; info = drm_get_format_info(adev->ddev, mode_cmd); cpp = info->cpp[0]; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c index 7531527067df..d878fe7fee51 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c @@ -192,14 +192,22 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f, * Used For polling fence. * Returns 0 on success, -ENOMEM on failure. */ -int amdgpu_fence_emit_polling(struct amdgpu_ring *ring, uint32_t *s) +int amdgpu_fence_emit_polling(struct amdgpu_ring *ring, uint32_t *s, + uint32_t timeout) { uint32_t seq; + signed long r; if (!s) return -EINVAL; seq = ++ring->fence_drv.sync_seq; + r = amdgpu_fence_wait_polling(ring, + seq - ring->fence_drv.num_fences_mask, + timeout); + if (r < 1) + return -ETIMEDOUT; + amdgpu_ring_emit_fence(ring, ring->fence_drv.gpu_addr, seq, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c new file mode 100644 index 000000000000..815c072ac4da --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c @@ -0,0 +1,185 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ +#include <linux/pci.h> + +#include "amdgpu.h" +#include "amdgpu_i2c.h" +#include "smu_v11_0_i2c.h" +#include "atom.h" + +#define I2C_PRODUCT_INFO_ADDR 0xAC +#define I2C_PRODUCT_INFO_ADDR_SIZE 0x2 +#define I2C_PRODUCT_INFO_OFFSET 0xC0 + +bool is_fru_eeprom_supported(struct amdgpu_device *adev) +{ + /* TODO: Gaming SKUs don't have the FRU EEPROM. + * Use this hack to address hangs on modprobe on gaming SKUs + * until a proper solution can be implemented by only supporting + * the explicit chip IDs for VG20 Server cards + * + * TODO: Add list of supported Arcturus DIDs once confirmed + */ + if ((adev->asic_type == CHIP_VEGA20 && adev->pdev->device == 0x66a0) || + (adev->asic_type == CHIP_VEGA20 && adev->pdev->device == 0x66a1) || + (adev->asic_type == CHIP_VEGA20 && adev->pdev->device == 0x66a4)) + return true; + return false; +} + +int amdgpu_fru_read_eeprom(struct amdgpu_device *adev, uint32_t addrptr, + unsigned char *buff) +{ + int ret, size; + struct i2c_msg msg = { + .addr = I2C_PRODUCT_INFO_ADDR, + .flags = I2C_M_RD, + .buf = buff, + }; + buff[0] = 0; + buff[1] = addrptr; + msg.len = I2C_PRODUCT_INFO_ADDR_SIZE + 1; + ret = i2c_transfer(&adev->pm.smu_i2c, &msg, 1); + + if (ret < 1) { + DRM_WARN("FRU: Failed to get size field"); + return ret; + } + + /* The size returned by the i2c requires subtraction of 0xC0 since the + * size apparently always reports as 0xC0+actual size. + */ + size = buff[2] - I2C_PRODUCT_INFO_OFFSET; + /* Add 1 since address field was 1 byte */ + buff[1] = addrptr + 1; + + msg.len = I2C_PRODUCT_INFO_ADDR_SIZE + size; + ret = i2c_transfer(&adev->pm.smu_i2c, &msg, 1); + + if (ret < 1) { + DRM_WARN("FRU: Failed to get data field"); + return ret; + } + + return size; +} + +int amdgpu_fru_get_product_info(struct amdgpu_device *adev) +{ + unsigned char buff[34]; + int addrptr = 0, size = 0; + + if (!is_fru_eeprom_supported(adev)) + return 0; + + /* If algo exists, it means that the i2c_adapter's initialized */ + if (!adev->pm.smu_i2c.algo) { + DRM_WARN("Cannot access FRU, EEPROM accessor not initialized"); + return 0; + } + + /* There's a lot of repetition here. This is due to the FRU having + * variable-length fields. To get the information, we have to find the + * size of each field, and then keep reading along and reading along + * until we get all of the data that we want. We use addrptr to track + * the address as we go + */ + + /* The first fields are all of size 1-byte, from 0-7 are offsets that + * contain information that isn't useful to us. + * Bytes 8-a are all 1-byte and refer to the size of the entire struct, + * and the language field, so just start from 0xb, manufacturer size + */ + addrptr = 0xb; + size = amdgpu_fru_read_eeprom(adev, addrptr, buff); + if (size < 1) { + DRM_ERROR("Failed to read FRU Manufacturer, ret:%d", size); + return size; + } + + /* Increment the addrptr by the size of the field, and 1 due to the + * size field being 1 byte. This pattern continues below. + */ + addrptr += size + 1; + size = amdgpu_fru_read_eeprom(adev, addrptr, buff); + if (size < 1) { + DRM_ERROR("Failed to read FRU product name, ret:%d", size); + return size; + } + + /* Product name should only be 32 characters. Any more, + * and something could be wrong. Cap it at 32 to be safe + */ + if (size > 32) { + DRM_WARN("FRU Product Number is larger than 32 characters. This is likely a mistake"); + size = 32; + } + /* Start at 2 due to buff using fields 0 and 1 for the address */ + memcpy(adev->product_name, &buff[2], size); + adev->product_name[size] = '\0'; + + addrptr += size + 1; + size = amdgpu_fru_read_eeprom(adev, addrptr, buff); + if (size < 1) { + DRM_ERROR("Failed to read FRU product number, ret:%d", size); + return size; + } + + /* Product number should only be 16 characters. Any more, + * and something could be wrong. Cap it at 16 to be safe + */ + if (size > 16) { + DRM_WARN("FRU Product Number is larger than 16 characters. This is likely a mistake"); + size = 16; + } + memcpy(adev->product_number, &buff[2], size); + adev->product_number[size] = '\0'; + + addrptr += size + 1; + size = amdgpu_fru_read_eeprom(adev, addrptr, buff); + + if (size < 1) { + DRM_ERROR("Failed to read FRU product version, ret:%d", size); + return size; + } + + addrptr += size + 1; + size = amdgpu_fru_read_eeprom(adev, addrptr, buff); + + if (size < 1) { + DRM_ERROR("Failed to read FRU serial number, ret:%d", size); + return size; + } + + /* Serial number should only be 16 characters. Any more, + * and something could be wrong. Cap it at 16 to be safe + */ + if (size > 16) { + DRM_WARN("FRU Serial Number is larger than 16 characters. This is likely a mistake"); + size = 16; + } + memcpy(adev->serial, &buff[2], size); + adev->serial[size] = '\0'; + + return 0; +} diff --git a/drivers/gpu/drm/amd/display/dc/dml/dml_common_defs.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.h index f78cbae9db88..968115c97e33 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dml_common_defs.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.h @@ -1,5 +1,5 @@ /* - * Copyright 2017 Advanced Micro Devices, Inc. + * Copyright 2020 Advanced Micro Devices, Inc. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -19,19 +19,11 @@ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR * OTHER DEALINGS IN THE SOFTWARE. * - * Authors: AMD - * */ -#ifndef __DC_COMMON_DEFS_H__ -#define __DC_COMMON_DEFS_H__ - -#include "dm_services.h" -#include "dc_features.h" -#include "display_mode_structs.h" -#include "display_mode_enums.h" - +#ifndef __AMDGPU_PRODINFO_H__ +#define __AMDGPU_PRODINFO_H__ -double dml_round(double a); +int amdgpu_fru_get_product_info(struct amdgpu_device *adev); -#endif /* __DC_COMMON_DEFS_H__ */ +#endif // __AMDGPU_PRODINFO_H__ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 8c0597f306ec..de9784b0c19b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -162,16 +162,17 @@ void amdgpu_gem_object_close(struct drm_gem_object *obj, struct amdgpu_bo_list_entry vm_pd; struct list_head list, duplicates; + struct dma_fence *fence = NULL; struct ttm_validate_buffer tv; struct ww_acquire_ctx ticket; struct amdgpu_bo_va *bo_va; - int r; + long r; INIT_LIST_HEAD(&list); INIT_LIST_HEAD(&duplicates); tv.bo = &bo->tbo; - tv.num_shared = 1; + tv.num_shared = 2; list_add(&tv.head, &list); amdgpu_vm_get_pd_bo(vm, &list, &vm_pd); @@ -179,28 +180,34 @@ void amdgpu_gem_object_close(struct drm_gem_object *obj, r = ttm_eu_reserve_buffers(&ticket, &list, false, &duplicates); if (r) { dev_err(adev->dev, "leaking bo va because " - "we fail to reserve bo (%d)\n", r); + "we fail to reserve bo (%ld)\n", r); return; } bo_va = amdgpu_vm_bo_find(vm, bo); - if (bo_va && --bo_va->ref_count == 0) { - amdgpu_vm_bo_rmv(adev, bo_va); - - if (amdgpu_vm_ready(vm)) { - struct dma_fence *fence = NULL; + if (!bo_va || --bo_va->ref_count) + goto out_unlock; - r = amdgpu_vm_clear_freed(adev, vm, &fence); - if (unlikely(r)) { - dev_err(adev->dev, "failed to clear page " - "tables on GEM object close (%d)\n", r); - } + amdgpu_vm_bo_rmv(adev, bo_va); + if (!amdgpu_vm_ready(vm)) + goto out_unlock; - if (fence) { - amdgpu_bo_fence(bo, fence, true); - dma_fence_put(fence); - } - } + fence = dma_resv_get_excl(bo->tbo.base.resv); + if (fence) { + amdgpu_bo_fence(bo, fence, true); + fence = NULL; } + + r = amdgpu_vm_clear_freed(adev, vm, &fence); + if (r || !fence) + goto out_unlock; + + amdgpu_bo_fence(bo, fence, true); + dma_fence_put(fence); + +out_unlock: + if (unlikely(r < 0)) + dev_err(adev->dev, "failed to clear page " + "tables on GEM object close (%ld)\n", r); ttm_eu_backoff_reservation(&ticket, &list); } @@ -227,7 +234,8 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data, AMDGPU_GEM_CREATE_CPU_GTT_USWC | AMDGPU_GEM_CREATE_VRAM_CLEARED | AMDGPU_GEM_CREATE_VM_ALWAYS_VALID | - AMDGPU_GEM_CREATE_EXPLICIT_SYNC)) + AMDGPU_GEM_CREATE_EXPLICIT_SYNC | + AMDGPU_GEM_CREATE_ENCRYPTED)) return -EINVAL; @@ -235,6 +243,11 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data, if (args->in.domains & ~AMDGPU_GEM_DOMAIN_MASK) return -EINVAL; + if (!amdgpu_is_tmz(adev) && (flags & AMDGPU_GEM_CREATE_ENCRYPTED)) { + DRM_NOTE_ONCE("Cannot allocate secure buffer since TMZ is disabled\n"); + return -EINVAL; + } + /* create a gem object to contain this object in */ if (args->in.domains & (AMDGPU_GEM_DOMAIN_GDS | AMDGPU_GEM_DOMAIN_GWS | AMDGPU_GEM_DOMAIN_OA)) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c index 6b9c9193cdfa..d612033a23ac 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c @@ -48,7 +48,7 @@ int amdgpu_gfx_mec_queue_to_bit(struct amdgpu_device *adev, int mec, return bit; } -void amdgpu_gfx_bit_to_mec_queue(struct amdgpu_device *adev, int bit, +void amdgpu_queue_mask_bit_to_mec_queue(struct amdgpu_device *adev, int bit, int *mec, int *pipe, int *queue) { *queue = bit % adev->gfx.mec.num_queue_per_pipe; @@ -274,7 +274,7 @@ static int amdgpu_gfx_kiq_acquire(struct amdgpu_device *adev, if (test_bit(queue_bit, adev->gfx.mec.queue_bitmap)) continue; - amdgpu_gfx_bit_to_mec_queue(adev, queue_bit, &mec, &pipe, &queue); + amdgpu_queue_mask_bit_to_mec_queue(adev, queue_bit, &mec, &pipe, &queue); /* * 1. Using pipes 2/3 from MEC 2 seems cause problems. @@ -304,10 +304,6 @@ int amdgpu_gfx_kiq_init_ring(struct amdgpu_device *adev, spin_lock_init(&kiq->ring_lock); - r = amdgpu_device_wb_get(adev, &kiq->reg_val_offs); - if (r) - return r; - ring->adev = NULL; ring->ring_obj = NULL; ring->use_doorbell = true; @@ -318,9 +314,11 @@ int amdgpu_gfx_kiq_init_ring(struct amdgpu_device *adev, return r; ring->eop_gpu_addr = kiq->eop_gpu_addr; + ring->no_scheduler = true; sprintf(ring->name, "kiq_%d.%d.%d", ring->me, ring->pipe, ring->queue); r = amdgpu_ring_init(adev, ring, 1024, - irq, AMDGPU_CP_KIQ_IRQ_DRIVER0); + irq, AMDGPU_CP_KIQ_IRQ_DRIVER0, + AMDGPU_RING_PRIO_DEFAULT); if (r) dev_warn(adev->dev, "(%d) failed to init kiq ring\n", r); @@ -329,7 +327,6 @@ int amdgpu_gfx_kiq_init_ring(struct amdgpu_device *adev, void amdgpu_gfx_kiq_free_ring(struct amdgpu_ring *ring) { - amdgpu_device_wb_free(ring->adev, ring->adev->gfx.kiq.reg_val_offs); amdgpu_ring_fini(ring); } @@ -488,6 +485,19 @@ int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev) return amdgpu_ring_test_helper(kiq_ring); } +int amdgpu_queue_mask_bit_to_set_resource_bit(struct amdgpu_device *adev, + int queue_bit) +{ + int mec, pipe, queue; + int set_resource_bit = 0; + + amdgpu_queue_mask_bit_to_mec_queue(adev, queue_bit, &mec, &pipe, &queue); + + set_resource_bit = mec * 4 * 8 + pipe * 8 + queue; + + return set_resource_bit; +} + int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev) { struct amdgpu_kiq *kiq = &adev->gfx.kiq; @@ -510,7 +520,7 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev) break; } - queue_mask |= (1ull << i); + queue_mask |= (1ull << amdgpu_queue_mask_bit_to_set_resource_bit(adev, i)); } DRM_INFO("kiq ring mec %d pipe %d q %d\n", kiq_ring->me, kiq_ring->pipe, @@ -670,16 +680,23 @@ uint32_t amdgpu_kiq_rreg(struct amdgpu_device *adev, uint32_t reg) { signed long r, cnt = 0; unsigned long flags; - uint32_t seq; + uint32_t seq, reg_val_offs = 0, value = 0; struct amdgpu_kiq *kiq = &adev->gfx.kiq; struct amdgpu_ring *ring = &kiq->ring; BUG_ON(!ring->funcs->emit_rreg); spin_lock_irqsave(&kiq->ring_lock, flags); + if (amdgpu_device_wb_get(adev, ®_val_offs)) { + pr_err("critical bug! too many kiq readers\n"); + goto failed_unlock; + } amdgpu_ring_alloc(ring, 32); - amdgpu_ring_emit_rreg(ring, reg); - amdgpu_fence_emit_polling(ring, &seq); + amdgpu_ring_emit_rreg(ring, reg, reg_val_offs); + r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT); + if (r) + goto failed_undo; + amdgpu_ring_commit(ring); spin_unlock_irqrestore(&kiq->ring_lock, flags); @@ -705,9 +722,18 @@ uint32_t amdgpu_kiq_rreg(struct amdgpu_device *adev, uint32_t reg) if (cnt > MAX_KIQ_REG_TRY) goto failed_kiq_read; - return adev->wb.wb[kiq->reg_val_offs]; + mb(); + value = adev->wb.wb[reg_val_offs]; + amdgpu_device_wb_free(adev, reg_val_offs); + return value; +failed_undo: + amdgpu_ring_undo(ring); +failed_unlock: + spin_unlock_irqrestore(&kiq->ring_lock, flags); failed_kiq_read: + if (reg_val_offs) + amdgpu_device_wb_free(adev, reg_val_offs); pr_err("failed to read reg:%x\n", reg); return ~0; } @@ -725,7 +751,10 @@ void amdgpu_kiq_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v) spin_lock_irqsave(&kiq->ring_lock, flags); amdgpu_ring_alloc(ring, 32); amdgpu_ring_emit_wreg(ring, reg, v); - amdgpu_fence_emit_polling(ring, &seq); + r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT); + if (r) + goto failed_undo; + amdgpu_ring_commit(ring); spin_unlock_irqrestore(&kiq->ring_lock, flags); @@ -754,6 +783,9 @@ void amdgpu_kiq_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v) return; +failed_undo: + amdgpu_ring_undo(ring); + spin_unlock_irqrestore(&kiq->ring_lock, flags); failed_kiq_write: pr_err("failed to write reg:%x\n", reg); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h index 5825692d07e4..d43c11671a38 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h @@ -103,7 +103,6 @@ struct amdgpu_kiq { struct amdgpu_ring ring; struct amdgpu_irq_src irq; const struct kiq_pm4_funcs *pmf; - uint32_t reg_val_offs; }; /* @@ -286,13 +285,8 @@ struct amdgpu_gfx { bool me_fw_write_wait; bool cp_fw_write_wait; struct amdgpu_ring gfx_ring[AMDGPU_MAX_GFX_RINGS]; - struct drm_gpu_scheduler *gfx_sched[AMDGPU_MAX_GFX_RINGS]; - uint32_t num_gfx_sched; unsigned num_gfx_rings; struct amdgpu_ring compute_ring[AMDGPU_MAX_COMPUTE_RINGS]; - struct drm_gpu_scheduler **compute_prio_sched[AMDGPU_GFX_PIPE_PRIO_MAX]; - struct drm_gpu_scheduler *compute_sched[AMDGPU_MAX_COMPUTE_RINGS]; - uint32_t num_compute_sched[AMDGPU_GFX_PIPE_PRIO_MAX]; unsigned num_compute_rings; struct amdgpu_irq_src eop_irq; struct amdgpu_irq_src priv_reg_irq; @@ -370,7 +364,7 @@ void amdgpu_gfx_graphics_queue_acquire(struct amdgpu_device *adev); int amdgpu_gfx_mec_queue_to_bit(struct amdgpu_device *adev, int mec, int pipe, int queue); -void amdgpu_gfx_bit_to_mec_queue(struct amdgpu_device *adev, int bit, +void amdgpu_queue_mask_bit_to_mec_queue(struct amdgpu_device *adev, int bit, int *mec, int *pipe, int *queue); bool amdgpu_gfx_is_mec_queue_enabled(struct amdgpu_device *adev, int mec, int pipe, int queue); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c index 5884ab590486..acabb57aa8af 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c @@ -136,8 +136,8 @@ uint64_t amdgpu_gmc_agp_addr(struct ttm_buffer_object *bo) /** * amdgpu_gmc_vram_location - try to find VRAM location * - * @adev: amdgpu device structure holding all necessary informations - * @mc: memory controller structure holding memory informations + * @adev: amdgpu device structure holding all necessary information + * @mc: memory controller structure holding memory information * @base: base address at which to put VRAM * * Function will try to place VRAM at base address provided @@ -165,8 +165,8 @@ void amdgpu_gmc_vram_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc, /** * amdgpu_gmc_gart_location - try to find GART location * - * @adev: amdgpu device structure holding all necessary informations - * @mc: memory controller structure holding memory informations + * @adev: amdgpu device structure holding all necessary information + * @mc: memory controller structure holding memory information * * Function will place try to place GART before or after VRAM. * @@ -207,8 +207,8 @@ void amdgpu_gmc_gart_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc) /** * amdgpu_gmc_agp_location - try to find AGP location - * @adev: amdgpu device structure holding all necessary informations - * @mc: memory controller structure holding memory informations + * @adev: amdgpu device structure holding all necessary information + * @mc: memory controller structure holding memory information * * Function will place try to find a place for the AGP BAR in the MC address * space. @@ -373,3 +373,38 @@ int amdgpu_gmc_allocate_vm_inv_eng(struct amdgpu_device *adev) return 0; } + +/** + * amdgpu_tmz_set -- check and set if a device supports TMZ + * @adev: amdgpu_device pointer + * + * Check and set if an the device @adev supports Trusted Memory + * Zones (TMZ). + */ +void amdgpu_gmc_tmz_set(struct amdgpu_device *adev) +{ + switch (adev->asic_type) { + case CHIP_RAVEN: + case CHIP_RENOIR: + case CHIP_NAVI10: + case CHIP_NAVI14: + case CHIP_NAVI12: + /* Don't enable it by default yet. + */ + if (amdgpu_tmz < 1) { + adev->gmc.tmz_enabled = false; + dev_info(adev->dev, + "Trusted Memory Zone (TMZ) feature disabled as experimental (default)\n"); + } else { + adev->gmc.tmz_enabled = true; + dev_info(adev->dev, + "Trusted Memory Zone (TMZ) feature enabled as experimental (cmd line)\n"); + } + break; + default: + adev->gmc.tmz_enabled = false; + dev_warn(adev->dev, + "Trusted Memory Zone (TMZ) feature not supported\n"); + break; + } +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h index 7546da0cc70c..2bd9423c1dab 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h @@ -213,6 +213,8 @@ struct amdgpu_gmc { } fault_hash[AMDGPU_GMC_FAULT_HASH_SIZE]; uint64_t last_fault:AMDGPU_GMC_FAULT_RING_ORDER; + bool tmz_enabled; + const struct amdgpu_gmc_funcs *gmc_funcs; struct amdgpu_xgmi xgmi; @@ -276,4 +278,6 @@ int amdgpu_gmc_ras_late_init(struct amdgpu_device *adev); void amdgpu_gmc_ras_fini(struct amdgpu_device *adev); int amdgpu_gmc_allocate_vm_inv_eng(struct amdgpu_device *adev); +extern void amdgpu_gmc_tmz_set(struct amdgpu_device *adev); + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c index ccbd7acfc4cb..b91853fd66d3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c @@ -61,12 +61,13 @@ * Returns 0 on success, error on failure. */ int amdgpu_ib_get(struct amdgpu_device *adev, struct amdgpu_vm *vm, - unsigned size, struct amdgpu_ib *ib) + unsigned size, enum amdgpu_ib_pool_type pool_type, + struct amdgpu_ib *ib) { int r; if (size) { - r = amdgpu_sa_bo_new(&adev->ring_tmp_bo, + r = amdgpu_sa_bo_new(&adev->ib_pools[pool_type], &ib->sa_bo, size, 256); if (r) { dev_err(adev->dev, "failed to get a new IB (%d)\n", r); @@ -131,6 +132,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, uint64_t fence_ctx; uint32_t status = 0, alloc_size; unsigned fence_flags = 0; + bool secure; unsigned i; int r = 0; @@ -159,6 +161,12 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, return -EINVAL; } + if ((ib->flags & AMDGPU_IB_FLAGS_SECURE) && + (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE)) { + dev_err(adev->dev, "secure submissions not supported on compute rings\n"); + return -EINVAL; + } + alloc_size = ring->funcs->emit_frame_size + num_ibs * ring->funcs->emit_ib_size; @@ -181,6 +189,9 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, dma_fence_put(tmp); } + if ((ib->flags & AMDGPU_IB_FLAG_EMIT_MEM_SYNC) && ring->funcs->emit_mem_sync) + ring->funcs->emit_mem_sync(ring); + if (ring->funcs->insert_start) ring->funcs->insert_start(ring); @@ -215,6 +226,14 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, amdgpu_ring_emit_cntxcntl(ring, status); } + /* Setup initial TMZiness and send it off. + */ + secure = false; + if (job && ring->funcs->emit_frame_cntl) { + secure = ib->flags & AMDGPU_IB_FLAGS_SECURE; + amdgpu_ring_emit_frame_cntl(ring, true, secure); + } + for (i = 0; i < num_ibs; ++i) { ib = &ibs[i]; @@ -226,12 +245,20 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, !amdgpu_sriov_vf(adev)) /* for SRIOV preemption, Preamble CE ib must be inserted anyway */ continue; + if (job && ring->funcs->emit_frame_cntl) { + if (secure != !!(ib->flags & AMDGPU_IB_FLAGS_SECURE)) { + amdgpu_ring_emit_frame_cntl(ring, false, secure); + secure = !secure; + amdgpu_ring_emit_frame_cntl(ring, true, secure); + } + } + amdgpu_ring_emit_ib(ring, job, ib, status); status &= ~AMDGPU_HAVE_CTX_SWITCH; } - if (ring->funcs->emit_tmz) - amdgpu_ring_emit_tmz(ring, false); + if (job && ring->funcs->emit_frame_cntl) + amdgpu_ring_emit_frame_cntl(ring, false, secure); #ifdef CONFIG_X86_64 if (!(adev->flags & AMD_IS_APU)) @@ -280,22 +307,32 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, */ int amdgpu_ib_pool_init(struct amdgpu_device *adev) { - int r; + unsigned size; + int r, i; - if (adev->ib_pool_ready) { + if (adev->ib_pool_ready) return 0; - } - r = amdgpu_sa_bo_manager_init(adev, &adev->ring_tmp_bo, - AMDGPU_IB_POOL_SIZE*64*1024, - AMDGPU_GPU_PAGE_SIZE, - AMDGPU_GEM_DOMAIN_GTT); - if (r) { - return r; - } + for (i = 0; i < AMDGPU_IB_POOL_MAX; i++) { + if (i == AMDGPU_IB_POOL_DIRECT) + size = PAGE_SIZE * 2; + else + size = AMDGPU_IB_POOL_SIZE; + + r = amdgpu_sa_bo_manager_init(adev, &adev->ib_pools[i], + size, AMDGPU_GPU_PAGE_SIZE, + AMDGPU_GEM_DOMAIN_GTT); + if (r) + goto error; + } adev->ib_pool_ready = true; return 0; + +error: + while (i--) + amdgpu_sa_bo_manager_fini(adev, &adev->ib_pools[i]); + return r; } /** @@ -308,10 +345,14 @@ int amdgpu_ib_pool_init(struct amdgpu_device *adev) */ void amdgpu_ib_pool_fini(struct amdgpu_device *adev) { - if (adev->ib_pool_ready) { - amdgpu_sa_bo_manager_fini(adev, &adev->ring_tmp_bo); - adev->ib_pool_ready = false; - } + int i; + + if (!adev->ib_pool_ready) + return; + + for (i = 0; i < AMDGPU_IB_POOL_MAX; i++) + amdgpu_sa_bo_manager_fini(adev, &adev->ib_pools[i]); + adev->ib_pool_ready = false; } /** @@ -326,9 +367,9 @@ void amdgpu_ib_pool_fini(struct amdgpu_device *adev) */ int amdgpu_ib_ring_tests(struct amdgpu_device *adev) { - unsigned i; - int r, ret = 0; long tmo_gfx, tmo_mm; + int r, ret = 0; + unsigned i; tmo_mm = tmo_gfx = AMDGPU_IB_TEST_TIMEOUT; if (amdgpu_sriov_vf(adev)) { @@ -406,10 +447,16 @@ static int amdgpu_debugfs_sa_info(struct seq_file *m, void *data) struct drm_device *dev = node->minor->dev; struct amdgpu_device *adev = dev->dev_private; - amdgpu_sa_bo_dump_debug_info(&adev->ring_tmp_bo, m); + seq_printf(m, "--------------------- DELAYED --------------------- \n"); + amdgpu_sa_bo_dump_debug_info(&adev->ib_pools[AMDGPU_IB_POOL_DELAYED], + m); + seq_printf(m, "-------------------- IMMEDIATE -------------------- \n"); + amdgpu_sa_bo_dump_debug_info(&adev->ib_pools[AMDGPU_IB_POOL_IMMEDIATE], + m); + seq_printf(m, "--------------------- DIRECT ---------------------- \n"); + amdgpu_sa_bo_dump_debug_info(&adev->ib_pools[AMDGPU_IB_POOL_DIRECT], m); return 0; - } static const struct drm_info_list amdgpu_debugfs_sa_list[] = { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c index 3a67f6c046d4..fe92dcd94d4a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c @@ -282,7 +282,7 @@ static int amdgpu_vmid_grab_reserved(struct amdgpu_vm *vm, !dma_fence_is_later(updates, (*id)->flushed_updates)) updates = NULL; - if ((*id)->owner != vm->direct.fence_context || + if ((*id)->owner != vm->immediate.fence_context || job->vm_pd_addr != (*id)->pd_gpu_addr || updates || !(*id)->last_flush || ((*id)->last_flush->context != fence_context && @@ -349,7 +349,7 @@ static int amdgpu_vmid_grab_used(struct amdgpu_vm *vm, struct dma_fence *flushed; /* Check all the prerequisites to using this VMID */ - if ((*id)->owner != vm->direct.fence_context) + if ((*id)->owner != vm->immediate.fence_context) continue; if ((*id)->pd_gpu_addr != job->vm_pd_addr) @@ -448,7 +448,7 @@ int amdgpu_vmid_grab(struct amdgpu_vm *vm, struct amdgpu_ring *ring, } id->pd_gpu_addr = job->vm_pd_addr; - id->owner = vm->direct.fence_context; + id->owner = vm->immediate.fence_context; if (job->vm_needs_flush) { dma_fence_put(id->last_flush); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c index 5ed4227f304b..0cc4c67f95f7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c @@ -260,7 +260,7 @@ int amdgpu_irq_init(struct amdgpu_device *adev) nvec = pci_alloc_irq_vectors(adev->pdev, 1, 1, flags); if (nvec > 0) { adev->irq.msi_enabled = true; - dev_dbg(adev->dev, "amdgpu: using MSI/MSI-X.\n"); + dev_dbg(adev->dev, "using MSI/MSI-X.\n"); } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c index 4981e443a884..47207188c569 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c @@ -33,6 +33,7 @@ static void amdgpu_job_timedout(struct drm_sched_job *s_job) struct amdgpu_ring *ring = to_amdgpu_ring(s_job->sched); struct amdgpu_job *job = to_amdgpu_job(s_job); struct amdgpu_task_info ti; + struct amdgpu_device *adev = ring->adev; memset(&ti, 0, sizeof(struct amdgpu_task_info)); @@ -49,10 +50,13 @@ static void amdgpu_job_timedout(struct drm_sched_job *s_job) DRM_ERROR("Process information: process %s pid %d thread %s pid %d\n", ti.process_name, ti.tgid, ti.task_name, ti.pid); - if (amdgpu_device_should_recover_gpu(ring->adev)) + if (amdgpu_device_should_recover_gpu(ring->adev)) { amdgpu_device_gpu_recover(ring->adev, job); - else + } else { drm_sched_suspend_timeout(&ring->sched); + if (amdgpu_sriov_vf(adev)) + adev->virt.tdr_debug = true; + } } int amdgpu_job_alloc(struct amdgpu_device *adev, unsigned num_ibs, @@ -87,7 +91,8 @@ int amdgpu_job_alloc(struct amdgpu_device *adev, unsigned num_ibs, } int amdgpu_job_alloc_with_ib(struct amdgpu_device *adev, unsigned size, - struct amdgpu_job **job) + enum amdgpu_ib_pool_type pool_type, + struct amdgpu_job **job) { int r; @@ -95,7 +100,7 @@ int amdgpu_job_alloc_with_ib(struct amdgpu_device *adev, unsigned size, if (r) return r; - r = amdgpu_ib_get(adev, NULL, size, &(*job)->ibs[0]); + r = amdgpu_ib_get(adev, NULL, size, pool_type, &(*job)->ibs[0]); if (r) kfree(*job); @@ -140,7 +145,6 @@ void amdgpu_job_free(struct amdgpu_job *job) int amdgpu_job_submit(struct amdgpu_job *job, struct drm_sched_entity *entity, void *owner, struct dma_fence **f) { - enum drm_sched_priority priority; int r; if (!f) @@ -152,7 +156,6 @@ int amdgpu_job_submit(struct amdgpu_job *job, struct drm_sched_entity *entity, *f = dma_fence_get(&job->base.s_fence->finished); amdgpu_job_free_resources(job); - priority = job->base.s_priority; drm_sched_entity_push_job(&job->base, entity); return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h index 3f7b8433d179..81caac9b958a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h @@ -38,6 +38,7 @@ #define AMDGPU_JOB_GET_VMID(job) ((job) ? (job)->vmid : 0) struct amdgpu_fence; +enum amdgpu_ib_pool_type; struct amdgpu_job { struct drm_sched_job base; @@ -61,14 +62,12 @@ struct amdgpu_job { /* user fence handling */ uint64_t uf_addr; uint64_t uf_sequence; - }; int amdgpu_job_alloc(struct amdgpu_device *adev, unsigned num_ibs, struct amdgpu_job **job, struct amdgpu_vm *vm); int amdgpu_job_alloc_with_ib(struct amdgpu_device *adev, unsigned size, - struct amdgpu_job **job); - + enum amdgpu_ib_pool_type pool, struct amdgpu_job **job); void amdgpu_job_free_resources(struct amdgpu_job *job); void amdgpu_job_free(struct amdgpu_job *job); int amdgpu_job_submit(struct amdgpu_job *job, struct drm_sched_entity *entity, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c index 5727f00afc8e..d31d65e6b039 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c @@ -144,7 +144,8 @@ static int amdgpu_jpeg_dec_set_reg(struct amdgpu_ring *ring, uint32_t handle, const unsigned ib_size_dw = 16; int i, r; - r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, &job); + r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, + AMDGPU_IB_POOL_DIRECT, &job); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h index bd9ef9cc86de..5131a0a1bc8a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h @@ -43,8 +43,6 @@ struct amdgpu_jpeg { uint8_t num_jpeg_inst; struct amdgpu_jpeg_inst inst[AMDGPU_MAX_JPEG_INSTANCES]; struct amdgpu_jpeg_reg internal; - struct drm_gpu_scheduler *jpeg_sched[AMDGPU_MAX_JPEG_INSTANCES]; - uint32_t num_jpeg_sched; unsigned harvest_config; struct delayed_work idle_work; enum amd_powergating_state cur_state; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index fd1dc3236eca..d7e17e34fee1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -183,18 +183,18 @@ int amdgpu_driver_load_kms(struct drm_device *dev, unsigned long flags) /* Call ACPI methods: require modeset init * but failure is not fatal */ - if (!r) { - acpi_status = amdgpu_acpi_init(adev); - if (acpi_status) - dev_dbg(&dev->pdev->dev, - "Error during ACPI methods call\n"); - } + + acpi_status = amdgpu_acpi_init(adev); + if (acpi_status) + dev_dbg(&dev->pdev->dev, "Error during ACPI methods call\n"); if (adev->runpm) { - dev_pm_set_driver_flags(dev->dev, DPM_FLAG_NEVER_SKIP); + /* only need to skip on ATPX */ + if (amdgpu_device_supports_boco(dev) && + !amdgpu_is_atpx_hybrid()) + dev_pm_set_driver_flags(dev->dev, DPM_FLAG_NO_DIRECT_COMPLETE); pm_runtime_use_autosuspend(dev->dev); pm_runtime_set_autosuspend_delay(dev->dev, 5000); - pm_runtime_set_active(dev->dev); pm_runtime_allow(dev->dev); pm_runtime_mark_last_busy(dev->dev); pm_runtime_put_autosuspend(dev->dev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h index 919bd566ba3c..edaac242ff85 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h @@ -77,7 +77,6 @@ struct amdgpu_nbio_funcs { u32 *flags); void (*ih_control)(struct amdgpu_device *adev); void (*init_registers)(struct amdgpu_device *adev); - void (*detect_hw_virt)(struct amdgpu_device *adev); void (*remap_hdp_registers)(struct amdgpu_device *adev); void (*handle_ras_controller_intr_no_bifring)(struct amdgpu_device *adev); void (*handle_ras_err_event_athub_intr_no_bifring)(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index a27b6a692119..5ac7b5561475 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -753,7 +753,7 @@ int amdgpu_bo_restore_shadow(struct amdgpu_bo *shadow, struct dma_fence **fence) return amdgpu_copy_buffer(ring, shadow_addr, parent_addr, amdgpu_bo_size(shadow), NULL, fence, - true, false); + true, false, false); } /** diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h index 32edd35d2ccf..afa5189dba7d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h @@ -229,6 +229,17 @@ static inline bool amdgpu_bo_explicit_sync(struct amdgpu_bo *bo) return bo->flags & AMDGPU_GEM_CREATE_EXPLICIT_SYNC; } +/** + * amdgpu_bo_encrypted - test if the BO is encrypted + * @bo: pointer to a buffer object + * + * Return true if the buffer object is encrypted, false otherwise. + */ +static inline bool amdgpu_bo_encrypted(struct amdgpu_bo *bo) +{ + return bo->flags & AMDGPU_GEM_CREATE_ENCRYPTED; +} + bool amdgpu_bo_is_amdgpu_bo(struct ttm_buffer_object *bo); void amdgpu_bo_placement_from_domain(struct amdgpu_bo *abo, u32 domain); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c index abe94a55ecad..16596a9ccabe 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c @@ -154,17 +154,17 @@ int amdgpu_dpm_read_sensor(struct amdgpu_device *adev, enum amd_pp_sensors senso * */ -static ssize_t amdgpu_get_dpm_state(struct device *dev, - struct device_attribute *attr, - char *buf) +static ssize_t amdgpu_get_power_dpm_state(struct device *dev, + struct device_attribute *attr, + char *buf) { struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = ddev->dev_private; enum amd_pm_state_type pm; int ret; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; ret = pm_runtime_get_sync(ddev->dev); if (ret < 0) @@ -189,18 +189,18 @@ static ssize_t amdgpu_get_dpm_state(struct device *dev, (pm == POWER_STATE_TYPE_BALANCED) ? "balanced" : "performance"); } -static ssize_t amdgpu_set_dpm_state(struct device *dev, - struct device_attribute *attr, - const char *buf, - size_t count) +static ssize_t amdgpu_set_power_dpm_state(struct device *dev, + struct device_attribute *attr, + const char *buf, + size_t count) { struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = ddev->dev_private; enum amd_pm_state_type state; int ret; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return -EINVAL; + if (adev->in_gpu_reset) + return -EPERM; if (strncmp("battery", buf, strlen("battery")) == 0) state = POWER_STATE_TYPE_BATTERY; @@ -294,17 +294,17 @@ static ssize_t amdgpu_set_dpm_state(struct device *dev, * */ -static ssize_t amdgpu_get_dpm_forced_performance_level(struct device *dev, - struct device_attribute *attr, - char *buf) +static ssize_t amdgpu_get_power_dpm_force_performance_level(struct device *dev, + struct device_attribute *attr, + char *buf) { struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = ddev->dev_private; enum amd_dpm_forced_level level = 0xff; int ret; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; ret = pm_runtime_get_sync(ddev->dev); if (ret < 0) @@ -332,10 +332,10 @@ static ssize_t amdgpu_get_dpm_forced_performance_level(struct device *dev, "unknown"); } -static ssize_t amdgpu_set_dpm_forced_performance_level(struct device *dev, - struct device_attribute *attr, - const char *buf, - size_t count) +static ssize_t amdgpu_set_power_dpm_force_performance_level(struct device *dev, + struct device_attribute *attr, + const char *buf, + size_t count) { struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = ddev->dev_private; @@ -343,8 +343,8 @@ static ssize_t amdgpu_set_dpm_forced_performance_level(struct device *dev, enum amd_dpm_forced_level current_level = 0xff; int ret = 0; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return -EINVAL; + if (adev->in_gpu_reset) + return -EPERM; if (strncmp("low", buf, strlen("low")) == 0) { level = AMD_DPM_FORCED_LEVEL_LOW; @@ -383,6 +383,15 @@ static ssize_t amdgpu_set_dpm_forced_performance_level(struct device *dev, return count; } + if (adev->asic_type == CHIP_RAVEN) { + if (!(adev->apu_flags & AMD_APU_IS_RAVEN2)) { + if (current_level != AMD_DPM_FORCED_LEVEL_MANUAL && level == AMD_DPM_FORCED_LEVEL_MANUAL) + amdgpu_gfx_off_ctrl(adev, false); + else if (current_level == AMD_DPM_FORCED_LEVEL_MANUAL && level != AMD_DPM_FORCED_LEVEL_MANUAL) + amdgpu_gfx_off_ctrl(adev, true); + } + } + /* profile_exit setting is valid only when current mode is in profile mode */ if (!(current_level & (AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD | AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK | @@ -436,6 +445,9 @@ static ssize_t amdgpu_get_pp_num_states(struct device *dev, struct pp_states_info data; int i, buf_len, ret; + if (adev->in_gpu_reset) + return -EPERM; + ret = pm_runtime_get_sync(ddev->dev); if (ret < 0) return ret; @@ -444,8 +456,11 @@ static ssize_t amdgpu_get_pp_num_states(struct device *dev, ret = smu_get_power_num_states(&adev->smu, &data); if (ret) return ret; - } else if (adev->powerplay.pp_funcs->get_pp_num_states) + } else if (adev->powerplay.pp_funcs->get_pp_num_states) { amdgpu_dpm_get_pp_num_states(adev, &data); + } else { + memset(&data, 0, sizeof(data)); + } pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); @@ -472,8 +487,8 @@ static ssize_t amdgpu_get_pp_cur_state(struct device *dev, enum amd_pm_state_type pm = 0; int i = 0, ret = 0; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; ret = pm_runtime_get_sync(ddev->dev); if (ret < 0) @@ -511,8 +526,8 @@ static ssize_t amdgpu_get_pp_force_state(struct device *dev, struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = ddev->dev_private; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; if (adev->pp_force_state_enabled) return amdgpu_get_pp_cur_state(dev, attr, buf); @@ -531,8 +546,8 @@ static ssize_t amdgpu_set_pp_force_state(struct device *dev, unsigned long idx; int ret; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return -EINVAL; + if (adev->in_gpu_reset) + return -EPERM; if (strlen(buf) == 1) adev->pp_force_state_enabled = false; @@ -589,8 +604,8 @@ static ssize_t amdgpu_get_pp_table(struct device *dev, char *table = NULL; int size, ret; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; ret = pm_runtime_get_sync(ddev->dev); if (ret < 0) @@ -631,8 +646,8 @@ static ssize_t amdgpu_set_pp_table(struct device *dev, struct amdgpu_device *adev = ddev->dev_private; int ret = 0; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return -EINVAL; + if (adev->in_gpu_reset) + return -EPERM; ret = pm_runtime_get_sync(ddev->dev); if (ret < 0) @@ -681,7 +696,7 @@ static ssize_t amdgpu_set_pp_table(struct device *dev, * default power levels, write "r" (reset) to the file to reset them. * * - * < For Vega20 > + * < For Vega20 and newer ASICs > * * Reading the file will display: * @@ -736,8 +751,8 @@ static ssize_t amdgpu_set_pp_od_clk_voltage(struct device *dev, const char delimiter[3] = {' ', '\n', '\0'}; uint32_t type; - if (amdgpu_sriov_vf(adev)) - return -EINVAL; + if (adev->in_gpu_reset) + return -EPERM; if (count > 127) return -EINVAL; @@ -828,8 +843,8 @@ static ssize_t amdgpu_get_pp_od_clk_voltage(struct device *dev, ssize_t size; int ret; - if (amdgpu_sriov_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; ret = pm_runtime_get_sync(ddev->dev); if (ret < 0) @@ -870,18 +885,18 @@ static ssize_t amdgpu_get_pp_od_clk_voltage(struct device *dev, * the corresponding bit from original ppfeature masks and input the * new ppfeature masks. */ -static ssize_t amdgpu_set_pp_feature_status(struct device *dev, - struct device_attribute *attr, - const char *buf, - size_t count) +static ssize_t amdgpu_set_pp_features(struct device *dev, + struct device_attribute *attr, + const char *buf, + size_t count) { struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = ddev->dev_private; uint64_t featuremask; int ret; - if (amdgpu_sriov_vf(adev)) - return -EINVAL; + if (adev->in_gpu_reset) + return -EPERM; ret = kstrtou64(buf, 0, &featuremask); if (ret) @@ -914,17 +929,17 @@ static ssize_t amdgpu_set_pp_feature_status(struct device *dev, return count; } -static ssize_t amdgpu_get_pp_feature_status(struct device *dev, - struct device_attribute *attr, - char *buf) +static ssize_t amdgpu_get_pp_features(struct device *dev, + struct device_attribute *attr, + char *buf) { struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = ddev->dev_private; ssize_t size; int ret; - if (amdgpu_sriov_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; ret = pm_runtime_get_sync(ddev->dev); if (ret < 0) @@ -982,8 +997,8 @@ static ssize_t amdgpu_get_pp_dpm_sclk(struct device *dev, ssize_t size; int ret; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; ret = pm_runtime_get_sync(ddev->dev); if (ret < 0) @@ -1048,8 +1063,8 @@ static ssize_t amdgpu_set_pp_dpm_sclk(struct device *dev, int ret; uint32_t mask = 0; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return -EINVAL; + if (adev->in_gpu_reset) + return -EPERM; ret = amdgpu_read_mask(buf, count, &mask); if (ret) @@ -1082,8 +1097,8 @@ static ssize_t amdgpu_get_pp_dpm_mclk(struct device *dev, ssize_t size; int ret; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; ret = pm_runtime_get_sync(ddev->dev); if (ret < 0) @@ -1112,8 +1127,8 @@ static ssize_t amdgpu_set_pp_dpm_mclk(struct device *dev, uint32_t mask = 0; int ret; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return -EINVAL; + if (adev->in_gpu_reset) + return -EPERM; ret = amdgpu_read_mask(buf, count, &mask); if (ret) @@ -1146,8 +1161,8 @@ static ssize_t amdgpu_get_pp_dpm_socclk(struct device *dev, ssize_t size; int ret; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; ret = pm_runtime_get_sync(ddev->dev); if (ret < 0) @@ -1176,8 +1191,8 @@ static ssize_t amdgpu_set_pp_dpm_socclk(struct device *dev, int ret; uint32_t mask = 0; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return -EINVAL; + if (adev->in_gpu_reset) + return -EPERM; ret = amdgpu_read_mask(buf, count, &mask); if (ret) @@ -1212,8 +1227,8 @@ static ssize_t amdgpu_get_pp_dpm_fclk(struct device *dev, ssize_t size; int ret; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; ret = pm_runtime_get_sync(ddev->dev); if (ret < 0) @@ -1242,8 +1257,8 @@ static ssize_t amdgpu_set_pp_dpm_fclk(struct device *dev, int ret; uint32_t mask = 0; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return -EINVAL; + if (adev->in_gpu_reset) + return -EPERM; ret = amdgpu_read_mask(buf, count, &mask); if (ret) @@ -1278,8 +1293,8 @@ static ssize_t amdgpu_get_pp_dpm_dcefclk(struct device *dev, ssize_t size; int ret; - if (amdgpu_sriov_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; ret = pm_runtime_get_sync(ddev->dev); if (ret < 0) @@ -1308,8 +1323,8 @@ static ssize_t amdgpu_set_pp_dpm_dcefclk(struct device *dev, int ret; uint32_t mask = 0; - if (amdgpu_sriov_vf(adev)) - return -EINVAL; + if (adev->in_gpu_reset) + return -EPERM; ret = amdgpu_read_mask(buf, count, &mask); if (ret) @@ -1344,8 +1359,8 @@ static ssize_t amdgpu_get_pp_dpm_pcie(struct device *dev, ssize_t size; int ret; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; ret = pm_runtime_get_sync(ddev->dev); if (ret < 0) @@ -1374,8 +1389,8 @@ static ssize_t amdgpu_set_pp_dpm_pcie(struct device *dev, int ret; uint32_t mask = 0; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return -EINVAL; + if (adev->in_gpu_reset) + return -EPERM; ret = amdgpu_read_mask(buf, count, &mask); if (ret) @@ -1410,8 +1425,8 @@ static ssize_t amdgpu_get_pp_sclk_od(struct device *dev, uint32_t value = 0; int ret; - if (amdgpu_sriov_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; ret = pm_runtime_get_sync(ddev->dev); if (ret < 0) @@ -1438,8 +1453,8 @@ static ssize_t amdgpu_set_pp_sclk_od(struct device *dev, int ret; long int value; - if (amdgpu_sriov_vf(adev)) - return -EINVAL; + if (adev->in_gpu_reset) + return -EPERM; ret = kstrtol(buf, 0, &value); @@ -1479,8 +1494,8 @@ static ssize_t amdgpu_get_pp_mclk_od(struct device *dev, uint32_t value = 0; int ret; - if (amdgpu_sriov_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; ret = pm_runtime_get_sync(ddev->dev); if (ret < 0) @@ -1507,8 +1522,8 @@ static ssize_t amdgpu_set_pp_mclk_od(struct device *dev, int ret; long int value; - if (amdgpu_sriov_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; ret = kstrtol(buf, 0, &value); @@ -1568,8 +1583,8 @@ static ssize_t amdgpu_get_pp_power_profile_mode(struct device *dev, ssize_t size; int ret; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; ret = pm_runtime_get_sync(ddev->dev); if (ret < 0) @@ -1606,15 +1621,15 @@ static ssize_t amdgpu_set_pp_power_profile_mode(struct device *dev, long int profile_mode = 0; const char delimiter[3] = {' ', '\n', '\0'}; + if (adev->in_gpu_reset) + return -EPERM; + tmp[0] = *(buf); tmp[1] = '\0'; ret = kstrtol(tmp, 0, &profile_mode); if (ret) return -EINVAL; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return -EINVAL; - if (profile_mode == PP_SMC_POWER_PROFILE_CUSTOM) { if (count < 2 || count > 127) return -EINVAL; @@ -1653,23 +1668,23 @@ static ssize_t amdgpu_set_pp_power_profile_mode(struct device *dev, } /** - * DOC: busy_percent + * DOC: gpu_busy_percent * * The amdgpu driver provides a sysfs API for reading how busy the GPU * is as a percentage. The file gpu_busy_percent is used for this. * The SMU firmware computes a percentage of load based on the * aggregate activity level in the IP cores. */ -static ssize_t amdgpu_get_busy_percent(struct device *dev, - struct device_attribute *attr, - char *buf) +static ssize_t amdgpu_get_gpu_busy_percent(struct device *dev, + struct device_attribute *attr, + char *buf) { struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = ddev->dev_private; int r, value, size = sizeof(value); - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; r = pm_runtime_get_sync(ddev->dev); if (r < 0) @@ -1696,16 +1711,16 @@ static ssize_t amdgpu_get_busy_percent(struct device *dev, * The SMU firmware computes a percentage of load based on the * aggregate activity level in the IP cores. */ -static ssize_t amdgpu_get_memory_busy_percent(struct device *dev, - struct device_attribute *attr, - char *buf) +static ssize_t amdgpu_get_mem_busy_percent(struct device *dev, + struct device_attribute *attr, + char *buf) { struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = ddev->dev_private; int r, value, size = sizeof(value); - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; r = pm_runtime_get_sync(ddev->dev); if (r < 0) @@ -1742,11 +1757,17 @@ static ssize_t amdgpu_get_pcie_bw(struct device *dev, { struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = ddev->dev_private; - uint64_t count0, count1; + uint64_t count0 = 0, count1 = 0; int ret; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; + + if (adev->flags & AMD_IS_APU) + return -ENODATA; + + if (!adev->asic_funcs->get_pcie_usage) + return -ENODATA; ret = pm_runtime_get_sync(ddev->dev); if (ret < 0) @@ -1778,8 +1799,8 @@ static ssize_t amdgpu_get_unique_id(struct device *dev, struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = ddev->dev_private; - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return 0; + if (adev->in_gpu_reset) + return -EPERM; if (adev->unique_id) return snprintf(buf, PAGE_SIZE, "%016llx\n", adev->unique_id); @@ -1787,57 +1808,185 @@ static ssize_t amdgpu_get_unique_id(struct device *dev, return 0; } -static DEVICE_ATTR(power_dpm_state, S_IRUGO | S_IWUSR, amdgpu_get_dpm_state, amdgpu_set_dpm_state); -static DEVICE_ATTR(power_dpm_force_performance_level, S_IRUGO | S_IWUSR, - amdgpu_get_dpm_forced_performance_level, - amdgpu_set_dpm_forced_performance_level); -static DEVICE_ATTR(pp_num_states, S_IRUGO, amdgpu_get_pp_num_states, NULL); -static DEVICE_ATTR(pp_cur_state, S_IRUGO, amdgpu_get_pp_cur_state, NULL); -static DEVICE_ATTR(pp_force_state, S_IRUGO | S_IWUSR, - amdgpu_get_pp_force_state, - amdgpu_set_pp_force_state); -static DEVICE_ATTR(pp_table, S_IRUGO | S_IWUSR, - amdgpu_get_pp_table, - amdgpu_set_pp_table); -static DEVICE_ATTR(pp_dpm_sclk, S_IRUGO | S_IWUSR, - amdgpu_get_pp_dpm_sclk, - amdgpu_set_pp_dpm_sclk); -static DEVICE_ATTR(pp_dpm_mclk, S_IRUGO | S_IWUSR, - amdgpu_get_pp_dpm_mclk, - amdgpu_set_pp_dpm_mclk); -static DEVICE_ATTR(pp_dpm_socclk, S_IRUGO | S_IWUSR, - amdgpu_get_pp_dpm_socclk, - amdgpu_set_pp_dpm_socclk); -static DEVICE_ATTR(pp_dpm_fclk, S_IRUGO | S_IWUSR, - amdgpu_get_pp_dpm_fclk, - amdgpu_set_pp_dpm_fclk); -static DEVICE_ATTR(pp_dpm_dcefclk, S_IRUGO | S_IWUSR, - amdgpu_get_pp_dpm_dcefclk, - amdgpu_set_pp_dpm_dcefclk); -static DEVICE_ATTR(pp_dpm_pcie, S_IRUGO | S_IWUSR, - amdgpu_get_pp_dpm_pcie, - amdgpu_set_pp_dpm_pcie); -static DEVICE_ATTR(pp_sclk_od, S_IRUGO | S_IWUSR, - amdgpu_get_pp_sclk_od, - amdgpu_set_pp_sclk_od); -static DEVICE_ATTR(pp_mclk_od, S_IRUGO | S_IWUSR, - amdgpu_get_pp_mclk_od, - amdgpu_set_pp_mclk_od); -static DEVICE_ATTR(pp_power_profile_mode, S_IRUGO | S_IWUSR, - amdgpu_get_pp_power_profile_mode, - amdgpu_set_pp_power_profile_mode); -static DEVICE_ATTR(pp_od_clk_voltage, S_IRUGO | S_IWUSR, - amdgpu_get_pp_od_clk_voltage, - amdgpu_set_pp_od_clk_voltage); -static DEVICE_ATTR(gpu_busy_percent, S_IRUGO, - amdgpu_get_busy_percent, NULL); -static DEVICE_ATTR(mem_busy_percent, S_IRUGO, - amdgpu_get_memory_busy_percent, NULL); -static DEVICE_ATTR(pcie_bw, S_IRUGO, amdgpu_get_pcie_bw, NULL); -static DEVICE_ATTR(pp_features, S_IRUGO | S_IWUSR, - amdgpu_get_pp_feature_status, - amdgpu_set_pp_feature_status); -static DEVICE_ATTR(unique_id, S_IRUGO, amdgpu_get_unique_id, NULL); +static struct amdgpu_device_attr amdgpu_device_attrs[] = { + AMDGPU_DEVICE_ATTR_RW(power_dpm_state, ATTR_FLAG_BASIC|ATTR_FLAG_ONEVF), + AMDGPU_DEVICE_ATTR_RW(power_dpm_force_performance_level, ATTR_FLAG_BASIC|ATTR_FLAG_ONEVF), + AMDGPU_DEVICE_ATTR_RO(pp_num_states, ATTR_FLAG_BASIC), + AMDGPU_DEVICE_ATTR_RO(pp_cur_state, ATTR_FLAG_BASIC), + AMDGPU_DEVICE_ATTR_RW(pp_force_state, ATTR_FLAG_BASIC), + AMDGPU_DEVICE_ATTR_RW(pp_table, ATTR_FLAG_BASIC), + AMDGPU_DEVICE_ATTR_RW(pp_dpm_sclk, ATTR_FLAG_BASIC|ATTR_FLAG_ONEVF), + AMDGPU_DEVICE_ATTR_RW(pp_dpm_mclk, ATTR_FLAG_BASIC|ATTR_FLAG_ONEVF), + AMDGPU_DEVICE_ATTR_RW(pp_dpm_socclk, ATTR_FLAG_BASIC|ATTR_FLAG_ONEVF), + AMDGPU_DEVICE_ATTR_RW(pp_dpm_fclk, ATTR_FLAG_BASIC|ATTR_FLAG_ONEVF), + AMDGPU_DEVICE_ATTR_RW(pp_dpm_dcefclk, ATTR_FLAG_BASIC), + AMDGPU_DEVICE_ATTR_RW(pp_dpm_pcie, ATTR_FLAG_BASIC), + AMDGPU_DEVICE_ATTR_RW(pp_sclk_od, ATTR_FLAG_BASIC), + AMDGPU_DEVICE_ATTR_RW(pp_mclk_od, ATTR_FLAG_BASIC), + AMDGPU_DEVICE_ATTR_RW(pp_power_profile_mode, ATTR_FLAG_BASIC), + AMDGPU_DEVICE_ATTR_RW(pp_od_clk_voltage, ATTR_FLAG_BASIC), + AMDGPU_DEVICE_ATTR_RO(gpu_busy_percent, ATTR_FLAG_BASIC), + AMDGPU_DEVICE_ATTR_RO(mem_busy_percent, ATTR_FLAG_BASIC), + AMDGPU_DEVICE_ATTR_RO(pcie_bw, ATTR_FLAG_BASIC), + AMDGPU_DEVICE_ATTR_RW(pp_features, ATTR_FLAG_BASIC), + AMDGPU_DEVICE_ATTR_RO(unique_id, ATTR_FLAG_BASIC), +}; + +static int default_attr_update(struct amdgpu_device *adev, struct amdgpu_device_attr *attr, + uint32_t mask, enum amdgpu_device_attr_states *states) +{ + struct device_attribute *dev_attr = &attr->dev_attr; + const char *attr_name = dev_attr->attr.name; + struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle; + enum amd_asic_type asic_type = adev->asic_type; + + if (!(attr->flags & mask)) { + *states = ATTR_STATE_UNSUPPORTED; + return 0; + } + +#define DEVICE_ATTR_IS(_name) (!strcmp(attr_name, #_name)) + + if (DEVICE_ATTR_IS(pp_dpm_socclk)) { + if (asic_type < CHIP_VEGA10) + *states = ATTR_STATE_UNSUPPORTED; + } else if (DEVICE_ATTR_IS(pp_dpm_dcefclk)) { + if (asic_type < CHIP_VEGA10 || asic_type == CHIP_ARCTURUS) + *states = ATTR_STATE_UNSUPPORTED; + } else if (DEVICE_ATTR_IS(pp_dpm_fclk)) { + if (asic_type < CHIP_VEGA20) + *states = ATTR_STATE_UNSUPPORTED; + } else if (DEVICE_ATTR_IS(pp_dpm_pcie)) { + if (asic_type == CHIP_ARCTURUS) + *states = ATTR_STATE_UNSUPPORTED; + } else if (DEVICE_ATTR_IS(pp_od_clk_voltage)) { + *states = ATTR_STATE_UNSUPPORTED; + if ((is_support_sw_smu(adev) && adev->smu.od_enabled) || + (!is_support_sw_smu(adev) && hwmgr->od_enabled)) + *states = ATTR_STATE_SUPPORTED; + } else if (DEVICE_ATTR_IS(mem_busy_percent)) { + if (adev->flags & AMD_IS_APU || asic_type == CHIP_VEGA10) + *states = ATTR_STATE_UNSUPPORTED; + } else if (DEVICE_ATTR_IS(pcie_bw)) { + /* PCIe Perf counters won't work on APU nodes */ + if (adev->flags & AMD_IS_APU) + *states = ATTR_STATE_UNSUPPORTED; + } else if (DEVICE_ATTR_IS(unique_id)) { + if (!adev->unique_id) + *states = ATTR_STATE_UNSUPPORTED; + } else if (DEVICE_ATTR_IS(pp_features)) { + if (adev->flags & AMD_IS_APU || asic_type < CHIP_VEGA10) + *states = ATTR_STATE_UNSUPPORTED; + } + + if (asic_type == CHIP_ARCTURUS) { + /* Arcturus does not support standalone mclk/socclk/fclk level setting */ + if (DEVICE_ATTR_IS(pp_dpm_mclk) || + DEVICE_ATTR_IS(pp_dpm_socclk) || + DEVICE_ATTR_IS(pp_dpm_fclk)) { + dev_attr->attr.mode &= ~S_IWUGO; + dev_attr->store = NULL; + } + } + +#undef DEVICE_ATTR_IS + + return 0; +} + + +static int amdgpu_device_attr_create(struct amdgpu_device *adev, + struct amdgpu_device_attr *attr, + uint32_t mask, struct list_head *attr_list) +{ + int ret = 0; + struct device_attribute *dev_attr = &attr->dev_attr; + const char *name = dev_attr->attr.name; + enum amdgpu_device_attr_states attr_states = ATTR_STATE_SUPPORTED; + struct amdgpu_device_attr_entry *attr_entry; + + int (*attr_update)(struct amdgpu_device *adev, struct amdgpu_device_attr *attr, + uint32_t mask, enum amdgpu_device_attr_states *states) = default_attr_update; + + BUG_ON(!attr); + + attr_update = attr->attr_update ? attr_update : default_attr_update; + + ret = attr_update(adev, attr, mask, &attr_states); + if (ret) { + dev_err(adev->dev, "failed to update device file %s, ret = %d\n", + name, ret); + return ret; + } + + if (attr_states == ATTR_STATE_UNSUPPORTED) + return 0; + + ret = device_create_file(adev->dev, dev_attr); + if (ret) { + dev_err(adev->dev, "failed to create device file %s, ret = %d\n", + name, ret); + } + + attr_entry = kmalloc(sizeof(*attr_entry), GFP_KERNEL); + if (!attr_entry) + return -ENOMEM; + + attr_entry->attr = attr; + INIT_LIST_HEAD(&attr_entry->entry); + + list_add_tail(&attr_entry->entry, attr_list); + + return ret; +} + +static void amdgpu_device_attr_remove(struct amdgpu_device *adev, struct amdgpu_device_attr *attr) +{ + struct device_attribute *dev_attr = &attr->dev_attr; + + device_remove_file(adev->dev, dev_attr); +} + +static void amdgpu_device_attr_remove_groups(struct amdgpu_device *adev, + struct list_head *attr_list); + +static int amdgpu_device_attr_create_groups(struct amdgpu_device *adev, + struct amdgpu_device_attr *attrs, + uint32_t counts, + uint32_t mask, + struct list_head *attr_list) +{ + int ret = 0; + uint32_t i = 0; + + for (i = 0; i < counts; i++) { + ret = amdgpu_device_attr_create(adev, &attrs[i], mask, attr_list); + if (ret) + goto failed; + } + + return 0; + +failed: + amdgpu_device_attr_remove_groups(adev, attr_list); + + return ret; +} + +static void amdgpu_device_attr_remove_groups(struct amdgpu_device *adev, + struct list_head *attr_list) +{ + struct amdgpu_device_attr_entry *entry, *entry_tmp; + + if (list_empty(attr_list)) + return ; + + list_for_each_entry_safe(entry, entry_tmp, attr_list, entry) { + amdgpu_device_attr_remove(adev, entry->attr); + list_del(&entry->entry); + kfree(entry); + } +} static ssize_t amdgpu_hwmon_show_temp(struct device *dev, struct device_attribute *attr, @@ -1847,6 +1996,9 @@ static ssize_t amdgpu_hwmon_show_temp(struct device *dev, int channel = to_sensor_dev_attr(attr)->index; int r, temp = 0, size = sizeof(temp); + if (adev->in_gpu_reset) + return -EPERM; + if (channel >= PP_TEMP_MAX) return -EINVAL; @@ -1978,6 +2130,9 @@ static ssize_t amdgpu_hwmon_get_pwm1_enable(struct device *dev, u32 pwm_mode = 0; int ret; + if (adev->in_gpu_reset) + return -EPERM; + ret = pm_runtime_get_sync(adev->ddev->dev); if (ret < 0) return ret; @@ -2009,6 +2164,9 @@ static ssize_t amdgpu_hwmon_set_pwm1_enable(struct device *dev, int err, ret; int value; + if (adev->in_gpu_reset) + return -EPERM; + err = kstrtoint(buf, 10, &value); if (err) return err; @@ -2058,6 +2216,9 @@ static ssize_t amdgpu_hwmon_set_pwm1(struct device *dev, u32 value; u32 pwm_mode; + if (adev->in_gpu_reset) + return -EPERM; + err = pm_runtime_get_sync(adev->ddev->dev); if (err < 0) return err; @@ -2107,6 +2268,9 @@ static ssize_t amdgpu_hwmon_get_pwm1(struct device *dev, int err; u32 speed = 0; + if (adev->in_gpu_reset) + return -EPERM; + err = pm_runtime_get_sync(adev->ddev->dev); if (err < 0) return err; @@ -2137,6 +2301,9 @@ static ssize_t amdgpu_hwmon_get_fan1_input(struct device *dev, int err; u32 speed = 0; + if (adev->in_gpu_reset) + return -EPERM; + err = pm_runtime_get_sync(adev->ddev->dev); if (err < 0) return err; @@ -2166,6 +2333,9 @@ static ssize_t amdgpu_hwmon_get_fan1_min(struct device *dev, u32 size = sizeof(min_rpm); int r; + if (adev->in_gpu_reset) + return -EPERM; + r = pm_runtime_get_sync(adev->ddev->dev); if (r < 0) return r; @@ -2191,6 +2361,9 @@ static ssize_t amdgpu_hwmon_get_fan1_max(struct device *dev, u32 size = sizeof(max_rpm); int r; + if (adev->in_gpu_reset) + return -EPERM; + r = pm_runtime_get_sync(adev->ddev->dev); if (r < 0) return r; @@ -2215,6 +2388,9 @@ static ssize_t amdgpu_hwmon_get_fan1_target(struct device *dev, int err; u32 rpm = 0; + if (adev->in_gpu_reset) + return -EPERM; + err = pm_runtime_get_sync(adev->ddev->dev); if (err < 0) return err; @@ -2244,6 +2420,9 @@ static ssize_t amdgpu_hwmon_set_fan1_target(struct device *dev, u32 value; u32 pwm_mode; + if (adev->in_gpu_reset) + return -EPERM; + err = pm_runtime_get_sync(adev->ddev->dev); if (err < 0) return err; @@ -2290,6 +2469,9 @@ static ssize_t amdgpu_hwmon_get_fan1_enable(struct device *dev, u32 pwm_mode = 0; int ret; + if (adev->in_gpu_reset) + return -EPERM; + ret = pm_runtime_get_sync(adev->ddev->dev); if (ret < 0) return ret; @@ -2322,6 +2504,9 @@ static ssize_t amdgpu_hwmon_set_fan1_enable(struct device *dev, int value; u32 pwm_mode; + if (adev->in_gpu_reset) + return -EPERM; + err = kstrtoint(buf, 10, &value); if (err) return err; @@ -2362,6 +2547,9 @@ static ssize_t amdgpu_hwmon_show_vddgfx(struct device *dev, u32 vddgfx; int r, size = sizeof(vddgfx); + if (adev->in_gpu_reset) + return -EPERM; + r = pm_runtime_get_sync(adev->ddev->dev); if (r < 0) return r; @@ -2394,6 +2582,9 @@ static ssize_t amdgpu_hwmon_show_vddnb(struct device *dev, u32 vddnb; int r, size = sizeof(vddnb); + if (adev->in_gpu_reset) + return -EPERM; + /* only APUs have vddnb */ if (!(adev->flags & AMD_IS_APU)) return -EINVAL; @@ -2431,6 +2622,9 @@ static ssize_t amdgpu_hwmon_show_power_avg(struct device *dev, int r, size = sizeof(u32); unsigned uw; + if (adev->in_gpu_reset) + return -EPERM; + r = pm_runtime_get_sync(adev->ddev->dev); if (r < 0) return r; @@ -2467,6 +2661,9 @@ static ssize_t amdgpu_hwmon_show_power_cap_max(struct device *dev, ssize_t size; int r; + if (adev->in_gpu_reset) + return -EPERM; + r = pm_runtime_get_sync(adev->ddev->dev); if (r < 0) return r; @@ -2496,6 +2693,9 @@ static ssize_t amdgpu_hwmon_show_power_cap(struct device *dev, ssize_t size; int r; + if (adev->in_gpu_reset) + return -EPERM; + r = pm_runtime_get_sync(adev->ddev->dev); if (r < 0) return r; @@ -2526,6 +2726,9 @@ static ssize_t amdgpu_hwmon_set_power_cap(struct device *dev, int err; u32 value; + if (adev->in_gpu_reset) + return -EPERM; + if (amdgpu_sriov_vf(adev)) return -EINVAL; @@ -2564,6 +2767,9 @@ static ssize_t amdgpu_hwmon_show_sclk(struct device *dev, uint32_t sclk; int r, size = sizeof(sclk); + if (adev->in_gpu_reset) + return -EPERM; + r = pm_runtime_get_sync(adev->ddev->dev); if (r < 0) return r; @@ -2596,6 +2802,9 @@ static ssize_t amdgpu_hwmon_show_mclk(struct device *dev, uint32_t mclk; int r, size = sizeof(mclk); + if (adev->in_gpu_reset) + return -EPERM; + r = pm_runtime_get_sync(adev->ddev->dev); if (r < 0) return r; @@ -3238,8 +3447,8 @@ int amdgpu_pm_load_smu_firmware(struct amdgpu_device *adev, uint32_t *smu_versio int amdgpu_pm_sysfs_init(struct amdgpu_device *adev) { - struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle; int ret; + uint32_t mask = 0; if (adev->pm.sysfs_initialized) return 0; @@ -3247,6 +3456,8 @@ int amdgpu_pm_sysfs_init(struct amdgpu_device *adev) if (adev->pm.dpm_enabled == 0) return 0; + INIT_LIST_HEAD(&adev->pm.pm_attr_list); + adev->pm.int_hwmon_dev = hwmon_device_register_with_groups(adev->dev, DRIVER_NAME, adev, hwmon_groups); @@ -3257,160 +3468,26 @@ int amdgpu_pm_sysfs_init(struct amdgpu_device *adev) return ret; } - ret = device_create_file(adev->dev, &dev_attr_power_dpm_state); - if (ret) { - DRM_ERROR("failed to create device file for dpm state\n"); - return ret; - } - ret = device_create_file(adev->dev, &dev_attr_power_dpm_force_performance_level); - if (ret) { - DRM_ERROR("failed to create device file for dpm state\n"); - return ret; - } - - - ret = device_create_file(adev->dev, &dev_attr_pp_num_states); - if (ret) { - DRM_ERROR("failed to create device file pp_num_states\n"); - return ret; - } - ret = device_create_file(adev->dev, &dev_attr_pp_cur_state); - if (ret) { - DRM_ERROR("failed to create device file pp_cur_state\n"); - return ret; - } - ret = device_create_file(adev->dev, &dev_attr_pp_force_state); - if (ret) { - DRM_ERROR("failed to create device file pp_force_state\n"); - return ret; - } - ret = device_create_file(adev->dev, &dev_attr_pp_table); - if (ret) { - DRM_ERROR("failed to create device file pp_table\n"); - return ret; - } - - ret = device_create_file(adev->dev, &dev_attr_pp_dpm_sclk); - if (ret) { - DRM_ERROR("failed to create device file pp_dpm_sclk\n"); - return ret; - } - - /* Arcturus does not support standalone mclk/socclk/fclk level setting */ - if (adev->asic_type == CHIP_ARCTURUS) { - dev_attr_pp_dpm_mclk.attr.mode &= ~S_IWUGO; - dev_attr_pp_dpm_mclk.store = NULL; - - dev_attr_pp_dpm_socclk.attr.mode &= ~S_IWUGO; - dev_attr_pp_dpm_socclk.store = NULL; - - dev_attr_pp_dpm_fclk.attr.mode &= ~S_IWUGO; - dev_attr_pp_dpm_fclk.store = NULL; + switch (amdgpu_virt_get_sriov_vf_mode(adev)) { + case SRIOV_VF_MODE_ONE_VF: + mask = ATTR_FLAG_ONEVF; + break; + case SRIOV_VF_MODE_MULTI_VF: + mask = 0; + break; + case SRIOV_VF_MODE_BARE_METAL: + default: + mask = ATTR_FLAG_MASK_ALL; + break; } - ret = device_create_file(adev->dev, &dev_attr_pp_dpm_mclk); - if (ret) { - DRM_ERROR("failed to create device file pp_dpm_mclk\n"); - return ret; - } - if (adev->asic_type >= CHIP_VEGA10) { - ret = device_create_file(adev->dev, &dev_attr_pp_dpm_socclk); - if (ret) { - DRM_ERROR("failed to create device file pp_dpm_socclk\n"); - return ret; - } - if (adev->asic_type != CHIP_ARCTURUS) { - ret = device_create_file(adev->dev, &dev_attr_pp_dpm_dcefclk); - if (ret) { - DRM_ERROR("failed to create device file pp_dpm_dcefclk\n"); - return ret; - } - } - } - if (adev->asic_type >= CHIP_VEGA20) { - ret = device_create_file(adev->dev, &dev_attr_pp_dpm_fclk); - if (ret) { - DRM_ERROR("failed to create device file pp_dpm_fclk\n"); - return ret; - } - } - if (adev->asic_type != CHIP_ARCTURUS) { - ret = device_create_file(adev->dev, &dev_attr_pp_dpm_pcie); - if (ret) { - DRM_ERROR("failed to create device file pp_dpm_pcie\n"); - return ret; - } - } - ret = device_create_file(adev->dev, &dev_attr_pp_sclk_od); - if (ret) { - DRM_ERROR("failed to create device file pp_sclk_od\n"); - return ret; - } - ret = device_create_file(adev->dev, &dev_attr_pp_mclk_od); - if (ret) { - DRM_ERROR("failed to create device file pp_mclk_od\n"); - return ret; - } - ret = device_create_file(adev->dev, - &dev_attr_pp_power_profile_mode); - if (ret) { - DRM_ERROR("failed to create device file " - "pp_power_profile_mode\n"); - return ret; - } - if ((is_support_sw_smu(adev) && adev->smu.od_enabled) || - (!is_support_sw_smu(adev) && hwmgr->od_enabled)) { - ret = device_create_file(adev->dev, - &dev_attr_pp_od_clk_voltage); - if (ret) { - DRM_ERROR("failed to create device file " - "pp_od_clk_voltage\n"); - return ret; - } - } - ret = device_create_file(adev->dev, - &dev_attr_gpu_busy_percent); - if (ret) { - DRM_ERROR("failed to create device file " - "gpu_busy_level\n"); - return ret; - } - /* APU does not have its own dedicated memory */ - if (!(adev->flags & AMD_IS_APU) && - (adev->asic_type != CHIP_VEGA10)) { - ret = device_create_file(adev->dev, - &dev_attr_mem_busy_percent); - if (ret) { - DRM_ERROR("failed to create device file " - "mem_busy_percent\n"); - return ret; - } - } - /* PCIe Perf counters won't work on APU nodes */ - if (!(adev->flags & AMD_IS_APU)) { - ret = device_create_file(adev->dev, &dev_attr_pcie_bw); - if (ret) { - DRM_ERROR("failed to create device file pcie_bw\n"); - return ret; - } - } - if (adev->unique_id) - ret = device_create_file(adev->dev, &dev_attr_unique_id); - if (ret) { - DRM_ERROR("failed to create device file unique_id\n"); + ret = amdgpu_device_attr_create_groups(adev, + amdgpu_device_attrs, + ARRAY_SIZE(amdgpu_device_attrs), + mask, + &adev->pm.pm_attr_list); + if (ret) return ret; - } - - if ((adev->asic_type >= CHIP_VEGA10) && - !(adev->flags & AMD_IS_APU)) { - ret = device_create_file(adev->dev, - &dev_attr_pp_features); - if (ret) { - DRM_ERROR("failed to create device file " - "pp_features\n"); - return ret; - } - } adev->pm.sysfs_initialized = true; @@ -3419,51 +3496,13 @@ int amdgpu_pm_sysfs_init(struct amdgpu_device *adev) void amdgpu_pm_sysfs_fini(struct amdgpu_device *adev) { - struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle; - if (adev->pm.dpm_enabled == 0) return; if (adev->pm.int_hwmon_dev) hwmon_device_unregister(adev->pm.int_hwmon_dev); - device_remove_file(adev->dev, &dev_attr_power_dpm_state); - device_remove_file(adev->dev, &dev_attr_power_dpm_force_performance_level); - - device_remove_file(adev->dev, &dev_attr_pp_num_states); - device_remove_file(adev->dev, &dev_attr_pp_cur_state); - device_remove_file(adev->dev, &dev_attr_pp_force_state); - device_remove_file(adev->dev, &dev_attr_pp_table); - - device_remove_file(adev->dev, &dev_attr_pp_dpm_sclk); - device_remove_file(adev->dev, &dev_attr_pp_dpm_mclk); - if (adev->asic_type >= CHIP_VEGA10) { - device_remove_file(adev->dev, &dev_attr_pp_dpm_socclk); - if (adev->asic_type != CHIP_ARCTURUS) - device_remove_file(adev->dev, &dev_attr_pp_dpm_dcefclk); - } - if (adev->asic_type != CHIP_ARCTURUS) - device_remove_file(adev->dev, &dev_attr_pp_dpm_pcie); - if (adev->asic_type >= CHIP_VEGA20) - device_remove_file(adev->dev, &dev_attr_pp_dpm_fclk); - device_remove_file(adev->dev, &dev_attr_pp_sclk_od); - device_remove_file(adev->dev, &dev_attr_pp_mclk_od); - device_remove_file(adev->dev, - &dev_attr_pp_power_profile_mode); - if ((is_support_sw_smu(adev) && adev->smu.od_enabled) || - (!is_support_sw_smu(adev) && hwmgr->od_enabled)) - device_remove_file(adev->dev, - &dev_attr_pp_od_clk_voltage); - device_remove_file(adev->dev, &dev_attr_gpu_busy_percent); - if (!(adev->flags & AMD_IS_APU) && - (adev->asic_type != CHIP_VEGA10)) - device_remove_file(adev->dev, &dev_attr_mem_busy_percent); - if (!(adev->flags & AMD_IS_APU)) - device_remove_file(adev->dev, &dev_attr_pcie_bw); - if (adev->unique_id) - device_remove_file(adev->dev, &dev_attr_unique_id); - if ((adev->asic_type >= CHIP_VEGA10) && - !(adev->flags & AMD_IS_APU)) - device_remove_file(adev->dev, &dev_attr_pp_features); + + amdgpu_device_attr_remove_groups(adev, &adev->pm.pm_attr_list); } void amdgpu_pm_compute_clocks(struct amdgpu_device *adev) @@ -3626,6 +3665,9 @@ static int amdgpu_debugfs_pm_info(struct seq_file *m, void *data) u32 flags = 0; int r; + if (adev->in_gpu_reset) + return -EPERM; + r = pm_runtime_get_sync(dev->dev); if (r < 0) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.h index 5db0ef86e84c..d9ae2b49a402 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.h @@ -30,6 +30,55 @@ struct cg_flag_name const char *name; }; +enum amdgpu_device_attr_flags { + ATTR_FLAG_BASIC = (1 << 0), + ATTR_FLAG_ONEVF = (1 << 16), +}; + +#define ATTR_FLAG_TYPE_MASK (0x0000ffff) +#define ATTR_FLAG_MODE_MASK (0xffff0000) +#define ATTR_FLAG_MASK_ALL (0xffffffff) + +enum amdgpu_device_attr_states { + ATTR_STATE_UNSUPPORTED = 0, + ATTR_STATE_SUPPORTED, +}; + +struct amdgpu_device_attr { + struct device_attribute dev_attr; + enum amdgpu_device_attr_flags flags; + int (*attr_update)(struct amdgpu_device *adev, struct amdgpu_device_attr *attr, + uint32_t mask, enum amdgpu_device_attr_states *states); + +}; + +struct amdgpu_device_attr_entry { + struct list_head entry; + struct amdgpu_device_attr *attr; +}; + +#define to_amdgpu_device_attr(_dev_attr) \ + container_of(_dev_attr, struct amdgpu_device_attr, dev_attr) + +#define __AMDGPU_DEVICE_ATTR(_name, _mode, _show, _store, _flags, ...) \ + { .dev_attr = __ATTR(_name, _mode, _show, _store), \ + .flags = _flags, \ + ##__VA_ARGS__, } + +#define AMDGPU_DEVICE_ATTR(_name, _mode, _flags, ...) \ + __AMDGPU_DEVICE_ATTR(_name, _mode, \ + amdgpu_get_##_name, amdgpu_set_##_name, \ + _flags, ##__VA_ARGS__) + +#define AMDGPU_DEVICE_ATTR_RW(_name, _flags, ...) \ + AMDGPU_DEVICE_ATTR(_name, S_IRUGO | S_IWUSR, \ + _flags, ##__VA_ARGS__) + +#define AMDGPU_DEVICE_ATTR_RO(_name, _flags, ...) \ + __AMDGPU_DEVICE_ATTR(_name, S_IRUGO, \ + amdgpu_get_##_name, NULL, \ + _flags, ##__VA_ARGS__) + void amdgpu_pm_acpi_event_handler(struct amdgpu_device *adev); int amdgpu_pm_sysfs_init(struct amdgpu_device *adev); int amdgpu_pm_virt_sysfs_init(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index deaa26808841..7301fdcfb8bc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -37,11 +37,11 @@ #include "amdgpu_ras.h" -static void psp_set_funcs(struct amdgpu_device *adev); - static int psp_sysfs_init(struct amdgpu_device *adev); static void psp_sysfs_fini(struct amdgpu_device *adev); +static int psp_load_smu_fw(struct psp_context *psp); + /* * Due to DF Cstate management centralized to PMFW, the firmware * loading sequence will be updated as below: @@ -80,8 +80,6 @@ static int psp_early_init(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; struct psp_context *psp = &adev->psp; - psp_set_funcs(adev); - switch (adev->asic_type) { case CHIP_VEGA10: case CHIP_VEGA12: @@ -201,6 +199,7 @@ psp_cmd_submit_buf(struct psp_context *psp, int index; int timeout = 2000; bool ras_intr = false; + bool skip_unsupport = false; mutex_lock(&psp->mutex); @@ -232,6 +231,9 @@ psp_cmd_submit_buf(struct psp_context *psp, amdgpu_asic_invalidate_hdp(psp->adev, NULL); } + /* We allow TEE_ERROR_NOT_SUPPORTED for VMR command in SRIOV */ + skip_unsupport = (psp->cmd_buf_mem->resp.status == 0xffff000a) && amdgpu_sriov_vf(psp->adev); + /* In some cases, psp response status is not 0 even there is no * problem while the command is submitted. Some version of PSP FW * doesn't write 0 to that field. @@ -239,7 +241,7 @@ psp_cmd_submit_buf(struct psp_context *psp, * during psp initialization to avoid breaking hw_init and it doesn't * return -EINVAL. */ - if ((psp->cmd_buf_mem->resp.status || !timeout) && !ras_intr) { + if (!skip_unsupport && (psp->cmd_buf_mem->resp.status || !timeout) && !ras_intr) { if (ucode) DRM_WARN("failed to load ucode id (%d) ", ucode->ucode_id); @@ -268,7 +270,7 @@ static void psp_prep_tmr_cmd_buf(struct psp_context *psp, struct psp_gfx_cmd_resp *cmd, uint64_t tmr_mc, uint32_t size) { - if (psp_support_vmr_ring(psp)) + if (amdgpu_sriov_vf(psp->adev)) cmd->cmd_id = GFX_CMD_ID_SETUP_VMR; else cmd->cmd_id = GFX_CMD_ID_SETUP_TMR; @@ -662,6 +664,121 @@ int psp_xgmi_initialize(struct psp_context *psp) return ret; } +int psp_xgmi_get_hive_id(struct psp_context *psp, uint64_t *hive_id) +{ + struct ta_xgmi_shared_memory *xgmi_cmd; + int ret; + + xgmi_cmd = (struct ta_xgmi_shared_memory*)psp->xgmi_context.xgmi_shared_buf; + memset(xgmi_cmd, 0, sizeof(struct ta_xgmi_shared_memory)); + + xgmi_cmd->cmd_id = TA_COMMAND_XGMI__GET_HIVE_ID; + + /* Invoke xgmi ta to get hive id */ + ret = psp_xgmi_invoke(psp, xgmi_cmd->cmd_id); + if (ret) + return ret; + + *hive_id = xgmi_cmd->xgmi_out_message.get_hive_id.hive_id; + + return 0; +} + +int psp_xgmi_get_node_id(struct psp_context *psp, uint64_t *node_id) +{ + struct ta_xgmi_shared_memory *xgmi_cmd; + int ret; + + xgmi_cmd = (struct ta_xgmi_shared_memory*)psp->xgmi_context.xgmi_shared_buf; + memset(xgmi_cmd, 0, sizeof(struct ta_xgmi_shared_memory)); + + xgmi_cmd->cmd_id = TA_COMMAND_XGMI__GET_NODE_ID; + + /* Invoke xgmi ta to get the node id */ + ret = psp_xgmi_invoke(psp, xgmi_cmd->cmd_id); + if (ret) + return ret; + + *node_id = xgmi_cmd->xgmi_out_message.get_node_id.node_id; + + return 0; +} + +int psp_xgmi_get_topology_info(struct psp_context *psp, + int number_devices, + struct psp_xgmi_topology_info *topology) +{ + struct ta_xgmi_shared_memory *xgmi_cmd; + struct ta_xgmi_cmd_get_topology_info_input *topology_info_input; + struct ta_xgmi_cmd_get_topology_info_output *topology_info_output; + int i; + int ret; + + if (!topology || topology->num_nodes > TA_XGMI__MAX_CONNECTED_NODES) + return -EINVAL; + + xgmi_cmd = (struct ta_xgmi_shared_memory*)psp->xgmi_context.xgmi_shared_buf; + memset(xgmi_cmd, 0, sizeof(struct ta_xgmi_shared_memory)); + + /* Fill in the shared memory with topology information as input */ + topology_info_input = &xgmi_cmd->xgmi_in_message.get_topology_info; + xgmi_cmd->cmd_id = TA_COMMAND_XGMI__GET_GET_TOPOLOGY_INFO; + topology_info_input->num_nodes = number_devices; + + for (i = 0; i < topology_info_input->num_nodes; i++) { + topology_info_input->nodes[i].node_id = topology->nodes[i].node_id; + topology_info_input->nodes[i].num_hops = topology->nodes[i].num_hops; + topology_info_input->nodes[i].is_sharing_enabled = topology->nodes[i].is_sharing_enabled; + topology_info_input->nodes[i].sdma_engine = topology->nodes[i].sdma_engine; + } + + /* Invoke xgmi ta to get the topology information */ + ret = psp_xgmi_invoke(psp, TA_COMMAND_XGMI__GET_GET_TOPOLOGY_INFO); + if (ret) + return ret; + + /* Read the output topology information from the shared memory */ + topology_info_output = &xgmi_cmd->xgmi_out_message.get_topology_info; + topology->num_nodes = xgmi_cmd->xgmi_out_message.get_topology_info.num_nodes; + for (i = 0; i < topology->num_nodes; i++) { + topology->nodes[i].node_id = topology_info_output->nodes[i].node_id; + topology->nodes[i].num_hops = topology_info_output->nodes[i].num_hops; + topology->nodes[i].is_sharing_enabled = topology_info_output->nodes[i].is_sharing_enabled; + topology->nodes[i].sdma_engine = topology_info_output->nodes[i].sdma_engine; + } + + return 0; +} + +int psp_xgmi_set_topology_info(struct psp_context *psp, + int number_devices, + struct psp_xgmi_topology_info *topology) +{ + struct ta_xgmi_shared_memory *xgmi_cmd; + struct ta_xgmi_cmd_get_topology_info_input *topology_info_input; + int i; + + if (!topology || topology->num_nodes > TA_XGMI__MAX_CONNECTED_NODES) + return -EINVAL; + + xgmi_cmd = (struct ta_xgmi_shared_memory*)psp->xgmi_context.xgmi_shared_buf; + memset(xgmi_cmd, 0, sizeof(struct ta_xgmi_shared_memory)); + + topology_info_input = &xgmi_cmd->xgmi_in_message.get_topology_info; + xgmi_cmd->cmd_id = TA_COMMAND_XGMI__SET_TOPOLOGY_INFO; + topology_info_input->num_nodes = number_devices; + + for (i = 0; i < topology_info_input->num_nodes; i++) { + topology_info_input->nodes[i].node_id = topology->nodes[i].node_id; + topology_info_input->nodes[i].num_hops = topology->nodes[i].num_hops; + topology_info_input->nodes[i].is_sharing_enabled = 1; + topology_info_input->nodes[i].sdma_engine = topology->nodes[i].sdma_engine; + } + + /* Invoke xgmi ta to set topology information */ + return psp_xgmi_invoke(psp, TA_COMMAND_XGMI__SET_TOPOLOGY_INFO); +} + // ras begin static int psp_ras_init_shared_buf(struct psp_context *psp) { @@ -744,13 +861,40 @@ static int psp_ras_unload(struct psp_context *psp) int psp_ras_invoke(struct psp_context *psp, uint32_t ta_cmd_id) { + struct ta_ras_shared_memory *ras_cmd; + int ret; + + ras_cmd = (struct ta_ras_shared_memory *)psp->ras.ras_shared_buf; + /* * TODO: bypass the loading in sriov for now */ if (amdgpu_sriov_vf(psp->adev)) return 0; - return psp_ta_invoke(psp, ta_cmd_id, psp->ras.session_id); + ret = psp_ta_invoke(psp, ta_cmd_id, psp->ras.session_id); + + if (amdgpu_ras_intr_triggered()) + return ret; + + if (ras_cmd->if_version > RAS_TA_HOST_IF_VER) + { + DRM_WARN("RAS: Unsupported Interface"); + return -EINVAL; + } + + if (!ret) { + if (ras_cmd->ras_out_message.flags.err_inject_switch_disable_flag) { + dev_warn(psp->adev->dev, "ECC switch disabled\n"); + + ras_cmd->ras_status = TA_RAS_STATUS__ERROR_RAS_NOT_AVAILABLE; + } + else if (ras_cmd->ras_out_message.flags.reg_access_failure_flag) + dev_warn(psp->adev->dev, + "RAS internal register access blocked\n"); + } + + return ret; } int psp_ras_enable_features(struct psp_context *psp, @@ -834,6 +978,33 @@ static int psp_ras_initialize(struct psp_context *psp) return 0; } + +int psp_ras_trigger_error(struct psp_context *psp, + struct ta_ras_trigger_error_input *info) +{ + struct ta_ras_shared_memory *ras_cmd; + int ret; + + if (!psp->ras.ras_initialized) + return -EINVAL; + + ras_cmd = (struct ta_ras_shared_memory *)psp->ras.ras_shared_buf; + memset(ras_cmd, 0, sizeof(struct ta_ras_shared_memory)); + + ras_cmd->cmd_id = TA_RAS_COMMAND__TRIGGER_ERROR; + ras_cmd->ras_in_message.trigger_error = *info; + + ret = psp_ras_invoke(psp, ras_cmd->cmd_id); + if (ret) + return -EINVAL; + + /* If err_event_athub occurs error inject was successful, however + return status from TA is no long reliable */ + if (amdgpu_ras_intr_triggered()) + return 0; + + return ras_cmd->ras_status; +} // ras end // HDCP start @@ -884,6 +1055,7 @@ static int psp_hdcp_load(struct psp_context *psp) if (!ret) { psp->hdcp_context.hdcp_initialized = true; psp->hdcp_context.session_id = cmd->resp.session_id; + mutex_init(&psp->hdcp_context.mutex); } kfree(cmd); @@ -1029,6 +1201,7 @@ static int psp_dtm_load(struct psp_context *psp) if (!ret) { psp->dtm_context.dtm_initialized = true; psp->dtm_context.session_id = cmd->resp.session_id; + mutex_init(&psp->dtm_context.mutex); } kfree(cmd); @@ -1169,16 +1342,20 @@ static int psp_hw_start(struct psp_context *psp) } /* - * For those ASICs with DF Cstate management centralized + * For ASICs with DF Cstate management centralized * to PMFW, TMR setup should be performed after PMFW * loaded and before other non-psp firmware loaded. */ - if (!psp->pmfw_centralized_cstate_management) { - ret = psp_tmr_load(psp); - if (ret) { - DRM_ERROR("PSP load tmr failed!\n"); + if (psp->pmfw_centralized_cstate_management) { + ret = psp_load_smu_fw(psp); + if (ret) return ret; - } + } + + ret = psp_tmr_load(psp); + if (ret) { + DRM_ERROR("PSP load tmr failed!\n"); + return ret; } return 0; @@ -1355,7 +1532,7 @@ static int psp_prep_load_ip_fw_cmd_buf(struct amdgpu_firmware_info *ucode, } static int psp_execute_np_fw_load(struct psp_context *psp, - struct amdgpu_firmware_info *ucode) + struct amdgpu_firmware_info *ucode) { int ret = 0; @@ -1369,64 +1546,96 @@ static int psp_execute_np_fw_load(struct psp_context *psp, return ret; } +static int psp_load_smu_fw(struct psp_context *psp) +{ + int ret; + struct amdgpu_device* adev = psp->adev; + struct amdgpu_firmware_info *ucode = + &adev->firmware.ucode[AMDGPU_UCODE_ID_SMC]; + struct amdgpu_ras *ras = psp->ras.ras; + + if (!ucode->fw || amdgpu_sriov_vf(psp->adev)) + return 0; + + + if (adev->in_gpu_reset && ras && ras->supported) { + ret = amdgpu_dpm_set_mp1_state(adev, PP_MP1_STATE_UNLOAD); + if (ret) { + DRM_WARN("Failed to set MP1 state prepare for reload\n"); + } + } + + ret = psp_execute_np_fw_load(psp, ucode); + + if (ret) + DRM_ERROR("PSP load smu failed!\n"); + + return ret; +} + +static bool fw_load_skip_check(struct psp_context *psp, + struct amdgpu_firmware_info *ucode) +{ + if (!ucode->fw) + return true; + + if (ucode->ucode_id == AMDGPU_UCODE_ID_SMC && + (psp_smu_reload_quirk(psp) || + psp->autoload_supported || + psp->pmfw_centralized_cstate_management)) + return true; + + if (amdgpu_sriov_vf(psp->adev) && + (ucode->ucode_id == AMDGPU_UCODE_ID_SDMA0 + || ucode->ucode_id == AMDGPU_UCODE_ID_SDMA1 + || ucode->ucode_id == AMDGPU_UCODE_ID_SDMA2 + || ucode->ucode_id == AMDGPU_UCODE_ID_SDMA3 + || ucode->ucode_id == AMDGPU_UCODE_ID_SDMA4 + || ucode->ucode_id == AMDGPU_UCODE_ID_SDMA5 + || ucode->ucode_id == AMDGPU_UCODE_ID_SDMA6 + || ucode->ucode_id == AMDGPU_UCODE_ID_SDMA7 + || ucode->ucode_id == AMDGPU_UCODE_ID_RLC_G + || ucode->ucode_id == AMDGPU_UCODE_ID_RLC_RESTORE_LIST_CNTL + || ucode->ucode_id == AMDGPU_UCODE_ID_RLC_RESTORE_LIST_GPM_MEM + || ucode->ucode_id == AMDGPU_UCODE_ID_RLC_RESTORE_LIST_SRM_MEM + || ucode->ucode_id == AMDGPU_UCODE_ID_SMC)) + /*skip ucode loading in SRIOV VF */ + return true; + + if (psp->autoload_supported && + (ucode->ucode_id == AMDGPU_UCODE_ID_CP_MEC1_JT || + ucode->ucode_id == AMDGPU_UCODE_ID_CP_MEC2_JT)) + /* skip mec JT when autoload is enabled */ + return true; + + return false; +} + static int psp_np_fw_load(struct psp_context *psp) { int i, ret; struct amdgpu_firmware_info *ucode; struct amdgpu_device* adev = psp->adev; - if (psp->autoload_supported || - psp->pmfw_centralized_cstate_management) { - ucode = &adev->firmware.ucode[AMDGPU_UCODE_ID_SMC]; - if (!ucode->fw || amdgpu_sriov_vf(adev)) - goto out; - - ret = psp_execute_np_fw_load(psp, ucode); + if (psp->autoload_supported && + !psp->pmfw_centralized_cstate_management) { + ret = psp_load_smu_fw(psp); if (ret) return ret; } - if (psp->pmfw_centralized_cstate_management) { - ret = psp_tmr_load(psp); - if (ret) { - DRM_ERROR("PSP load tmr failed!\n"); - return ret; - } - } - -out: for (i = 0; i < adev->firmware.max_ucodes; i++) { ucode = &adev->firmware.ucode[i]; - if (!ucode->fw) - continue; if (ucode->ucode_id == AMDGPU_UCODE_ID_SMC && - (psp_smu_reload_quirk(psp) || - psp->autoload_supported || - psp->pmfw_centralized_cstate_management)) - continue; - - if (amdgpu_sriov_vf(adev) && - (ucode->ucode_id == AMDGPU_UCODE_ID_SDMA0 - || ucode->ucode_id == AMDGPU_UCODE_ID_SDMA1 - || ucode->ucode_id == AMDGPU_UCODE_ID_SDMA2 - || ucode->ucode_id == AMDGPU_UCODE_ID_SDMA3 - || ucode->ucode_id == AMDGPU_UCODE_ID_SDMA4 - || ucode->ucode_id == AMDGPU_UCODE_ID_SDMA5 - || ucode->ucode_id == AMDGPU_UCODE_ID_SDMA6 - || ucode->ucode_id == AMDGPU_UCODE_ID_SDMA7 - || ucode->ucode_id == AMDGPU_UCODE_ID_RLC_G - || ucode->ucode_id == AMDGPU_UCODE_ID_RLC_RESTORE_LIST_CNTL - || ucode->ucode_id == AMDGPU_UCODE_ID_RLC_RESTORE_LIST_GPM_MEM - || ucode->ucode_id == AMDGPU_UCODE_ID_RLC_RESTORE_LIST_SRM_MEM - || ucode->ucode_id == AMDGPU_UCODE_ID_SMC)) - /*skip ucode loading in SRIOV VF */ + !fw_load_skip_check(psp, ucode)) { + ret = psp_load_smu_fw(psp); + if (ret) + return ret; continue; + } - if (psp->autoload_supported && - (ucode->ucode_id == AMDGPU_UCODE_ID_CP_MEC1_JT || - ucode->ucode_id == AMDGPU_UCODE_ID_CP_MEC2_JT)) - /* skip mec JT when autoload is enabled */ + if (fw_load_skip_check(psp, ucode)) continue; psp_print_fw_hdr(psp, ucode); @@ -1438,17 +1647,12 @@ out: /* Start rlc autoload after psp recieved all the gfx firmware */ if (psp->autoload_supported && ucode->ucode_id == (amdgpu_sriov_vf(adev) ? AMDGPU_UCODE_ID_CP_MEC2 : AMDGPU_UCODE_ID_RLC_G)) { - ret = psp_rlc_autoload(psp); + ret = psp_rlc_autoload_start(psp); if (ret) { DRM_ERROR("Failed to start rlc autoload\n"); return ret; } } -#if 0 - /* check if firmware loaded sucessfully */ - if (!amdgpu_psp_check_fw_loading_status(adev, i)) - return -EINVAL; -#endif } return 0; @@ -1806,19 +2010,110 @@ int psp_ring_cmd_submit(struct psp_context *psp, return 0; } -static bool psp_check_fw_loading_status(struct amdgpu_device *adev, - enum AMDGPU_UCODE_ID ucode_type) +int psp_init_asd_microcode(struct psp_context *psp, + const char *chip_name) { - struct amdgpu_firmware_info *ucode = NULL; + struct amdgpu_device *adev = psp->adev; + char fw_name[30]; + const struct psp_firmware_header_v1_0 *asd_hdr; + int err = 0; - if (!adev->firmware.fw_size) - return false; + if (!chip_name) { + dev_err(adev->dev, "invalid chip name for asd microcode\n"); + return -EINVAL; + } - ucode = &adev->firmware.ucode[ucode_type]; - if (!ucode->fw || !ucode->ucode_size) - return false; + snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_asd.bin", chip_name); + err = request_firmware(&adev->psp.asd_fw, fw_name, adev->dev); + if (err) + goto out; + + err = amdgpu_ucode_validate(adev->psp.asd_fw); + if (err) + goto out; + + asd_hdr = (const struct psp_firmware_header_v1_0 *)adev->psp.asd_fw->data; + adev->psp.asd_fw_version = le32_to_cpu(asd_hdr->header.ucode_version); + adev->psp.asd_feature_version = le32_to_cpu(asd_hdr->ucode_feature_version); + adev->psp.asd_ucode_size = le32_to_cpu(asd_hdr->header.ucode_size_bytes); + adev->psp.asd_start_addr = (uint8_t *)asd_hdr + + le32_to_cpu(asd_hdr->header.ucode_array_offset_bytes); + return 0; +out: + dev_err(adev->dev, "fail to initialize asd microcode\n"); + release_firmware(adev->psp.asd_fw); + adev->psp.asd_fw = NULL; + return err; +} + +int psp_init_sos_microcode(struct psp_context *psp, + const char *chip_name) +{ + struct amdgpu_device *adev = psp->adev; + char fw_name[30]; + const struct psp_firmware_header_v1_0 *sos_hdr; + const struct psp_firmware_header_v1_1 *sos_hdr_v1_1; + const struct psp_firmware_header_v1_2 *sos_hdr_v1_2; + int err = 0; + + if (!chip_name) { + dev_err(adev->dev, "invalid chip name for sos microcode\n"); + return -EINVAL; + } + + snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sos.bin", chip_name); + err = request_firmware(&adev->psp.sos_fw, fw_name, adev->dev); + if (err) + goto out; + + err = amdgpu_ucode_validate(adev->psp.sos_fw); + if (err) + goto out; + + sos_hdr = (const struct psp_firmware_header_v1_0 *)adev->psp.sos_fw->data; + amdgpu_ucode_print_psp_hdr(&sos_hdr->header); + + switch (sos_hdr->header.header_version_major) { + case 1: + adev->psp.sos_fw_version = le32_to_cpu(sos_hdr->header.ucode_version); + adev->psp.sos_feature_version = le32_to_cpu(sos_hdr->ucode_feature_version); + adev->psp.sos_bin_size = le32_to_cpu(sos_hdr->sos_size_bytes); + adev->psp.sys_bin_size = le32_to_cpu(sos_hdr->sos_offset_bytes); + adev->psp.sys_start_addr = (uint8_t *)sos_hdr + + le32_to_cpu(sos_hdr->header.ucode_array_offset_bytes); + adev->psp.sos_start_addr = (uint8_t *)adev->psp.sys_start_addr + + le32_to_cpu(sos_hdr->sos_offset_bytes); + if (sos_hdr->header.header_version_minor == 1) { + sos_hdr_v1_1 = (const struct psp_firmware_header_v1_1 *)adev->psp.sos_fw->data; + adev->psp.toc_bin_size = le32_to_cpu(sos_hdr_v1_1->toc_size_bytes); + adev->psp.toc_start_addr = (uint8_t *)adev->psp.sys_start_addr + + le32_to_cpu(sos_hdr_v1_1->toc_offset_bytes); + adev->psp.kdb_bin_size = le32_to_cpu(sos_hdr_v1_1->kdb_size_bytes); + adev->psp.kdb_start_addr = (uint8_t *)adev->psp.sys_start_addr + + le32_to_cpu(sos_hdr_v1_1->kdb_offset_bytes); + } + if (sos_hdr->header.header_version_minor == 2) { + sos_hdr_v1_2 = (const struct psp_firmware_header_v1_2 *)adev->psp.sos_fw->data; + adev->psp.kdb_bin_size = le32_to_cpu(sos_hdr_v1_2->kdb_size_bytes); + adev->psp.kdb_start_addr = (uint8_t *)adev->psp.sys_start_addr + + le32_to_cpu(sos_hdr_v1_2->kdb_offset_bytes); + } + break; + default: + dev_err(adev->dev, + "unsupported psp sos firmware\n"); + err = -EINVAL; + goto out; + } + + return 0; +out: + dev_err(adev->dev, + "failed to init sos firmware\n"); + release_firmware(adev->psp.sos_fw); + adev->psp.sos_fw = NULL; - return psp_compare_sram_data(&adev->psp, ucode, ucode_type); + return err; } static int psp_set_clockgating_state(void *handle, @@ -1957,16 +2252,6 @@ static void psp_sysfs_fini(struct amdgpu_device *adev) device_remove_file(adev->dev, &dev_attr_usbc_pd_fw); } -static const struct amdgpu_psp_funcs psp_funcs = { - .check_fw_loading_status = psp_check_fw_loading_status, -}; - -static void psp_set_funcs(struct amdgpu_device *adev) -{ - if (NULL == adev->firmware.funcs) - adev->firmware.funcs = &psp_funcs; -} - const struct amdgpu_ip_block_version psp_v3_1_ip_block = { .type = AMD_IP_BLOCK_TYPE_PSP, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h index 297435c0c7c1..2a56ad996d83 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h @@ -93,22 +93,8 @@ struct psp_funcs enum psp_ring_type ring_type); int (*ring_destroy)(struct psp_context *psp, enum psp_ring_type ring_type); - bool (*compare_sram_data)(struct psp_context *psp, - struct amdgpu_firmware_info *ucode, - enum AMDGPU_UCODE_ID ucode_type); bool (*smu_reload_quirk)(struct psp_context *psp); int (*mode1_reset)(struct psp_context *psp); - int (*xgmi_get_node_id)(struct psp_context *psp, uint64_t *node_id); - int (*xgmi_get_hive_id)(struct psp_context *psp, uint64_t *hive_id); - int (*xgmi_get_topology_info)(struct psp_context *psp, int number_devices, - struct psp_xgmi_topology_info *topology); - int (*xgmi_set_topology_info)(struct psp_context *psp, int number_devices, - struct psp_xgmi_topology_info *topology); - bool (*support_vmr_ring)(struct psp_context *psp); - int (*ras_trigger_error)(struct psp_context *psp, - struct ta_ras_trigger_error_input *info); - int (*ras_cure_posion)(struct psp_context *psp, uint64_t *mode_ptr); - int (*rlc_autoload_start)(struct psp_context *psp); int (*mem_training_init)(struct psp_context *psp); void (*mem_training_fini)(struct psp_context *psp); int (*mem_training)(struct psp_context *psp, uint32_t ops); @@ -161,6 +147,7 @@ struct psp_hdcp_context { struct amdgpu_bo *hdcp_shared_bo; uint64_t hdcp_shared_mc_addr; void *hdcp_shared_buf; + struct mutex mutex; }; struct psp_dtm_context { @@ -169,6 +156,7 @@ struct psp_dtm_context { struct amdgpu_bo *dtm_shared_bo; uint64_t dtm_shared_mc_addr; void *dtm_shared_buf; + struct mutex mutex; }; #define MEM_TRAIN_SYSTEM_SIGNATURE 0x54534942 @@ -306,8 +294,6 @@ struct amdgpu_psp_funcs { #define psp_ring_create(psp, type) (psp)->funcs->ring_create((psp), (type)) #define psp_ring_stop(psp, type) (psp)->funcs->ring_stop((psp), (type)) #define psp_ring_destroy(psp, type) ((psp)->funcs->ring_destroy((psp), (type))) -#define psp_compare_sram_data(psp, ucode, type) \ - (psp)->funcs->compare_sram_data((psp), (ucode), (type)) #define psp_init_microcode(psp) \ ((psp)->funcs->init_microcode ? (psp)->funcs->init_microcode((psp)) : 0) #define psp_bootloader_load_kdb(psp) \ @@ -318,22 +304,8 @@ struct amdgpu_psp_funcs { ((psp)->funcs->bootloader_load_sos ? (psp)->funcs->bootloader_load_sos((psp)) : 0) #define psp_smu_reload_quirk(psp) \ ((psp)->funcs->smu_reload_quirk ? (psp)->funcs->smu_reload_quirk((psp)) : false) -#define psp_support_vmr_ring(psp) \ - ((psp)->funcs->support_vmr_ring ? (psp)->funcs->support_vmr_ring((psp)) : false) #define psp_mode1_reset(psp) \ ((psp)->funcs->mode1_reset ? (psp)->funcs->mode1_reset((psp)) : false) -#define psp_xgmi_get_node_id(psp, node_id) \ - ((psp)->funcs->xgmi_get_node_id ? (psp)->funcs->xgmi_get_node_id((psp), (node_id)) : -EINVAL) -#define psp_xgmi_get_hive_id(psp, hive_id) \ - ((psp)->funcs->xgmi_get_hive_id ? (psp)->funcs->xgmi_get_hive_id((psp), (hive_id)) : -EINVAL) -#define psp_xgmi_get_topology_info(psp, num_device, topology) \ - ((psp)->funcs->xgmi_get_topology_info ? \ - (psp)->funcs->xgmi_get_topology_info((psp), (num_device), (topology)) : -EINVAL) -#define psp_xgmi_set_topology_info(psp, num_device, topology) \ - ((psp)->funcs->xgmi_set_topology_info ? \ - (psp)->funcs->xgmi_set_topology_info((psp), (num_device), (topology)) : -EINVAL) -#define psp_rlc_autoload(psp) \ - ((psp)->funcs->rlc_autoload_start ? (psp)->funcs->rlc_autoload_start((psp)) : 0) #define psp_mem_training_init(psp) \ ((psp)->funcs->mem_training_init ? (psp)->funcs->mem_training_init((psp)) : 0) #define psp_mem_training_fini(psp) \ @@ -341,15 +313,6 @@ struct amdgpu_psp_funcs { #define psp_mem_training(psp, ops) \ ((psp)->funcs->mem_training ? (psp)->funcs->mem_training((psp), (ops)) : 0) -#define amdgpu_psp_check_fw_loading_status(adev, i) (adev)->firmware.funcs->check_fw_loading_status((adev), (i)) - -#define psp_ras_trigger_error(psp, info) \ - ((psp)->funcs->ras_trigger_error ? \ - (psp)->funcs->ras_trigger_error((psp), (info)) : -EINVAL) -#define psp_ras_cure_posion(psp, addr) \ - ((psp)->funcs->ras_cure_posion ? \ - (psp)->funcs->ras_cure_posion(psp, (addr)) : -EINVAL) - #define psp_ring_get_wptr(psp) (psp)->funcs->ring_get_wptr((psp)) #define psp_ring_set_wptr(psp, value) (psp)->funcs->ring_set_wptr((psp), (value)) @@ -377,10 +340,21 @@ int psp_update_vcn_sram(struct amdgpu_device *adev, int inst_idx, int psp_xgmi_initialize(struct psp_context *psp); int psp_xgmi_terminate(struct psp_context *psp); int psp_xgmi_invoke(struct psp_context *psp, uint32_t ta_cmd_id); +int psp_xgmi_get_hive_id(struct psp_context *psp, uint64_t *hive_id); +int psp_xgmi_get_node_id(struct psp_context *psp, uint64_t *node_id); +int psp_xgmi_get_topology_info(struct psp_context *psp, + int number_devices, + struct psp_xgmi_topology_info *topology); +int psp_xgmi_set_topology_info(struct psp_context *psp, + int number_devices, + struct psp_xgmi_topology_info *topology); int psp_ras_invoke(struct psp_context *psp, uint32_t ta_cmd_id); int psp_ras_enable_features(struct psp_context *psp, union ta_ras_cmd_input *info, bool enable); +int psp_ras_trigger_error(struct psp_context *psp, + struct ta_ras_trigger_error_input *info); + int psp_hdcp_invoke(struct psp_context *psp, uint32_t ta_cmd_id); int psp_dtm_invoke(struct psp_context *psp, uint32_t ta_cmd_id); @@ -393,4 +367,8 @@ int psp_ring_cmd_submit(struct psp_context *psp, uint64_t cmd_buf_mc_addr, uint64_t fence_mc_addr, int index); +int psp_init_asd_microcode(struct psp_context *psp, + const char *chip_name); +int psp_init_sos_microcode(struct psp_context *psp, + const char *chip_name); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index ab379b44679c..50fe08bf2f72 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -80,6 +80,20 @@ atomic_t amdgpu_ras_in_intr = ATOMIC_INIT(0); static bool amdgpu_ras_check_bad_page(struct amdgpu_device *adev, uint64_t addr); +void amdgpu_ras_set_error_query_ready(struct amdgpu_device *adev, bool ready) +{ + if (adev && amdgpu_ras_get_context(adev)) + amdgpu_ras_get_context(adev)->error_query_ready = ready; +} + +bool amdgpu_ras_get_error_query_ready(struct amdgpu_device *adev) +{ + if (adev && amdgpu_ras_get_context(adev)) + return amdgpu_ras_get_context(adev)->error_query_ready; + + return false; +} + static ssize_t amdgpu_ras_debugfs_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { @@ -281,8 +295,9 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, const char __user * struct ras_debug_if data; int ret = 0; - if (amdgpu_ras_intr_triggered()) { - DRM_WARN("RAS WARN: error injection currently inaccessible\n"); + if (!amdgpu_ras_get_error_query_ready(adev)) { + dev_warn(adev->dev, "RAS WARN: error injection " + "currently inaccessible\n"); return size; } @@ -310,7 +325,8 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, const char __user * /* umc ce/ue error injection for a bad page is not allowed */ if ((data.head.block == AMDGPU_RAS_BLOCK__UMC) && amdgpu_ras_check_bad_page(adev, data.inject.address)) { - DRM_WARN("RAS WARN: 0x%llx has been marked as bad before error injection!\n", + dev_warn(adev->dev, "RAS WARN: 0x%llx has been marked " + "as bad before error injection!\n", data.inject.address); break; } @@ -399,7 +415,7 @@ static ssize_t amdgpu_ras_sysfs_read(struct device *dev, .head = obj->head, }; - if (amdgpu_ras_intr_triggered()) + if (!amdgpu_ras_get_error_query_ready(obj->adev)) return snprintf(buf, PAGE_SIZE, "Query currently inaccessible\n"); @@ -486,6 +502,29 @@ struct ras_manager *amdgpu_ras_find_obj(struct amdgpu_device *adev, } /* obj end */ +void amdgpu_ras_parse_status_code(struct amdgpu_device* adev, + const char* invoke_type, + const char* block_name, + enum ta_ras_status ret) +{ + switch (ret) { + case TA_RAS_STATUS__SUCCESS: + return; + case TA_RAS_STATUS__ERROR_RAS_NOT_AVAILABLE: + dev_warn(adev->dev, + "RAS WARN: %s %s currently unavailable\n", + invoke_type, + block_name); + break; + default: + dev_err(adev->dev, + "RAS ERROR: %s %s error failed ret 0x%X\n", + invoke_type, + block_name, + ret); + } +} + /* feature ctl begin */ static int amdgpu_ras_is_feature_allowed(struct amdgpu_device *adev, struct ras_common_if *head) @@ -549,19 +588,23 @@ int amdgpu_ras_feature_enable(struct amdgpu_device *adev, struct ras_common_if *head, bool enable) { struct amdgpu_ras *con = amdgpu_ras_get_context(adev); - union ta_ras_cmd_input info; + union ta_ras_cmd_input *info; int ret; if (!con) return -EINVAL; + info = kzalloc(sizeof(union ta_ras_cmd_input), GFP_KERNEL); + if (!info) + return -ENOMEM; + if (!enable) { - info.disable_features = (struct ta_ras_disable_features_input) { + info->disable_features = (struct ta_ras_disable_features_input) { .block_id = amdgpu_ras_block_to_ta(head->block), .error_type = amdgpu_ras_error_to_ta(head->type), }; } else { - info.enable_features = (struct ta_ras_enable_features_input) { + info->enable_features = (struct ta_ras_enable_features_input) { .block_id = amdgpu_ras_block_to_ta(head->block), .error_type = amdgpu_ras_error_to_ta(head->type), }; @@ -570,26 +613,33 @@ int amdgpu_ras_feature_enable(struct amdgpu_device *adev, /* Do not enable if it is not allowed. */ WARN_ON(enable && !amdgpu_ras_is_feature_allowed(adev, head)); /* Are we alerady in that state we are going to set? */ - if (!(!!enable ^ !!amdgpu_ras_is_feature_enabled(adev, head))) - return 0; + if (!(!!enable ^ !!amdgpu_ras_is_feature_enabled(adev, head))) { + ret = 0; + goto out; + } if (!amdgpu_ras_intr_triggered()) { - ret = psp_ras_enable_features(&adev->psp, &info, enable); + ret = psp_ras_enable_features(&adev->psp, info, enable); if (ret) { - DRM_ERROR("RAS ERROR: %s %s feature failed ret %d\n", - enable ? "enable":"disable", - ras_block_str(head->block), - ret); + amdgpu_ras_parse_status_code(adev, + enable ? "enable":"disable", + ras_block_str(head->block), + (enum ta_ras_status)ret); if (ret == TA_RAS_STATUS__RESET_NEEDED) - return -EAGAIN; - return -EINVAL; + ret = -EAGAIN; + else + ret = -EINVAL; + + goto out; } } /* setup the obj */ __amdgpu_ras_feature_enable(adev, head, enable); - - return 0; + ret = 0; +out: + kfree(info); + return ret; } /* Only used in device probe stage and called only once. */ @@ -618,7 +668,8 @@ int amdgpu_ras_feature_enable_on_boot(struct amdgpu_device *adev, if (ret == -EINVAL) { ret = __amdgpu_ras_feature_enable(adev, head, 1); if (!ret) - DRM_INFO("RAS INFO: %s setup object\n", + dev_info(adev->dev, + "RAS INFO: %s setup object\n", ras_block_str(head->block)); } } else { @@ -744,17 +795,48 @@ int amdgpu_ras_error_query(struct amdgpu_device *adev, info->ce_count = obj->err_data.ce_count; if (err_data.ce_count) { - dev_info(adev->dev, "%ld correctable errors detected in %s block\n", - obj->err_data.ce_count, ras_block_str(info->head.block)); + dev_info(adev->dev, "%ld correctable hardware errors " + "detected in %s block, no user " + "action is needed.\n", + obj->err_data.ce_count, + ras_block_str(info->head.block)); } if (err_data.ue_count) { - dev_info(adev->dev, "%ld uncorrectable errors detected in %s block\n", - obj->err_data.ue_count, ras_block_str(info->head.block)); + dev_info(adev->dev, "%ld uncorrectable hardware errors " + "detected in %s block\n", + obj->err_data.ue_count, + ras_block_str(info->head.block)); } return 0; } +/* Trigger XGMI/WAFL error */ +int amdgpu_ras_error_inject_xgmi(struct amdgpu_device *adev, + struct ta_ras_trigger_error_input *block_info) +{ + int ret; + + if (amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_DISALLOW)) + dev_warn(adev->dev, "Failed to disallow df cstate"); + + if (amdgpu_dpm_allow_xgmi_power_down(adev, false)) + dev_warn(adev->dev, "Failed to disallow XGMI power down"); + + ret = psp_ras_trigger_error(&adev->psp, block_info); + + if (amdgpu_ras_intr_triggered()) + return ret; + + if (amdgpu_dpm_allow_xgmi_power_down(adev, true)) + dev_warn(adev->dev, "Failed to allow XGMI power down"); + + if (amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_DISALLOW)) + dev_warn(adev->dev, "Failed to allow df cstate"); + + return ret; +} + /* wrapper of psp_ras_trigger_error */ int amdgpu_ras_error_inject(struct amdgpu_device *adev, struct ras_inject_if *info) @@ -788,20 +870,22 @@ int amdgpu_ras_error_inject(struct amdgpu_device *adev, break; case AMDGPU_RAS_BLOCK__UMC: case AMDGPU_RAS_BLOCK__MMHUB: - case AMDGPU_RAS_BLOCK__XGMI_WAFL: case AMDGPU_RAS_BLOCK__PCIE_BIF: ret = psp_ras_trigger_error(&adev->psp, &block_info); break; + case AMDGPU_RAS_BLOCK__XGMI_WAFL: + ret = amdgpu_ras_error_inject_xgmi(adev, &block_info); + break; default: - DRM_INFO("%s error injection is not supported yet\n", + dev_info(adev->dev, "%s error injection is not supported yet\n", ras_block_str(info->head.block)); ret = -EINVAL; } - if (ret) - DRM_ERROR("RAS ERROR: inject %s error failed ret %d\n", - ras_block_str(info->head.block), - ret); + amdgpu_ras_parse_status_code(adev, + "inject", + ras_block_str(info->head.block), + (enum ta_ras_status)ret); return ret; } @@ -1430,9 +1514,10 @@ static void amdgpu_ras_do_recovery(struct work_struct *work) struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev, false); /* Build list of devices to query RAS related errors */ - if (hive && adev->gmc.xgmi.num_physical_nodes > 1) { + if (hive && adev->gmc.xgmi.num_physical_nodes > 1) device_list_handle = &hive->device_list; - } else { + else { + INIT_LIST_HEAD(&device_list); list_add_tail(&adev->gmc.xgmi.head, &device_list); device_list_handle = &device_list; } @@ -1535,7 +1620,7 @@ static int amdgpu_ras_save_bad_pages(struct amdgpu_device *adev) &data->bps[control->num_recs], true, save_count)) { - DRM_ERROR("Failed to save EEPROM table data!"); + dev_err(adev->dev, "Failed to save EEPROM table data!"); return -EIO; } @@ -1563,7 +1648,7 @@ static int amdgpu_ras_load_bad_pages(struct amdgpu_device *adev) if (amdgpu_ras_eeprom_process_recods(control, bps, false, control->num_recs)) { - DRM_ERROR("Failed to load EEPROM table records!"); + dev_err(adev->dev, "Failed to load EEPROM table records!"); ret = -EIO; goto out; } @@ -1637,7 +1722,8 @@ int amdgpu_ras_reserve_bad_pages(struct amdgpu_device *adev) AMDGPU_GPU_PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM, &bo, NULL)) - DRM_WARN("RAS WARN: reserve vram for retired page %llx fail\n", bp); + dev_warn(adev->dev, "RAS WARN: reserve vram for " + "retired page %llx fail\n", bp); data->bps_bo[i] = bo; data->last_reserved = i + 1; @@ -1725,7 +1811,7 @@ free: kfree(*data); con->eh_data = NULL; out: - DRM_WARN("Failed to initialize ras recovery!\n"); + dev_warn(adev->dev, "Failed to initialize ras recovery!\n"); return ret; } @@ -1787,18 +1873,18 @@ static void amdgpu_ras_check_supported(struct amdgpu_device *adev, return; if (amdgpu_atomfirmware_mem_ecc_supported(adev)) { - DRM_INFO("HBM ECC is active.\n"); + dev_info(adev->dev, "HBM ECC is active.\n"); *hw_supported |= (1 << AMDGPU_RAS_BLOCK__UMC | 1 << AMDGPU_RAS_BLOCK__DF); } else - DRM_INFO("HBM ECC is not presented.\n"); + dev_info(adev->dev, "HBM ECC is not presented.\n"); if (amdgpu_atomfirmware_sram_ecc_supported(adev)) { - DRM_INFO("SRAM ECC is active.\n"); + dev_info(adev->dev, "SRAM ECC is active.\n"); *hw_supported |= ~(1 << AMDGPU_RAS_BLOCK__UMC | 1 << AMDGPU_RAS_BLOCK__DF); } else - DRM_INFO("SRAM ECC is not presented.\n"); + dev_info(adev->dev, "SRAM ECC is not presented.\n"); /* hw_supported needs to be aligned with RAS block mask. */ *hw_supported &= AMDGPU_RAS_BLOCK_MASK; @@ -1855,7 +1941,7 @@ int amdgpu_ras_init(struct amdgpu_device *adev) if (amdgpu_ras_fs_init(adev)) goto fs_out; - DRM_INFO("RAS INFO: ras initialized successfully, " + dev_info(adev->dev, "RAS INFO: ras initialized successfully, " "hardware ability[%x] ras_mask[%x]\n", con->hw_supported, con->supported); return 0; @@ -2037,7 +2123,8 @@ void amdgpu_ras_global_ras_isr(struct amdgpu_device *adev) return; if (atomic_cmpxchg(&amdgpu_ras_in_intr, 0, 1) == 0) { - DRM_WARN("RAS event of type ERREVENT_ATHUB_INTERRUPT detected!\n"); + dev_info(adev->dev, "uncorrectable hardware error" + "(ERREVENT_ATHUB_INTERRUPT) detected!\n"); amdgpu_ras_reset_gpu(adev); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h index 55c3eceb390d..e7df5d8429f8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h @@ -334,6 +334,8 @@ struct amdgpu_ras { uint32_t flags; bool reboot; struct amdgpu_ras_eeprom_control eeprom_control; + + bool error_query_ready; }; struct ras_fs_data { @@ -629,4 +631,6 @@ static inline void amdgpu_ras_intr_cleared(void) void amdgpu_ras_global_ras_isr(struct amdgpu_device *adev); +void amdgpu_ras_set_error_query_ready(struct amdgpu_device *adev, bool ready); + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c index a7e1d0425ed0..13ea8ebc421c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c @@ -162,11 +162,13 @@ void amdgpu_ring_undo(struct amdgpu_ring *ring) * Returns 0 on success, error on failure. */ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring, - unsigned max_dw, struct amdgpu_irq_src *irq_src, - unsigned irq_type) + unsigned int max_dw, struct amdgpu_irq_src *irq_src, + unsigned int irq_type, unsigned int hw_prio) { int r, i; int sched_hw_submission = amdgpu_sched_hw_submission; + u32 *num_sched; + u32 hw_ip; /* Set the hw submission limit higher for KIQ because * it's used for a number of gfx/compute tasks by both @@ -258,6 +260,13 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring, ring->priority = DRM_SCHED_PRIORITY_NORMAL; mutex_init(&ring->priority_mutex); + if (!ring->no_scheduler) { + hw_ip = ring->funcs->type; + num_sched = &adev->gpu_sched[hw_ip][hw_prio].num_scheds; + adev->gpu_sched[hw_ip][hw_prio].sched[(*num_sched)++] = + &ring->sched; + } + for (i = 0; i < DRM_SCHED_PRIORITY_MAX; ++i) atomic_set(&ring->num_jobs[i], 0); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h index 9a443013d70d..be218754629a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h @@ -30,11 +30,15 @@ /* max number of rings */ #define AMDGPU_MAX_RINGS 28 +#define AMDGPU_MAX_HWIP_RINGS 8 #define AMDGPU_MAX_GFX_RINGS 2 #define AMDGPU_MAX_COMPUTE_RINGS 8 #define AMDGPU_MAX_VCE_RINGS 3 #define AMDGPU_MAX_UVD_ENC_RINGS 2 +#define AMDGPU_RING_PRIO_DEFAULT 1 +#define AMDGPU_RING_PRIO_MAX AMDGPU_GFX_PIPE_PRIO_MAX + /* some special values for the owner field */ #define AMDGPU_FENCE_OWNER_UNDEFINED ((void *)0ul) #define AMDGPU_FENCE_OWNER_VM ((void *)1ul) @@ -46,17 +50,30 @@ #define to_amdgpu_ring(s) container_of((s), struct amdgpu_ring, sched) +#define AMDGPU_IB_POOL_SIZE (1024 * 1024) + enum amdgpu_ring_type { - AMDGPU_RING_TYPE_GFX, - AMDGPU_RING_TYPE_COMPUTE, - AMDGPU_RING_TYPE_SDMA, - AMDGPU_RING_TYPE_UVD, - AMDGPU_RING_TYPE_VCE, - AMDGPU_RING_TYPE_KIQ, - AMDGPU_RING_TYPE_UVD_ENC, - AMDGPU_RING_TYPE_VCN_DEC, - AMDGPU_RING_TYPE_VCN_ENC, - AMDGPU_RING_TYPE_VCN_JPEG + AMDGPU_RING_TYPE_GFX = AMDGPU_HW_IP_GFX, + AMDGPU_RING_TYPE_COMPUTE = AMDGPU_HW_IP_COMPUTE, + AMDGPU_RING_TYPE_SDMA = AMDGPU_HW_IP_DMA, + AMDGPU_RING_TYPE_UVD = AMDGPU_HW_IP_UVD, + AMDGPU_RING_TYPE_VCE = AMDGPU_HW_IP_VCE, + AMDGPU_RING_TYPE_UVD_ENC = AMDGPU_HW_IP_UVD_ENC, + AMDGPU_RING_TYPE_VCN_DEC = AMDGPU_HW_IP_VCN_DEC, + AMDGPU_RING_TYPE_VCN_ENC = AMDGPU_HW_IP_VCN_ENC, + AMDGPU_RING_TYPE_VCN_JPEG = AMDGPU_HW_IP_VCN_JPEG, + AMDGPU_RING_TYPE_KIQ +}; + +enum amdgpu_ib_pool_type { + /* Normal submissions to the top of the pipeline. */ + AMDGPU_IB_POOL_DELAYED, + /* Immediate submissions to the bottom of the pipeline. */ + AMDGPU_IB_POOL_IMMEDIATE, + /* Direct submission to the ring buffer during init and reset. */ + AMDGPU_IB_POOL_DIRECT, + + AMDGPU_IB_POOL_MAX }; struct amdgpu_device; @@ -65,6 +82,11 @@ struct amdgpu_ib; struct amdgpu_cs_parser; struct amdgpu_job; +struct amdgpu_sched { + u32 num_scheds; + struct drm_gpu_scheduler *sched[AMDGPU_MAX_HWIP_RINGS]; +}; + /* * Fences. */ @@ -96,7 +118,8 @@ void amdgpu_fence_driver_suspend(struct amdgpu_device *adev); void amdgpu_fence_driver_resume(struct amdgpu_device *adev); int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **fence, unsigned flags); -int amdgpu_fence_emit_polling(struct amdgpu_ring *ring, uint32_t *s); +int amdgpu_fence_emit_polling(struct amdgpu_ring *ring, uint32_t *s, + uint32_t timeout); bool amdgpu_fence_process(struct amdgpu_ring *ring); int amdgpu_fence_wait_empty(struct amdgpu_ring *ring); signed long amdgpu_fence_wait_polling(struct amdgpu_ring *ring, @@ -159,17 +182,20 @@ struct amdgpu_ring_funcs { void (*end_use)(struct amdgpu_ring *ring); void (*emit_switch_buffer) (struct amdgpu_ring *ring); void (*emit_cntxcntl) (struct amdgpu_ring *ring, uint32_t flags); - void (*emit_rreg)(struct amdgpu_ring *ring, uint32_t reg); + void (*emit_rreg)(struct amdgpu_ring *ring, uint32_t reg, + uint32_t reg_val_offs); void (*emit_wreg)(struct amdgpu_ring *ring, uint32_t reg, uint32_t val); void (*emit_reg_wait)(struct amdgpu_ring *ring, uint32_t reg, uint32_t val, uint32_t mask); void (*emit_reg_write_reg_wait)(struct amdgpu_ring *ring, uint32_t reg0, uint32_t reg1, uint32_t ref, uint32_t mask); - void (*emit_tmz)(struct amdgpu_ring *ring, bool start); + void (*emit_frame_cntl)(struct amdgpu_ring *ring, bool start, + bool secure); /* Try to soft recover the ring to make the fence signal */ void (*soft_recovery)(struct amdgpu_ring *ring, unsigned vmid); int (*preempt_ib)(struct amdgpu_ring *ring); + void (*emit_mem_sync)(struct amdgpu_ring *ring); }; struct amdgpu_ring { @@ -214,12 +240,12 @@ struct amdgpu_ring { unsigned vm_inv_eng; struct dma_fence *vmid_wait; bool has_compute_vm_bug; + bool no_scheduler; atomic_t num_jobs[DRM_SCHED_PRIORITY_MAX]; struct mutex priority_mutex; /* protected by priority_mutex */ int priority; - bool has_high_prio; #if defined(CONFIG_DEBUG_FS) struct dentry *ent; @@ -241,11 +267,11 @@ struct amdgpu_ring { #define amdgpu_ring_emit_hdp_flush(r) (r)->funcs->emit_hdp_flush((r)) #define amdgpu_ring_emit_switch_buffer(r) (r)->funcs->emit_switch_buffer((r)) #define amdgpu_ring_emit_cntxcntl(r, d) (r)->funcs->emit_cntxcntl((r), (d)) -#define amdgpu_ring_emit_rreg(r, d) (r)->funcs->emit_rreg((r), (d)) +#define amdgpu_ring_emit_rreg(r, d, o) (r)->funcs->emit_rreg((r), (d), (o)) #define amdgpu_ring_emit_wreg(r, d, v) (r)->funcs->emit_wreg((r), (d), (v)) #define amdgpu_ring_emit_reg_wait(r, d, v, m) (r)->funcs->emit_reg_wait((r), (d), (v), (m)) #define amdgpu_ring_emit_reg_write_reg_wait(r, d0, d1, v, m) (r)->funcs->emit_reg_write_reg_wait((r), (d0), (d1), (v), (m)) -#define amdgpu_ring_emit_tmz(r, b) (r)->funcs->emit_tmz((r), (b)) +#define amdgpu_ring_emit_frame_cntl(r, b, s) (r)->funcs->emit_frame_cntl((r), (b), (s)) #define amdgpu_ring_pad_ib(r, ib) ((r)->funcs->pad_ib((r), (ib))) #define amdgpu_ring_init_cond_exec(r) (r)->funcs->init_cond_exec((r)) #define amdgpu_ring_patch_cond_exec(r,o) (r)->funcs->patch_cond_exec((r),(o)) @@ -257,8 +283,8 @@ void amdgpu_ring_generic_pad_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib); void amdgpu_ring_commit(struct amdgpu_ring *ring); void amdgpu_ring_undo(struct amdgpu_ring *ring); int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring, - unsigned ring_size, struct amdgpu_irq_src *irq_src, - unsigned irq_type); + unsigned int ring_size, struct amdgpu_irq_src *irq_src, + unsigned int irq_type, unsigned int prio); void amdgpu_ring_fini(struct amdgpu_ring *ring); void amdgpu_ring_emit_reg_write_reg_wait_helper(struct amdgpu_ring *ring, uint32_t reg0, uint32_t val0, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h index 4b352206354b..e5b8fb8e75c5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h @@ -61,8 +61,6 @@ struct amdgpu_sdma_ras_funcs { struct amdgpu_sdma { struct amdgpu_sdma_instance instance[AMDGPU_MAX_SDMA_INSTANCES]; - struct drm_gpu_scheduler *sdma_sched[AMDGPU_MAX_SDMA_INSTANCES]; - uint32_t num_sdma_sched; struct amdgpu_irq_src trap_irq; struct amdgpu_irq_src illegal_inst_irq; struct amdgpu_irq_src ecc_irq; @@ -91,7 +89,8 @@ struct amdgpu_buffer_funcs { /* dst addr in bytes */ uint64_t dst_offset, /* number of byte to transfer */ - uint32_t byte_count); + uint32_t byte_count, + bool tmz); /* maximum bytes in a single operation */ uint32_t fill_max_bytes; @@ -109,7 +108,7 @@ struct amdgpu_buffer_funcs { uint32_t byte_count); }; -#define amdgpu_emit_copy_buffer(adev, ib, s, d, b) (adev)->mman.buffer_funcs->emit_copy_buffer((ib), (s), (d), (b)) +#define amdgpu_emit_copy_buffer(adev, ib, s, d, b, t) (adev)->mman.buffer_funcs->emit_copy_buffer((ib), (s), (d), (b), (t)) #define amdgpu_emit_fill_buffer(adev, ib, s, d, b) (adev)->mman.buffer_funcs->emit_fill_buffer((ib), (s), (d), (b)) struct amdgpu_sdma_instance * diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c index b86392253696..b87ca171986a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c @@ -249,6 +249,11 @@ int amdgpu_sync_resv(struct amdgpu_device *adev, struct amdgpu_sync *sync, owner != AMDGPU_FENCE_OWNER_UNDEFINED) continue; + /* Never sync to VM updates either. */ + if (fence_owner == AMDGPU_FENCE_OWNER_VM && + owner != AMDGPU_FENCE_OWNER_UNDEFINED) + continue; + /* Ignore fences depending on the sync mode */ switch (mode) { case AMDGPU_SYNC_ALWAYS: diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c index b158230af8db..2f4d5ca9894f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c @@ -44,7 +44,7 @@ static void amdgpu_do_test_moves(struct amdgpu_device *adev) /* Number of tests = * (Total GTT - IB pool - writeback page - ring buffers) / test size */ - n = adev->gmc.gart_size - AMDGPU_IB_POOL_SIZE*64*1024; + n = adev->gmc.gart_size - AMDGPU_IB_POOL_SIZE; for (i = 0; i < AMDGPU_MAX_RINGS; ++i) if (adev->rings[i]) n -= adev->rings[i]->ring_size; @@ -124,7 +124,7 @@ static void amdgpu_do_test_moves(struct amdgpu_device *adev) amdgpu_bo_kunmap(gtt_obj[i]); r = amdgpu_copy_buffer(ring, gart_addr, vram_addr, - size, NULL, &fence, false, false); + size, NULL, &fence, false, false, false); if (r) { DRM_ERROR("Failed GTT->VRAM copy %d\n", i); @@ -170,7 +170,7 @@ static void amdgpu_do_test_moves(struct amdgpu_device *adev) amdgpu_bo_kunmap(vram_obj); r = amdgpu_copy_buffer(ring, vram_addr, gart_addr, - size, NULL, &fence, false, false); + size, NULL, &fence, false, false, false); if (r) { DRM_ERROR("Failed VRAM->GTT copy %d\n", i); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h index 63e734a125fb..5da20fc166d9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h @@ -35,7 +35,7 @@ #define AMDGPU_JOB_GET_TIMELINE_NAME(job) \ job->base.s_fence->finished.ops->get_timeline_name(&job->base.s_fence->finished) -TRACE_EVENT(amdgpu_mm_rreg, +TRACE_EVENT(amdgpu_device_rreg, TP_PROTO(unsigned did, uint32_t reg, uint32_t value), TP_ARGS(did, reg, value), TP_STRUCT__entry( @@ -54,7 +54,7 @@ TRACE_EVENT(amdgpu_mm_rreg, (unsigned long)__entry->value) ); -TRACE_EVENT(amdgpu_mm_wreg, +TRACE_EVENT(amdgpu_device_wreg, TP_PROTO(unsigned did, uint32_t reg, uint32_t value), TP_ARGS(did, reg, value), TP_STRUCT__entry( diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index c2e13e1ec53e..38d2a7fb5698 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -62,11 +62,6 @@ #define AMDGPU_TTM_VRAM_MAX_DW_READ (size_t)128 -static int amdgpu_map_buffer(struct ttm_buffer_object *bo, - struct ttm_mem_reg *mem, unsigned num_pages, - uint64_t offset, unsigned window, - struct amdgpu_ring *ring, - uint64_t *addr); /** * amdgpu_init_mem_type - Initialize a memory manager for a specific type of @@ -275,7 +270,7 @@ static uint64_t amdgpu_mm_node_addr(struct ttm_buffer_object *bo, * */ static struct drm_mm_node *amdgpu_find_mm_node(struct ttm_mem_reg *mem, - unsigned long *offset) + uint64_t *offset) { struct drm_mm_node *mm_node = mem->mm_node; @@ -287,91 +282,191 @@ static struct drm_mm_node *amdgpu_find_mm_node(struct ttm_mem_reg *mem, } /** + * amdgpu_ttm_map_buffer - Map memory into the GART windows + * @bo: buffer object to map + * @mem: memory object to map + * @mm_node: drm_mm node object to map + * @num_pages: number of pages to map + * @offset: offset into @mm_node where to start + * @window: which GART window to use + * @ring: DMA ring to use for the copy + * @tmz: if we should setup a TMZ enabled mapping + * @addr: resulting address inside the MC address space + * + * Setup one of the GART windows to access a specific piece of memory or return + * the physical address for local memory. + */ +static int amdgpu_ttm_map_buffer(struct ttm_buffer_object *bo, + struct ttm_mem_reg *mem, + struct drm_mm_node *mm_node, + unsigned num_pages, uint64_t offset, + unsigned window, struct amdgpu_ring *ring, + bool tmz, uint64_t *addr) +{ + struct amdgpu_device *adev = ring->adev; + struct amdgpu_job *job; + unsigned num_dw, num_bytes; + struct dma_fence *fence; + uint64_t src_addr, dst_addr; + void *cpu_addr; + uint64_t flags; + unsigned int i; + int r; + + BUG_ON(adev->mman.buffer_funcs->copy_max_bytes < + AMDGPU_GTT_MAX_TRANSFER_SIZE * 8); + + /* Map only what can't be accessed directly */ + if (!tmz && mem->start != AMDGPU_BO_INVALID_OFFSET) { + *addr = amdgpu_mm_node_addr(bo, mm_node, mem) + offset; + return 0; + } + + *addr = adev->gmc.gart_start; + *addr += (u64)window * AMDGPU_GTT_MAX_TRANSFER_SIZE * + AMDGPU_GPU_PAGE_SIZE; + *addr += offset & ~PAGE_MASK; + + num_dw = ALIGN(adev->mman.buffer_funcs->copy_num_dw, 8); + num_bytes = num_pages * 8; + + r = amdgpu_job_alloc_with_ib(adev, num_dw * 4 + num_bytes, + AMDGPU_IB_POOL_DELAYED, &job); + if (r) + return r; + + src_addr = num_dw * 4; + src_addr += job->ibs[0].gpu_addr; + + dst_addr = amdgpu_bo_gpu_offset(adev->gart.bo); + dst_addr += window * AMDGPU_GTT_MAX_TRANSFER_SIZE * 8; + amdgpu_emit_copy_buffer(adev, &job->ibs[0], src_addr, + dst_addr, num_bytes, false); + + amdgpu_ring_pad_ib(ring, &job->ibs[0]); + WARN_ON(job->ibs[0].length_dw > num_dw); + + flags = amdgpu_ttm_tt_pte_flags(adev, bo->ttm, mem); + if (tmz) + flags |= AMDGPU_PTE_TMZ; + + cpu_addr = &job->ibs[0].ptr[num_dw]; + + if (mem->mem_type == TTM_PL_TT) { + struct ttm_dma_tt *dma; + dma_addr_t *dma_address; + + dma = container_of(bo->ttm, struct ttm_dma_tt, ttm); + dma_address = &dma->dma_address[offset >> PAGE_SHIFT]; + r = amdgpu_gart_map(adev, 0, num_pages, dma_address, flags, + cpu_addr); + if (r) + goto error_free; + } else { + dma_addr_t dma_address; + + dma_address = (mm_node->start << PAGE_SHIFT) + offset; + dma_address += adev->vm_manager.vram_base_offset; + + for (i = 0; i < num_pages; ++i) { + r = amdgpu_gart_map(adev, i << PAGE_SHIFT, 1, + &dma_address, flags, cpu_addr); + if (r) + goto error_free; + + dma_address += PAGE_SIZE; + } + } + + r = amdgpu_job_submit(job, &adev->mman.entity, + AMDGPU_FENCE_OWNER_UNDEFINED, &fence); + if (r) + goto error_free; + + dma_fence_put(fence); + + return r; + +error_free: + amdgpu_job_free(job); + return r; +} + +/** * amdgpu_copy_ttm_mem_to_mem - Helper function for copy + * @adev: amdgpu device + * @src: buffer/address where to read from + * @dst: buffer/address where to write to + * @size: number of bytes to copy + * @tmz: if a secure copy should be used + * @resv: resv object to sync to + * @f: Returns the last fence if multiple jobs are submitted. * * The function copies @size bytes from {src->mem + src->offset} to * {dst->mem + dst->offset}. src->bo and dst->bo could be same BO for a * move and different for a BO to BO copy. * - * @f: Returns the last fence if multiple jobs are submitted. */ int amdgpu_ttm_copy_mem_to_mem(struct amdgpu_device *adev, - struct amdgpu_copy_mem *src, - struct amdgpu_copy_mem *dst, - uint64_t size, + const struct amdgpu_copy_mem *src, + const struct amdgpu_copy_mem *dst, + uint64_t size, bool tmz, struct dma_resv *resv, struct dma_fence **f) { + const uint32_t GTT_MAX_BYTES = (AMDGPU_GTT_MAX_TRANSFER_SIZE * + AMDGPU_GPU_PAGE_SIZE); + + uint64_t src_node_size, dst_node_size, src_offset, dst_offset; struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring; struct drm_mm_node *src_mm, *dst_mm; - uint64_t src_node_start, dst_node_start, src_node_size, - dst_node_size, src_page_offset, dst_page_offset; struct dma_fence *fence = NULL; int r = 0; - const uint64_t GTT_MAX_BYTES = (AMDGPU_GTT_MAX_TRANSFER_SIZE * - AMDGPU_GPU_PAGE_SIZE); if (!adev->mman.buffer_funcs_enabled) { DRM_ERROR("Trying to move memory with ring turned off.\n"); return -EINVAL; } - src_mm = amdgpu_find_mm_node(src->mem, &src->offset); - src_node_start = amdgpu_mm_node_addr(src->bo, src_mm, src->mem) + - src->offset; - src_node_size = (src_mm->size << PAGE_SHIFT) - src->offset; - src_page_offset = src_node_start & (PAGE_SIZE - 1); + src_offset = src->offset; + src_mm = amdgpu_find_mm_node(src->mem, &src_offset); + src_node_size = (src_mm->size << PAGE_SHIFT) - src_offset; - dst_mm = amdgpu_find_mm_node(dst->mem, &dst->offset); - dst_node_start = amdgpu_mm_node_addr(dst->bo, dst_mm, dst->mem) + - dst->offset; - dst_node_size = (dst_mm->size << PAGE_SHIFT) - dst->offset; - dst_page_offset = dst_node_start & (PAGE_SIZE - 1); + dst_offset = dst->offset; + dst_mm = amdgpu_find_mm_node(dst->mem, &dst_offset); + dst_node_size = (dst_mm->size << PAGE_SHIFT) - dst_offset; mutex_lock(&adev->mman.gtt_window_lock); while (size) { - unsigned long cur_size; - uint64_t from = src_node_start, to = dst_node_start; + uint32_t src_page_offset = src_offset & ~PAGE_MASK; + uint32_t dst_page_offset = dst_offset & ~PAGE_MASK; struct dma_fence *next; + uint32_t cur_size; + uint64_t from, to; /* Copy size cannot exceed GTT_MAX_BYTES. So if src or dst * begins at an offset, then adjust the size accordingly */ - cur_size = min3(min(src_node_size, dst_node_size), size, - GTT_MAX_BYTES); - if (cur_size + src_page_offset > GTT_MAX_BYTES || - cur_size + dst_page_offset > GTT_MAX_BYTES) - cur_size -= max(src_page_offset, dst_page_offset); - - /* Map only what needs to be accessed. Map src to window 0 and - * dst to window 1 - */ - if (src->mem->start == AMDGPU_BO_INVALID_OFFSET) { - r = amdgpu_map_buffer(src->bo, src->mem, - PFN_UP(cur_size + src_page_offset), - src_node_start, 0, ring, - &from); - if (r) - goto error; - /* Adjust the offset because amdgpu_map_buffer returns - * start of mapped page - */ - from += src_page_offset; - } + cur_size = max(src_page_offset, dst_page_offset); + cur_size = min(min3(src_node_size, dst_node_size, size), + (uint64_t)(GTT_MAX_BYTES - cur_size)); + + /* Map src to window 0 and dst to window 1. */ + r = amdgpu_ttm_map_buffer(src->bo, src->mem, src_mm, + PFN_UP(cur_size + src_page_offset), + src_offset, 0, ring, tmz, &from); + if (r) + goto error; - if (dst->mem->start == AMDGPU_BO_INVALID_OFFSET) { - r = amdgpu_map_buffer(dst->bo, dst->mem, - PFN_UP(cur_size + dst_page_offset), - dst_node_start, 1, ring, - &to); - if (r) - goto error; - to += dst_page_offset; - } + r = amdgpu_ttm_map_buffer(dst->bo, dst->mem, dst_mm, + PFN_UP(cur_size + dst_page_offset), + dst_offset, 1, ring, tmz, &to); + if (r) + goto error; r = amdgpu_copy_buffer(ring, from, to, cur_size, - resv, &next, false, true); + resv, &next, false, true, tmz); if (r) goto error; @@ -384,23 +479,20 @@ int amdgpu_ttm_copy_mem_to_mem(struct amdgpu_device *adev, src_node_size -= cur_size; if (!src_node_size) { - src_node_start = amdgpu_mm_node_addr(src->bo, ++src_mm, - src->mem); - src_node_size = (src_mm->size << PAGE_SHIFT); - src_page_offset = 0; + ++src_mm; + src_node_size = src_mm->size << PAGE_SHIFT; + src_offset = 0; } else { - src_node_start += cur_size; - src_page_offset = src_node_start & (PAGE_SIZE - 1); + src_offset += cur_size; } + dst_node_size -= cur_size; if (!dst_node_size) { - dst_node_start = amdgpu_mm_node_addr(dst->bo, ++dst_mm, - dst->mem); - dst_node_size = (dst_mm->size << PAGE_SHIFT); - dst_page_offset = 0; + ++dst_mm; + dst_node_size = dst_mm->size << PAGE_SHIFT; + dst_offset = 0; } else { - dst_node_start += cur_size; - dst_page_offset = dst_node_start & (PAGE_SIZE - 1); + dst_offset += cur_size; } } error: @@ -423,6 +515,7 @@ static int amdgpu_move_blit(struct ttm_buffer_object *bo, struct ttm_mem_reg *old_mem) { struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev); + struct amdgpu_bo *abo = ttm_to_amdgpu_bo(bo); struct amdgpu_copy_mem src, dst; struct dma_fence *fence = NULL; int r; @@ -436,14 +529,14 @@ static int amdgpu_move_blit(struct ttm_buffer_object *bo, r = amdgpu_ttm_copy_mem_to_mem(adev, &src, &dst, new_mem->num_pages << PAGE_SHIFT, + amdgpu_bo_encrypted(abo), bo->base.resv, &fence); if (r) goto error; /* clear the space being freed */ if (old_mem->mem_type == TTM_PL_VRAM && - (ttm_to_amdgpu_bo(bo)->flags & - AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE)) { + (abo->flags & AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE)) { struct dma_fence *wipe_fence = NULL; r = amdgpu_fill_buffer(ttm_to_amdgpu_bo(bo), AMDGPU_POISON, @@ -740,8 +833,8 @@ static void amdgpu_ttm_io_mem_free(struct ttm_bo_device *bdev, struct ttm_mem_re static unsigned long amdgpu_ttm_io_mem_pfn(struct ttm_buffer_object *bo, unsigned long page_offset) { + uint64_t offset = (page_offset << PAGE_SHIFT); struct drm_mm_node *mm; - unsigned long offset = (page_offset << PAGE_SHIFT); mm = amdgpu_find_mm_node(&bo->mem, &offset); return (bo->mem.bus.base >> PAGE_SHIFT) + mm->start + @@ -785,18 +878,6 @@ struct amdgpu_ttm_tt { }; #ifdef CONFIG_DRM_AMDGPU_USERPTR -/* flags used by HMM internal, not related to CPU/GPU PTE flags */ -static const uint64_t hmm_range_flags[HMM_PFN_FLAG_MAX] = { - (1 << 0), /* HMM_PFN_VALID */ - (1 << 1), /* HMM_PFN_WRITE */ -}; - -static const uint64_t hmm_range_values[HMM_PFN_VALUE_MAX] = { - 0xfffffffffffffffeUL, /* HMM_PFN_ERROR */ - 0, /* HMM_PFN_NONE */ - 0xfffffffffffffffcUL /* HMM_PFN_SPECIAL */ -}; - /** * amdgpu_ttm_tt_get_user_pages - get device accessible pages that back user * memory and start HMM tracking CPU page table update @@ -835,23 +916,20 @@ int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages) goto out; } range->notifier = &bo->notifier; - range->flags = hmm_range_flags; - range->values = hmm_range_values; - range->pfn_shift = PAGE_SHIFT; range->start = bo->notifier.interval_tree.start; range->end = bo->notifier.interval_tree.last + 1; - range->default_flags = hmm_range_flags[HMM_PFN_VALID]; + range->default_flags = HMM_PFN_REQ_FAULT; if (!amdgpu_ttm_tt_is_readonly(ttm)) - range->default_flags |= range->flags[HMM_PFN_WRITE]; + range->default_flags |= HMM_PFN_REQ_WRITE; - range->pfns = kvmalloc_array(ttm->num_pages, sizeof(*range->pfns), - GFP_KERNEL); - if (unlikely(!range->pfns)) { + range->hmm_pfns = kvmalloc_array(ttm->num_pages, + sizeof(*range->hmm_pfns), GFP_KERNEL); + if (unlikely(!range->hmm_pfns)) { r = -ENOMEM; goto out_free_ranges; } - down_read(&mm->mmap_sem); + mmap_read_lock(mm); vma = find_vma(mm, start); if (unlikely(!vma || start < vma->vm_start)) { r = -EFAULT; @@ -862,36 +940,32 @@ int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages) r = -EPERM; goto out_unlock; } - up_read(&mm->mmap_sem); + mmap_read_unlock(mm); timeout = jiffies + msecs_to_jiffies(HMM_RANGE_DEFAULT_TIMEOUT); retry: range->notifier_seq = mmu_interval_read_begin(&bo->notifier); - down_read(&mm->mmap_sem); + mmap_read_lock(mm); r = hmm_range_fault(range); - up_read(&mm->mmap_sem); - if (unlikely(r <= 0)) { + mmap_read_unlock(mm); + if (unlikely(r)) { /* * FIXME: This timeout should encompass the retry from * mmu_interval_read_retry() as well. */ - if ((r == 0 || r == -EBUSY) && !time_after(jiffies, timeout)) + if (r == -EBUSY && !time_after(jiffies, timeout)) goto retry; goto out_free_pfns; } - for (i = 0; i < ttm->num_pages; i++) { - /* FIXME: The pages cannot be touched outside the notifier_lock */ - pages[i] = hmm_device_entry_to_page(range, range->pfns[i]); - if (unlikely(!pages[i])) { - pr_err("Page fault failed for pfn[%lu] = 0x%llx\n", - i, range->pfns[i]); - r = -ENOMEM; - - goto out_free_pfns; - } - } + /* + * Due to default_flags, all pages are HMM_PFN_VALID or + * hmm_range_fault() fails. FIXME: The pages cannot be touched outside + * the notifier_lock, and mmu_interval_read_retry() must be done first. + */ + for (i = 0; i < ttm->num_pages; i++) + pages[i] = hmm_pfn_to_page(range->hmm_pfns[i]); gtt->range = range; mmput(mm); @@ -899,9 +973,9 @@ retry: return 0; out_unlock: - up_read(&mm->mmap_sem); + mmap_read_unlock(mm); out_free_pfns: - kvfree(range->pfns); + kvfree(range->hmm_pfns); out_free_ranges: kfree(range); out: @@ -926,7 +1000,7 @@ bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm) DRM_DEBUG_DRIVER("user_pages_done 0x%llx pages 0x%lx\n", gtt->userptr, ttm->num_pages); - WARN_ONCE(!gtt->range || !gtt->range->pfns, + WARN_ONCE(!gtt->range || !gtt->range->hmm_pfns, "No user pages to check\n"); if (gtt->range) { @@ -936,7 +1010,7 @@ bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm) */ r = mmu_interval_read_retry(gtt->range->notifier, gtt->range->notifier_seq); - kvfree(gtt->range->pfns); + kvfree(gtt->range->hmm_pfns); kfree(gtt->range); gtt->range = NULL; } @@ -1027,8 +1101,7 @@ static void amdgpu_ttm_tt_unpin_userptr(struct ttm_tt *ttm) for (i = 0; i < ttm->num_pages; i++) { if (ttm->pages[i] != - hmm_device_entry_to_page(gtt->range, - gtt->range->pfns[i])) + hmm_pfn_to_page(gtt->range->hmm_pfns[i])) break; } @@ -1046,6 +1119,9 @@ int amdgpu_ttm_gart_bind(struct amdgpu_device *adev, struct amdgpu_ttm_tt *gtt = (void *)ttm; int r; + if (amdgpu_bo_encrypted(abo)) + flags |= AMDGPU_PTE_TMZ; + if (abo->flags & AMDGPU_GEM_CREATE_CP_MQD_GFX9) { uint64_t page_idx = 1; @@ -1555,6 +1631,9 @@ static bool amdgpu_ttm_bo_eviction_valuable(struct ttm_buffer_object *bo, switch (bo->mem.mem_type) { case TTM_PL_TT: + if (amdgpu_bo_is_amdgpu_bo(bo) && + amdgpu_bo_encrypted(ttm_to_amdgpu_bo(bo))) + return false; return true; case TTM_PL_VRAM: @@ -1603,8 +1682,9 @@ static int amdgpu_ttm_access_memory(struct ttm_buffer_object *bo, if (bo->mem.mem_type != TTM_PL_VRAM) return -EIO; - nodes = amdgpu_find_mm_node(&abo->tbo.mem, &offset); - pos = (nodes->start << PAGE_SHIFT) + offset; + pos = offset; + nodes = amdgpu_find_mm_node(&abo->tbo.mem, &pos); + pos += (nodes->start << PAGE_SHIFT); while (len && pos < adev->gmc.mc_vram_size) { uint64_t aligned_pos = pos & ~(uint64_t)3; @@ -1873,17 +1953,19 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) return r; /* - * reserve one TMR (64K) memory at the top of VRAM which holds + * reserve TMR memory at the top of VRAM which holds * IP Discovery data and is protected by PSP. */ - r = amdgpu_bo_create_kernel_at(adev, - adev->gmc.real_vram_size - DISCOVERY_TMR_SIZE, - DISCOVERY_TMR_SIZE, - AMDGPU_GEM_DOMAIN_VRAM, - &adev->discovery_memory, - NULL); - if (r) - return r; + if (adev->discovery_tmr_size > 0) { + r = amdgpu_bo_create_kernel_at(adev, + adev->gmc.real_vram_size - adev->discovery_tmr_size, + adev->discovery_tmr_size, + AMDGPU_GEM_DOMAIN_VRAM, + &adev->discovery_memory, + NULL); + if (r) + return r; + } DRM_INFO("amdgpu: %uM of VRAM memory ready\n", (unsigned) (adev->gmc.real_vram_size / (1024 * 1024))); @@ -2031,75 +2113,14 @@ int amdgpu_mmap(struct file *filp, struct vm_area_struct *vma) return ttm_bo_mmap(filp, vma, &adev->mman.bdev); } -static int amdgpu_map_buffer(struct ttm_buffer_object *bo, - struct ttm_mem_reg *mem, unsigned num_pages, - uint64_t offset, unsigned window, - struct amdgpu_ring *ring, - uint64_t *addr) -{ - struct amdgpu_ttm_tt *gtt = (void *)bo->ttm; - struct amdgpu_device *adev = ring->adev; - struct ttm_tt *ttm = bo->ttm; - struct amdgpu_job *job; - unsigned num_dw, num_bytes; - dma_addr_t *dma_address; - struct dma_fence *fence; - uint64_t src_addr, dst_addr; - uint64_t flags; - int r; - - BUG_ON(adev->mman.buffer_funcs->copy_max_bytes < - AMDGPU_GTT_MAX_TRANSFER_SIZE * 8); - - *addr = adev->gmc.gart_start; - *addr += (u64)window * AMDGPU_GTT_MAX_TRANSFER_SIZE * - AMDGPU_GPU_PAGE_SIZE; - - num_dw = ALIGN(adev->mman.buffer_funcs->copy_num_dw, 8); - num_bytes = num_pages * 8; - - r = amdgpu_job_alloc_with_ib(adev, num_dw * 4 + num_bytes, &job); - if (r) - return r; - - src_addr = num_dw * 4; - src_addr += job->ibs[0].gpu_addr; - - dst_addr = amdgpu_bo_gpu_offset(adev->gart.bo); - dst_addr += window * AMDGPU_GTT_MAX_TRANSFER_SIZE * 8; - amdgpu_emit_copy_buffer(adev, &job->ibs[0], src_addr, - dst_addr, num_bytes); - - amdgpu_ring_pad_ib(ring, &job->ibs[0]); - WARN_ON(job->ibs[0].length_dw > num_dw); - - dma_address = >t->ttm.dma_address[offset >> PAGE_SHIFT]; - flags = amdgpu_ttm_tt_pte_flags(adev, ttm, mem); - r = amdgpu_gart_map(adev, 0, num_pages, dma_address, flags, - &job->ibs[0].ptr[num_dw]); - if (r) - goto error_free; - - r = amdgpu_job_submit(job, &adev->mman.entity, - AMDGPU_FENCE_OWNER_UNDEFINED, &fence); - if (r) - goto error_free; - - dma_fence_put(fence); - - return r; - -error_free: - amdgpu_job_free(job); - return r; -} - int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset, uint64_t dst_offset, uint32_t byte_count, struct dma_resv *resv, struct dma_fence **fence, bool direct_submit, - bool vm_needs_flush) + bool vm_needs_flush, bool tmz) { + enum amdgpu_ib_pool_type pool = direct_submit ? AMDGPU_IB_POOL_DIRECT : + AMDGPU_IB_POOL_DELAYED; struct amdgpu_device *adev = ring->adev; struct amdgpu_job *job; @@ -2117,7 +2138,7 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset, num_loops = DIV_ROUND_UP(byte_count, max_bytes); num_dw = ALIGN(num_loops * adev->mman.buffer_funcs->copy_num_dw, 8); - r = amdgpu_job_alloc_with_ib(adev, num_dw * 4, &job); + r = amdgpu_job_alloc_with_ib(adev, num_dw * 4, pool, &job); if (r) return r; @@ -2139,7 +2160,7 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset, uint32_t cur_size_in_bytes = min(byte_count, max_bytes); amdgpu_emit_copy_buffer(adev, &job->ibs[0], src_offset, - dst_offset, cur_size_in_bytes); + dst_offset, cur_size_in_bytes, tmz); src_offset += cur_size_in_bytes; dst_offset += cur_size_in_bytes; @@ -2206,7 +2227,8 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo, /* for IB padding */ num_dw += 64; - r = amdgpu_job_alloc_with_ib(adev, num_dw * 4, &job); + r = amdgpu_job_alloc_with_ib(adev, num_dw * 4, AMDGPU_IB_POOL_DELAYED, + &job); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h index b6086ab21069..17c8d0d7bcc3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h @@ -97,11 +97,11 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset, uint64_t dst_offset, uint32_t byte_count, struct dma_resv *resv, struct dma_fence **fence, bool direct_submit, - bool vm_needs_flush); + bool vm_needs_flush, bool tmz); int amdgpu_ttm_copy_mem_to_mem(struct amdgpu_device *adev, - struct amdgpu_copy_mem *src, - struct amdgpu_copy_mem *dst, - uint64_t size, + const struct amdgpu_copy_mem *src, + const struct amdgpu_copy_mem *dst, + uint64_t size, bool tmz, struct dma_resv *resv, struct dma_fence **f); int amdgpu_fill_buffer(struct amdgpu_bo *bo, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c index 9ef312428231..65bb25e31d45 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c @@ -403,8 +403,8 @@ FW_VERSION_ATTR(mec_fw_version, 0444, gfx.mec_fw_version); FW_VERSION_ATTR(mec2_fw_version, 0444, gfx.mec2_fw_version); FW_VERSION_ATTR(sos_fw_version, 0444, psp.sos_fw_version); FW_VERSION_ATTR(asd_fw_version, 0444, psp.asd_fw_version); -FW_VERSION_ATTR(ta_ras_fw_version, 0444, psp.ta_fw_version); -FW_VERSION_ATTR(ta_xgmi_fw_version, 0444, psp.ta_fw_version); +FW_VERSION_ATTR(ta_ras_fw_version, 0444, psp.ta_ras_ucode_version); +FW_VERSION_ATTR(ta_xgmi_fw_version, 0444, psp.ta_xgmi_ucode_version); FW_VERSION_ATTR(smc_fw_version, 0444, pm.fw_version); FW_VERSION_ATTR(sdma_fw_version, 0444, sdma.instance[0].fw_version); FW_VERSION_ATTR(sdma2_fw_version, 0444, sdma.instance[1].fw_version); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c index 9dd51f0d2c11..af1b1ccf613c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c @@ -110,7 +110,8 @@ int amdgpu_umc_process_ras_data_cb(struct amdgpu_device *adev, * even NOMEM error is encountered */ if(!err_data->err_addr) - DRM_WARN("Failed to alloc memory for umc error address record!\n"); + dev_warn(adev->dev, "Failed to alloc memory for " + "umc error address record!\n"); /* umc query_ras_error_address is also responsible for clearing * error status @@ -120,10 +121,14 @@ int amdgpu_umc_process_ras_data_cb(struct amdgpu_device *adev, /* only uncorrectable error needs gpu reset */ if (err_data->ue_count) { + dev_info(adev->dev, "%ld uncorrectable hardware errors " + "detected in UMC block\n", + err_data->ue_count); + if (err_data->err_addr_cnt && amdgpu_ras_add_bad_pages(adev, err_data->err_addr, err_data->err_addr_cnt)) - DRM_WARN("Failed to add ras bad page!\n"); + dev_warn(adev->dev, "Failed to add ras bad page!\n"); amdgpu_ras_reset_gpu(adev); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c index 5fd32ad1c575..5100ebe8858d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c @@ -1056,7 +1056,8 @@ static int amdgpu_uvd_send_msg(struct amdgpu_ring *ring, struct amdgpu_bo *bo, goto err; } - r = amdgpu_job_alloc_with_ib(adev, 64, &job); + r = amdgpu_job_alloc_with_ib(adev, 64, direct ? AMDGPU_IB_POOL_DIRECT : + AMDGPU_IB_POOL_DELAYED, &job); if (r) goto err; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c index 59ddba137946..ecaa2d7483b2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c @@ -446,7 +446,8 @@ static int amdgpu_vce_get_create_msg(struct amdgpu_ring *ring, uint32_t handle, uint64_t addr; int i, r; - r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, &job); + r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, + AMDGPU_IB_POOL_DIRECT, &job); if (r) return r; @@ -524,7 +525,9 @@ static int amdgpu_vce_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle, struct dma_fence *f = NULL; int i, r; - r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, &job); + r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, + direct ? AMDGPU_IB_POOL_DIRECT : + AMDGPU_IB_POOL_DELAYED, &job); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c index a41272fbcba2..2badbc0355f2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c @@ -56,19 +56,23 @@ static void amdgpu_vcn_idle_work_handler(struct work_struct *work); int amdgpu_vcn_sw_init(struct amdgpu_device *adev) { - unsigned long bo_size; + unsigned long bo_size, fw_shared_bo_size; const char *fw_name; const struct common_firmware_header *hdr; unsigned char fw_check; int i, r; INIT_DELAYED_WORK(&adev->vcn.idle_work, amdgpu_vcn_idle_work_handler); + mutex_init(&adev->vcn.vcn_pg_lock); + atomic_set(&adev->vcn.total_submission_cnt, 0); + for (i = 0; i < adev->vcn.num_vcn_inst; i++) + atomic_set(&adev->vcn.inst[i].dpg_enc_submission_cnt, 0); switch (adev->asic_type) { case CHIP_RAVEN: - if (adev->rev_id >= 8) + if (adev->apu_flags & AMD_APU_IS_RAVEN2) fw_name = FIRMWARE_RAVEN2; - else if (adev->pdev->device == 0x15d8) + else if (adev->apu_flags & AMD_APU_IS_PICASSO) fw_name = FIRMWARE_PICASSO; else fw_name = FIRMWARE_RAVEN; @@ -178,6 +182,17 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev) return r; } } + + r = amdgpu_bo_create_kernel(adev, AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_fw_shared)), + PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM, &adev->vcn.inst[i].fw_shared_bo, + &adev->vcn.inst[i].fw_shared_gpu_addr, &adev->vcn.inst[i].fw_shared_cpu_addr); + if (r) { + dev_err(adev->dev, "VCN %d (%d) failed to allocate firmware shared bo\n", i, r); + return r; + } + + fw_shared_bo_size = amdgpu_bo_size(adev->vcn.inst[i].fw_shared_bo); + adev->vcn.inst[i].saved_shm_bo = kvmalloc(fw_shared_bo_size, GFP_KERNEL); } return 0; @@ -192,6 +207,12 @@ int amdgpu_vcn_sw_fini(struct amdgpu_device *adev) for (j = 0; j < adev->vcn.num_vcn_inst; ++j) { if (adev->vcn.harvest_config & (1 << j)) continue; + + kvfree(adev->vcn.inst[j].saved_shm_bo); + amdgpu_bo_free_kernel(&adev->vcn.inst[j].fw_shared_bo, + &adev->vcn.inst[j].fw_shared_gpu_addr, + (void **)&adev->vcn.inst[j].fw_shared_cpu_addr); + if (adev->vcn.indirect_sram) { amdgpu_bo_free_kernel(&adev->vcn.inst[j].dpg_sram_bo, &adev->vcn.inst[j].dpg_sram_gpu_addr, @@ -210,6 +231,7 @@ int amdgpu_vcn_sw_fini(struct amdgpu_device *adev) } release_firmware(adev->vcn.fw); + mutex_destroy(&adev->vcn.vcn_pg_lock); return 0; } @@ -236,6 +258,17 @@ int amdgpu_vcn_suspend(struct amdgpu_device *adev) return -ENOMEM; memcpy_fromio(adev->vcn.inst[i].saved_bo, ptr, size); + + if (adev->vcn.inst[i].fw_shared_bo == NULL) + return 0; + + if (!adev->vcn.inst[i].saved_shm_bo) + return -ENOMEM; + + size = amdgpu_bo_size(adev->vcn.inst[i].fw_shared_bo); + ptr = adev->vcn.inst[i].fw_shared_cpu_addr; + + memcpy_fromio(adev->vcn.inst[i].saved_shm_bo, ptr, size); } return 0; } @@ -273,6 +306,17 @@ int amdgpu_vcn_resume(struct amdgpu_device *adev) } memset_io(ptr, 0, size); } + + if (adev->vcn.inst[i].fw_shared_bo == NULL) + return -EINVAL; + + size = amdgpu_bo_size(adev->vcn.inst[i].fw_shared_bo); + ptr = adev->vcn.inst[i].fw_shared_cpu_addr; + + if (adev->vcn.inst[i].saved_shm_bo != NULL) + memcpy_toio(ptr, adev->vcn.inst[i].saved_shm_bo, size); + else + memset_io(ptr, 0, size); } return 0; } @@ -295,7 +339,8 @@ static void amdgpu_vcn_idle_work_handler(struct work_struct *work) if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) { struct dpg_pause_state new_state; - if (fence[j]) + if (fence[j] || + unlikely(atomic_read(&adev->vcn.inst[j].dpg_enc_submission_cnt))) new_state.fw_based = VCN_DPG_STATE__PAUSE; else new_state.fw_based = VCN_DPG_STATE__UNPAUSE; @@ -307,8 +352,7 @@ static void amdgpu_vcn_idle_work_handler(struct work_struct *work) fences += fence[j]; } - if (fences == 0) { - amdgpu_gfx_off_ctrl(adev, true); + if (!fences && !atomic_read(&adev->vcn.total_submission_cnt)) { amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VCN, AMD_PG_STATE_GATE); } else { @@ -319,36 +363,46 @@ static void amdgpu_vcn_idle_work_handler(struct work_struct *work) void amdgpu_vcn_ring_begin_use(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; - bool set_clocks = !cancel_delayed_work_sync(&adev->vcn.idle_work); - if (set_clocks) { - amdgpu_gfx_off_ctrl(adev, false); - amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VCN, - AMD_PG_STATE_UNGATE); - } + atomic_inc(&adev->vcn.total_submission_cnt); + cancel_delayed_work_sync(&adev->vcn.idle_work); + + mutex_lock(&adev->vcn.vcn_pg_lock); + amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VCN, + AMD_PG_STATE_UNGATE); if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) { struct dpg_pause_state new_state; - unsigned int fences = 0; - unsigned int i; - for (i = 0; i < adev->vcn.num_enc_rings; ++i) { - fences += amdgpu_fence_count_emitted(&adev->vcn.inst[ring->me].ring_enc[i]); - } - if (fences) + if (ring->funcs->type == AMDGPU_RING_TYPE_VCN_ENC) { + atomic_inc(&adev->vcn.inst[ring->me].dpg_enc_submission_cnt); new_state.fw_based = VCN_DPG_STATE__PAUSE; - else - new_state.fw_based = VCN_DPG_STATE__UNPAUSE; + } else { + unsigned int fences = 0; + unsigned int i; - if (ring->funcs->type == AMDGPU_RING_TYPE_VCN_ENC) - new_state.fw_based = VCN_DPG_STATE__PAUSE; + for (i = 0; i < adev->vcn.num_enc_rings; ++i) + fences += amdgpu_fence_count_emitted(&adev->vcn.inst[ring->me].ring_enc[i]); + + if (fences || atomic_read(&adev->vcn.inst[ring->me].dpg_enc_submission_cnt)) + new_state.fw_based = VCN_DPG_STATE__PAUSE; + else + new_state.fw_based = VCN_DPG_STATE__UNPAUSE; + } adev->vcn.pause_dpg_mode(adev, ring->me, &new_state); } + mutex_unlock(&adev->vcn.vcn_pg_lock); } void amdgpu_vcn_ring_end_use(struct amdgpu_ring *ring) { + if (ring->adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG && + ring->funcs->type == AMDGPU_RING_TYPE_VCN_ENC) + atomic_dec(&ring->adev->vcn.inst[ring->me].dpg_enc_submission_cnt); + + atomic_dec(&ring->adev->vcn.total_submission_cnt); + schedule_delayed_work(&ring->adev->vcn.idle_work, VCN_IDLE_TIMEOUT); } @@ -390,7 +444,8 @@ static int amdgpu_vcn_dec_send_msg(struct amdgpu_ring *ring, uint64_t addr; int i, r; - r = amdgpu_job_alloc_with_ib(adev, 64, &job); + r = amdgpu_job_alloc_with_ib(adev, 64, + AMDGPU_IB_POOL_DIRECT, &job); if (r) goto err; @@ -557,7 +612,8 @@ static int amdgpu_vcn_enc_get_create_msg(struct amdgpu_ring *ring, uint32_t hand uint64_t addr; int i, r; - r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, &job); + r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, + AMDGPU_IB_POOL_DIRECT, &job); if (r) return r; @@ -610,7 +666,8 @@ static int amdgpu_vcn_enc_get_destroy_msg(struct amdgpu_ring *ring, uint32_t han uint64_t addr; int i, r; - r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, &job); + r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, + AMDGPU_IB_POOL_DIRECT, &job); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h index 6fe057329de2..90aa12b22725 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h @@ -132,6 +132,13 @@ } \ } while (0) +#define AMDGPU_VCN_MULTI_QUEUE_FLAG (1 << 8) + +enum fw_queue_mode { + FW_QUEUE_RING_RESET = 1, + FW_QUEUE_DPG_HOLD_OFF = 2, +}; + enum engine_status_constants { UVD_PGFSM_STATUS__UVDM_UVDU_PWR_ON = 0x2AAAA0, UVD_PGFSM_STATUS__UVDM_UVDU_PWR_ON_2_0 = 0xAAAA0, @@ -179,10 +186,15 @@ struct amdgpu_vcn_inst { struct amdgpu_irq_src irq; struct amdgpu_vcn_reg external; struct amdgpu_bo *dpg_sram_bo; + struct amdgpu_bo *fw_shared_bo; struct dpg_pause_state pause_state; void *dpg_sram_cpu_addr; uint64_t dpg_sram_gpu_addr; uint32_t *dpg_sram_curr_addr; + atomic_t dpg_enc_submission_cnt; + void *fw_shared_cpu_addr; + uint64_t fw_shared_gpu_addr; + void *saved_shm_bo; }; struct amdgpu_vcn { @@ -196,16 +208,28 @@ struct amdgpu_vcn { uint8_t num_vcn_inst; struct amdgpu_vcn_inst inst[AMDGPU_MAX_VCN_INSTANCES]; struct amdgpu_vcn_reg internal; - struct drm_gpu_scheduler *vcn_enc_sched[AMDGPU_MAX_VCN_ENC_RINGS]; - struct drm_gpu_scheduler *vcn_dec_sched[AMDGPU_MAX_VCN_INSTANCES]; - uint32_t num_vcn_enc_sched; - uint32_t num_vcn_dec_sched; + struct mutex vcn_pg_lock; + atomic_t total_submission_cnt; unsigned harvest_config; int (*pause_dpg_mode)(struct amdgpu_device *adev, int inst_idx, struct dpg_pause_state *new_state); }; +struct amdgpu_fw_shared_multi_queue { + uint8_t decode_queue_mode; + uint8_t encode_generalpurpose_queue_mode; + uint8_t encode_lowlatency_queue_mode; + uint8_t encode_realtime_queue_mode; + uint8_t padding[4]; +}; + +struct amdgpu_fw_shared { + uint32_t present_flag_0; + uint8_t pad[53]; + struct amdgpu_fw_shared_multi_queue multi_queue; +} __attribute__((__packed__)); + int amdgpu_vcn_sw_init(struct amdgpu_device *adev); int amdgpu_vcn_sw_fini(struct amdgpu_device *adev); int amdgpu_vcn_suspend(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c index adc813cde8e2..f3b38c9e04ca 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c @@ -38,7 +38,8 @@ bool amdgpu_virt_mmio_blocked(struct amdgpu_device *adev) void amdgpu_virt_init_setting(struct amdgpu_device *adev) { /* enable virtual display */ - adev->mode_info.num_crtc = 1; + if (adev->mode_info.num_crtc == 0) + adev->mode_info.num_crtc = 1; adev->enable_virtual_display = true; adev->ddev->driver->driver_features &= ~DRIVER_ATOMIC; adev->cg_flags = 0; @@ -59,7 +60,10 @@ void amdgpu_virt_kiq_reg_write_reg_wait(struct amdgpu_device *adev, amdgpu_ring_alloc(ring, 32); amdgpu_ring_emit_reg_write_reg_wait(ring, reg0, reg1, ref, mask); - amdgpu_fence_emit_polling(ring, &seq); + r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT); + if (r) + goto failed_undo; + amdgpu_ring_commit(ring); spin_unlock_irqrestore(&kiq->ring_lock, flags); @@ -81,6 +85,9 @@ void amdgpu_virt_kiq_reg_write_reg_wait(struct amdgpu_device *adev, return; +failed_undo: + amdgpu_ring_undo(ring); + spin_unlock_irqrestore(&kiq->ring_lock, flags); failed_kiq: pr_err("failed to write reg %x wait reg %x\n", reg0, reg1); } @@ -152,6 +159,19 @@ int amdgpu_virt_reset_gpu(struct amdgpu_device *adev) return 0; } +void amdgpu_virt_request_init_data(struct amdgpu_device *adev) +{ + struct amdgpu_virt *virt = &adev->virt; + + if (virt->ops && virt->ops->req_init_data) + virt->ops->req_init_data(adev); + + if (adev->virt.req_init_data_ver > 0) + DRM_INFO("host supports REQ_INIT_DATA handshake\n"); + else + DRM_WARN("host doesn't support REQ_INIT_DATA handshake\n"); +} + /** * amdgpu_virt_wait_reset() - wait for reset gpu completed * @amdgpu: amdgpu device. @@ -287,3 +307,82 @@ void amdgpu_virt_init_data_exchange(struct amdgpu_device *adev) } } } + +void amdgpu_detect_virtualization(struct amdgpu_device *adev) +{ + uint32_t reg; + + switch (adev->asic_type) { + case CHIP_TONGA: + case CHIP_FIJI: + reg = RREG32(mmBIF_IOV_FUNC_IDENTIFIER); + break; + case CHIP_VEGA10: + case CHIP_VEGA20: + case CHIP_NAVI10: + case CHIP_NAVI12: + case CHIP_ARCTURUS: + reg = RREG32(mmRCC_IOV_FUNC_IDENTIFIER); + break; + default: /* other chip doesn't support SRIOV */ + reg = 0; + break; + } + + if (reg & 1) + adev->virt.caps |= AMDGPU_SRIOV_CAPS_IS_VF; + + if (reg & 0x80000000) + adev->virt.caps |= AMDGPU_SRIOV_CAPS_ENABLE_IOV; + + if (!reg) { + if (is_virtual_machine()) /* passthrough mode exclus sriov mod */ + adev->virt.caps |= AMDGPU_PASSTHROUGH_MODE; + } +} + +bool amdgpu_virt_access_debugfs_is_mmio(struct amdgpu_device *adev) +{ + return amdgpu_sriov_is_debug(adev) ? true : false; +} + +bool amdgpu_virt_access_debugfs_is_kiq(struct amdgpu_device *adev) +{ + return amdgpu_sriov_is_normal(adev) ? true : false; +} + +int amdgpu_virt_enable_access_debugfs(struct amdgpu_device *adev) +{ + if (!amdgpu_sriov_vf(adev) || + amdgpu_virt_access_debugfs_is_kiq(adev)) + return 0; + + if (amdgpu_virt_access_debugfs_is_mmio(adev)) + adev->virt.caps &= ~AMDGPU_SRIOV_CAPS_RUNTIME; + else + return -EPERM; + + return 0; +} + +void amdgpu_virt_disable_access_debugfs(struct amdgpu_device *adev) +{ + if (amdgpu_sriov_vf(adev)) + adev->virt.caps |= AMDGPU_SRIOV_CAPS_RUNTIME; +} + +enum amdgpu_sriov_vf_mode amdgpu_virt_get_sriov_vf_mode(struct amdgpu_device *adev) +{ + enum amdgpu_sriov_vf_mode mode; + + if (amdgpu_sriov_vf(adev)) { + if (amdgpu_sriov_is_pp_one_vf(adev)) + mode = SRIOV_VF_MODE_ONE_VF; + else + mode = SRIOV_VF_MODE_MULTI_VF; + } else { + mode = SRIOV_VF_MODE_BARE_METAL; + } + + return mode; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h index f0128f745bd2..b90e822cebd7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h @@ -30,6 +30,17 @@ #define AMDGPU_PASSTHROUGH_MODE (1 << 3) /* thw whole GPU is pass through for VM */ #define AMDGPU_SRIOV_CAPS_RUNTIME (1 << 4) /* is out of full access mode */ +/* all asic after AI use this offset */ +#define mmRCC_IOV_FUNC_IDENTIFIER 0xDE5 +/* tonga/fiji use this offset */ +#define mmBIF_IOV_FUNC_IDENTIFIER 0x1503 + +enum amdgpu_sriov_vf_mode { + SRIOV_VF_MODE_BARE_METAL = 0, + SRIOV_VF_MODE_ONE_VF, + SRIOV_VF_MODE_MULTI_VF, +}; + struct amdgpu_mm_table { struct amdgpu_bo *bo; uint32_t *cpu_addr; @@ -54,6 +65,7 @@ struct amdgpu_vf_error_buffer { struct amdgpu_virt_ops { int (*req_full_gpu)(struct amdgpu_device *adev, bool init); int (*rel_full_gpu)(struct amdgpu_device *adev, bool init); + int (*req_init_data)(struct amdgpu_device *adev); int (*reset_gpu)(struct amdgpu_device *adev); int (*wait_reset)(struct amdgpu_device *adev); void (*trans_msg)(struct amdgpu_device *adev, u32 req, u32 data1, u32 data2, u32 data3); @@ -83,6 +95,8 @@ enum AMDGIM_FEATURE_FLAG { AMDGIM_FEATURE_GIM_LOAD_UCODES = 0x2, /* VRAM LOST by GIM */ AMDGIM_FEATURE_GIM_FLR_VRAMLOST = 0x4, + /* MM bandwidth */ + AMDGIM_FEATURE_GIM_MM_BW_MGR = 0x8, /* PP ONE VF MODE in GIM */ AMDGIM_FEATURE_PP_ONE_VF = (1 << 4), }; @@ -256,6 +270,8 @@ struct amdgpu_virt { struct amdgpu_virt_fw_reserve fw_reserve; uint32_t gim_feature; uint32_t reg_access_mode; + int req_init_data_ver; + bool tdr_debug; }; #define amdgpu_sriov_enabled(adev) \ @@ -287,6 +303,10 @@ static inline bool is_virtual_machine(void) #define amdgpu_sriov_is_pp_one_vf(adev) \ ((adev)->virt.gim_feature & AMDGIM_FEATURE_PP_ONE_VF) +#define amdgpu_sriov_is_debug(adev) \ + ((!adev->in_gpu_reset) && adev->virt.tdr_debug) +#define amdgpu_sriov_is_normal(adev) \ + ((!adev->in_gpu_reset) && (!adev->virt.tdr_debug)) bool amdgpu_virt_mmio_blocked(struct amdgpu_device *adev); void amdgpu_virt_init_setting(struct amdgpu_device *adev); @@ -296,6 +316,7 @@ void amdgpu_virt_kiq_reg_write_reg_wait(struct amdgpu_device *adev, int amdgpu_virt_request_full_gpu(struct amdgpu_device *adev, bool init); int amdgpu_virt_release_full_gpu(struct amdgpu_device *adev, bool init); int amdgpu_virt_reset_gpu(struct amdgpu_device *adev); +void amdgpu_virt_request_init_data(struct amdgpu_device *adev); int amdgpu_virt_wait_reset(struct amdgpu_device *adev); int amdgpu_virt_alloc_mm_table(struct amdgpu_device *adev); void amdgpu_virt_free_mm_table(struct amdgpu_device *adev); @@ -303,4 +324,11 @@ int amdgpu_virt_fw_reserve_get_checksum(void *obj, unsigned long obj_size, unsigned int key, unsigned int chksum); void amdgpu_virt_init_data_exchange(struct amdgpu_device *adev); +void amdgpu_detect_virtualization(struct amdgpu_device *adev); + +bool amdgpu_virt_can_access_debugfs(struct amdgpu_device *adev); +int amdgpu_virt_enable_access_debugfs(struct amdgpu_device *adev); +void amdgpu_virt_disable_access_debugfs(struct amdgpu_device *adev); + +enum amdgpu_sriov_vf_mode amdgpu_virt_get_sriov_vf_mode(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index 6d9252a27916..7417754e9141 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -82,7 +82,7 @@ struct amdgpu_prt_cb { struct dma_fence_cb cb; }; -/** +/* * vm eviction_lock can be taken in MMU notifiers. Make sure no reclaim-FS * happens while holding this lock anywhere to prevent deadlocks when * an MMU notifier runs in reclaim-FS context. @@ -726,7 +726,7 @@ bool amdgpu_vm_ready(struct amdgpu_vm *vm) * @adev: amdgpu_device pointer * @vm: VM to clear BO from * @bo: BO to clear - * @direct: use a direct update + * @immediate: use an immediate update * * Root PD needs to be reserved when calling this. * @@ -736,7 +736,7 @@ bool amdgpu_vm_ready(struct amdgpu_vm *vm) static int amdgpu_vm_clear_bo(struct amdgpu_device *adev, struct amdgpu_vm *vm, struct amdgpu_bo *bo, - bool direct) + bool immediate) { struct ttm_operation_ctx ctx = { true, false }; unsigned level = adev->vm_manager.root_level; @@ -795,7 +795,7 @@ static int amdgpu_vm_clear_bo(struct amdgpu_device *adev, memset(¶ms, 0, sizeof(params)); params.adev = adev; params.vm = vm; - params.direct = direct; + params.immediate = immediate; r = vm->update_funcs->prepare(¶ms, NULL, AMDGPU_SYNC_EXPLICIT); if (r) @@ -850,11 +850,11 @@ static int amdgpu_vm_clear_bo(struct amdgpu_device *adev, * @adev: amdgpu_device pointer * @vm: requesting vm * @level: the page table level - * @direct: use a direct update + * @immediate: use a immediate update * @bp: resulting BO allocation parameters */ static void amdgpu_vm_bo_param(struct amdgpu_device *adev, struct amdgpu_vm *vm, - int level, bool direct, + int level, bool immediate, struct amdgpu_bo_param *bp) { memset(bp, 0, sizeof(*bp)); @@ -870,7 +870,7 @@ static void amdgpu_vm_bo_param(struct amdgpu_device *adev, struct amdgpu_vm *vm, else if (!vm->root.base.bo || vm->root.base.bo->shadow) bp->flags |= AMDGPU_GEM_CREATE_SHADOW; bp->type = ttm_bo_type_kernel; - bp->no_wait_gpu = direct; + bp->no_wait_gpu = immediate; if (vm->root.base.bo) bp->resv = vm->root.base.bo->tbo.base.resv; } @@ -881,7 +881,7 @@ static void amdgpu_vm_bo_param(struct amdgpu_device *adev, struct amdgpu_vm *vm, * @adev: amdgpu_device pointer * @vm: VM to allocate page tables for * @cursor: Which page table to allocate - * @direct: use a direct update + * @immediate: use an immediate update * * Make sure a specific page table or directory is allocated. * @@ -892,7 +892,7 @@ static void amdgpu_vm_bo_param(struct amdgpu_device *adev, struct amdgpu_vm *vm, static int amdgpu_vm_alloc_pts(struct amdgpu_device *adev, struct amdgpu_vm *vm, struct amdgpu_vm_pt_cursor *cursor, - bool direct) + bool immediate) { struct amdgpu_vm_pt *entry = cursor->entry; struct amdgpu_bo_param bp; @@ -913,7 +913,7 @@ static int amdgpu_vm_alloc_pts(struct amdgpu_device *adev, if (entry->base.bo) return 0; - amdgpu_vm_bo_param(adev, vm, cursor->level, direct, &bp); + amdgpu_vm_bo_param(adev, vm, cursor->level, immediate, &bp); r = amdgpu_bo_create(adev, &bp, &pt); if (r) @@ -925,7 +925,7 @@ static int amdgpu_vm_alloc_pts(struct amdgpu_device *adev, pt->parent = amdgpu_bo_ref(cursor->parent->base.bo); amdgpu_vm_bo_base_init(&entry->base, vm, pt); - r = amdgpu_vm_clear_bo(adev, vm, pt, direct); + r = amdgpu_vm_clear_bo(adev, vm, pt, immediate); if (r) goto error_free_pt; @@ -1276,7 +1276,7 @@ static void amdgpu_vm_invalidate_pds(struct amdgpu_device *adev, * * @adev: amdgpu_device pointer * @vm: requested vm - * @direct: submit directly to the paging queue + * @immediate: submit immediately to the paging queue * * Makes sure all directories are up to date. * @@ -1284,7 +1284,7 @@ static void amdgpu_vm_invalidate_pds(struct amdgpu_device *adev, * 0 for success, error for failure. */ int amdgpu_vm_update_pdes(struct amdgpu_device *adev, - struct amdgpu_vm *vm, bool direct) + struct amdgpu_vm *vm, bool immediate) { struct amdgpu_vm_update_params params; int r; @@ -1295,7 +1295,7 @@ int amdgpu_vm_update_pdes(struct amdgpu_device *adev, memset(¶ms, 0, sizeof(params)); params.adev = adev; params.vm = vm; - params.direct = direct; + params.immediate = immediate; r = vm->update_funcs->prepare(¶ms, NULL, AMDGPU_SYNC_EXPLICIT); if (r) @@ -1446,20 +1446,24 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params, uint64_t incr, entry_end, pe_start; struct amdgpu_bo *pt; - if (flags & (AMDGPU_PTE_VALID | AMDGPU_PTE_PRT)) { + if (!params->unlocked) { /* make sure that the page tables covering the * address range are actually allocated */ r = amdgpu_vm_alloc_pts(params->adev, params->vm, - &cursor, params->direct); + &cursor, params->immediate); if (r) return r; } shift = amdgpu_vm_level_shift(adev, cursor.level); parent_shift = amdgpu_vm_level_shift(adev, cursor.level - 1); - if (adev->asic_type < CHIP_VEGA10 && - (flags & AMDGPU_PTE_VALID)) { + if (params->unlocked) { + /* Unlocked updates are only allowed on the leaves */ + if (amdgpu_vm_pt_descendant(adev, &cursor)) + continue; + } else if (adev->asic_type < CHIP_VEGA10 && + (flags & AMDGPU_PTE_VALID)) { /* No huge page support before GMC v9 */ if (cursor.level != AMDGPU_VM_PTB) { if (!amdgpu_vm_pt_descendant(adev, &cursor)) @@ -1557,7 +1561,8 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params, * * @adev: amdgpu_device pointer * @vm: requested vm - * @direct: direct submission in a page fault + * @immediate: immediate submission in a page fault + * @unlocked: unlocked invalidation during MM callback * @resv: fences we need to sync to * @start: start of mapped range * @last: last mapped entry @@ -1572,8 +1577,8 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params, * 0 for success, -EINVAL for failure. */ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev, - struct amdgpu_vm *vm, bool direct, - struct dma_resv *resv, + struct amdgpu_vm *vm, bool immediate, + bool unlocked, struct dma_resv *resv, uint64_t start, uint64_t last, uint64_t flags, uint64_t addr, dma_addr_t *pages_addr, @@ -1586,8 +1591,9 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev, memset(¶ms, 0, sizeof(params)); params.adev = adev; params.vm = vm; - params.direct = direct; + params.immediate = immediate; params.pages_addr = pages_addr; + params.unlocked = unlocked; /* Implicitly sync to command submissions in the same VM before * unmapping. Sync to moving fences before mapping. @@ -1603,11 +1609,12 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev, goto error_unlock; } - if (flags & (AMDGPU_PTE_VALID | AMDGPU_PTE_PRT)) { - struct amdgpu_bo *root = vm->root.base.bo; + if (!unlocked && !dma_fence_is_signaled(vm->last_unlocked)) { + struct dma_fence *tmp = dma_fence_get_stub(); - if (!dma_fence_is_signaled(vm->last_direct)) - amdgpu_bo_fence(root, vm->last_direct, true); + amdgpu_bo_fence(vm->root.base.bo, vm->last_unlocked, true); + swap(vm->last_unlocked, tmp); + dma_fence_put(tmp); } r = vm->update_funcs->prepare(¶ms, resv, sync_mode); @@ -1721,7 +1728,7 @@ static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev, } last = min((uint64_t)mapping->last, start + max_entries - 1); - r = amdgpu_vm_bo_update_mapping(adev, vm, false, resv, + r = amdgpu_vm_bo_update_mapping(adev, vm, false, false, resv, start, last, flags, addr, dma_addr, fence); if (r) @@ -1784,6 +1791,10 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va, if (bo) { flags = amdgpu_ttm_tt_pte_flags(adev, bo->tbo.ttm, mem); + + if (amdgpu_bo_encrypted(bo)) + flags |= AMDGPU_PTE_TMZ; + bo_adev = amdgpu_ttm_adev(bo->tbo.bdev); } else { flags = 0x0; @@ -2014,7 +2025,7 @@ int amdgpu_vm_clear_freed(struct amdgpu_device *adev, mapping->start < AMDGPU_GMC_HOLE_START) init_pte_value = AMDGPU_PTE_DEFAULT_ATC; - r = amdgpu_vm_bo_update_mapping(adev, vm, false, resv, + r = amdgpu_vm_bo_update_mapping(adev, vm, false, false, resv, mapping->start, mapping->last, init_pte_value, 0, NULL, &f); amdgpu_vm_free_mapping(adev, vm, mapping, f); @@ -2124,11 +2135,8 @@ struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev, if (bo && amdgpu_xgmi_same_hive(adev, amdgpu_ttm_adev(bo->tbo.bdev)) && (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM)) { bo_va->is_xgmi = true; - mutex_lock(&adev->vm_manager.lock_pstate); /* Power up XGMI if it can be potentially used */ - if (++adev->vm_manager.xgmi_map_counter == 1) - amdgpu_xgmi_set_pstate(adev, 1); - mutex_unlock(&adev->vm_manager.lock_pstate); + amdgpu_xgmi_set_pstate(adev, AMDGPU_XGMI_PSTATE_MAX_VEGA20); } return bo_va; @@ -2551,12 +2559,8 @@ void amdgpu_vm_bo_rmv(struct amdgpu_device *adev, dma_fence_put(bo_va->last_pt_update); - if (bo && bo_va->is_xgmi) { - mutex_lock(&adev->vm_manager.lock_pstate); - if (--adev->vm_manager.xgmi_map_counter == 0) - amdgpu_xgmi_set_pstate(adev, 0); - mutex_unlock(&adev->vm_manager.lock_pstate); - } + if (bo && bo_va->is_xgmi) + amdgpu_xgmi_set_pstate(adev, AMDGPU_XGMI_PSTATE_MIN); kfree(bo_va); } @@ -2585,7 +2589,7 @@ bool amdgpu_vm_evictable(struct amdgpu_bo *bo) return false; /* Don't evict VM page tables while they are updated */ - if (!dma_fence_is_signaled(bo_base->vm->last_direct)) { + if (!dma_fence_is_signaled(bo_base->vm->last_unlocked)) { amdgpu_vm_eviction_unlock(bo_base->vm); return false; } @@ -2762,7 +2766,7 @@ long amdgpu_vm_wait_idle(struct amdgpu_vm *vm, long timeout) if (timeout <= 0) return timeout; - return dma_fence_wait_timeout(vm->last_direct, true, timeout); + return dma_fence_wait_timeout(vm->last_unlocked, true, timeout); } /** @@ -2798,7 +2802,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, /* create scheduler entities for page table updates */ - r = drm_sched_entity_init(&vm->direct, DRM_SCHED_PRIORITY_NORMAL, + r = drm_sched_entity_init(&vm->immediate, DRM_SCHED_PRIORITY_NORMAL, adev->vm_manager.vm_pte_scheds, adev->vm_manager.vm_pte_num_scheds, NULL); if (r) @@ -2808,7 +2812,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, adev->vm_manager.vm_pte_scheds, adev->vm_manager.vm_pte_num_scheds, NULL); if (r) - goto error_free_direct; + goto error_free_immediate; vm->pte_support_ats = false; vm->is_compute_context = false; @@ -2834,7 +2838,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, else vm->update_funcs = &amdgpu_vm_sdma_funcs; vm->last_update = NULL; - vm->last_direct = dma_fence_get_stub(); + vm->last_unlocked = dma_fence_get_stub(); mutex_init(&vm->eviction_lock); vm->evicting = false; @@ -2888,11 +2892,11 @@ error_free_root: vm->root.base.bo = NULL; error_free_delayed: - dma_fence_put(vm->last_direct); + dma_fence_put(vm->last_unlocked); drm_sched_entity_destroy(&vm->delayed); -error_free_direct: - drm_sched_entity_destroy(&vm->direct); +error_free_immediate: + drm_sched_entity_destroy(&vm->immediate); return r; } @@ -2996,10 +3000,17 @@ int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm, !amdgpu_gmc_vram_full_visible(&adev->gmc)), "CPU update of VM recommended only for large BAR system\n"); - if (vm->use_cpu_for_update) + if (vm->use_cpu_for_update) { + /* Sync with last SDMA update/clear before switching to CPU */ + r = amdgpu_bo_sync_wait(vm->root.base.bo, + AMDGPU_FENCE_OWNER_UNDEFINED, true); + if (r) + goto free_idr; + vm->update_funcs = &amdgpu_vm_cpu_funcs; - else + } else { vm->update_funcs = &amdgpu_vm_sdma_funcs; + } dma_fence_put(vm->last_update); vm->last_update = NULL; vm->is_compute_context = true; @@ -3089,8 +3100,8 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm) vm->pasid = 0; } - dma_fence_wait(vm->last_direct, false); - dma_fence_put(vm->last_direct); + dma_fence_wait(vm->last_unlocked, false); + dma_fence_put(vm->last_unlocked); list_for_each_entry_safe(mapping, tmp, &vm->freed, list) { if (mapping->flags & AMDGPU_PTE_PRT && prt_fini_needed) { @@ -3107,7 +3118,7 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm) amdgpu_bo_unref(&root); WARN_ON(vm->root.base.bo); - drm_sched_entity_destroy(&vm->direct); + drm_sched_entity_destroy(&vm->immediate); drm_sched_entity_destroy(&vm->delayed); if (!RB_EMPTY_ROOT(&vm->va.rb_root)) { @@ -3166,9 +3177,6 @@ void amdgpu_vm_manager_init(struct amdgpu_device *adev) idr_init(&adev->vm_manager.pasid_idr); spin_lock_init(&adev->vm_manager.pasid_lock); - - adev->vm_manager.xgmi_map_counter = 0; - mutex_init(&adev->vm_manager.lock_pstate); } /** @@ -3343,8 +3351,8 @@ bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, unsigned int pasid, value = 0; } - r = amdgpu_vm_bo_update_mapping(adev, vm, true, NULL, addr, addr + 1, - flags, value, NULL, NULL); + r = amdgpu_vm_bo_update_mapping(adev, vm, true, false, NULL, addr, + addr + 1, flags, value, NULL, NULL); if (r) goto error_unlock; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h index 06fe30e1492d..c8e68d7890bf 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h @@ -54,6 +54,9 @@ struct amdgpu_bo_list_entry; #define AMDGPU_PTE_SYSTEM (1ULL << 1) #define AMDGPU_PTE_SNOOPED (1ULL << 2) +/* RV+ */ +#define AMDGPU_PTE_TMZ (1ULL << 3) + /* VI only */ #define AMDGPU_PTE_EXECUTABLE (1ULL << 4) @@ -203,9 +206,14 @@ struct amdgpu_vm_update_params { struct amdgpu_vm *vm; /** - * @direct: if changes should be made directly + * @immediate: if changes should be made immediately */ - bool direct; + bool immediate; + + /** + * @unlocked: true if the root BO is not locked + */ + bool unlocked; /** * @pages_addr: @@ -271,11 +279,11 @@ struct amdgpu_vm { struct dma_fence *last_update; /* Scheduler entities for page table updates */ - struct drm_sched_entity direct; + struct drm_sched_entity immediate; struct drm_sched_entity delayed; - /* Last submission to the scheduler entities */ - struct dma_fence *last_direct; + /* Last unlocked submission to the scheduler entities */ + struct dma_fence *last_unlocked; unsigned int pasid; /* dedicated to vm */ @@ -349,10 +357,6 @@ struct amdgpu_vm_manager { */ struct idr pasid_idr; spinlock_t pasid_lock; - - /* counter of mapped memory through xgmi */ - uint32_t xgmi_map_counter; - struct mutex lock_pstate; }; #define amdgpu_vm_copy_pte(adev, ib, pe, src, count) ((adev)->vm_manager.vm_pte_funcs->copy_pte((ib), (pe), (src), (count))) @@ -380,7 +384,7 @@ int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm, void *param); int amdgpu_vm_flush(struct amdgpu_ring *ring, struct amdgpu_job *job, bool need_pipe_sync); int amdgpu_vm_update_pdes(struct amdgpu_device *adev, - struct amdgpu_vm *vm, bool direct); + struct amdgpu_vm *vm, bool immediate); int amdgpu_vm_clear_freed(struct amdgpu_device *adev, struct amdgpu_vm *vm, struct dma_fence **fence); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c index e38516304070..39c704a1fb0e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c @@ -84,7 +84,7 @@ static int amdgpu_vm_cpu_update(struct amdgpu_vm_update_params *p, pe += (unsigned long)amdgpu_bo_kptr(bo); - trace_amdgpu_vm_set_ptes(pe, addr, count, incr, flags, p->direct); + trace_amdgpu_vm_set_ptes(pe, addr, count, incr, flags, p->immediate); for (i = 0; i < count; i++) { value = p->pages_addr ? diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c index 2a7a6f62d627..e06412f5066e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c @@ -61,10 +61,12 @@ static int amdgpu_vm_sdma_prepare(struct amdgpu_vm_update_params *p, struct dma_resv *resv, enum amdgpu_sync_mode sync_mode) { + enum amdgpu_ib_pool_type pool = p->immediate ? AMDGPU_IB_POOL_IMMEDIATE + : AMDGPU_IB_POOL_DELAYED; unsigned int ndw = AMDGPU_VM_SDMA_MIN_NUM_DW; int r; - r = amdgpu_job_alloc_with_ib(p->adev, ndw * 4, &p->job); + r = amdgpu_job_alloc_with_ib(p->adev, ndw * 4, pool, &p->job); if (r) return r; @@ -90,11 +92,11 @@ static int amdgpu_vm_sdma_commit(struct amdgpu_vm_update_params *p, { struct amdgpu_ib *ib = p->job->ibs; struct drm_sched_entity *entity; - struct dma_fence *f, *tmp; struct amdgpu_ring *ring; + struct dma_fence *f; int r; - entity = p->direct ? &p->vm->direct : &p->vm->delayed; + entity = p->immediate ? &p->vm->immediate : &p->vm->delayed; ring = container_of(entity->rq->sched, struct amdgpu_ring, sched); WARN_ON(ib->length_dw == 0); @@ -104,15 +106,16 @@ static int amdgpu_vm_sdma_commit(struct amdgpu_vm_update_params *p, if (r) goto error; - if (p->direct) { - tmp = dma_fence_get(f); - swap(p->vm->last_direct, tmp); + if (p->unlocked) { + struct dma_fence *tmp = dma_fence_get(f); + + swap(p->vm->last_unlocked, f); dma_fence_put(tmp); } else { - dma_resv_add_shared_fence(p->vm->root.base.bo->tbo.base.resv, f); + amdgpu_bo_fence(p->vm->root.base.bo, f, true); } - if (fence && !p->direct) + if (fence && !p->immediate) swap(*fence, f); dma_fence_put(f); return 0; @@ -141,8 +144,8 @@ static void amdgpu_vm_sdma_copy_ptes(struct amdgpu_vm_update_params *p, src += p->num_dw_left * 4; - pe += amdgpu_bo_gpu_offset_no_check(bo); - trace_amdgpu_vm_copy_ptes(pe, src, count, p->direct); + pe += amdgpu_gmc_sign_extend(amdgpu_bo_gpu_offset_no_check(bo)); + trace_amdgpu_vm_copy_ptes(pe, src, count, p->immediate); amdgpu_vm_copy_pte(p->adev, ib, pe, src, count); } @@ -168,8 +171,8 @@ static void amdgpu_vm_sdma_set_ptes(struct amdgpu_vm_update_params *p, { struct amdgpu_ib *ib = p->job->ibs; - pe += amdgpu_bo_gpu_offset_no_check(bo); - trace_amdgpu_vm_set_ptes(pe, addr, count, incr, flags, p->direct); + pe += amdgpu_gmc_sign_extend(amdgpu_bo_gpu_offset_no_check(bo)); + trace_amdgpu_vm_set_ptes(pe, addr, count, incr, flags, p->immediate); if (count < 3) { amdgpu_vm_write_pte(p->adev, ib, pe, addr | flags, count, incr); @@ -198,6 +201,8 @@ static int amdgpu_vm_sdma_update(struct amdgpu_vm_update_params *p, uint64_t addr, unsigned count, uint32_t incr, uint64_t flags) { + enum amdgpu_ib_pool_type pool = p->immediate ? AMDGPU_IB_POOL_IMMEDIATE + : AMDGPU_IB_POOL_DELAYED; unsigned int i, ndw, nptes; uint64_t *pte; int r; @@ -223,7 +228,8 @@ static int amdgpu_vm_sdma_update(struct amdgpu_vm_update_params *p, ndw = max(ndw, AMDGPU_VM_SDMA_MIN_NUM_DW); ndw = min(ndw, AMDGPU_VM_SDMA_MAX_NUM_DW); - r = amdgpu_job_alloc_with_ib(p->adev, ndw * 4, &p->job); + r = amdgpu_job_alloc_with_ib(p->adev, ndw * 4, pool, + &p->job); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c index 128a667ed8fa..d399e5893170 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c @@ -149,6 +149,15 @@ static DEVICE_ATTR(mem_info_vis_vram_used, S_IRUGO, static DEVICE_ATTR(mem_info_vram_vendor, S_IRUGO, amdgpu_mem_info_vram_vendor, NULL); +static const struct attribute *amdgpu_vram_mgr_attributes[] = { + &dev_attr_mem_info_vram_total.attr, + &dev_attr_mem_info_vis_vram_total.attr, + &dev_attr_mem_info_vram_used.attr, + &dev_attr_mem_info_vis_vram_used.attr, + &dev_attr_mem_info_vram_vendor.attr, + NULL +}; + /** * amdgpu_vram_mgr_init - init VRAM manager and DRM MM * @@ -173,31 +182,9 @@ static int amdgpu_vram_mgr_init(struct ttm_mem_type_manager *man, man->priv = mgr; /* Add the two VRAM-related sysfs files */ - ret = device_create_file(adev->dev, &dev_attr_mem_info_vram_total); - if (ret) { - DRM_ERROR("Failed to create device file mem_info_vram_total\n"); - return ret; - } - ret = device_create_file(adev->dev, &dev_attr_mem_info_vis_vram_total); - if (ret) { - DRM_ERROR("Failed to create device file mem_info_vis_vram_total\n"); - return ret; - } - ret = device_create_file(adev->dev, &dev_attr_mem_info_vram_used); - if (ret) { - DRM_ERROR("Failed to create device file mem_info_vram_used\n"); - return ret; - } - ret = device_create_file(adev->dev, &dev_attr_mem_info_vis_vram_used); - if (ret) { - DRM_ERROR("Failed to create device file mem_info_vis_vram_used\n"); - return ret; - } - ret = device_create_file(adev->dev, &dev_attr_mem_info_vram_vendor); - if (ret) { - DRM_ERROR("Failed to create device file mem_info_vram_vendor\n"); - return ret; - } + ret = sysfs_create_files(&adev->dev->kobj, amdgpu_vram_mgr_attributes); + if (ret) + DRM_ERROR("Failed to register sysfs\n"); return 0; } @@ -220,11 +207,7 @@ static int amdgpu_vram_mgr_fini(struct ttm_mem_type_manager *man) spin_unlock(&mgr->lock); kfree(mgr); man->priv = NULL; - device_remove_file(adev->dev, &dev_attr_mem_info_vram_total); - device_remove_file(adev->dev, &dev_attr_mem_info_vis_vram_total); - device_remove_file(adev->dev, &dev_attr_mem_info_vram_used); - device_remove_file(adev->dev, &dev_attr_mem_info_vis_vram_used); - device_remove_file(adev->dev, &dev_attr_mem_info_vram_vendor); + sysfs_remove_files(&adev->dev->kobj, amdgpu_vram_mgr_attributes); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c index 95b3327168ac..91837a991319 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c @@ -325,9 +325,18 @@ success: static void amdgpu_xgmi_sysfs_rem_dev_info(struct amdgpu_device *adev, struct amdgpu_hive_info *hive) { + char node[10]; + memset(node, 0, sizeof(node)); + device_remove_file(adev->dev, &dev_attr_xgmi_device_id); - sysfs_remove_link(&adev->dev->kobj, adev->ddev->unique); - sysfs_remove_link(hive->kobj, adev->ddev->unique); + device_remove_file(adev->dev, &dev_attr_xgmi_error); + + if (adev != hive->adev) + sysfs_remove_link(&adev->dev->kobj,"xgmi_hive_info"); + + sprintf(node, "node%d", hive->number_devices); + sysfs_remove_link(hive->kobj, node); + } @@ -373,7 +382,13 @@ struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev, int lo if (lock) mutex_lock(&tmp->hive_lock); - tmp->pstate = -1; + tmp->pstate = AMDGPU_XGMI_PSTATE_UNKNOWN; + tmp->hi_req_gpu = NULL; + /* + * hive pstate on boot is high in vega20 so we have to go to low + * pstate on after boot. + */ + tmp->hi_req_count = AMDGPU_MAX_XGMI_DEVICE_PER_HIVE; mutex_unlock(&xgmi_mutex); return tmp; @@ -383,56 +398,59 @@ int amdgpu_xgmi_set_pstate(struct amdgpu_device *adev, int pstate) { int ret = 0; struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev, 0); - struct amdgpu_device *tmp_adev; - bool update_hive_pstate = true; - bool is_high_pstate = pstate && adev->asic_type == CHIP_VEGA20; + struct amdgpu_device *request_adev = hive->hi_req_gpu ? + hive->hi_req_gpu : adev; + bool is_hi_req = pstate == AMDGPU_XGMI_PSTATE_MAX_VEGA20; + bool init_low = hive->pstate == AMDGPU_XGMI_PSTATE_UNKNOWN; - if (!hive) + /* fw bug so temporarily disable pstate switching */ + return 0; + + if (!hive || adev->asic_type != CHIP_VEGA20) return 0; mutex_lock(&hive->hive_lock); - if (hive->pstate == pstate) { - adev->pstate = is_high_pstate ? pstate : adev->pstate; + if (is_hi_req) + hive->hi_req_count++; + else + hive->hi_req_count--; + + /* + * Vega20 only needs single peer to request pstate high for the hive to + * go high but all peers must request pstate low for the hive to go low + */ + if (hive->pstate == pstate || + (!is_hi_req && hive->hi_req_count && !init_low)) goto out; - } - dev_dbg(adev->dev, "Set xgmi pstate %d.\n", pstate); + dev_dbg(request_adev->dev, "Set xgmi pstate %d.\n", pstate); - ret = amdgpu_dpm_set_xgmi_pstate(adev, pstate); + ret = amdgpu_dpm_set_xgmi_pstate(request_adev, pstate); if (ret) { - dev_err(adev->dev, + dev_err(request_adev->dev, "XGMI: Set pstate failure on device %llx, hive %llx, ret %d", - adev->gmc.xgmi.node_id, - adev->gmc.xgmi.hive_id, ret); + request_adev->gmc.xgmi.node_id, + request_adev->gmc.xgmi.hive_id, ret); goto out; } - /* Update device pstate */ - adev->pstate = pstate; - - /* - * Update the hive pstate only all devices of the hive - * are in the same pstate - */ - list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) { - if (tmp_adev->pstate != adev->pstate) { - update_hive_pstate = false; - break; - } - } - if (update_hive_pstate || is_high_pstate) + if (init_low) + hive->pstate = hive->hi_req_count ? + hive->pstate : AMDGPU_XGMI_PSTATE_MIN; + else { hive->pstate = pstate; - + hive->hi_req_gpu = pstate != AMDGPU_XGMI_PSTATE_MIN ? + adev : NULL; + } out: mutex_unlock(&hive->hive_lock); - return ret; } int amdgpu_xgmi_update_topology(struct amdgpu_hive_info *hive, struct amdgpu_device *adev) { - int ret = -EINVAL; + int ret; /* Each psp need to set the latest topology */ ret = psp_xgmi_set_topology_info(&adev->psp, @@ -507,9 +525,6 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev) goto exit; } - /* Set default device pstate */ - adev->pstate = -1; - top_info = &adev->psp.xgmi_context.top_info; list_add_tail(&adev->gmc.xgmi.head, &hive->device_list); @@ -577,14 +592,14 @@ int amdgpu_xgmi_remove_device(struct amdgpu_device *adev) if (!hive) return -EINVAL; - if (!(hive->number_devices--)) { + task_barrier_rem_task(&hive->tb); + amdgpu_xgmi_sysfs_rem_dev_info(adev, hive); + mutex_unlock(&hive->hive_lock); + + if(!(--hive->number_devices)){ amdgpu_xgmi_sysfs_destroy(adev, hive); mutex_destroy(&hive->hive_lock); mutex_destroy(&hive->reset_lock); - } else { - task_barrier_rem_task(&hive->tb); - amdgpu_xgmi_sysfs_rem_dev_info(adev, hive); - mutex_unlock(&hive->hive_lock); } return psp_xgmi_terminate(&adev->psp); @@ -604,6 +619,8 @@ int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev) adev->gmc.xgmi.num_physical_nodes == 0) return 0; + amdgpu_xgmi_reset_ras_error_count(adev); + if (!adev->gmc.xgmi.ras_if) { adev->gmc.xgmi.ras_if = kmalloc(sizeof(struct ras_common_if), GFP_KERNEL); if (!adev->gmc.xgmi.ras_if) @@ -641,31 +658,34 @@ void amdgpu_xgmi_ras_fini(struct amdgpu_device *adev) uint64_t amdgpu_xgmi_get_relative_phy_addr(struct amdgpu_device *adev, uint64_t addr) { - uint32_t df_inst_id; - uint64_t dram_base_addr = 0; - const struct amdgpu_df_funcs *df_funcs = adev->df.funcs; - - if ((!df_funcs) || - (!df_funcs->get_df_inst_id) || - (!df_funcs->get_dram_base_addr)) { - dev_warn(adev->dev, - "XGMI: relative phy_addr algorithm is not supported\n"); - return addr; - } - - if (amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_DISALLOW)) { - dev_warn(adev->dev, - "failed to disable DF-Cstate, DF register may not be accessible\n"); - return addr; - } + struct amdgpu_xgmi *xgmi = &adev->gmc.xgmi; + return (addr + xgmi->physical_node_id * xgmi->node_segment_size); +} - df_inst_id = df_funcs->get_df_inst_id(adev); - dram_base_addr = df_funcs->get_dram_base_addr(adev, df_inst_id); +static void pcs_clear_status(struct amdgpu_device *adev, uint32_t pcs_status_reg) +{ + WREG32_PCIE(pcs_status_reg, 0xFFFFFFFF); + WREG32_PCIE(pcs_status_reg, 0); +} - if (amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_ALLOW)) - dev_warn(adev->dev, "failed to enable DF-Cstate\n"); +void amdgpu_xgmi_reset_ras_error_count(struct amdgpu_device *adev) +{ + uint32_t i; - return addr + dram_base_addr; + switch (adev->asic_type) { + case CHIP_ARCTURUS: + for (i = 0; i < ARRAY_SIZE(xgmi_pcs_err_status_reg_arct); i++) + pcs_clear_status(adev, + xgmi_pcs_err_status_reg_arct[i]); + break; + case CHIP_VEGA20: + for (i = 0; i < ARRAY_SIZE(xgmi_pcs_err_status_reg_vg20); i++) + pcs_clear_status(adev, + xgmi_pcs_err_status_reg_vg20[i]); + break; + default: + break; + } } static int amdgpu_xgmi_query_pcs_error_status(struct amdgpu_device *adev, @@ -758,6 +778,8 @@ int amdgpu_xgmi_query_ras_error_count(struct amdgpu_device *adev, break; } + amdgpu_xgmi_reset_ras_error_count(adev); + err_data->ue_count += ue_cnt; err_data->ce_count += ce_cnt; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h index 4a92067fe595..6999eab16a72 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h @@ -25,6 +25,7 @@ #include <drm/task_barrier.h> #include "amdgpu_psp.h" + struct amdgpu_hive_info { uint64_t hive_id; struct list_head device_list; @@ -33,8 +34,14 @@ struct amdgpu_hive_info { struct kobject *kobj; struct device_attribute dev_attr; struct amdgpu_device *adev; - int pstate; /*0 -- low , 1 -- high , -1 unknown*/ + int hi_req_count; + struct amdgpu_device *hi_req_gpu; struct task_barrier tb; + enum { + AMDGPU_XGMI_PSTATE_MIN, + AMDGPU_XGMI_PSTATE_MAX_VEGA20, + AMDGPU_XGMI_PSTATE_UNKNOWN + } pstate; }; struct amdgpu_pcs_ras_field { @@ -56,6 +63,7 @@ uint64_t amdgpu_xgmi_get_relative_phy_addr(struct amdgpu_device *adev, uint64_t addr); int amdgpu_xgmi_query_ras_error_count(struct amdgpu_device *adev, void *ras_error_status); +void amdgpu_xgmi_reset_ras_error_count(struct amdgpu_device *adev); static inline bool amdgpu_xgmi_same_hive(struct amdgpu_device *adev, struct amdgpu_device *bo_adev) diff --git a/drivers/gpu/drm/amd/amdgpu/atom.c b/drivers/gpu/drm/amd/amdgpu/atom.c index cae426c7c086..4cfc786699c7 100644 --- a/drivers/gpu/drm/amd/amdgpu/atom.c +++ b/drivers/gpu/drm/amd/amdgpu/atom.c @@ -54,6 +54,8 @@ #define PLL_INDEX 2 #define PLL_DATA 3 +#define ATOM_CMD_TIMEOUT_SEC 20 + typedef struct { struct atom_context *ctx; uint32_t *ps, *ws; @@ -744,8 +746,9 @@ static void atom_op_jump(atom_exec_context *ctx, int *ptr, int arg) cjiffies = jiffies; if (time_after(cjiffies, ctx->last_jump_jiffies)) { cjiffies -= ctx->last_jump_jiffies; - if ((jiffies_to_msecs(cjiffies) > 10000)) { - DRM_ERROR("atombios stuck in loop for more than 10secs aborting\n"); + if ((jiffies_to_msecs(cjiffies) > ATOM_CMD_TIMEOUT_SEC*1000)) { + DRM_ERROR("atombios stuck in loop for more than %dsecs aborting\n", + ATOM_CMD_TIMEOUT_SEC); ctx->abort = true; } } else { diff --git a/drivers/gpu/drm/amd/amdgpu/cik.c b/drivers/gpu/drm/amd/amdgpu/cik.c index 006f21ef7ddf..fe306d0f73f7 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik.c +++ b/drivers/gpu/drm/amd/amdgpu/cik.c @@ -1358,8 +1358,6 @@ static int cik_asic_reset(struct amdgpu_device *adev) int r; if (cik_asic_reset_method(adev) == AMD_RESET_METHOD_BACO) { - if (!adev->in_suspend) - amdgpu_inc_vram_lost(adev); r = amdgpu_dpm_baco_reset(adev); } else { r = cik_asic_pci_config_reset(adev); @@ -1811,12 +1809,6 @@ static uint32_t cik_get_rev_id(struct amdgpu_device *adev) >> CC_DRM_ID_STRAPS__ATI_REV_ID__SHIFT; } -static void cik_detect_hw_virtualization(struct amdgpu_device *adev) -{ - if (is_virtual_machine()) /* passthrough mode */ - adev->virt.caps |= AMDGPU_PASSTHROUGH_MODE; -} - static void cik_flush_hdp(struct amdgpu_device *adev, struct amdgpu_ring *ring) { if (!ring || !ring->funcs->emit_wreg) { @@ -2179,8 +2171,6 @@ static const struct amdgpu_ip_block_version cik_common_ip_block = int cik_set_ip_blocks(struct amdgpu_device *adev) { - cik_detect_hw_virtualization(adev); - switch (adev->asic_type) { case CHIP_BONAIRE: amdgpu_device_ip_block_add(adev, &cik_common_ip_block); diff --git a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c index 580d3f93d670..20f108818b2b 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c @@ -320,8 +320,6 @@ static void cik_sdma_gfx_stop(struct amdgpu_device *adev) WREG32(mmSDMA0_GFX_RB_CNTL + sdma_offsets[i], rb_cntl); WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], 0); } - sdma0->sched.ready = false; - sdma1->sched.ready = false; } /** @@ -679,7 +677,8 @@ static int cik_sdma_ring_test_ib(struct amdgpu_ring *ring, long timeout) tmp = 0xCAFEDEAD; adev->wb.wb[index] = cpu_to_le32(tmp); memset(&ib, 0, sizeof(ib)); - r = amdgpu_ib_get(adev, NULL, 256, &ib); + r = amdgpu_ib_get(adev, NULL, 256, + AMDGPU_IB_POOL_DIRECT, &ib); if (r) goto err0; @@ -980,7 +979,8 @@ static int cik_sdma_sw_init(void *handle) &adev->sdma.trap_irq, (i == 0) ? AMDGPU_SDMA_IRQ_INSTANCE0 : - AMDGPU_SDMA_IRQ_INSTANCE1); + AMDGPU_SDMA_IRQ_INSTANCE1, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } @@ -1313,7 +1313,8 @@ static void cik_sdma_set_irq_funcs(struct amdgpu_device *adev) static void cik_sdma_emit_copy_buffer(struct amdgpu_ib *ib, uint64_t src_offset, uint64_t dst_offset, - uint32_t byte_count) + uint32_t byte_count, + bool tmz) { ib->ptr[ib->length_dw++] = SDMA_PACKET(SDMA_OPCODE_COPY, SDMA_COPY_SUB_OPCODE_LINEAR, 0); ib->ptr[ib->length_dw++] = byte_count; diff --git a/drivers/gpu/drm/amd/amdgpu/cikd.h b/drivers/gpu/drm/amd/amdgpu/cikd.h index cee6e8a3ad9c..5f3f6ebfb387 100644 --- a/drivers/gpu/drm/amd/amdgpu/cikd.h +++ b/drivers/gpu/drm/amd/amdgpu/cikd.h @@ -450,7 +450,7 @@ # define PACKET3_DMA_DATA_CMD_SAIC (1 << 28) # define PACKET3_DMA_DATA_CMD_DAIC (1 << 29) # define PACKET3_DMA_DATA_CMD_RAW_WAIT (1 << 30) -#define PACKET3_AQUIRE_MEM 0x58 +#define PACKET3_ACQUIRE_MEM 0x58 #define PACKET3_REWIND 0x59 #define PACKET3_LOAD_UCONFIG_REG 0x5E #define PACKET3_LOAD_SH_REG 0x5F diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c index 47a6e3d0aa96..84b45a019a36 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c @@ -2303,9 +2303,9 @@ static void dce_v10_0_hide_cursor(struct drm_crtc *crtc) struct amdgpu_device *adev = crtc->dev->dev_private; u32 tmp; - tmp = RREG32_IDX(mmCUR_CONTROL + amdgpu_crtc->crtc_offset); + tmp = RREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset); tmp = REG_SET_FIELD(tmp, CUR_CONTROL, CURSOR_EN, 0); - WREG32_IDX(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, tmp); + WREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, tmp); } static void dce_v10_0_show_cursor(struct drm_crtc *crtc) @@ -2319,10 +2319,10 @@ static void dce_v10_0_show_cursor(struct drm_crtc *crtc) WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset, lower_32_bits(amdgpu_crtc->cursor_addr)); - tmp = RREG32_IDX(mmCUR_CONTROL + amdgpu_crtc->crtc_offset); + tmp = RREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset); tmp = REG_SET_FIELD(tmp, CUR_CONTROL, CURSOR_EN, 1); tmp = REG_SET_FIELD(tmp, CUR_CONTROL, CURSOR_MODE, 2); - WREG32_IDX(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, tmp); + WREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, tmp); } static int dce_v10_0_cursor_move_locked(struct drm_crtc *crtc, diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c index 7d16fb5a9482..ec61532e2f83 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c @@ -2382,9 +2382,9 @@ static void dce_v11_0_hide_cursor(struct drm_crtc *crtc) struct amdgpu_device *adev = crtc->dev->dev_private; u32 tmp; - tmp = RREG32_IDX(mmCUR_CONTROL + amdgpu_crtc->crtc_offset); + tmp = RREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset); tmp = REG_SET_FIELD(tmp, CUR_CONTROL, CURSOR_EN, 0); - WREG32_IDX(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, tmp); + WREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, tmp); } static void dce_v11_0_show_cursor(struct drm_crtc *crtc) @@ -2398,10 +2398,10 @@ static void dce_v11_0_show_cursor(struct drm_crtc *crtc) WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset, lower_32_bits(amdgpu_crtc->cursor_addr)); - tmp = RREG32_IDX(mmCUR_CONTROL + amdgpu_crtc->crtc_offset); + tmp = RREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset); tmp = REG_SET_FIELD(tmp, CUR_CONTROL, CURSOR_EN, 1); tmp = REG_SET_FIELD(tmp, CUR_CONTROL, CURSOR_MODE, 2); - WREG32_IDX(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, tmp); + WREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, tmp); } static int dce_v11_0_cursor_move_locked(struct drm_crtc *crtc, diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c index 1ab33aa3980e..cbddead3dafb 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c @@ -2194,9 +2194,9 @@ static void dce_v6_0_hide_cursor(struct drm_crtc *crtc) struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc); struct amdgpu_device *adev = crtc->dev->dev_private; - WREG32_IDX(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, - (CURSOR_24_8_PRE_MULT << CUR_CONTROL__CURSOR_MODE__SHIFT) | - (CURSOR_URGENT_1_2 << CUR_CONTROL__CURSOR_URGENT_CONTROL__SHIFT)); + WREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, + (CURSOR_24_8_PRE_MULT << CUR_CONTROL__CURSOR_MODE__SHIFT) | + (CURSOR_URGENT_1_2 << CUR_CONTROL__CURSOR_URGENT_CONTROL__SHIFT)); } @@ -2211,10 +2211,10 @@ static void dce_v6_0_show_cursor(struct drm_crtc *crtc) WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset, lower_32_bits(amdgpu_crtc->cursor_addr)); - WREG32_IDX(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, - CUR_CONTROL__CURSOR_EN_MASK | - (CURSOR_24_8_PRE_MULT << CUR_CONTROL__CURSOR_MODE__SHIFT) | - (CURSOR_URGENT_1_2 << CUR_CONTROL__CURSOR_URGENT_CONTROL__SHIFT)); + WREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, + CUR_CONTROL__CURSOR_EN_MASK | + (CURSOR_24_8_PRE_MULT << CUR_CONTROL__CURSOR_MODE__SHIFT) | + (CURSOR_URGENT_1_2 << CUR_CONTROL__CURSOR_URGENT_CONTROL__SHIFT)); } diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c index 8a8f1aa49212..fa0ad50b628c 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c @@ -2205,9 +2205,9 @@ static void dce_v8_0_hide_cursor(struct drm_crtc *crtc) struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc); struct amdgpu_device *adev = crtc->dev->dev_private; - WREG32_IDX(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, - (CURSOR_24_8_PRE_MULT << CUR_CONTROL__CURSOR_MODE__SHIFT) | - (CURSOR_URGENT_1_2 << CUR_CONTROL__CURSOR_URGENT_CONTROL__SHIFT)); + WREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, + (CURSOR_24_8_PRE_MULT << CUR_CONTROL__CURSOR_MODE__SHIFT) | + (CURSOR_URGENT_1_2 << CUR_CONTROL__CURSOR_URGENT_CONTROL__SHIFT)); } static void dce_v8_0_show_cursor(struct drm_crtc *crtc) @@ -2220,10 +2220,10 @@ static void dce_v8_0_show_cursor(struct drm_crtc *crtc) WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset, lower_32_bits(amdgpu_crtc->cursor_addr)); - WREG32_IDX(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, - CUR_CONTROL__CURSOR_EN_MASK | - (CURSOR_24_8_PRE_MULT << CUR_CONTROL__CURSOR_MODE__SHIFT) | - (CURSOR_URGENT_1_2 << CUR_CONTROL__CURSOR_URGENT_CONTROL__SHIFT)); + WREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, + CUR_CONTROL__CURSOR_EN_MASK | + (CURSOR_24_8_PRE_MULT << CUR_CONTROL__CURSOR_MODE__SHIFT) | + (CURSOR_URGENT_1_2 << CUR_CONTROL__CURSOR_URGENT_CONTROL__SHIFT)); } static int dce_v8_0_cursor_move_locked(struct drm_crtc *crtc, diff --git a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c index 13e12be667fc..d5ff7b6331ff 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c @@ -172,8 +172,9 @@ static void dce_virtual_crtc_disable(struct drm_crtc *crtc) { struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc); - dce_virtual_crtc_dpms(crtc, DRM_MODE_DPMS_OFF); + drm_crtc_vblank_off(crtc); + amdgpu_crtc->enabled = false; amdgpu_crtc->pll_id = ATOM_PPLL_INVALID; amdgpu_crtc->encoder = NULL; amdgpu_crtc->connector = NULL; @@ -286,7 +287,7 @@ static int dce_virtual_get_modes(struct drm_connector *connector) static const struct mode_size { int w; int h; - } common_modes[17] = { + } common_modes[21] = { { 640, 480}, { 720, 480}, { 800, 600}, @@ -303,10 +304,14 @@ static int dce_virtual_get_modes(struct drm_connector *connector) {1680, 1050}, {1600, 1200}, {1920, 1080}, - {1920, 1200} + {1920, 1200}, + {4096, 3112}, + {3656, 2664}, + {3840, 2160}, + {4096, 2160}, }; - for (i = 0; i < 17; i++) { + for (i = 0; i < 21; i++) { mode = drm_cvt_mode(dev, common_modes[i].w, common_modes[i].h, 60, false, false, false); drm_mode_probed_add(connector, mode); } diff --git a/drivers/gpu/drm/amd/amdgpu/df_v3_6.c b/drivers/gpu/drm/amd/amdgpu/df_v3_6.c index 5a1bd8ed1a6c..a7b8292cefee 100644 --- a/drivers/gpu/drm/amd/amdgpu/df_v3_6.c +++ b/drivers/gpu/drm/amd/amdgpu/df_v3_6.c @@ -686,58 +686,6 @@ static void df_v3_6_pmc_get_count(struct amdgpu_device *adev, } } -static uint64_t df_v3_6_get_dram_base_addr(struct amdgpu_device *adev, - uint32_t df_inst) -{ - uint32_t base_addr_reg_val = 0; - uint64_t base_addr = 0; - - base_addr_reg_val = RREG32_PCIE(smnDF_CS_UMC_AON0_DramBaseAddress0 + - df_inst * DF_3_6_SMN_REG_INST_DIST); - - if (REG_GET_FIELD(base_addr_reg_val, - DF_CS_UMC_AON0_DramBaseAddress0, - AddrRngVal) == 0) { - DRM_WARN("address range not valid"); - return 0; - } - - base_addr = REG_GET_FIELD(base_addr_reg_val, - DF_CS_UMC_AON0_DramBaseAddress0, - DramBaseAddr); - - return base_addr << 28; -} - -static uint32_t df_v3_6_get_df_inst_id(struct amdgpu_device *adev) -{ - uint32_t xgmi_node_id = 0; - uint32_t df_inst_id = 0; - - /* Walk through DF dst nodes to find current XGMI node */ - for (df_inst_id = 0; df_inst_id < DF_3_6_INST_CNT; df_inst_id++) { - - xgmi_node_id = RREG32_PCIE(smnDF_CS_UMC_AON0_DramLimitAddress0 + - df_inst_id * DF_3_6_SMN_REG_INST_DIST); - xgmi_node_id = REG_GET_FIELD(xgmi_node_id, - DF_CS_UMC_AON0_DramLimitAddress0, - DstFabricID); - - /* TODO: establish reason dest fabric id is offset by 7 */ - xgmi_node_id = xgmi_node_id >> 7; - - if (adev->gmc.xgmi.physical_node_id == xgmi_node_id) - break; - } - - if (df_inst_id == DF_3_6_INST_CNT) { - DRM_WARN("cant match df dst id with gpu node"); - return 0; - } - - return df_inst_id; -} - const struct amdgpu_df_funcs df_v3_6_funcs = { .sw_init = df_v3_6_sw_init, .sw_fini = df_v3_6_sw_fini, @@ -752,6 +700,4 @@ const struct amdgpu_df_funcs df_v3_6_funcs = { .pmc_get_count = df_v3_6_pmc_get_count, .get_fica = df_v3_6_get_fica, .set_fica = df_v3_6_set_fica, - .get_dram_base_addr = df_v3_6_get_dram_base_addr, - .get_df_inst_id = df_v3_6_get_df_inst_id }; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index d78059fd2c72..bd5dd4f64311 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -138,6 +138,1062 @@ static const struct soc15_reg_golden golden_settings_gc_10_0_nv10[] = /* Pending on emulation bring up */ }; +static const struct soc15_reg_golden golden_settings_gc_rlc_spm_10_0_nv10[] = +{ + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xe0000000, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x28), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x9), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1b), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1b), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x20), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xf), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xcc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xf), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xd0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xf), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xd4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xf), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x24), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xf), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x24), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xf), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x11), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xf), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x11), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x88), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x8c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xbc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x90), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x94), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x98), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x9c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xac), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x9), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x38), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x3c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x40), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x44), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x48), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x5), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x4c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x9), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x70), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x9), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x74), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x9), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x78), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x7c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x80), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xf), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x84), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x50), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x54), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x5), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x58), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x9), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x5c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x60), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x64), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x9), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x68), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x9), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x6c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x1c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x3), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x20), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x20), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x24), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x24), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x28), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x28), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x2c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1b), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x2c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1b), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x30), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x30), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x34), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x34), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x38), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x38), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x3c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x3c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x50), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x50), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x54), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x54), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x58), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x58), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x5c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x5c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1d), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1d), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x48), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x48), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x4c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x4c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x40), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x40), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x44), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1b), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x44), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1b), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x17), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x17), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x60), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x60), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x64), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xf), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x64), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xf), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x70), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x70), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x74), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x74), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x68), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x68), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x6c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x6c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x78), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x78), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x7c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x7c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x88), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x88), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x8c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x8c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x80), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x80), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x84), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x84), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x90), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x90), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x94), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x94), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x3), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x3), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x98), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x98), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x9c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x9c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xac), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xac), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x13), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x13), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xbc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xbc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xd0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xd0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xd4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xd4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xcc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xcc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xe8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xe8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xec), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xec), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xf0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xf0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xf4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xf4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xf8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xf8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xfc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xfc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x100), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x100), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x104), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x104), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xe0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x13), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xe0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x13), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x118), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x13), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x118), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x13), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x11c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x13), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x11c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x13), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x120), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x13), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x120), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x13), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x124), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x13), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x124), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x13), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xdc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1b), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xdc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1b), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x110), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x110), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x114), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x114), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x108), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x108), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x10c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x10c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xd8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x17), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xd8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x17), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x128), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xf), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x128), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xf), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x12c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x12c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x138), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x138), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x13c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x13c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x130), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x130), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x134), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x134), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x140), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x140), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x144), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x144), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x150), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x150), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x154), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x154), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x148), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x148), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x14c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x14c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x158), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x5), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x158), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x5), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x15c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x2), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x15c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x2), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x168), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x168), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x16c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x16c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x160), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x160), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x164), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x2), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x164), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x2), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x170), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x11), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x170), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x11), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x174), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x174), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x180), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xf), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x180), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xf), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x184), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x184), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x178), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x11), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x178), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x11), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x17c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x17c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x188), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x188), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x18c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x18c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x198), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x9), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x198), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x9), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x19c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x3), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x19c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x3), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x190), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x190), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x194), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x194), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x30), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x11), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x34), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1d), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1d), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1f), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1f), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x2c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLOBALS_SAMPLE_SKEW, 0x000000FF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLOBALS_MUXSEL_SKEW, 0x000000FF, 0x20), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLE_SKEW, 0x000000FF, 0x5), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLE_SKEW, 0x000000FF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_MUXSEL_SKEW, 0x000000FF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_MUXSEL_SKEW, 0x000000FF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_DESER_START_SKEW, 0x000000FF, 0x33), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xFFFFFFFF, 0xe0000000) +}; + static const struct soc15_reg_golden golden_settings_gc_10_1_1[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL_4, 0xffffffff, 0x003c0014), @@ -272,14 +1328,1694 @@ static const struct soc15_reg_golden golden_settings_gc_10_1_nv14[] = /* Pending on emulation bring up */ }; +static const struct soc15_reg_golden golden_settings_gc_rlc_spm_10_1_nv14[] = +{ + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xE0000000L, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x28), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x20), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xf), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x80), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x84), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x88), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x8c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x24), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x24), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x20), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x60), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x64), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x2), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x68), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x2), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x6c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x2), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x70), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x2), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x74), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x2), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x78), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x7c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x2), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x38), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x3c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x40), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x3), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x44), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x9), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x48), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x4c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x3), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x50), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x3), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x54), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x3), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x58), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x5c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x9), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x1c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x2), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x20), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x24), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x28), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x2c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x30), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x34), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x38), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x3c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x50), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1e), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x54), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1e), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x58), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1e), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x5c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1e), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x20), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x48), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x4c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x40), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1e), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x44), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1e), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x60), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x64), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x70), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x74), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x68), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x6c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x78), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x7c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x88), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x8c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x80), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x84), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x90), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x94), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x98), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x9c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xac), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xbc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xd0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xd4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xcc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xd8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xdc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xe8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xec), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xe0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xe4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x100), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x104), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x108), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x10c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x110), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x114), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x118), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x11c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xf8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x17), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x130), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x134), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x138), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x13c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xf4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x20), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x128), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x12c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x120), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x124), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xf0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x140), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x144), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x150), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x154), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x148), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x14c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x158), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x15c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x168), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x16c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x160), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x164), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x170), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x174), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x180), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x184), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x178), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x17c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x188), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x18c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x198), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x19c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x190), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x194), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x1a0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x1a4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x1b0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x1b4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x1a8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x1ac), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x1b8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x1bc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x1c8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x1cc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x1c0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x1c4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x30), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x34), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x20), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x20), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x2c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLOBALS_SAMPLE_SKEW, 0x000000FF, 0x26), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLOBALS_MUXSEL_SKEW, 0x000000FF, 0x28), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLE_SKEW, 0x000000FF, 0xf), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLE_SKEW, 0x000000FF, 0x15), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_MUXSEL_SKEW, 0x000000FF, 0x1f), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_MUXSEL_SKEW, 0x000000FF, 0x25), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_DESER_START_SKEW, 0x000000FF, 0x3b), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xFFFFFFFF, 0xe0000000) +}; + static const struct soc15_reg_golden golden_settings_gc_10_1_2_nv12[] = { /* Pending on emulation bring up */ }; +static const struct soc15_reg_golden golden_settings_gc_rlc_spm_10_1_2_nv12[] = +{ + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xe0000000L, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x28), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x5), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1b), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1b), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x20), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xcc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xd0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xd4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x24), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x24), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x11), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x11), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x88), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x8c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x2), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xbc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x90), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x2), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x94), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x98), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x5), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x9c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xac), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x5), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x9), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x38), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x3c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x40), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x44), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x48), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x4c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x70), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x74), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x78), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x7c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x80), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x84), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x50), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x54), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x3), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x58), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x5c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x60), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x64), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x68), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x6c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x1c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x3), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x11), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x20), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x20), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x24), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x24), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x28), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x28), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x2c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x2c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x30), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1d), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x30), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1d), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x34), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x34), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x38), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x38), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x3c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x3c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x18), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x50), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x50), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x54), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x54), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x58), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x58), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x5c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x5c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x48), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x48), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x4c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x4c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x40), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x40), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x44), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x44), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1a), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1b), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1b), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x60), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x15), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x60), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x15), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x64), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x13), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x64), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x13), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x70), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x70), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x74), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x74), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x68), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x15), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x68), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x15), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x6c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x13), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x6c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x13), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x78), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x78), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x7c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x7c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x88), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x88), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x8c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x8c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x80), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x80), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x84), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x84), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x90), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x9), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x90), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x9), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x94), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x94), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x98), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x9), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x98), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x9), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x9c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x9c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xa8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xac), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xac), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xbc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xbc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xb4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x11), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x11), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xd0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xd0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xd4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xd4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x11), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xc8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x11), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xcc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xcc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xe8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xe8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xec), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xec), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x16), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xf0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x15), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xf0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x15), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xf4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xf4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xf8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xf8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xfc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x17), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xfc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x17), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x100), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x13), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x100), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x13), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x104), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x11), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x104), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x11), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xe0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xe0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x118), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x15), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x118), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x15), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x11c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x15), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x11c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x15), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x120), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x15), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x120), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x15), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x124), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x15), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x124), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x15), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xdc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xdc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x110), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x15), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x110), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x15), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x114), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x114), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x14), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x108), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x108), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x10c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x10c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x19), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xd8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1b), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0xd8), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1b), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x128), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x128), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x12c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x12c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x138), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x138), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x13c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x13c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x130), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x130), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x12), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x134), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xf), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x134), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xf), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x140), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x140), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x144), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x144), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x150), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x150), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x154), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x154), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x148), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x148), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x14c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x14c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x7), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x158), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x5), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x158), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x5), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x15c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x15c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x168), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x168), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xa), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x16c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x9), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x16c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x9), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x160), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x5), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x160), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x5), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x164), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x164), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x170), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x170), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x174), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x174), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x180), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x180), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x184), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x184), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x178), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x178), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x17c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x17c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x188), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x188), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x18c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x5), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x18c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x5), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x198), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x198), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xc), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x19c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x19c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x190), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x190), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xe), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x194), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x194), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x30), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xd), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x34), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x11), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1d), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1d), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1f), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x4), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0x1f), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x2c), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_DATA, 0xFFFFFFFF, 0xb), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLOBALS_SAMPLE_SKEW, 0x000000FF, 0x1f), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLOBALS_MUXSEL_SKEW, 0x000000FF, 0x22), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLE_SKEW, 0x000000FF, 0x1), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_SAMPLE_SKEW, 0x000000FF, 0x6), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_MUXSEL_SKEW, 0x000000FF, 0x10), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x10000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_SE_MUXSEL_SKEW, 0x000000FF, 0x15), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_DESER_START_SKEW, 0x000000FF, 0x35), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xFFFFFFFF, 0xe0000000) +}; + #define DEFAULT_SH_MEM_CONFIG \ ((SH_MEM_ADDRESS_MODE_64 << SH_MEM_CONFIG__ADDRESS_MODE__SHIFT) | \ - (SH_MEM_ALIGNMENT_MODE_DWORD << SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT) | \ + (SH_MEM_ALIGNMENT_MODE_UNALIGNED << SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT) | \ (SH_MEM_RETRY_MODE_ALL << SH_MEM_CONFIG__RETRY_MODE__SHIFT) | \ (3 << SH_MEM_CONFIG__INITIAL_INST_PREFETCH__SHIFT)) @@ -301,7 +3037,7 @@ static int gfx_v10_0_rlc_backdoor_autoload_enable(struct amdgpu_device *adev); static int gfx_v10_0_wait_for_rlc_autoload_complete(struct amdgpu_device *adev); static void gfx_v10_0_ring_emit_ce_meta(struct amdgpu_ring *ring, bool resume); static void gfx_v10_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume); -static void gfx_v10_0_ring_emit_tmz(struct amdgpu_ring *ring, bool start); +static void gfx_v10_0_ring_emit_frame_cntl(struct amdgpu_ring *ring, bool start, bool secure); static void gfx10_kiq_set_resources(struct amdgpu_ring *kiq_ring, uint64_t queue_mask) { @@ -431,6 +3167,9 @@ static void gfx_v10_0_init_golden_registers(struct amdgpu_device *adev) soc15_program_register_sequence(adev, golden_settings_gc_10_0_nv10, (const u32)ARRAY_SIZE(golden_settings_gc_10_0_nv10)); + soc15_program_register_sequence(adev, + golden_settings_gc_rlc_spm_10_0_nv10, + (const u32)ARRAY_SIZE(golden_settings_gc_rlc_spm_10_0_nv10)); break; case CHIP_NAVI14: soc15_program_register_sequence(adev, @@ -439,6 +3178,9 @@ static void gfx_v10_0_init_golden_registers(struct amdgpu_device *adev) soc15_program_register_sequence(adev, golden_settings_gc_10_1_nv14, (const u32)ARRAY_SIZE(golden_settings_gc_10_1_nv14)); + soc15_program_register_sequence(adev, + golden_settings_gc_rlc_spm_10_1_nv14, + (const u32)ARRAY_SIZE(golden_settings_gc_rlc_spm_10_1_nv14)); break; case CHIP_NAVI12: soc15_program_register_sequence(adev, @@ -447,6 +3189,9 @@ static void gfx_v10_0_init_golden_registers(struct amdgpu_device *adev) soc15_program_register_sequence(adev, golden_settings_gc_10_1_2_nv12, (const u32)ARRAY_SIZE(golden_settings_gc_10_1_2_nv12)); + soc15_program_register_sequence(adev, + golden_settings_gc_rlc_spm_10_1_2_nv12, + (const u32)ARRAY_SIZE(golden_settings_gc_rlc_spm_10_1_2_nv12)); break; default: break; @@ -557,7 +3302,8 @@ static int gfx_v10_0_ring_test_ib(struct amdgpu_ring *ring, long timeout) gpu_addr = adev->wb.gpu_addr + (index * 4); adev->wb.wb[index] = cpu_to_le32(0xCAFEDEAD); memset(&ib, 0, sizeof(ib)); - r = amdgpu_ib_get(adev, NULL, 16, &ib); + r = amdgpu_ib_get(adev, NULL, 16, + AMDGPU_IB_POOL_DIRECT, &ib); if (r) goto err1; @@ -1298,7 +4044,8 @@ static int gfx_v10_0_gfx_ring_init(struct amdgpu_device *adev, int ring_id, irq_type = AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP + ring->pipe; r = amdgpu_ring_init(adev, ring, 1024, - &adev->gfx.eop_irq, irq_type); + &adev->gfx.eop_irq, irq_type, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; return 0; @@ -1309,7 +4056,8 @@ static int gfx_v10_0_compute_ring_init(struct amdgpu_device *adev, int ring_id, { int r; unsigned irq_type; - struct amdgpu_ring *ring = &adev->gfx.compute_ring[ring_id]; + struct amdgpu_ring *ring; + unsigned int hw_prio; ring = &adev->gfx.compute_ring[ring_id]; @@ -1328,10 +4076,11 @@ static int gfx_v10_0_compute_ring_init(struct amdgpu_device *adev, int ring_id, irq_type = AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP + ((ring->me - 1) * adev->gfx.mec.num_pipe_per_mec) + ring->pipe; - + hw_prio = amdgpu_gfx_is_high_priority_compute_queue(adev, ring->queue) ? + AMDGPU_GFX_PIPE_PRIO_HIGH : AMDGPU_GFX_PIPE_PRIO_NORMAL; /* type-2 packets are deprecated on MEC, use type-3 instead */ r = amdgpu_ring_init(adev, ring, 1024, - &adev->gfx.eop_irq, irq_type); + &adev->gfx.eop_irq, irq_type, hw_prio); if (r) return r; @@ -1829,9 +4578,9 @@ static int gfx_v10_0_init_csb(struct amdgpu_device *adev) /* csib */ WREG32_SOC15_RLC(GC, 0, mmRLC_CSIB_ADDR_HI, - adev->gfx.rlc.clear_state_gpu_addr >> 32); + adev->gfx.rlc.clear_state_gpu_addr >> 32); WREG32_SOC15_RLC(GC, 0, mmRLC_CSIB_ADDR_LO, - adev->gfx.rlc.clear_state_gpu_addr & 0xfffffffc); + adev->gfx.rlc.clear_state_gpu_addr & 0xfffffffc); WREG32_SOC15_RLC(GC, 0, mmRLC_CSIB_LENGTH, adev->gfx.rlc.clear_state_size); return 0; @@ -2441,10 +5190,6 @@ static int gfx_v10_0_cp_gfx_enable(struct amdgpu_device *adev, bool enable) tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, ME_HALT, enable ? 0 : 1); tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, PFP_HALT, enable ? 0 : 1); tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, CE_HALT, enable ? 0 : 1); - if (!enable) { - for (i = 0; i < adev->gfx.num_gfx_rings; i++) - adev->gfx.gfx_ring[i].sched.ready = false; - } WREG32_SOC15_RLC(GC, 0, mmCP_ME_CNTL, tmp); for (i = 0; i < adev->usec_timeout; i++) { @@ -2923,16 +5668,12 @@ static int gfx_v10_0_cp_gfx_resume(struct amdgpu_device *adev) static void gfx_v10_0_cp_compute_enable(struct amdgpu_device *adev, bool enable) { - int i; - if (enable) { WREG32_SOC15(GC, 0, mmCP_MEC_CNTL, 0); } else { WREG32_SOC15(GC, 0, mmCP_MEC_CNTL, (CP_MEC_CNTL__MEC_ME1_HALT_MASK | CP_MEC_CNTL__MEC_ME2_HALT_MASK)); - for (i = 0; i < adev->gfx.num_compute_rings; i++) - adev->gfx.compute_ring[i].sched.ready = false; adev->gfx.kiq.ring.sched.ready = false; } udelay(50); @@ -3268,11 +6009,8 @@ static void gfx_v10_0_compute_mqd_set_priority(struct amdgpu_ring *ring, struct if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) { if (amdgpu_gfx_is_high_priority_compute_queue(adev, ring->queue)) { mqd->cp_hqd_pipe_priority = AMDGPU_GFX_PIPE_PRIO_HIGH; - ring->has_high_prio = true; mqd->cp_hqd_queue_priority = AMDGPU_GFX_QUEUE_PRIORITY_MAXIMUM; - } else { - ring->has_high_prio = false; } } } @@ -3802,14 +6540,16 @@ static int gfx_v10_0_hw_init(void *handle) * loaded firstly, so in direct type, it has to load smc ucode * here before rlc. */ - r = smu_load_microcode(&adev->smu); - if (r) - return r; + if (adev->smu.ppt_funcs != NULL) { + r = smu_load_microcode(&adev->smu); + if (r) + return r; - r = smu_check_fw_status(&adev->smu); - if (r) { - pr_err("SMC firmware status is not correct\n"); - return r; + r = smu_check_fw_status(&adev->smu); + if (r) { + pr_err("SMC firmware status is not correct\n"); + return r; + } } } @@ -4273,7 +7013,7 @@ static int gfx_v10_0_update_gfx_clock_gating(struct amdgpu_device *adev, /* === CGCG /CGLS for GFX 3D Only === */ gfx_v10_0_update_3d_clock_gating(adev, enable); /* === MGCG + MGLS === */ - /* gfx_v10_0_update_medium_grain_clock_gating(adev, enable); */ + gfx_v10_0_update_medium_grain_clock_gating(adev, enable); } if (adev->cg_flags & @@ -4292,14 +7032,21 @@ static int gfx_v10_0_update_gfx_clock_gating(struct amdgpu_device *adev, static void gfx_v10_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) { - u32 data; + u32 reg, data; - data = RREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL); + reg = SOC15_REG_OFFSET(GC, 0, mmRLC_SPM_MC_CNTL); + if (amdgpu_sriov_is_pp_one_vf(adev)) + data = RREG32_NO_KIQ(reg); + else + data = RREG32(reg); data &= ~RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK; data |= (vmid & RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK) << RLC_SPM_MC_CNTL__RLC_SPM_VMID__SHIFT; - WREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL, data); + if (amdgpu_sriov_is_pp_one_vf(adev)) + WREG32_SOC15_NO_KIQ(GC, 0, mmRLC_SPM_MC_CNTL, data); + else + WREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL, data); } static bool gfx_v10_0_check_rlcg_range(struct amdgpu_device *adev, @@ -4341,6 +7088,20 @@ static const struct amdgpu_rlc_funcs gfx_v10_0_rlc_funcs = { .reset = gfx_v10_0_rlc_reset, .start = gfx_v10_0_rlc_start, .update_spm_vmid = gfx_v10_0_update_spm_vmid, +}; + +static const struct amdgpu_rlc_funcs gfx_v10_0_rlc_funcs_sriov = { + .is_rlc_enabled = gfx_v10_0_is_rlc_enabled, + .set_safe_mode = gfx_v10_0_set_safe_mode, + .unset_safe_mode = gfx_v10_0_unset_safe_mode, + .init = gfx_v10_0_rlc_init, + .get_csb_size = gfx_v10_0_get_csb_size, + .get_csb_buffer = gfx_v10_0_get_csb_buffer, + .resume = gfx_v10_0_rlc_resume, + .stop = gfx_v10_0_rlc_stop, + .reset = gfx_v10_0_rlc_reset, + .start = gfx_v10_0_rlc_start, + .update_spm_vmid = gfx_v10_0_update_spm_vmid, .rlcg_wreg = gfx_v10_rlcg_wreg, .is_rlcg_access_range = gfx_v10_0_is_rlcg_access_range, }; @@ -4350,14 +7111,14 @@ static int gfx_v10_0_set_powergating_state(void *handle, { struct amdgpu_device *adev = (struct amdgpu_device *)handle; bool enable = (state == AMD_PG_STATE_GATE); + + if (amdgpu_sriov_vf(adev)) + return 0; + switch (adev->asic_type) { case CHIP_NAVI10: case CHIP_NAVI14: - if (!enable) { - amdgpu_gfx_off_ctrl(adev, false); - cancel_delayed_work_sync(&adev->gfx.gfx_off_delay_work); - } else - amdgpu_gfx_off_ctrl(adev, true); + amdgpu_gfx_off_ctrl(adev, enable); break; default: break; @@ -4370,6 +7131,9 @@ static int gfx_v10_0_set_clockgating_state(void *handle, { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + if (amdgpu_sriov_vf(adev)) + return 0; + switch (adev->asic_type) { case CHIP_NAVI10: case CHIP_NAVI14: @@ -4682,7 +7446,8 @@ static void gfx_v10_0_ring_emit_sb(struct amdgpu_ring *ring) amdgpu_ring_write(ring, 0); } -static void gfx_v10_0_ring_emit_cntxcntl(struct amdgpu_ring *ring, uint32_t flags) +static void gfx_v10_0_ring_emit_cntxcntl(struct amdgpu_ring *ring, + uint32_t flags) { uint32_t dw2 = 0; @@ -4690,8 +7455,6 @@ static void gfx_v10_0_ring_emit_cntxcntl(struct amdgpu_ring *ring, uint32_t flag gfx_v10_0_ring_emit_ce_meta(ring, (!amdgpu_sriov_vf(ring->adev) && flags & AMDGPU_IB_PREEMPTED) ? true : false); - gfx_v10_0_ring_emit_tmz(ring, true); - dw2 |= 0x80000000; /* set load_enable otherwise this package is just NOPs */ if (flags & AMDGPU_HAVE_CTX_SWITCH) { /* set load_global_config & load_global_uconfig */ @@ -4848,16 +7611,19 @@ static void gfx_v10_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume) sizeof(de_payload) >> 2); } -static void gfx_v10_0_ring_emit_tmz(struct amdgpu_ring *ring, bool start) +static void gfx_v10_0_ring_emit_frame_cntl(struct amdgpu_ring *ring, bool start, + bool secure) { + uint32_t v = secure ? FRAME_TMZ : 0; + amdgpu_ring_write(ring, PACKET3(PACKET3_FRAME_CONTROL, 0)); - amdgpu_ring_write(ring, FRAME_CMD(start ? 0 : 1)); /* frame_end */ + amdgpu_ring_write(ring, v | FRAME_CMD(start ? 0 : 1)); } -static void gfx_v10_0_ring_emit_rreg(struct amdgpu_ring *ring, uint32_t reg) +static void gfx_v10_0_ring_emit_rreg(struct amdgpu_ring *ring, uint32_t reg, + uint32_t reg_val_offs) { struct amdgpu_device *adev = ring->adev; - struct amdgpu_kiq *kiq = &adev->gfx.kiq; amdgpu_ring_write(ring, PACKET3(PACKET3_COPY_DATA, 4)); amdgpu_ring_write(ring, 0 | /* src: register*/ @@ -4866,9 +7632,9 @@ static void gfx_v10_0_ring_emit_rreg(struct amdgpu_ring *ring, uint32_t reg) amdgpu_ring_write(ring, reg); amdgpu_ring_write(ring, 0); amdgpu_ring_write(ring, lower_32_bits(adev->wb.gpu_addr + - kiq->reg_val_offs * 4)); + reg_val_offs * 4)); amdgpu_ring_write(ring, upper_32_bits(adev->wb.gpu_addr + - kiq->reg_val_offs * 4)); + reg_val_offs * 4)); } static void gfx_v10_0_ring_emit_wreg(struct amdgpu_ring *ring, uint32_t reg, @@ -4918,6 +7684,19 @@ static void gfx_v10_0_ring_emit_reg_write_reg_wait(struct amdgpu_ring *ring, ref, mask); } +static void gfx_v10_0_ring_soft_recovery(struct amdgpu_ring *ring, + unsigned vmid) +{ + struct amdgpu_device *adev = ring->adev; + uint32_t value = 0; + + value = REG_SET_FIELD(value, SQ_CMD, CMD, 0x03); + value = REG_SET_FIELD(value, SQ_CMD, MODE, 0x01); + value = REG_SET_FIELD(value, SQ_CMD, CHECK_VMID, 1); + value = REG_SET_FIELD(value, SQ_CMD, VM_ID, vmid); + WREG32_SOC15(GC, 0, mmSQ_CMD, value); +} + static void gfx_v10_0_set_gfx_eop_interrupt_state(struct amdgpu_device *adev, uint32_t me, uint32_t pipe, @@ -5241,6 +8020,29 @@ static int gfx_v10_0_kiq_irq(struct amdgpu_device *adev, return 0; } +static void gfx_v10_0_emit_mem_sync(struct amdgpu_ring *ring) +{ + const unsigned int gcr_cntl = + PACKET3_ACQUIRE_MEM_GCR_CNTL_GL2_INV(1) | + PACKET3_ACQUIRE_MEM_GCR_CNTL_GL2_WB(1) | + PACKET3_ACQUIRE_MEM_GCR_CNTL_GLM_INV(1) | + PACKET3_ACQUIRE_MEM_GCR_CNTL_GLM_WB(1) | + PACKET3_ACQUIRE_MEM_GCR_CNTL_GL1_INV(1) | + PACKET3_ACQUIRE_MEM_GCR_CNTL_GLV_INV(1) | + PACKET3_ACQUIRE_MEM_GCR_CNTL_GLK_INV(1) | + PACKET3_ACQUIRE_MEM_GCR_CNTL_GLI_INV(1); + + /* ACQUIRE_MEM - make one or more surfaces valid for use by the subsequent operations */ + amdgpu_ring_write(ring, PACKET3(PACKET3_ACQUIRE_MEM, 6)); + amdgpu_ring_write(ring, 0); /* CP_COHER_CNTL */ + amdgpu_ring_write(ring, 0xffffffff); /* CP_COHER_SIZE */ + amdgpu_ring_write(ring, 0xffffff); /* CP_COHER_SIZE_HI */ + amdgpu_ring_write(ring, 0); /* CP_COHER_BASE */ + amdgpu_ring_write(ring, 0); /* CP_COHER_BASE_HI */ + amdgpu_ring_write(ring, 0x0000000A); /* POLL_INTERVAL */ + amdgpu_ring_write(ring, gcr_cntl); /* GCR_CNTL */ +} + static const struct amd_ip_funcs gfx_v10_0_ip_funcs = { .name = "gfx_v10_0", .early_init = gfx_v10_0_early_init, @@ -5288,7 +8090,8 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_gfx = { 3 + /* CNTX_CTRL */ 5 + /* HDP_INVL */ 8 + 8 + /* FENCE x2 */ - 2, /* SWITCH_BUFFER */ + 2 + /* SWITCH_BUFFER */ + 8, /* gfx_v10_0_emit_mem_sync */ .emit_ib_size = 4, /* gfx_v10_0_ring_emit_ib_gfx */ .emit_ib = gfx_v10_0_ring_emit_ib_gfx, .emit_fence = gfx_v10_0_ring_emit_fence, @@ -5305,10 +8108,12 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_gfx = { .init_cond_exec = gfx_v10_0_ring_emit_init_cond_exec, .patch_cond_exec = gfx_v10_0_ring_emit_patch_cond_exec, .preempt_ib = gfx_v10_0_ring_preempt_ib, - .emit_tmz = gfx_v10_0_ring_emit_tmz, + .emit_frame_cntl = gfx_v10_0_ring_emit_frame_cntl, .emit_wreg = gfx_v10_0_ring_emit_wreg, .emit_reg_wait = gfx_v10_0_ring_emit_reg_wait, .emit_reg_write_reg_wait = gfx_v10_0_ring_emit_reg_write_reg_wait, + .soft_recovery = gfx_v10_0_ring_soft_recovery, + .emit_mem_sync = gfx_v10_0_emit_mem_sync, }; static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_compute = { @@ -5328,7 +8133,8 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_compute = { SOC15_FLUSH_GPU_TLB_NUM_WREG * 5 + SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 7 + 2 + /* gfx_v10_0_ring_emit_vm_flush */ - 8 + 8 + 8, /* gfx_v10_0_ring_emit_fence x3 for user fence, vm fence */ + 8 + 8 + 8 + /* gfx_v10_0_ring_emit_fence x3 for user fence, vm fence */ + 8, /* gfx_v10_0_emit_mem_sync */ .emit_ib_size = 7, /* gfx_v10_0_ring_emit_ib_compute */ .emit_ib = gfx_v10_0_ring_emit_ib_compute, .emit_fence = gfx_v10_0_ring_emit_fence, @@ -5343,6 +8149,7 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_compute = { .emit_wreg = gfx_v10_0_ring_emit_wreg, .emit_reg_wait = gfx_v10_0_ring_emit_reg_wait, .emit_reg_write_reg_wait = gfx_v10_0_ring_emit_reg_write_reg_wait, + .emit_mem_sync = gfx_v10_0_emit_mem_sync, }; static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_kiq = { @@ -5429,9 +8236,11 @@ static void gfx_v10_0_set_rlc_funcs(struct amdgpu_device *adev) switch (adev->asic_type) { case CHIP_NAVI10: case CHIP_NAVI14: - case CHIP_NAVI12: adev->gfx.rlc.funcs = &gfx_v10_0_rlc_funcs; break; + case CHIP_NAVI12: + adev->gfx.rlc.funcs = &gfx_v10_0_rlc_funcs_sriov; + break; default: break; } diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c index 31f44d05e606..79c52c7a02e3 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c @@ -1914,7 +1914,8 @@ static int gfx_v6_0_ring_test_ib(struct amdgpu_ring *ring, long timeout) WREG32(scratch, 0xCAFEDEAD); memset(&ib, 0, sizeof(ib)); - r = amdgpu_ib_get(adev, NULL, 256, &ib); + r = amdgpu_ib_get(adev, NULL, 256, + AMDGPU_IB_POOL_DIRECT, &ib); if (r) goto err1; @@ -1950,7 +1951,6 @@ err1: static void gfx_v6_0_cp_gfx_enable(struct amdgpu_device *adev, bool enable) { - int i; if (enable) { WREG32(mmCP_ME_CNTL, 0); } else { @@ -1958,10 +1958,6 @@ static void gfx_v6_0_cp_gfx_enable(struct amdgpu_device *adev, bool enable) CP_ME_CNTL__PFP_HALT_MASK | CP_ME_CNTL__CE_HALT_MASK)); WREG32(mmSCRATCH_UMSK, 0); - for (i = 0; i < adev->gfx.num_gfx_rings; i++) - adev->gfx.gfx_ring[i].sched.ready = false; - for (i = 0; i < adev->gfx.num_compute_rings; i++) - adev->gfx.compute_ring[i].sched.ready = false; } udelay(50); } @@ -3114,7 +3110,9 @@ static int gfx_v6_0_sw_init(void *handle) ring->ring_obj = NULL; sprintf(ring->name, "gfx"); r = amdgpu_ring_init(adev, ring, 1024, - &adev->gfx.eop_irq, AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP); + &adev->gfx.eop_irq, + AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } @@ -3136,7 +3134,8 @@ static int gfx_v6_0_sw_init(void *handle) sprintf(ring->name, "comp_%d.%d.%d", ring->me, ring->pipe, ring->queue); irq_type = AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP + ring->pipe; r = amdgpu_ring_init(adev, ring, 1024, - &adev->gfx.eop_irq, irq_type); + &adev->gfx.eop_irq, irq_type, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } @@ -3466,6 +3465,18 @@ static int gfx_v6_0_set_powergating_state(void *handle, return 0; } +static void gfx_v6_0_emit_mem_sync(struct amdgpu_ring *ring) +{ + amdgpu_ring_write(ring, PACKET3(PACKET3_SURFACE_SYNC, 3)); + amdgpu_ring_write(ring, PACKET3_TCL1_ACTION_ENA | + PACKET3_TC_ACTION_ENA | + PACKET3_SH_KCACHE_ACTION_ENA | + PACKET3_SH_ICACHE_ACTION_ENA); /* CP_COHER_CNTL */ + amdgpu_ring_write(ring, 0xffffffff); /* CP_COHER_SIZE */ + amdgpu_ring_write(ring, 0); /* CP_COHER_BASE */ + amdgpu_ring_write(ring, 0x0000000A); /* poll interval */ +} + static const struct amd_ip_funcs gfx_v6_0_ip_funcs = { .name = "gfx_v6_0", .early_init = gfx_v6_0_early_init, @@ -3496,7 +3507,8 @@ static const struct amdgpu_ring_funcs gfx_v6_0_ring_funcs_gfx = { 14 + 14 + 14 + /* gfx_v6_0_ring_emit_fence x3 for user fence, vm fence */ 7 + 4 + /* gfx_v6_0_ring_emit_pipeline_sync */ SI_FLUSH_GPU_TLB_NUM_WREG * 5 + 7 + 6 + /* gfx_v6_0_ring_emit_vm_flush */ - 3 + 2, /* gfx_v6_ring_emit_cntxcntl including vgt flush */ + 3 + 2 + /* gfx_v6_ring_emit_cntxcntl including vgt flush */ + 5, /* SURFACE_SYNC */ .emit_ib_size = 6, /* gfx_v6_0_ring_emit_ib */ .emit_ib = gfx_v6_0_ring_emit_ib, .emit_fence = gfx_v6_0_ring_emit_fence, @@ -3507,6 +3519,7 @@ static const struct amdgpu_ring_funcs gfx_v6_0_ring_funcs_gfx = { .insert_nop = amdgpu_ring_insert_nop, .emit_cntxcntl = gfx_v6_ring_emit_cntxcntl, .emit_wreg = gfx_v6_0_ring_emit_wreg, + .emit_mem_sync = gfx_v6_0_emit_mem_sync, }; static const struct amdgpu_ring_funcs gfx_v6_0_ring_funcs_compute = { @@ -3520,7 +3533,8 @@ static const struct amdgpu_ring_funcs gfx_v6_0_ring_funcs_compute = { 5 + 5 + /* hdp flush / invalidate */ 7 + /* gfx_v6_0_ring_emit_pipeline_sync */ SI_FLUSH_GPU_TLB_NUM_WREG * 5 + 7 + /* gfx_v6_0_ring_emit_vm_flush */ - 14 + 14 + 14, /* gfx_v6_0_ring_emit_fence x3 for user fence, vm fence */ + 14 + 14 + 14 + /* gfx_v6_0_ring_emit_fence x3 for user fence, vm fence */ + 5, /* SURFACE_SYNC */ .emit_ib_size = 6, /* gfx_v6_0_ring_emit_ib */ .emit_ib = gfx_v6_0_ring_emit_ib, .emit_fence = gfx_v6_0_ring_emit_fence, @@ -3530,6 +3544,7 @@ static const struct amdgpu_ring_funcs gfx_v6_0_ring_funcs_compute = { .test_ib = gfx_v6_0_ring_test_ib, .insert_nop = amdgpu_ring_insert_nop, .emit_wreg = gfx_v6_0_ring_emit_wreg, + .emit_mem_sync = gfx_v6_0_emit_mem_sync, }; static void gfx_v6_0_set_ring_funcs(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index 733d398c61cc..0cc011f9190d 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -2364,7 +2364,8 @@ static int gfx_v7_0_ring_test_ib(struct amdgpu_ring *ring, long timeout) WREG32(scratch, 0xCAFEDEAD); memset(&ib, 0, sizeof(ib)); - r = amdgpu_ib_get(adev, NULL, 256, &ib); + r = amdgpu_ib_get(adev, NULL, 256, + AMDGPU_IB_POOL_DIRECT, &ib); if (r) goto err1; @@ -2431,15 +2432,12 @@ err1: */ static void gfx_v7_0_cp_gfx_enable(struct amdgpu_device *adev, bool enable) { - int i; - - if (enable) { + if (enable) WREG32(mmCP_ME_CNTL, 0); - } else { - WREG32(mmCP_ME_CNTL, (CP_ME_CNTL__ME_HALT_MASK | CP_ME_CNTL__PFP_HALT_MASK | CP_ME_CNTL__CE_HALT_MASK)); - for (i = 0; i < adev->gfx.num_gfx_rings; i++) - adev->gfx.gfx_ring[i].sched.ready = false; - } + else + WREG32(mmCP_ME_CNTL, (CP_ME_CNTL__ME_HALT_MASK | + CP_ME_CNTL__PFP_HALT_MASK | + CP_ME_CNTL__CE_HALT_MASK)); udelay(50); } @@ -2700,15 +2698,11 @@ static void gfx_v7_0_ring_set_wptr_compute(struct amdgpu_ring *ring) */ static void gfx_v7_0_cp_compute_enable(struct amdgpu_device *adev, bool enable) { - int i; - - if (enable) { + if (enable) WREG32(mmCP_MEC_CNTL, 0); - } else { - WREG32(mmCP_MEC_CNTL, (CP_MEC_CNTL__MEC_ME1_HALT_MASK | CP_MEC_CNTL__MEC_ME2_HALT_MASK)); - for (i = 0; i < adev->gfx.num_compute_rings; i++) - adev->gfx.compute_ring[i].sched.ready = false; - } + else + WREG32(mmCP_MEC_CNTL, (CP_MEC_CNTL__MEC_ME1_HALT_MASK | + CP_MEC_CNTL__MEC_ME2_HALT_MASK)); udelay(50); } @@ -4439,7 +4433,8 @@ static int gfx_v7_0_compute_ring_init(struct amdgpu_device *adev, int ring_id, /* type-2 packets are deprecated on MEC, use type-3 instead */ r = amdgpu_ring_init(adev, ring, 1024, - &adev->gfx.eop_irq, irq_type); + &adev->gfx.eop_irq, irq_type, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; @@ -4511,7 +4506,9 @@ static int gfx_v7_0_sw_init(void *handle) ring->ring_obj = NULL; sprintf(ring->name, "gfx"); r = amdgpu_ring_init(adev, ring, 1024, - &adev->gfx.eop_irq, AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP); + &adev->gfx.eop_irq, + AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } @@ -5001,6 +4998,32 @@ static int gfx_v7_0_set_powergating_state(void *handle, return 0; } +static void gfx_v7_0_emit_mem_sync(struct amdgpu_ring *ring) +{ + amdgpu_ring_write(ring, PACKET3(PACKET3_SURFACE_SYNC, 3)); + amdgpu_ring_write(ring, PACKET3_TCL1_ACTION_ENA | + PACKET3_TC_ACTION_ENA | + PACKET3_SH_KCACHE_ACTION_ENA | + PACKET3_SH_ICACHE_ACTION_ENA); /* CP_COHER_CNTL */ + amdgpu_ring_write(ring, 0xffffffff); /* CP_COHER_SIZE */ + amdgpu_ring_write(ring, 0); /* CP_COHER_BASE */ + amdgpu_ring_write(ring, 0x0000000A); /* poll interval */ +} + +static void gfx_v7_0_emit_mem_sync_compute(struct amdgpu_ring *ring) +{ + amdgpu_ring_write(ring, PACKET3(PACKET3_ACQUIRE_MEM, 5)); + amdgpu_ring_write(ring, PACKET3_TCL1_ACTION_ENA | + PACKET3_TC_ACTION_ENA | + PACKET3_SH_KCACHE_ACTION_ENA | + PACKET3_SH_ICACHE_ACTION_ENA); /* CP_COHER_CNTL */ + amdgpu_ring_write(ring, 0xffffffff); /* CP_COHER_SIZE */ + amdgpu_ring_write(ring, 0xff); /* CP_COHER_SIZE_HI */ + amdgpu_ring_write(ring, 0); /* CP_COHER_BASE */ + amdgpu_ring_write(ring, 0); /* CP_COHER_BASE_HI */ + amdgpu_ring_write(ring, 0x0000000A); /* poll interval */ +} + static const struct amd_ip_funcs gfx_v7_0_ip_funcs = { .name = "gfx_v7_0", .early_init = gfx_v7_0_early_init, @@ -5033,7 +5056,8 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_gfx = { 12 + 12 + 12 + /* gfx_v7_0_ring_emit_fence_gfx x3 for user fence, vm fence */ 7 + 4 + /* gfx_v7_0_ring_emit_pipeline_sync */ CIK_FLUSH_GPU_TLB_NUM_WREG * 5 + 7 + 6 + /* gfx_v7_0_ring_emit_vm_flush */ - 3 + 4, /* gfx_v7_ring_emit_cntxcntl including vgt flush*/ + 3 + 4 + /* gfx_v7_ring_emit_cntxcntl including vgt flush*/ + 5, /* SURFACE_SYNC */ .emit_ib_size = 4, /* gfx_v7_0_ring_emit_ib_gfx */ .emit_ib = gfx_v7_0_ring_emit_ib_gfx, .emit_fence = gfx_v7_0_ring_emit_fence_gfx, @@ -5048,6 +5072,7 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_gfx = { .emit_cntxcntl = gfx_v7_ring_emit_cntxcntl, .emit_wreg = gfx_v7_0_ring_emit_wreg, .soft_recovery = gfx_v7_0_ring_soft_recovery, + .emit_mem_sync = gfx_v7_0_emit_mem_sync, }; static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_compute = { @@ -5064,7 +5089,8 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_compute = { 5 + /* hdp invalidate */ 7 + /* gfx_v7_0_ring_emit_pipeline_sync */ CIK_FLUSH_GPU_TLB_NUM_WREG * 5 + 7 + /* gfx_v7_0_ring_emit_vm_flush */ - 7 + 7 + 7, /* gfx_v7_0_ring_emit_fence_compute x3 for user fence, vm fence */ + 7 + 7 + 7 + /* gfx_v7_0_ring_emit_fence_compute x3 for user fence, vm fence */ + 7, /* gfx_v7_0_emit_mem_sync_compute */ .emit_ib_size = 7, /* gfx_v7_0_ring_emit_ib_compute */ .emit_ib = gfx_v7_0_ring_emit_ib_compute, .emit_fence = gfx_v7_0_ring_emit_fence_compute, @@ -5077,6 +5103,7 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_compute = { .insert_nop = amdgpu_ring_insert_nop, .pad_ib = amdgpu_ring_generic_pad_ib, .emit_wreg = gfx_v7_0_ring_emit_wreg, + .emit_mem_sync = gfx_v7_0_emit_mem_sync_compute, }; static void gfx_v7_0_set_ring_funcs(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index fc32586ef80b..1d4128227ffd 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -888,7 +888,8 @@ static int gfx_v8_0_ring_test_ib(struct amdgpu_ring *ring, long timeout) gpu_addr = adev->wb.gpu_addr + (index * 4); adev->wb.wb[index] = cpu_to_le32(0xCAFEDEAD); memset(&ib, 0, sizeof(ib)); - r = amdgpu_ib_get(adev, NULL, 16, &ib); + r = amdgpu_ib_get(adev, NULL, 16, + AMDGPU_IB_POOL_DIRECT, &ib); if (r) goto err1; @@ -1550,7 +1551,8 @@ static int gfx_v8_0_do_edc_gpr_workarounds(struct amdgpu_device *adev) /* allocate an indirect buffer to put the commands in */ memset(&ib, 0, sizeof(ib)); - r = amdgpu_ib_get(adev, NULL, total_size, &ib); + r = amdgpu_ib_get(adev, NULL, total_size, + AMDGPU_IB_POOL_DIRECT, &ib); if (r) { DRM_ERROR("amdgpu: failed to get ib (%d).\n", r); return r; @@ -1892,6 +1894,7 @@ static int gfx_v8_0_compute_ring_init(struct amdgpu_device *adev, int ring_id, int r; unsigned irq_type; struct amdgpu_ring *ring = &adev->gfx.compute_ring[ring_id]; + unsigned int hw_prio; ring = &adev->gfx.compute_ring[ring_id]; @@ -1911,9 +1914,11 @@ static int gfx_v8_0_compute_ring_init(struct amdgpu_device *adev, int ring_id, + ((ring->me - 1) * adev->gfx.mec.num_pipe_per_mec) + ring->pipe; + hw_prio = amdgpu_gfx_is_high_priority_compute_queue(adev, ring->queue) ? + AMDGPU_GFX_PIPE_PRIO_HIGH : AMDGPU_RING_PRIO_DEFAULT; /* type-2 packets are deprecated on MEC, use type-3 instead */ r = amdgpu_ring_init(adev, ring, 1024, - &adev->gfx.eop_irq, irq_type); + &adev->gfx.eop_irq, irq_type, hw_prio); if (r) return r; @@ -2017,7 +2022,8 @@ static int gfx_v8_0_sw_init(void *handle) } r = amdgpu_ring_init(adev, ring, 1024, &adev->gfx.eop_irq, - AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP); + AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } @@ -4120,7 +4126,6 @@ static int gfx_v8_0_rlc_resume(struct amdgpu_device *adev) static void gfx_v8_0_cp_gfx_enable(struct amdgpu_device *adev, bool enable) { - int i; u32 tmp = RREG32(mmCP_ME_CNTL); if (enable) { @@ -4131,8 +4136,6 @@ static void gfx_v8_0_cp_gfx_enable(struct amdgpu_device *adev, bool enable) tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, ME_HALT, 1); tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, PFP_HALT, 1); tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, CE_HALT, 1); - for (i = 0; i < adev->gfx.num_gfx_rings; i++) - adev->gfx.gfx_ring[i].sched.ready = false; } WREG32(mmCP_ME_CNTL, tmp); udelay(50); @@ -4320,14 +4323,10 @@ static int gfx_v8_0_cp_gfx_resume(struct amdgpu_device *adev) static void gfx_v8_0_cp_compute_enable(struct amdgpu_device *adev, bool enable) { - int i; - if (enable) { WREG32(mmCP_MEC_CNTL, 0); } else { WREG32(mmCP_MEC_CNTL, (CP_MEC_CNTL__MEC_ME1_HALT_MASK | CP_MEC_CNTL__MEC_ME2_HALT_MASK)); - for (i = 0; i < adev->gfx.num_compute_rings; i++) - adev->gfx.compute_ring[i].sched.ready = false; adev->gfx.kiq.ring.sched.ready = false; } udelay(50); @@ -4437,11 +4436,8 @@ static void gfx_v8_0_mqd_set_priority(struct amdgpu_ring *ring, struct vi_mqd *m if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) { if (amdgpu_gfx_is_high_priority_compute_queue(adev, ring->queue)) { mqd->cp_hqd_pipe_priority = AMDGPU_GFX_PIPE_PRIO_HIGH; - ring->has_high_prio = true; mqd->cp_hqd_queue_priority = AMDGPU_GFX_QUEUE_PRIORITY_MAXIMUM; - } else { - ring->has_high_prio = false; } } } @@ -5619,12 +5615,18 @@ static void gfx_v8_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) { u32 data; - data = RREG32(mmRLC_SPM_VMID); + if (amdgpu_sriov_is_pp_one_vf(adev)) + data = RREG32_NO_KIQ(mmRLC_SPM_VMID); + else + data = RREG32(mmRLC_SPM_VMID); data &= ~RLC_SPM_VMID__RLC_SPM_VMID_MASK; data |= (vmid & RLC_SPM_VMID__RLC_SPM_VMID_MASK) << RLC_SPM_VMID__RLC_SPM_VMID__SHIFT; - WREG32(mmRLC_SPM_VMID, data); + if (amdgpu_sriov_is_pp_one_vf(adev)) + WREG32_NO_KIQ(mmRLC_SPM_VMID, data); + else + WREG32(mmRLC_SPM_VMID, data); } static const struct amdgpu_rlc_funcs iceland_rlc_funcs = { @@ -6387,10 +6389,10 @@ static void gfx_v8_0_ring_emit_patch_cond_exec(struct amdgpu_ring *ring, unsigne ring->ring[offset] = (ring->ring_size >> 2) - offset + cur; } -static void gfx_v8_0_ring_emit_rreg(struct amdgpu_ring *ring, uint32_t reg) +static void gfx_v8_0_ring_emit_rreg(struct amdgpu_ring *ring, uint32_t reg, + uint32_t reg_val_offs) { struct amdgpu_device *adev = ring->adev; - struct amdgpu_kiq *kiq = &adev->gfx.kiq; amdgpu_ring_write(ring, PACKET3(PACKET3_COPY_DATA, 4)); amdgpu_ring_write(ring, 0 | /* src: register*/ @@ -6399,9 +6401,9 @@ static void gfx_v8_0_ring_emit_rreg(struct amdgpu_ring *ring, uint32_t reg) amdgpu_ring_write(ring, reg); amdgpu_ring_write(ring, 0); amdgpu_ring_write(ring, lower_32_bits(adev->wb.gpu_addr + - kiq->reg_val_offs * 4)); + reg_val_offs * 4)); amdgpu_ring_write(ring, upper_32_bits(adev->wb.gpu_addr + - kiq->reg_val_offs * 4)); + reg_val_offs * 4)); } static void gfx_v8_0_ring_emit_wreg(struct amdgpu_ring *ring, uint32_t reg, @@ -6815,6 +6817,34 @@ static int gfx_v8_0_sq_irq(struct amdgpu_device *adev, return 0; } +static void gfx_v8_0_emit_mem_sync(struct amdgpu_ring *ring) +{ + amdgpu_ring_write(ring, PACKET3(PACKET3_SURFACE_SYNC, 3)); + amdgpu_ring_write(ring, PACKET3_TCL1_ACTION_ENA | + PACKET3_TC_ACTION_ENA | + PACKET3_SH_KCACHE_ACTION_ENA | + PACKET3_SH_ICACHE_ACTION_ENA | + PACKET3_TC_WB_ACTION_ENA); /* CP_COHER_CNTL */ + amdgpu_ring_write(ring, 0xffffffff); /* CP_COHER_SIZE */ + amdgpu_ring_write(ring, 0); /* CP_COHER_BASE */ + amdgpu_ring_write(ring, 0x0000000A); /* poll interval */ +} + +static void gfx_v8_0_emit_mem_sync_compute(struct amdgpu_ring *ring) +{ + amdgpu_ring_write(ring, PACKET3(PACKET3_ACQUIRE_MEM, 5)); + amdgpu_ring_write(ring, PACKET3_TCL1_ACTION_ENA | + PACKET3_TC_ACTION_ENA | + PACKET3_SH_KCACHE_ACTION_ENA | + PACKET3_SH_ICACHE_ACTION_ENA | + PACKET3_TC_WB_ACTION_ENA); /* CP_COHER_CNTL */ + amdgpu_ring_write(ring, 0xffffffff); /* CP_COHER_SIZE */ + amdgpu_ring_write(ring, 0xff); /* CP_COHER_SIZE_HI */ + amdgpu_ring_write(ring, 0); /* CP_COHER_BASE */ + amdgpu_ring_write(ring, 0); /* CP_COHER_BASE_HI */ + amdgpu_ring_write(ring, 0x0000000A); /* poll interval */ +} + static const struct amd_ip_funcs gfx_v8_0_ip_funcs = { .name = "gfx_v8_0", .early_init = gfx_v8_0_early_init, @@ -6861,7 +6891,8 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_gfx = { 3 + /* CNTX_CTRL */ 5 + /* HDP_INVL */ 12 + 12 + /* FENCE x2 */ - 2, /* SWITCH_BUFFER */ + 2 + /* SWITCH_BUFFER */ + 5, /* SURFACE_SYNC */ .emit_ib_size = 4, /* gfx_v8_0_ring_emit_ib_gfx */ .emit_ib = gfx_v8_0_ring_emit_ib_gfx, .emit_fence = gfx_v8_0_ring_emit_fence_gfx, @@ -6879,6 +6910,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_gfx = { .patch_cond_exec = gfx_v8_0_ring_emit_patch_cond_exec, .emit_wreg = gfx_v8_0_ring_emit_wreg, .soft_recovery = gfx_v8_0_ring_soft_recovery, + .emit_mem_sync = gfx_v8_0_emit_mem_sync, }; static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_compute = { @@ -6895,7 +6927,8 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_compute = { 5 + /* hdp_invalidate */ 7 + /* gfx_v8_0_ring_emit_pipeline_sync */ VI_FLUSH_GPU_TLB_NUM_WREG * 5 + 7 + /* gfx_v8_0_ring_emit_vm_flush */ - 7 + 7 + 7, /* gfx_v8_0_ring_emit_fence_compute x3 for user fence, vm fence */ + 7 + 7 + 7 + /* gfx_v8_0_ring_emit_fence_compute x3 for user fence, vm fence */ + 7, /* gfx_v8_0_emit_mem_sync_compute */ .emit_ib_size = 7, /* gfx_v8_0_ring_emit_ib_compute */ .emit_ib = gfx_v8_0_ring_emit_ib_compute, .emit_fence = gfx_v8_0_ring_emit_fence_compute, @@ -6908,6 +6941,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_compute = { .insert_nop = amdgpu_ring_insert_nop, .pad_ib = amdgpu_ring_generic_pad_ib, .emit_wreg = gfx_v8_0_ring_emit_wreg, + .emit_mem_sync = gfx_v8_0_emit_mem_sync_compute, }; static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_kiq = { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index e6b113ed2f40..711e9dd19705 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -50,18 +50,14 @@ #include "gfx_v9_4.h" +#include "asic_reg/pwr/pwr_10_0_offset.h" +#include "asic_reg/pwr/pwr_10_0_sh_mask.h" + #define GFX9_NUM_GFX_RINGS 1 #define GFX9_MEC_HPD_SIZE 4096 #define RLCG_UCODE_LOADING_START_ADDRESS 0x00002000L #define RLC_SAVE_RESTORE_ADDR_STARTING_OFFSET 0x00000000L -#define mmPWR_MISC_CNTL_STATUS 0x0183 -#define mmPWR_MISC_CNTL_STATUS_BASE_IDX 0 -#define PWR_MISC_CNTL_STATUS__PWR_GFX_RLC_CGPG_EN__SHIFT 0x0 -#define PWR_MISC_CNTL_STATUS__PWR_GFXOFF_STATUS__SHIFT 0x1 -#define PWR_MISC_CNTL_STATUS__PWR_GFX_RLC_CGPG_EN_MASK 0x00000001L -#define PWR_MISC_CNTL_STATUS__PWR_GFXOFF_STATUS_MASK 0x00000006L - #define mmGCEA_PROBE_MAP 0x070c #define mmGCEA_PROBE_MAP_BASE_IDX 0 @@ -511,8 +507,8 @@ static const struct soc15_reg_golden golden_settings_gc_9_0[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmSH_MEM_CONFIG, 0x00001000, 0x00001000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSPI_RESOURCE_RESERVE_CU_0, 0x0007ffff, 0x00000800), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSPI_RESOURCE_RESERVE_CU_1, 0x0007ffff, 0x00000800), - SOC15_REG_GOLDEN_VALUE(GC, 0, mmSPI_RESOURCE_RESERVE_EN_CU_0, 0x01ffffff, 0x0000ff87), - SOC15_REG_GOLDEN_VALUE(GC, 0, mmSPI_RESOURCE_RESERVE_EN_CU_1, 0x01ffffff, 0x0000ff8f), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSPI_RESOURCE_RESERVE_EN_CU_0, 0x01ffffff, 0x00ffff87), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSPI_RESOURCE_RESERVE_EN_CU_1, 0x01ffffff, 0x00ffff8f), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQC_CONFIG, 0x03000000, 0x020a2000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmTA_CNTL_AUX, 0xfffffeef, 0x010b0000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmTCP_CHAN_STEER_HI, 0xffffffff, 0x4a2c0e68), @@ -963,7 +959,7 @@ static void gfx_v9_0_init_golden_registers(struct amdgpu_device *adev) case CHIP_RAVEN: soc15_program_register_sequence(adev, golden_settings_gc_9_1, ARRAY_SIZE(golden_settings_gc_9_1)); - if (adev->rev_id >= 8) + if (adev->apu_flags & AMD_APU_IS_RAVEN2) soc15_program_register_sequence(adev, golden_settings_gc_9_1_rv2, ARRAY_SIZE(golden_settings_gc_9_1_rv2)); @@ -1082,7 +1078,8 @@ static int gfx_v9_0_ring_test_ib(struct amdgpu_ring *ring, long timeout) gpu_addr = adev->wb.gpu_addr + (index * 4); adev->wb.wb[index] = cpu_to_le32(0xCAFEDEAD); memset(&ib, 0, sizeof(ib)); - r = amdgpu_ib_get(adev, NULL, 16, &ib); + r = amdgpu_ib_get(adev, NULL, 16, + AMDGPU_IB_POOL_DIRECT, &ib); if (r) goto err1; @@ -1234,6 +1231,10 @@ struct amdgpu_gfxoff_quirk { static const struct amdgpu_gfxoff_quirk amdgpu_gfxoff_quirk_list[] = { /* https://bugzilla.kernel.org/show_bug.cgi?id=204689 */ { 0x1002, 0x15dd, 0x1002, 0x15dd, 0xc8 }, + /* https://bugzilla.kernel.org/show_bug.cgi?id=207171 */ + { 0x1002, 0x15dd, 0x103c, 0x83e7, 0xd3 }, + /* GFXOFF is unstable on C6 parts with a VBIOS 113-RAVEN-114 */ + { 0x1002, 0x15dd, 0x1002, 0x15dd, 0xc6 }, { 0, 0, 0, 0, 0 }, }; @@ -1273,7 +1274,8 @@ static void gfx_v9_0_check_if_need_gfxoff(struct amdgpu_device *adev) case CHIP_VEGA20: break; case CHIP_RAVEN: - if (!(adev->rev_id >= 0x8 || adev->pdev->device == 0x15d8) && + if (!((adev->apu_flags & AMD_APU_IS_RAVEN2) || + (adev->apu_flags & AMD_APU_IS_PICASSO)) && ((!is_raven_kicker(adev) && adev->gfx.rlc_fw_version < 531) || (adev->gfx.rlc_feature_version < 1) || @@ -1616,9 +1618,9 @@ static int gfx_v9_0_init_microcode(struct amdgpu_device *adev) chip_name = "vega20"; break; case CHIP_RAVEN: - if (adev->rev_id >= 8) + if (adev->apu_flags & AMD_APU_IS_RAVEN2) chip_name = "raven2"; - else if (adev->pdev->device == 0x15d8) + else if (adev->apu_flags & AMD_APU_IS_PICASSO) chip_name = "picasso"; else chip_name = "raven"; @@ -2118,7 +2120,7 @@ static int gfx_v9_0_gpu_early_init(struct amdgpu_device *adev) adev->gfx.config.sc_prim_fifo_size_backend = 0x100; adev->gfx.config.sc_hiz_tile_fifo_size = 0x30; adev->gfx.config.sc_earlyz_tile_fifo_size = 0x4C0; - if (adev->rev_id >= 8) + if (adev->apu_flags & AMD_APU_IS_RAVEN2) gb_addr_config = RAVEN2_GB_ADDR_CONFIG_GOLDEN; else gb_addr_config = RAVEN_GB_ADDR_CONFIG_GOLDEN; @@ -2195,6 +2197,7 @@ static int gfx_v9_0_compute_ring_init(struct amdgpu_device *adev, int ring_id, int r; unsigned irq_type; struct amdgpu_ring *ring = &adev->gfx.compute_ring[ring_id]; + unsigned int hw_prio; ring = &adev->gfx.compute_ring[ring_id]; @@ -2213,10 +2216,11 @@ static int gfx_v9_0_compute_ring_init(struct amdgpu_device *adev, int ring_id, irq_type = AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP + ((ring->me - 1) * adev->gfx.mec.num_pipe_per_mec) + ring->pipe; - + hw_prio = amdgpu_gfx_is_high_priority_compute_queue(adev, ring->queue) ? + AMDGPU_GFX_PIPE_PRIO_HIGH : AMDGPU_GFX_PIPE_PRIO_NORMAL; /* type-2 packets are deprecated on MEC, use type-3 instead */ r = amdgpu_ring_init(adev, ring, 1024, - &adev->gfx.eop_irq, irq_type); + &adev->gfx.eop_irq, irq_type, hw_prio); if (r) return r; @@ -2310,7 +2314,9 @@ static int gfx_v9_0_sw_init(void *handle) ring->use_doorbell = true; ring->doorbell_index = adev->doorbell_index.gfx_ring0 << 1; r = amdgpu_ring_init(adev, ring, 1024, - &adev->gfx.eop_irq, AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP); + &adev->gfx.eop_irq, + AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } @@ -2528,7 +2534,7 @@ static void gfx_v9_0_init_sq_config(struct amdgpu_device *adev) break; default: break; - }; + } } static void gfx_v9_0_constants_init(struct amdgpu_device *adev) @@ -2963,8 +2969,7 @@ static void gfx_v9_0_init_pg(struct amdgpu_device *adev) */ if (adev->gfx.rlc.is_rlc_v2_1) { if (adev->asic_type == CHIP_VEGA12 || - (adev->asic_type == CHIP_RAVEN && - adev->rev_id >= 8)) + (adev->apu_flags & AMD_APU_IS_RAVEN2)) gfx_v9_1_init_rlc_save_restore_list(adev); gfx_v9_0_enable_save_restore_machine(adev); } @@ -3100,16 +3105,11 @@ static int gfx_v9_0_rlc_resume(struct amdgpu_device *adev) static void gfx_v9_0_cp_gfx_enable(struct amdgpu_device *adev, bool enable) { - int i; u32 tmp = RREG32_SOC15(GC, 0, mmCP_ME_CNTL); tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, ME_HALT, enable ? 0 : 1); tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, PFP_HALT, enable ? 0 : 1); tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, CE_HALT, enable ? 0 : 1); - if (!enable) { - for (i = 0; i < adev->gfx.num_gfx_rings; i++) - adev->gfx.gfx_ring[i].sched.ready = false; - } WREG32_SOC15_RLC(GC, 0, mmCP_ME_CNTL, tmp); udelay(50); } @@ -3305,15 +3305,11 @@ static int gfx_v9_0_cp_gfx_resume(struct amdgpu_device *adev) static void gfx_v9_0_cp_compute_enable(struct amdgpu_device *adev, bool enable) { - int i; - if (enable) { WREG32_SOC15_RLC(GC, 0, mmCP_MEC_CNTL, 0); } else { WREG32_SOC15_RLC(GC, 0, mmCP_MEC_CNTL, (CP_MEC_CNTL__MEC_ME1_HALT_MASK | CP_MEC_CNTL__MEC_ME2_HALT_MASK)); - for (i = 0; i < adev->gfx.num_compute_rings; i++) - adev->gfx.compute_ring[i].sched.ready = false; adev->gfx.kiq.ring.sched.ready = false; } udelay(50); @@ -3383,11 +3379,8 @@ static void gfx_v9_0_mqd_set_priority(struct amdgpu_ring *ring, struct v9_mqd *m if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) { if (amdgpu_gfx_is_high_priority_compute_queue(adev, ring->queue)) { mqd->cp_hqd_pipe_priority = AMDGPU_GFX_PIPE_PRIO_HIGH; - ring->has_high_prio = true; mqd->cp_hqd_queue_priority = AMDGPU_GFX_QUEUE_PRIORITY_MAXIMUM; - } else { - ring->has_high_prio = false; } } } @@ -4054,13 +4047,18 @@ static uint64_t gfx_v9_0_kiq_read_clock(struct amdgpu_device *adev) { signed long r, cnt = 0; unsigned long flags; - uint32_t seq; + uint32_t seq, reg_val_offs = 0; + uint64_t value = 0; struct amdgpu_kiq *kiq = &adev->gfx.kiq; struct amdgpu_ring *ring = &kiq->ring; BUG_ON(!ring->funcs->emit_rreg); spin_lock_irqsave(&kiq->ring_lock, flags); + if (amdgpu_device_wb_get(adev, ®_val_offs)) { + pr_err("critical bug! too many kiq readers\n"); + goto failed_unlock; + } amdgpu_ring_alloc(ring, 32); amdgpu_ring_write(ring, PACKET3(PACKET3_COPY_DATA, 4)); amdgpu_ring_write(ring, 9 | /* src: register*/ @@ -4070,10 +4068,13 @@ static uint64_t gfx_v9_0_kiq_read_clock(struct amdgpu_device *adev) amdgpu_ring_write(ring, 0); amdgpu_ring_write(ring, 0); amdgpu_ring_write(ring, lower_32_bits(adev->wb.gpu_addr + - kiq->reg_val_offs * 4)); + reg_val_offs * 4)); amdgpu_ring_write(ring, upper_32_bits(adev->wb.gpu_addr + - kiq->reg_val_offs * 4)); - amdgpu_fence_emit_polling(ring, &seq); + reg_val_offs * 4)); + r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT); + if (r) + goto failed_undo; + amdgpu_ring_commit(ring); spin_unlock_irqrestore(&kiq->ring_lock, flags); @@ -4099,10 +4100,19 @@ static uint64_t gfx_v9_0_kiq_read_clock(struct amdgpu_device *adev) if (cnt > MAX_KIQ_REG_TRY) goto failed_kiq_read; - return (uint64_t)adev->wb.wb[kiq->reg_val_offs] | - (uint64_t)adev->wb.wb[kiq->reg_val_offs + 1 ] << 32ULL; + mb(); + value = (uint64_t)adev->wb.wb[reg_val_offs] | + (uint64_t)adev->wb.wb[reg_val_offs + 1 ] << 32ULL; + amdgpu_device_wb_free(adev, reg_val_offs); + return value; +failed_undo: + amdgpu_ring_undo(ring); +failed_unlock: + spin_unlock_irqrestore(&kiq->ring_lock, flags); failed_kiq_read: + if (reg_val_offs) + amdgpu_device_wb_free(adev, reg_val_offs); pr_err("failed to read gpu clock\n"); return ~0; } @@ -4487,7 +4497,8 @@ static int gfx_v9_0_do_edc_gpr_workarounds(struct amdgpu_device *adev) /* allocate an indirect buffer to put the commands in */ memset(&ib, 0, sizeof(ib)); - r = amdgpu_ib_get(adev, NULL, total_size, &ib); + r = amdgpu_ib_get(adev, NULL, total_size, + AMDGPU_IB_POOL_DIRECT, &ib); if (r) { DRM_ERROR("amdgpu: failed to get ib (%d).\n", r); return r; @@ -4958,14 +4969,21 @@ static int gfx_v9_0_update_gfx_clock_gating(struct amdgpu_device *adev, static void gfx_v9_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) { - u32 data; + u32 reg, data; - data = RREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL); + reg = SOC15_REG_OFFSET(GC, 0, mmRLC_SPM_MC_CNTL); + if (amdgpu_sriov_is_pp_one_vf(adev)) + data = RREG32_NO_KIQ(reg); + else + data = RREG32(reg); data &= ~RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK; data |= (vmid & RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK) << RLC_SPM_MC_CNTL__RLC_SPM_VMID__SHIFT; - WREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL, data); + if (amdgpu_sriov_is_pp_one_vf(adev)) + WREG32_SOC15_NO_KIQ(GC, 0, mmRLC_SPM_MC_CNTL, data); + else + WREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL, data); } static bool gfx_v9_0_check_rlcg_range(struct amdgpu_device *adev, @@ -5023,10 +5041,9 @@ static int gfx_v9_0_set_powergating_state(void *handle, switch (adev->asic_type) { case CHIP_RAVEN: case CHIP_RENOIR: - if (!enable) { + if (!enable) amdgpu_gfx_off_ctrl(adev, false); - cancel_delayed_work_sync(&adev->gfx.gfx_off_delay_work); - } + if (adev->pg_flags & AMD_PG_SUPPORT_RLC_SMU_HS) { gfx_v9_0_enable_sck_slow_down_on_power_up(adev, true); gfx_v9_0_enable_sck_slow_down_on_power_down(adev, true); @@ -5050,12 +5067,7 @@ static int gfx_v9_0_set_powergating_state(void *handle, amdgpu_gfx_off_ctrl(adev, true); break; case CHIP_VEGA12: - if (!enable) { - amdgpu_gfx_off_ctrl(adev, false); - cancel_delayed_work_sync(&adev->gfx.gfx_off_delay_work); - } else { - amdgpu_gfx_off_ctrl(adev, true); - } + amdgpu_gfx_off_ctrl(adev, enable); break; default: break; @@ -5426,10 +5438,13 @@ static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring) amdgpu_ring_write_multiple(ring, (void *)&de_payload, sizeof(de_payload) >> 2); } -static void gfx_v9_0_ring_emit_tmz(struct amdgpu_ring *ring, bool start) +static void gfx_v9_0_ring_emit_frame_cntl(struct amdgpu_ring *ring, bool start, + bool secure) { + uint32_t v = secure ? FRAME_TMZ : 0; + amdgpu_ring_write(ring, PACKET3(PACKET3_FRAME_CONTROL, 0)); - amdgpu_ring_write(ring, FRAME_CMD(start ? 0 : 1)); /* frame_end */ + amdgpu_ring_write(ring, v | FRAME_CMD(start ? 0 : 1)); } static void gfx_v9_ring_emit_cntxcntl(struct amdgpu_ring *ring, uint32_t flags) @@ -5439,8 +5454,6 @@ static void gfx_v9_ring_emit_cntxcntl(struct amdgpu_ring *ring, uint32_t flags) if (amdgpu_sriov_vf(ring->adev)) gfx_v9_0_ring_emit_ce_meta(ring); - gfx_v9_0_ring_emit_tmz(ring, true); - dw2 |= 0x80000000; /* set load_enable otherwise this package is just NOPs */ if (flags & AMDGPU_HAVE_CTX_SWITCH) { /* set load_global_config & load_global_uconfig */ @@ -5491,10 +5504,10 @@ static void gfx_v9_0_ring_emit_patch_cond_exec(struct amdgpu_ring *ring, unsigne ring->ring[offset] = (ring->ring_size>>2) - offset + cur; } -static void gfx_v9_0_ring_emit_rreg(struct amdgpu_ring *ring, uint32_t reg) +static void gfx_v9_0_ring_emit_rreg(struct amdgpu_ring *ring, uint32_t reg, + uint32_t reg_val_offs) { struct amdgpu_device *adev = ring->adev; - struct amdgpu_kiq *kiq = &adev->gfx.kiq; amdgpu_ring_write(ring, PACKET3(PACKET3_COPY_DATA, 4)); amdgpu_ring_write(ring, 0 | /* src: register*/ @@ -5503,9 +5516,9 @@ static void gfx_v9_0_ring_emit_rreg(struct amdgpu_ring *ring, uint32_t reg) amdgpu_ring_write(ring, reg); amdgpu_ring_write(ring, 0); amdgpu_ring_write(ring, lower_32_bits(adev->wb.gpu_addr + - kiq->reg_val_offs * 4)); + reg_val_offs * 4)); amdgpu_ring_write(ring, upper_32_bits(adev->wb.gpu_addr + - kiq->reg_val_offs * 4)); + reg_val_offs * 4)); } static void gfx_v9_0_ring_emit_wreg(struct amdgpu_ring *ring, uint32_t reg, @@ -6406,15 +6419,15 @@ static int gfx_v9_0_query_utc_edc_status(struct amdgpu_device *adev, sec_count = REG_GET_FIELD(data, VM_L2_MEM_ECC_CNT, SEC_COUNT); if (sec_count) { - DRM_INFO("Instance[%d]: SubBlock %s, SEC %d\n", i, - vml2_mems[i], sec_count); + dev_info(adev->dev, "Instance[%d]: SubBlock %s, " + "SEC %d\n", i, vml2_mems[i], sec_count); err_data->ce_count += sec_count; } ded_count = REG_GET_FIELD(data, VM_L2_MEM_ECC_CNT, DED_COUNT); if (ded_count) { - DRM_INFO("Instance[%d]: SubBlock %s, DED %d\n", i, - vml2_mems[i], ded_count); + dev_info(adev->dev, "Instance[%d]: SubBlock %s, " + "DED %d\n", i, vml2_mems[i], ded_count); err_data->ue_count += ded_count; } } @@ -6426,16 +6439,16 @@ static int gfx_v9_0_query_utc_edc_status(struct amdgpu_device *adev, sec_count = REG_GET_FIELD(data, VM_L2_WALKER_MEM_ECC_CNT, SEC_COUNT); if (sec_count) { - DRM_INFO("Instance[%d]: SubBlock %s, SEC %d\n", i, - vml2_walker_mems[i], sec_count); + dev_info(adev->dev, "Instance[%d]: SubBlock %s, " + "SEC %d\n", i, vml2_walker_mems[i], sec_count); err_data->ce_count += sec_count; } ded_count = REG_GET_FIELD(data, VM_L2_WALKER_MEM_ECC_CNT, DED_COUNT); if (ded_count) { - DRM_INFO("Instance[%d]: SubBlock %s, DED %d\n", i, - vml2_walker_mems[i], ded_count); + dev_info(adev->dev, "Instance[%d]: SubBlock %s, " + "DED %d\n", i, vml2_walker_mems[i], ded_count); err_data->ue_count += ded_count; } } @@ -6446,8 +6459,9 @@ static int gfx_v9_0_query_utc_edc_status(struct amdgpu_device *adev, sec_count = (data & 0x00006000L) >> 0xd; if (sec_count) { - DRM_INFO("Instance[%d]: SubBlock %s, SEC %d\n", i, - atc_l2_cache_2m_mems[i], sec_count); + dev_info(adev->dev, "Instance[%d]: SubBlock %s, " + "SEC %d\n", i, atc_l2_cache_2m_mems[i], + sec_count); err_data->ce_count += sec_count; } } @@ -6458,15 +6472,17 @@ static int gfx_v9_0_query_utc_edc_status(struct amdgpu_device *adev, sec_count = (data & 0x00006000L) >> 0xd; if (sec_count) { - DRM_INFO("Instance[%d]: SubBlock %s, SEC %d\n", i, - atc_l2_cache_4k_mems[i], sec_count); + dev_info(adev->dev, "Instance[%d]: SubBlock %s, " + "SEC %d\n", i, atc_l2_cache_4k_mems[i], + sec_count); err_data->ce_count += sec_count; } ded_count = (data & 0x00018000L) >> 0xf; if (ded_count) { - DRM_INFO("Instance[%d]: SubBlock %s, DED %d\n", i, - atc_l2_cache_4k_mems[i], ded_count); + dev_info(adev->dev, "Instance[%d]: SubBlock %s, " + "DED %d\n", i, atc_l2_cache_4k_mems[i], + ded_count); err_data->ue_count += ded_count; } } @@ -6479,7 +6495,8 @@ static int gfx_v9_0_query_utc_edc_status(struct amdgpu_device *adev, return 0; } -static int gfx_v9_0_ras_error_count(const struct soc15_reg_entry *reg, +static int gfx_v9_0_ras_error_count(struct amdgpu_device *adev, + const struct soc15_reg_entry *reg, uint32_t se_id, uint32_t inst_id, uint32_t value, uint32_t *sec_count, uint32_t *ded_count) { @@ -6496,7 +6513,8 @@ static int gfx_v9_0_ras_error_count(const struct soc15_reg_entry *reg, gfx_v9_0_ras_fields[i].sec_count_mask) >> gfx_v9_0_ras_fields[i].sec_count_shift; if (sec_cnt) { - DRM_INFO("GFX SubBlock %s, Instance[%d][%d], SEC %d\n", + dev_info(adev->dev, "GFX SubBlock %s, " + "Instance[%d][%d], SEC %d\n", gfx_v9_0_ras_fields[i].name, se_id, inst_id, sec_cnt); @@ -6507,7 +6525,8 @@ static int gfx_v9_0_ras_error_count(const struct soc15_reg_entry *reg, gfx_v9_0_ras_fields[i].ded_count_mask) >> gfx_v9_0_ras_fields[i].ded_count_shift; if (ded_cnt) { - DRM_INFO("GFX SubBlock %s, Instance[%d][%d], DED %d\n", + dev_info(adev->dev, "GFX SubBlock %s, " + "Instance[%d][%d], DED %d\n", gfx_v9_0_ras_fields[i].name, se_id, inst_id, ded_cnt); @@ -6596,9 +6615,10 @@ static int gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev, reg_value = RREG32(SOC15_REG_ENTRY_OFFSET(gfx_v9_0_edc_counter_regs[i])); if (reg_value) - gfx_v9_0_ras_error_count(&gfx_v9_0_edc_counter_regs[i], - j, k, reg_value, - &sec_count, &ded_count); + gfx_v9_0_ras_error_count(adev, + &gfx_v9_0_edc_counter_regs[i], + j, k, reg_value, + &sec_count, &ded_count); } } } @@ -6614,6 +6634,25 @@ static int gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev, return 0; } +static void gfx_v9_0_emit_mem_sync(struct amdgpu_ring *ring) +{ + const unsigned int cp_coher_cntl = + PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_SH_ICACHE_ACTION_ENA(1) | + PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_SH_KCACHE_ACTION_ENA(1) | + PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_TC_ACTION_ENA(1) | + PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_TCL1_ACTION_ENA(1) | + PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_TC_WB_ACTION_ENA(1); + + /* ACQUIRE_MEM -make one or more surfaces valid for use by the subsequent operations */ + amdgpu_ring_write(ring, PACKET3(PACKET3_ACQUIRE_MEM, 5)); + amdgpu_ring_write(ring, cp_coher_cntl); /* CP_COHER_CNTL */ + amdgpu_ring_write(ring, 0xffffffff); /* CP_COHER_SIZE */ + amdgpu_ring_write(ring, 0xffffff); /* CP_COHER_SIZE_HI */ + amdgpu_ring_write(ring, 0); /* CP_COHER_BASE */ + amdgpu_ring_write(ring, 0); /* CP_COHER_BASE_HI */ + amdgpu_ring_write(ring, 0x0000000A); /* POLL_INTERVAL */ +} + static const struct amd_ip_funcs gfx_v9_0_ip_funcs = { .name = "gfx_v9_0", .early_init = gfx_v9_0_early_init, @@ -6660,7 +6699,8 @@ static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_gfx = { 3 + /* CNTX_CTRL */ 5 + /* HDP_INVL */ 8 + 8 + /* FENCE x2 */ - 2, /* SWITCH_BUFFER */ + 2 + /* SWITCH_BUFFER */ + 7, /* gfx_v9_0_emit_mem_sync */ .emit_ib_size = 4, /* gfx_v9_0_ring_emit_ib_gfx */ .emit_ib = gfx_v9_0_ring_emit_ib_gfx, .emit_fence = gfx_v9_0_ring_emit_fence, @@ -6676,11 +6716,12 @@ static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_gfx = { .emit_cntxcntl = gfx_v9_ring_emit_cntxcntl, .init_cond_exec = gfx_v9_0_ring_emit_init_cond_exec, .patch_cond_exec = gfx_v9_0_ring_emit_patch_cond_exec, - .emit_tmz = gfx_v9_0_ring_emit_tmz, + .emit_frame_cntl = gfx_v9_0_ring_emit_frame_cntl, .emit_wreg = gfx_v9_0_ring_emit_wreg, .emit_reg_wait = gfx_v9_0_ring_emit_reg_wait, .emit_reg_write_reg_wait = gfx_v9_0_ring_emit_reg_write_reg_wait, .soft_recovery = gfx_v9_0_ring_soft_recovery, + .emit_mem_sync = gfx_v9_0_emit_mem_sync, }; static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_compute = { @@ -6700,7 +6741,8 @@ static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_compute = { SOC15_FLUSH_GPU_TLB_NUM_WREG * 5 + SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 7 + 2 + /* gfx_v9_0_ring_emit_vm_flush */ - 8 + 8 + 8, /* gfx_v9_0_ring_emit_fence x3 for user fence, vm fence */ + 8 + 8 + 8 + /* gfx_v9_0_ring_emit_fence x3 for user fence, vm fence */ + 7, /* gfx_v9_0_emit_mem_sync */ .emit_ib_size = 7, /* gfx_v9_0_ring_emit_ib_compute */ .emit_ib = gfx_v9_0_ring_emit_ib_compute, .emit_fence = gfx_v9_0_ring_emit_fence, @@ -6715,6 +6757,7 @@ static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_compute = { .emit_wreg = gfx_v9_0_ring_emit_wreg, .emit_reg_wait = gfx_v9_0_ring_emit_reg_wait, .emit_reg_write_reg_wait = gfx_v9_0_ring_emit_reg_write_reg_wait, + .emit_mem_sync = gfx_v9_0_emit_mem_sync, }; static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_kiq = { @@ -6838,7 +6881,7 @@ static void gfx_v9_0_set_gds_init(struct amdgpu_device *adev) adev->gds.gds_compute_max_wave_id = 0x27f; break; case CHIP_RAVEN: - if (adev->rev_id >= 0x8) + if (adev->apu_flags & AMD_APU_IS_RAVEN2) adev->gds.gds_compute_max_wave_id = 0x77; /* raven2 */ else adev->gds.gds_compute_max_wave_id = 0x15f; /* raven1 */ diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c index dce945ef21a5..46351db36922 100644..100755 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c @@ -732,7 +732,8 @@ static int gfx_v9_4_query_utc_edc_status(struct amdgpu_device *adev, sec_count = REG_GET_FIELD(data, VML2_WALKER_MEM_ECC_CNTL, SEC_COUNT); if (sec_count) { - DRM_INFO("Instance[%d]: SubBlock %s, SEC %d\n", i, + dev_info(adev->dev, + "Instance[%d]: SubBlock %s, SEC %d\n", i, vml2_walker_mems[i], sec_count); err_data->ce_count += sec_count; } @@ -740,7 +741,8 @@ static int gfx_v9_4_query_utc_edc_status(struct amdgpu_device *adev, ded_count = REG_GET_FIELD(data, VML2_WALKER_MEM_ECC_CNTL, DED_COUNT); if (ded_count) { - DRM_INFO("Instance[%d]: SubBlock %s, DED %d\n", i, + dev_info(adev->dev, + "Instance[%d]: SubBlock %s, DED %d\n", i, vml2_walker_mems[i], ded_count); err_data->ue_count += ded_count; } @@ -752,14 +754,16 @@ static int gfx_v9_4_query_utc_edc_status(struct amdgpu_device *adev, sec_count = REG_GET_FIELD(data, UTCL2_MEM_ECC_CNTL, SEC_COUNT); if (sec_count) { - DRM_INFO("Instance[%d]: SubBlock %s, SEC %d\n", i, + dev_info(adev->dev, + "Instance[%d]: SubBlock %s, SEC %d\n", i, utcl2_router_mems[i], sec_count); err_data->ce_count += sec_count; } ded_count = REG_GET_FIELD(data, UTCL2_MEM_ECC_CNTL, DED_COUNT); if (ded_count) { - DRM_INFO("Instance[%d]: SubBlock %s, DED %d\n", i, + dev_info(adev->dev, + "Instance[%d]: SubBlock %s, DED %d\n", i, utcl2_router_mems[i], ded_count); err_data->ue_count += ded_count; } @@ -772,7 +776,8 @@ static int gfx_v9_4_query_utc_edc_status(struct amdgpu_device *adev, sec_count = REG_GET_FIELD(data, ATC_L2_CACHE_2M_DSM_CNTL, SEC_COUNT); if (sec_count) { - DRM_INFO("Instance[%d]: SubBlock %s, SEC %d\n", i, + dev_info(adev->dev, + "Instance[%d]: SubBlock %s, SEC %d\n", i, atc_l2_cache_2m_mems[i], sec_count); err_data->ce_count += sec_count; } @@ -780,7 +785,8 @@ static int gfx_v9_4_query_utc_edc_status(struct amdgpu_device *adev, ded_count = REG_GET_FIELD(data, ATC_L2_CACHE_2M_DSM_CNTL, DED_COUNT); if (ded_count) { - DRM_INFO("Instance[%d]: SubBlock %s, DED %d\n", i, + dev_info(adev->dev, + "Instance[%d]: SubBlock %s, DED %d\n", i, atc_l2_cache_2m_mems[i], ded_count); err_data->ue_count += ded_count; } @@ -793,7 +799,8 @@ static int gfx_v9_4_query_utc_edc_status(struct amdgpu_device *adev, sec_count = REG_GET_FIELD(data, ATC_L2_CACHE_4K_DSM_CNTL, SEC_COUNT); if (sec_count) { - DRM_INFO("Instance[%d]: SubBlock %s, SEC %d\n", i, + dev_info(adev->dev, + "Instance[%d]: SubBlock %s, SEC %d\n", i, atc_l2_cache_4k_mems[i], sec_count); err_data->ce_count += sec_count; } @@ -801,7 +808,8 @@ static int gfx_v9_4_query_utc_edc_status(struct amdgpu_device *adev, ded_count = REG_GET_FIELD(data, ATC_L2_CACHE_4K_DSM_CNTL, DED_COUNT); if (ded_count) { - DRM_INFO("Instance[%d]: SubBlock %s, DED %d\n", i, + dev_info(adev->dev, + "Instance[%d]: SubBlock %s, DED %d\n", i, atc_l2_cache_4k_mems[i], ded_count); err_data->ue_count += ded_count; } @@ -816,7 +824,8 @@ static int gfx_v9_4_query_utc_edc_status(struct amdgpu_device *adev, return 0; } -static int gfx_v9_4_ras_error_count(const struct soc15_reg_entry *reg, +static int gfx_v9_4_ras_error_count(struct amdgpu_device *adev, + const struct soc15_reg_entry *reg, uint32_t se_id, uint32_t inst_id, uint32_t value, uint32_t *sec_count, uint32_t *ded_count) @@ -833,7 +842,8 @@ static int gfx_v9_4_ras_error_count(const struct soc15_reg_entry *reg, sec_cnt = (value & gfx_v9_4_ras_fields[i].sec_count_mask) >> gfx_v9_4_ras_fields[i].sec_count_shift; if (sec_cnt) { - DRM_INFO("GFX SubBlock %s, Instance[%d][%d], SEC %d\n", + dev_info(adev->dev, + "GFX SubBlock %s, Instance[%d][%d], SEC %d\n", gfx_v9_4_ras_fields[i].name, se_id, inst_id, sec_cnt); *sec_count += sec_cnt; @@ -842,7 +852,8 @@ static int gfx_v9_4_ras_error_count(const struct soc15_reg_entry *reg, ded_cnt = (value & gfx_v9_4_ras_fields[i].ded_count_mask) >> gfx_v9_4_ras_fields[i].ded_count_shift; if (ded_cnt) { - DRM_INFO("GFX SubBlock %s, Instance[%d][%d], DED %d\n", + dev_info(adev->dev, + "GFX SubBlock %s, Instance[%d][%d], DED %d\n", gfx_v9_4_ras_fields[i].name, se_id, inst_id, ded_cnt); *ded_count += ded_cnt; @@ -876,7 +887,7 @@ int gfx_v9_4_query_ras_error_count(struct amdgpu_device *adev, reg_value = RREG32(SOC15_REG_ENTRY_OFFSET( gfx_v9_4_edc_counter_regs[i])); if (reg_value) - gfx_v9_4_ras_error_count( + gfx_v9_4_ras_error_count(adev, &gfx_v9_4_edc_counter_regs[i], j, k, reg_value, &sec_count, &ded_count); diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c index 1a2f18b908fe..6682b843bafe 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c @@ -80,7 +80,7 @@ static void gfxhub_v1_0_init_system_aperture_regs(struct amdgpu_device *adev) WREG32_SOC15_RLC(GC, 0, mmMC_VM_SYSTEM_APERTURE_LOW_ADDR, min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18); - if (adev->asic_type == CHIP_RAVEN && adev->rev_id >= 0x8) + if (adev->apu_flags & AMD_APU_IS_RAVEN2) /* * Raven2 has a HW issue that it is unable to use the * vram which is out of MC_VM_SYSTEM_APERTURE_HIGH_ADDR. diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c index 9775eca6fe43..ba2b7ac0c02d 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c @@ -170,6 +170,9 @@ static int gmc_v10_0_process_interrupt(struct amdgpu_device *adev, dev_err(adev->dev, "GCVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n", status); + dev_err(adev->dev, "\t Faulty UTCL2 client ID: 0x%lx\n", + REG_GET_FIELD(status, + GCVM_L2_PROTECTION_FAULT_STATUS, CID)); dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, MORE_FAULTS)); @@ -369,7 +372,8 @@ static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, * translation. Avoid this by doing the invalidation from the SDMA * itself. */ - r = amdgpu_job_alloc_with_ib(adev, 16 * 4, &job); + r = amdgpu_job_alloc_with_ib(adev, 16 * 4, AMDGPU_IB_POOL_IMMEDIATE, + &job); if (r) goto error_alloc; @@ -423,7 +427,13 @@ static int gmc_v10_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, amdgpu_ring_alloc(ring, kiq->pmf->invalidate_tlbs_size + 8); kiq->pmf->kiq_invalidate_tlbs(ring, pasid, flush_type, all_hub); - amdgpu_fence_emit_polling(ring, &seq); + r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT); + if (r) { + amdgpu_ring_undo(ring); + spin_unlock(&adev->gfx.kiq.ring_lock); + return -ETIME; + } + amdgpu_ring_commit(ring); spin_unlock(&adev->gfx.kiq.ring_lock); r = amdgpu_fence_wait_polling(ring, seq, adev->usec_timeout); @@ -676,17 +686,23 @@ static void gmc_v10_0_vram_gtt_location(struct amdgpu_device *adev, */ static int gmc_v10_0_mc_init(struct amdgpu_device *adev) { - /* Could aper size report 0 ? */ - adev->gmc.aper_base = pci_resource_start(adev->pdev, 0); - adev->gmc.aper_size = pci_resource_len(adev->pdev, 0); + int r; /* size in MB on si */ adev->gmc.mc_vram_size = adev->nbio.funcs->get_memsize(adev) * 1024ULL * 1024ULL; adev->gmc.real_vram_size = adev->gmc.mc_vram_size; - adev->gmc.visible_vram_size = adev->gmc.aper_size; + + if (!(adev->flags & AMD_IS_APU)) { + r = amdgpu_device_resize_fb_bar(adev); + if (r) + return r; + } + adev->gmc.aper_base = pci_resource_start(adev->pdev, 0); + adev->gmc.aper_size = pci_resource_len(adev->pdev, 0); /* In case the PCI BAR is larger than the actual amount of vram */ + adev->gmc.visible_vram_size = adev->gmc.aper_size; if (adev->gmc.visible_vram_size > adev->gmc.real_vram_size) adev->gmc.visible_vram_size = adev->gmc.real_vram_size; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c index b205039350b6..a75e472b4a81 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c @@ -61,17 +61,6 @@ MODULE_FIRMWARE("amdgpu/si58_mc.bin"); #define MC_SEQ_MISC0__MT__HBM 0x60000000 #define MC_SEQ_MISC0__MT__DDR3 0xB0000000 - -static const u32 crtc_offsets[6] = -{ - SI_CRTC0_REGISTER_OFFSET, - SI_CRTC1_REGISTER_OFFSET, - SI_CRTC2_REGISTER_OFFSET, - SI_CRTC3_REGISTER_OFFSET, - SI_CRTC4_REGISTER_OFFSET, - SI_CRTC5_REGISTER_OFFSET -}; - static void gmc_v6_0_mc_stop(struct amdgpu_device *adev) { u32 blackout; @@ -858,7 +847,7 @@ static int gmc_v6_0_sw_init(void *handle) r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(44)); if (r) { - dev_warn(adev->dev, "amdgpu: No suitable DMA available.\n"); + dev_warn(adev->dev, "No suitable DMA available.\n"); return r; } adev->need_swiotlb = drm_need_swiotlb(44); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c index 9da9596a3638..bcd4baecfe11 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c @@ -762,6 +762,7 @@ static void gmc_v7_0_gart_disable(struct amdgpu_device *adev) * @adev: amdgpu_device pointer * @status: VM_CONTEXT1_PROTECTION_FAULT_STATUS register value * @addr: VM_CONTEXT1_PROTECTION_FAULT_ADDR register value + * @mc_client: VM_CONTEXT1_PROTECTION_FAULT_MCCLIENT register value * * Print human readable fault information (CIK). */ @@ -1019,7 +1020,7 @@ static int gmc_v7_0_sw_init(void *handle) r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(40)); if (r) { - pr_warn("amdgpu: No suitable DMA available\n"); + pr_warn("No suitable DMA available\n"); return r; } adev->need_swiotlb = drm_need_swiotlb(40); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index 27d83204fa2b..26976e50e2a2 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -1005,6 +1005,7 @@ static void gmc_v8_0_gart_disable(struct amdgpu_device *adev) * @adev: amdgpu_device pointer * @status: VM_CONTEXT1_PROTECTION_FAULT_STATUS register value * @addr: VM_CONTEXT1_PROTECTION_FAULT_ADDR register value + * @mc_client: VM_CONTEXT1_PROTECTION_FAULT_MCCLIENT register value * * Print human readable fault information (VI). */ @@ -1144,7 +1145,7 @@ static int gmc_v8_0_sw_init(void *handle) r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(40)); if (r) { - pr_warn("amdgpu: No suitable DMA available\n"); + pr_warn("No suitable DMA available\n"); return r; } adev->need_swiotlb = drm_need_swiotlb(40); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 8606f877478f..11e93a82131d 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -362,6 +362,9 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev, dev_err(adev->dev, "VM_L2_PROTECTION_FAULT_STATUS:0x%08X\n", status); + dev_err(adev->dev, "\t Faulty UTCL2 client ID: 0x%lx\n", + REG_GET_FIELD(status, + VM_L2_PROTECTION_FAULT_STATUS, CID)); dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n", REG_GET_FIELD(status, VM_L2_PROTECTION_FAULT_STATUS, MORE_FAULTS)); @@ -438,9 +441,8 @@ static bool gmc_v9_0_use_invalidate_semaphore(struct amdgpu_device *adev, return ((vmhub == AMDGPU_MMHUB_0 || vmhub == AMDGPU_MMHUB_1) && (!amdgpu_sriov_vf(adev)) && - (!(adev->asic_type == CHIP_RAVEN && - adev->rev_id < 0x8 && - adev->pdev->device == 0x15d8))); + (!(!(adev->apu_flags & AMD_APU_IS_RAVEN2) && + (adev->apu_flags & AMD_APU_IS_PICASSO)))); } static bool gmc_v9_0_get_atc_vmid_pasid_mapping_info(struct amdgpu_device *adev, @@ -618,7 +620,13 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, pasid, 2, all_hub); kiq->pmf->kiq_invalidate_tlbs(ring, pasid, flush_type, all_hub); - amdgpu_fence_emit_polling(ring, &seq); + r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT); + if (r) { + amdgpu_ring_undo(ring); + spin_unlock(&adev->gfx.kiq.ring_lock); + return -ETIME; + } + amdgpu_ring_commit(ring); spin_unlock(&adev->gfx.kiq.ring_lock); r = amdgpu_fence_wait_polling(ring, seq, adev->usec_timeout); diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c index 0debfd9f428c..b10c95cad9a2 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c @@ -480,7 +480,8 @@ int jpeg_v1_0_sw_init(void *handle) ring = &adev->jpeg.inst->ring_dec; sprintf(ring->name, "jpeg_dec"); - r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0); + r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, + 0, AMDGPU_RING_PRIO_DEFAULT); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c index 6173951db7b4..e67d09cb1b03 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c @@ -106,7 +106,8 @@ static int jpeg_v2_0_sw_init(void *handle) ring->use_doorbell = true; ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1; sprintf(ring->name, "jpeg_dec"); - r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0); + r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, + 0, AMDGPU_RING_PRIO_DEFAULT); if (r) return r; @@ -169,14 +170,11 @@ static int jpeg_v2_0_hw_init(void *handle) static int jpeg_v2_0_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec; if (adev->jpeg.cur_state != AMD_PG_STATE_GATE && RREG32_SOC15(JPEG, 0, mmUVD_JRBC_STATUS)) jpeg_v2_0_set_powergating_state(adev, AMD_PG_STATE_GATE); - ring->sched.ready = false; - return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c index c04c2078a7c1..713c32560445 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c @@ -118,7 +118,8 @@ static int jpeg_v2_5_sw_init(void *handle) ring->use_doorbell = true; ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1 + 8 * i; sprintf(ring->name, "jpeg_dec_%d", i); - r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst[i].irq, 0); + r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst[i].irq, + 0, AMDGPU_RING_PRIO_DEFAULT); if (r) return r; @@ -267,7 +268,6 @@ static void jpeg_v2_5_disable_clock_gating(struct amdgpu_device* adev, int inst) data = RREG32_SOC15(JPEG, inst, mmJPEG_CGC_GATE); data &= ~(JPEG_CGC_GATE__JPEG_DEC_MASK | JPEG_CGC_GATE__JPEG2_DEC_MASK - | JPEG_CGC_GATE__JPEG_ENC_MASK | JPEG_CGC_GATE__JMCIF_MASK | JPEG_CGC_GATE__JRBBM_MASK); WREG32_SOC15(JPEG, inst, mmJPEG_CGC_GATE, data); diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c index 396c2a624de0..405767208a4d 100644..100755 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c @@ -96,7 +96,7 @@ static void mmhub_v1_0_init_system_aperture_regs(struct amdgpu_device *adev) WREG32_SOC15(MMHUB, 0, mmMC_VM_SYSTEM_APERTURE_LOW_ADDR, min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18); - if (adev->asic_type == CHIP_RAVEN && adev->rev_id >= 0x8) + if (adev->apu_flags & AMD_APU_IS_RAVEN2) /* * Raven2 has a HW issue that it is unable to use the vram which * is out of MC_VM_SYSTEM_APERTURE_HIGH_ADDR. So here is the @@ -690,7 +690,8 @@ static const struct soc15_reg_entry mmhub_v1_0_edc_cnt_regs[] = { { SOC15_REG_ENTRY(MMHUB, 0, mmMMEA1_EDC_CNT2_VG20), 0, 0, 0}, }; -static int mmhub_v1_0_get_ras_error_count(const struct soc15_reg_entry *reg, +static int mmhub_v1_0_get_ras_error_count(struct amdgpu_device *adev, + const struct soc15_reg_entry *reg, uint32_t value, uint32_t *sec_count, uint32_t *ded_count) { uint32_t i; @@ -704,7 +705,8 @@ static int mmhub_v1_0_get_ras_error_count(const struct soc15_reg_entry *reg, mmhub_v1_0_ras_fields[i].sec_count_mask) >> mmhub_v1_0_ras_fields[i].sec_count_shift; if (sec_cnt) { - DRM_INFO("MMHUB SubBlock %s, SEC %d\n", + dev_info(adev->dev, + "MMHUB SubBlock %s, SEC %d\n", mmhub_v1_0_ras_fields[i].name, sec_cnt); *sec_count += sec_cnt; @@ -714,7 +716,8 @@ static int mmhub_v1_0_get_ras_error_count(const struct soc15_reg_entry *reg, mmhub_v1_0_ras_fields[i].ded_count_mask) >> mmhub_v1_0_ras_fields[i].ded_count_shift; if (ded_cnt) { - DRM_INFO("MMHUB SubBlock %s, DED %d\n", + dev_info(adev->dev, + "MMHUB SubBlock %s, DED %d\n", mmhub_v1_0_ras_fields[i].name, ded_cnt); *ded_count += ded_cnt; @@ -739,7 +742,8 @@ static void mmhub_v1_0_query_ras_error_count(struct amdgpu_device *adev, reg_value = RREG32(SOC15_REG_ENTRY_OFFSET(mmhub_v1_0_edc_cnt_regs[i])); if (reg_value) - mmhub_v1_0_get_ras_error_count(&mmhub_v1_0_edc_cnt_regs[i], + mmhub_v1_0_get_ras_error_count(adev, + &mmhub_v1_0_edc_cnt_regs[i], reg_value, &sec_count, &ded_count); } diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h index 37dbe0f2142f..83b453f5d717 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h @@ -26,7 +26,7 @@ #define AI_MAILBOX_POLL_ACK_TIMEDOUT 500 #define AI_MAILBOX_POLL_MSG_TIMEDOUT 12000 -#define AI_MAILBOX_POLL_FLR_TIMEDOUT 500 +#define AI_MAILBOX_POLL_FLR_TIMEDOUT 5000 enum idh_request { IDH_REQ_GPU_INIT_ACCESS = 1, @@ -46,7 +46,8 @@ enum idh_event { IDH_SUCCESS, IDH_FAIL, IDH_QUERY_ALIVE, - IDH_EVENT_MAX + + IDH_TEXT_MESSAGE = 255, }; extern const struct amdgpu_virt_ops xgpu_ai_virt_ops; diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c index 237fa5e16b7c..ce2bf1fb79ed 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c @@ -30,7 +30,6 @@ #include "navi10_ih.h" #include "soc15_common.h" #include "mxgpu_nv.h" -#include "mxgpu_ai.h" static void xgpu_nv_mailbox_send_ack(struct amdgpu_device *adev) { @@ -53,8 +52,7 @@ static void xgpu_nv_mailbox_set_valid(struct amdgpu_device *adev, bool val) */ static enum idh_event xgpu_nv_mailbox_peek_msg(struct amdgpu_device *adev) { - return RREG32_NO_KIQ(SOC15_REG_OFFSET(NBIO, 0, - mmBIF_BX_PF_MAILBOX_MSGBUF_RCV_DW0)); + return RREG32_NO_KIQ(mmMAILBOX_MSGBUF_RCV_DW0); } @@ -63,8 +61,7 @@ static int xgpu_nv_mailbox_rcv_msg(struct amdgpu_device *adev, { u32 reg; - reg = RREG32_NO_KIQ(SOC15_REG_OFFSET(NBIO, 0, - mmBIF_BX_PF_MAILBOX_MSGBUF_RCV_DW0)); + reg = RREG32_NO_KIQ(mmMAILBOX_MSGBUF_RCV_DW0); if (reg != event) return -ENOENT; @@ -110,7 +107,6 @@ static int xgpu_nv_poll_msg(struct amdgpu_device *adev, enum idh_event event) timeout -= 10; } while (timeout > 1); - pr_err("Doesn't get msg:%d from pf, error=%d\n", event, r); return -ETIME; } @@ -118,7 +114,6 @@ static int xgpu_nv_poll_msg(struct amdgpu_device *adev, enum idh_event event) static void xgpu_nv_mailbox_trans_msg (struct amdgpu_device *adev, enum idh_request req, u32 data1, u32 data2, u32 data3) { - u32 reg; int r; uint8_t trn; @@ -137,19 +132,10 @@ static void xgpu_nv_mailbox_trans_msg (struct amdgpu_device *adev, } } while (trn); - reg = RREG32_NO_KIQ(SOC15_REG_OFFSET(NBIO, 0, - mmBIF_BX_PF_MAILBOX_MSGBUF_TRN_DW0)); - reg = REG_SET_FIELD(reg, BIF_BX_PF_MAILBOX_MSGBUF_TRN_DW0, - MSGBUF_DATA, req); - WREG32_NO_KIQ(SOC15_REG_OFFSET(NBIO, 0, mmBIF_BX_PF_MAILBOX_MSGBUF_TRN_DW0), - reg); - WREG32_NO_KIQ(SOC15_REG_OFFSET(NBIO, 0, mmBIF_BX_PF_MAILBOX_MSGBUF_TRN_DW1), - data1); - WREG32_NO_KIQ(SOC15_REG_OFFSET(NBIO, 0, mmBIF_BX_PF_MAILBOX_MSGBUF_TRN_DW2), - data2); - WREG32_NO_KIQ(SOC15_REG_OFFSET(NBIO, 0, mmBIF_BX_PF_MAILBOX_MSGBUF_TRN_DW3), - data3); - + WREG32_NO_KIQ(mmMAILBOX_MSGBUF_TRN_DW0, req); + WREG32_NO_KIQ(mmMAILBOX_MSGBUF_TRN_DW1, data1); + WREG32_NO_KIQ(mmMAILBOX_MSGBUF_TRN_DW2, data2); + WREG32_NO_KIQ(mmMAILBOX_MSGBUF_TRN_DW3, data3); xgpu_nv_mailbox_set_valid(adev, true); /* start to poll ack */ @@ -164,23 +150,48 @@ static int xgpu_nv_send_access_requests(struct amdgpu_device *adev, enum idh_request req) { int r; + enum idh_event event = -1; xgpu_nv_mailbox_trans_msg(adev, req, 0, 0, 0); - /* start to check msg if request is idh_req_gpu_init_access */ - if (req == IDH_REQ_GPU_INIT_ACCESS || - req == IDH_REQ_GPU_FINI_ACCESS || - req == IDH_REQ_GPU_RESET_ACCESS) { - r = xgpu_nv_poll_msg(adev, IDH_READY_TO_ACCESS_GPU); + switch (req) { + case IDH_REQ_GPU_INIT_ACCESS: + case IDH_REQ_GPU_FINI_ACCESS: + case IDH_REQ_GPU_RESET_ACCESS: + event = IDH_READY_TO_ACCESS_GPU; + break; + case IDH_REQ_GPU_INIT_DATA: + event = IDH_REQ_GPU_INIT_DATA_READY; + break; + default: + break; + } + + if (event != -1) { + r = xgpu_nv_poll_msg(adev, event); if (r) { - pr_err("Doesn't get READY_TO_ACCESS_GPU from pf, give up\n"); - return r; + if (req != IDH_REQ_GPU_INIT_DATA) { + pr_err("Doesn't get msg:%d from pf, error=%d\n", event, r); + return r; + } + else /* host doesn't support REQ_GPU_INIT_DATA handshake */ + adev->virt.req_init_data_ver = 0; + } else { + if (req == IDH_REQ_GPU_INIT_DATA) + { + adev->virt.req_init_data_ver = + RREG32_NO_KIQ(mmMAILBOX_MSGBUF_RCV_DW1); + + /* assume V1 in case host doesn't set version number */ + if (adev->virt.req_init_data_ver < 1) + adev->virt.req_init_data_ver = 1; + } } + /* Retrieve checksum from mailbox2 */ if (req == IDH_REQ_GPU_INIT_ACCESS || req == IDH_REQ_GPU_RESET_ACCESS) { adev->virt.fw_reserve.checksum_key = - RREG32_NO_KIQ(SOC15_REG_OFFSET(NBIO, 0, - mmBIF_BX_PF_MAILBOX_MSGBUF_RCV_DW2)); + RREG32_NO_KIQ(mmMAILBOX_MSGBUF_RCV_DW2); } } @@ -213,6 +224,11 @@ static int xgpu_nv_release_full_gpu_access(struct amdgpu_device *adev, return r; } +static int xgpu_nv_request_init_data(struct amdgpu_device *adev) +{ + return xgpu_nv_send_access_requests(adev, IDH_REQ_GPU_INIT_DATA); +} + static int xgpu_nv_mailbox_ack_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) @@ -226,11 +242,14 @@ static int xgpu_nv_set_mailbox_ack_irq(struct amdgpu_device *adev, unsigned type, enum amdgpu_interrupt_state state) { - u32 tmp = RREG32_NO_KIQ(SOC15_REG_OFFSET(NBIO, 0, mmBIF_BX_PF_MAILBOX_INT_CNTL)); + u32 tmp = RREG32_NO_KIQ(mmMAILBOX_INT_CNTL); + + if (state == AMDGPU_IRQ_STATE_ENABLE) + tmp |= 2; + else + tmp &= ~2; - tmp = REG_SET_FIELD(tmp, BIF_BX_PF_MAILBOX_INT_CNTL, ACK_INT_EN, - (state == AMDGPU_IRQ_STATE_ENABLE) ? 1 : 0); - WREG32_NO_KIQ(SOC15_REG_OFFSET(NBIO, 0, mmBIF_BX_PF_MAILBOX_INT_CNTL), tmp); + WREG32_NO_KIQ(mmMAILBOX_INT_CNTL, tmp); return 0; } @@ -282,11 +301,14 @@ static int xgpu_nv_set_mailbox_rcv_irq(struct amdgpu_device *adev, unsigned type, enum amdgpu_interrupt_state state) { - u32 tmp = RREG32_NO_KIQ(SOC15_REG_OFFSET(NBIO, 0, mmBIF_BX_PF_MAILBOX_INT_CNTL)); + u32 tmp = RREG32_NO_KIQ(mmMAILBOX_INT_CNTL); + + if (state == AMDGPU_IRQ_STATE_ENABLE) + tmp |= 1; + else + tmp &= ~1; - tmp = REG_SET_FIELD(tmp, BIF_BX_PF_MAILBOX_INT_CNTL, VALID_INT_EN, - (state == AMDGPU_IRQ_STATE_ENABLE) ? 1 : 0); - WREG32_NO_KIQ(SOC15_REG_OFFSET(NBIO, 0, mmBIF_BX_PF_MAILBOX_INT_CNTL), tmp); + WREG32_NO_KIQ(mmMAILBOX_INT_CNTL, tmp); return 0; } @@ -378,6 +400,7 @@ void xgpu_nv_mailbox_put_irq(struct amdgpu_device *adev) const struct amdgpu_virt_ops xgpu_nv_virt_ops = { .req_full_gpu = xgpu_nv_request_full_gpu_access, .rel_full_gpu = xgpu_nv_release_full_gpu_access, + .req_init_data = xgpu_nv_request_init_data, .reset_gpu = xgpu_nv_request_reset, .wait_reset = NULL, .trans_msg = xgpu_nv_mailbox_trans_msg, diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.h b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.h index 99b15f6865cb..52605e14a1a5 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.h +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.h @@ -25,8 +25,32 @@ #define __MXGPU_NV_H__ #define NV_MAILBOX_POLL_ACK_TIMEDOUT 500 -#define NV_MAILBOX_POLL_MSG_TIMEDOUT 12000 -#define NV_MAILBOX_POLL_FLR_TIMEDOUT 500 +#define NV_MAILBOX_POLL_MSG_TIMEDOUT 6000 +#define NV_MAILBOX_POLL_FLR_TIMEDOUT 5000 + +enum idh_request { + IDH_REQ_GPU_INIT_ACCESS = 1, + IDH_REL_GPU_INIT_ACCESS, + IDH_REQ_GPU_FINI_ACCESS, + IDH_REL_GPU_FINI_ACCESS, + IDH_REQ_GPU_RESET_ACCESS, + IDH_REQ_GPU_INIT_DATA, + + IDH_LOG_VF_ERROR = 200, +}; + +enum idh_event { + IDH_CLR_MSG_BUF = 0, + IDH_READY_TO_ACCESS_GPU, + IDH_FLR_NOTIFICATION, + IDH_FLR_NOTIFICATION_CMPL, + IDH_SUCCESS, + IDH_FAIL, + IDH_QUERY_ALIVE, + IDH_REQ_GPU_INIT_DATA_READY, + + IDH_TEXT_MESSAGE = 255, +}; extern const struct amdgpu_virt_ops xgpu_nv_virt_ops; @@ -35,7 +59,21 @@ int xgpu_nv_mailbox_add_irq_id(struct amdgpu_device *adev); int xgpu_nv_mailbox_get_irq(struct amdgpu_device *adev); void xgpu_nv_mailbox_put_irq(struct amdgpu_device *adev); -#define NV_MAIBOX_CONTROL_TRN_OFFSET_BYTE (SOC15_REG_OFFSET(NBIO, 0, mmBIF_BX_PF_MAILBOX_CONTROL) * 4) -#define NV_MAIBOX_CONTROL_RCV_OFFSET_BYTE (SOC15_REG_OFFSET(NBIO, 0, mmBIF_BX_PF_MAILBOX_CONTROL) * 4 + 1) +#define mmMAILBOX_CONTROL 0xE5E + +#define NV_MAIBOX_CONTROL_TRN_OFFSET_BYTE (mmMAILBOX_CONTROL * 4) +#define NV_MAIBOX_CONTROL_RCV_OFFSET_BYTE (NV_MAIBOX_CONTROL_TRN_OFFSET_BYTE + 1) + +#define mmMAILBOX_MSGBUF_TRN_DW0 0xE56 +#define mmMAILBOX_MSGBUF_TRN_DW1 0xE57 +#define mmMAILBOX_MSGBUF_TRN_DW2 0xE58 +#define mmMAILBOX_MSGBUF_TRN_DW3 0xE59 + +#define mmMAILBOX_MSGBUF_RCV_DW0 0xE5A +#define mmMAILBOX_MSGBUF_RCV_DW1 0xE5B +#define mmMAILBOX_MSGBUF_RCV_DW2 0xE5C +#define mmMAILBOX_MSGBUF_RCV_DW3 0xE5D + +#define mmMAILBOX_INT_CNTL 0xE5F #endif diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.h b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.h index f13dc6cc158f..713ee66a4d3e 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.h +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.h @@ -43,7 +43,8 @@ enum idh_event { IDH_READY_TO_ACCESS_GPU, IDH_FLR_NOTIFICATION, IDH_FLR_NOTIFICATION_CMPL, - IDH_EVENT_MAX + + IDH_TEXT_MESSAGE = 255 }; extern const struct amdgpu_virt_ops xgpu_vi_virt_ops; diff --git a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c index e08245a446fc..f97857ed3c7e 100644 --- a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c @@ -49,8 +49,48 @@ static void navi10_ih_enable_interrupts(struct amdgpu_device *adev) ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, RB_ENABLE, 1); ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, ENABLE_INTR, 1); - WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL, ih_rb_cntl); + if (amdgpu_sriov_vf(adev) && adev->asic_type < CHIP_NAVI10) { + if (psp_reg_program(&adev->psp, PSP_REG_IH_RB_CNTL, ih_rb_cntl)) { + DRM_ERROR("PSP program IH_RB_CNTL failed!\n"); + return; + } + } else { + WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL, ih_rb_cntl); + } + adev->irq.ih.enabled = true; + + if (adev->irq.ih1.ring_size) { + ih_rb_cntl = RREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING1); + ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL_RING1, + RB_ENABLE, 1); + if (amdgpu_sriov_vf(adev) && adev->asic_type < CHIP_NAVI10) { + if (psp_reg_program(&adev->psp, PSP_REG_IH_RB_CNTL_RING1, + ih_rb_cntl)) { + DRM_ERROR("program IH_RB_CNTL_RING1 failed!\n"); + return; + } + } else { + WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING1, ih_rb_cntl); + } + adev->irq.ih1.enabled = true; + } + + if (adev->irq.ih2.ring_size) { + ih_rb_cntl = RREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING2); + ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL_RING2, + RB_ENABLE, 1); + if (amdgpu_sriov_vf(adev) && adev->asic_type < CHIP_NAVI10) { + if (psp_reg_program(&adev->psp, PSP_REG_IH_RB_CNTL_RING2, + ih_rb_cntl)) { + DRM_ERROR("program IH_RB_CNTL_RING2 failed!\n"); + return; + } + } else { + WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING2, ih_rb_cntl); + } + adev->irq.ih2.enabled = true; + } } /** @@ -66,12 +106,61 @@ static void navi10_ih_disable_interrupts(struct amdgpu_device *adev) ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, RB_ENABLE, 0); ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, ENABLE_INTR, 0); - WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL, ih_rb_cntl); + if (amdgpu_sriov_vf(adev) && adev->asic_type < CHIP_NAVI10) { + if (psp_reg_program(&adev->psp, PSP_REG_IH_RB_CNTL, ih_rb_cntl)) { + DRM_ERROR("PSP program IH_RB_CNTL failed!\n"); + return; + } + } else { + WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL, ih_rb_cntl); + } + /* set rptr, wptr to 0 */ WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR, 0); WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR, 0); adev->irq.ih.enabled = false; adev->irq.ih.rptr = 0; + + if (adev->irq.ih1.ring_size) { + ih_rb_cntl = RREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING1); + ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL_RING1, + RB_ENABLE, 0); + if (amdgpu_sriov_vf(adev) && adev->asic_type < CHIP_NAVI10) { + if (psp_reg_program(&adev->psp, PSP_REG_IH_RB_CNTL_RING1, + ih_rb_cntl)) { + DRM_ERROR("program IH_RB_CNTL_RING1 failed!\n"); + return; + } + } else { + WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING1, ih_rb_cntl); + } + /* set rptr, wptr to 0 */ + WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR_RING1, 0); + WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_RING1, 0); + adev->irq.ih1.enabled = false; + adev->irq.ih1.rptr = 0; + } + + if (adev->irq.ih2.ring_size) { + ih_rb_cntl = RREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING2); + ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL_RING2, + RB_ENABLE, 0); + if (amdgpu_sriov_vf(adev) && adev->asic_type < CHIP_NAVI10) { + if (psp_reg_program(&adev->psp, PSP_REG_IH_RB_CNTL_RING2, + ih_rb_cntl)) { + DRM_ERROR("program IH_RB_CNTL_RING2 failed!\n"); + return; + } + } else { + WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING2, ih_rb_cntl); + } + /* set rptr, wptr to 0 */ + WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR_RING2, 0); + WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_RING2, 0); + adev->irq.ih2.enabled = false; + adev->irq.ih2.rptr = 0; + } + } static uint32_t navi10_ih_rb_cntl(struct amdgpu_ih_ring *ih, uint32_t ih_rb_cntl) @@ -97,6 +186,43 @@ static uint32_t navi10_ih_rb_cntl(struct amdgpu_ih_ring *ih, uint32_t ih_rb_cntl return ih_rb_cntl; } +static uint32_t navi10_ih_doorbell_rptr(struct amdgpu_ih_ring *ih) +{ + u32 ih_doorbell_rtpr = 0; + + if (ih->use_doorbell) { + ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr, + IH_DOORBELL_RPTR, OFFSET, + ih->doorbell_index); + ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr, + IH_DOORBELL_RPTR, + ENABLE, 1); + } else { + ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr, + IH_DOORBELL_RPTR, + ENABLE, 0); + } + return ih_doorbell_rtpr; +} + +static void navi10_ih_reroute_ih(struct amdgpu_device *adev) +{ + uint32_t tmp; + + /* Reroute to IH ring 1 for VMC */ + WREG32_SOC15(OSSSYS, 0, mmIH_CLIENT_CFG_INDEX, 0x12); + tmp = RREG32_SOC15(OSSSYS, 0, mmIH_CLIENT_CFG_DATA); + tmp = REG_SET_FIELD(tmp, IH_CLIENT_CFG_DATA, CLIENT_TYPE, 1); + tmp = REG_SET_FIELD(tmp, IH_CLIENT_CFG_DATA, RING_ID, 1); + WREG32_SOC15(OSSSYS, 0, mmIH_CLIENT_CFG_DATA, tmp); + + /* Reroute IH ring 1 for UMC */ + WREG32_SOC15(OSSSYS, 0, mmIH_CLIENT_CFG_INDEX, 0x1B); + tmp = RREG32_SOC15(OSSSYS, 0, mmIH_CLIENT_CFG_DATA); + tmp = REG_SET_FIELD(tmp, IH_CLIENT_CFG_DATA, RING_ID, 1); + WREG32_SOC15(OSSSYS, 0, mmIH_CLIENT_CFG_DATA, tmp); +} + /** * navi10_ih_irq_init - init and enable the interrupt ring * @@ -111,7 +237,7 @@ static uint32_t navi10_ih_rb_cntl(struct amdgpu_ih_ring *ih, uint32_t ih_rb_cntl static int navi10_ih_irq_init(struct amdgpu_device *adev) { struct amdgpu_ih_ring *ih = &adev->irq.ih; - u32 ih_rb_cntl, ih_doorbell_rtpr, ih_chicken; + u32 ih_rb_cntl, ih_chicken; u32 tmp; /* disable irqs */ @@ -127,6 +253,15 @@ static int navi10_ih_irq_init(struct amdgpu_device *adev) ih_rb_cntl = navi10_ih_rb_cntl(ih, ih_rb_cntl); ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, RPTR_REARM, !!adev->irq.msi_enabled); + if (amdgpu_sriov_vf(adev) && adev->asic_type < CHIP_NAVI10) { + if (psp_reg_program(&adev->psp, PSP_REG_IH_RB_CNTL, ih_rb_cntl)) { + DRM_ERROR("PSP program IH_RB_CNTL failed!\n"); + return -ETIMEDOUT; + } + } else { + WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL, ih_rb_cntl); + } + navi10_ih_reroute_ih(adev); if (unlikely(adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT)) { if (ih->use_bus_addr) { @@ -137,8 +272,6 @@ static int navi10_ih_irq_init(struct amdgpu_device *adev) } } - WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL, ih_rb_cntl); - /* set the writeback address whether it's enabled or not */ WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_ADDR_LO, lower_32_bits(ih->wptr_addr)); @@ -149,22 +282,68 @@ static int navi10_ih_irq_init(struct amdgpu_device *adev) WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR, 0); WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR, 0); - ih_doorbell_rtpr = RREG32_SOC15(OSSSYS, 0, mmIH_DOORBELL_RPTR); - if (ih->use_doorbell) { - ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr, - IH_DOORBELL_RPTR, OFFSET, - ih->doorbell_index); - ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr, - IH_DOORBELL_RPTR, ENABLE, 1); - } else { - ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr, - IH_DOORBELL_RPTR, ENABLE, 0); - } - WREG32_SOC15(OSSSYS, 0, mmIH_DOORBELL_RPTR, ih_doorbell_rtpr); + WREG32_SOC15(OSSSYS, 0, mmIH_DOORBELL_RPTR, + navi10_ih_doorbell_rptr(ih)); adev->nbio.funcs->ih_doorbell_range(adev, ih->use_doorbell, ih->doorbell_index); + ih = &adev->irq.ih1; + if (ih->ring_size) { + WREG32_SOC15(OSSSYS, 0, mmIH_RB_BASE_RING1, ih->gpu_addr >> 8); + WREG32_SOC15(OSSSYS, 0, mmIH_RB_BASE_HI_RING1, + (ih->gpu_addr >> 40) & 0xff); + + ih_rb_cntl = RREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING1); + ih_rb_cntl = navi10_ih_rb_cntl(ih, ih_rb_cntl); + ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, + WPTR_OVERFLOW_ENABLE, 0); + ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, + RB_FULL_DRAIN_ENABLE, 1); + if (amdgpu_sriov_vf(adev) && adev->asic_type < CHIP_NAVI10) { + if (psp_reg_program(&adev->psp, PSP_REG_IH_RB_CNTL_RING1, + ih_rb_cntl)) { + DRM_ERROR("program IH_RB_CNTL_RING1 failed!\n"); + return -ETIMEDOUT; + } + } else { + WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING1, ih_rb_cntl); + } + /* set rptr, wptr to 0 */ + WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_RING1, 0); + WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR_RING1, 0); + + WREG32_SOC15(OSSSYS, 0, mmIH_DOORBELL_RPTR_RING1, + navi10_ih_doorbell_rptr(ih)); + } + + ih = &adev->irq.ih2; + if (ih->ring_size) { + WREG32_SOC15(OSSSYS, 0, mmIH_RB_BASE_RING2, ih->gpu_addr >> 8); + WREG32_SOC15(OSSSYS, 0, mmIH_RB_BASE_HI_RING2, + (ih->gpu_addr >> 40) & 0xff); + + ih_rb_cntl = RREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING2); + ih_rb_cntl = navi10_ih_rb_cntl(ih, ih_rb_cntl); + + if (amdgpu_sriov_vf(adev) && adev->asic_type < CHIP_NAVI10) { + if (psp_reg_program(&adev->psp, PSP_REG_IH_RB_CNTL_RING2, + ih_rb_cntl)) { + DRM_ERROR("program IH_RB_CNTL_RING2 failed!\n"); + return -ETIMEDOUT; + } + } else { + WREG32_SOC15(OSSSYS, 0, mmIH_RB_CNTL_RING2, ih_rb_cntl); + } + /* set rptr, wptr to 0 */ + WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_RING2, 0); + WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR_RING2, 0); + + WREG32_SOC15(OSSSYS, 0, mmIH_DOORBELL_RPTR_RING2, + navi10_ih_doorbell_rptr(ih)); + } + + tmp = RREG32_SOC15(OSSSYS, 0, mmIH_STORM_CLIENT_LIST_CNTL); tmp = REG_SET_FIELD(tmp, IH_STORM_CLIENT_LIST_CNTL, CLIENT18_IS_STORM_CLIENT, 1); @@ -217,7 +396,15 @@ static u32 navi10_ih_get_wptr(struct amdgpu_device *adev, if (!REG_GET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW)) goto out; - reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_WPTR); + if (ih == &adev->irq.ih) + reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_WPTR); + else if (ih == &adev->irq.ih1) + reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_WPTR_RING1); + else if (ih == &adev->irq.ih2) + reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_WPTR_RING2); + else + BUG(); + wptr = RREG32_NO_KIQ(reg); if (!REG_GET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW)) goto out; @@ -233,7 +420,15 @@ static u32 navi10_ih_get_wptr(struct amdgpu_device *adev, wptr, ih->rptr, tmp); ih->rptr = tmp; - reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_CNTL); + if (ih == &adev->irq.ih) + reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_CNTL); + else if (ih == &adev->irq.ih1) + reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_CNTL_RING1); + else if (ih == &adev->irq.ih2) + reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_RB_CNTL_RING2); + else + BUG(); + tmp = RREG32_NO_KIQ(reg); tmp = REG_SET_FIELD(tmp, IH_RB_CNTL, WPTR_OVERFLOW_CLEAR, 1); WREG32_NO_KIQ(reg, tmp); @@ -333,8 +528,52 @@ static void navi10_ih_set_rptr(struct amdgpu_device *adev, if (amdgpu_sriov_vf(adev)) navi10_ih_irq_rearm(adev, ih); - } else + } else if (ih == &adev->irq.ih) { WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR, ih->rptr); + } else if (ih == &adev->irq.ih1) { + WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR_RING1, ih->rptr); + } else if (ih == &adev->irq.ih2) { + WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR_RING2, ih->rptr); + } +} + +/** + * navi10_ih_self_irq - dispatch work for ring 1 and 2 + * + * @adev: amdgpu_device pointer + * @source: irq source + * @entry: IV with WPTR update + * + * Update the WPTR from the IV and schedule work to handle the entries. + */ +static int navi10_ih_self_irq(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + struct amdgpu_iv_entry *entry) +{ + uint32_t wptr = cpu_to_le32(entry->src_data[0]); + + switch (entry->ring_id) { + case 1: + *adev->irq.ih1.wptr_cpu = wptr; + schedule_work(&adev->irq.ih1_work); + break; + case 2: + *adev->irq.ih2.wptr_cpu = wptr; + schedule_work(&adev->irq.ih2_work); + break; + default: break; + } + return 0; +} + +static const struct amdgpu_irq_src_funcs navi10_ih_self_irq_funcs = { + .process = navi10_ih_self_irq, +}; + +static void navi10_ih_set_self_irq_funcs(struct amdgpu_device *adev) +{ + adev->irq.self_irq.num_types = 0; + adev->irq.self_irq.funcs = &navi10_ih_self_irq_funcs; } static int navi10_ih_early_init(void *handle) @@ -342,6 +581,7 @@ static int navi10_ih_early_init(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; navi10_ih_set_interrupt_funcs(adev); + navi10_ih_set_self_irq_funcs(adev); return 0; } @@ -351,6 +591,12 @@ static int navi10_ih_sw_init(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; bool use_bus_addr; + r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_IH, 0, + &adev->irq.self_irq); + + if (r) + return r; + /* use gpu virtual address for ih ring * until ih_checken is programmed to allow * use bus address for ih ring by psp bl */ @@ -363,6 +609,20 @@ static int navi10_ih_sw_init(void *handle) adev->irq.ih.use_doorbell = true; adev->irq.ih.doorbell_index = adev->doorbell_index.ih << 1; + r = amdgpu_ih_ring_init(adev, &adev->irq.ih1, PAGE_SIZE, true); + if (r) + return r; + + adev->irq.ih1.use_doorbell = true; + adev->irq.ih1.doorbell_index = (adev->doorbell_index.ih + 1) << 1; + + r = amdgpu_ih_ring_init(adev, &adev->irq.ih2, PAGE_SIZE, true); + if (r) + return r; + + adev->irq.ih2.use_doorbell = true; + adev->irq.ih2.doorbell_index = (adev->doorbell_index.ih + 2) << 1; + r = amdgpu_irq_init(adev); return r; @@ -373,6 +633,8 @@ static int navi10_ih_sw_fini(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; amdgpu_irq_fini(adev); + amdgpu_ih_ring_fini(adev, &adev->irq.ih2); + amdgpu_ih_ring_fini(adev, &adev->irq.ih1); amdgpu_ih_ring_fini(adev, &adev->irq.ih); return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/navi10_sdma_pkt_open.h b/drivers/gpu/drm/amd/amdgpu/navi10_sdma_pkt_open.h index 074a9a09c0a7..a5b60c9a2418 100644 --- a/drivers/gpu/drm/amd/amdgpu/navi10_sdma_pkt_open.h +++ b/drivers/gpu/drm/amd/amdgpu/navi10_sdma_pkt_open.h @@ -73,6 +73,22 @@ #define SDMA_OP_AQL_COPY 0 #define SDMA_OP_AQL_BARRIER_OR 0 +#define SDMA_GCR_RANGE_IS_PA (1 << 18) +#define SDMA_GCR_SEQ(x) (((x) & 0x3) << 16) +#define SDMA_GCR_GL2_WB (1 << 15) +#define SDMA_GCR_GL2_INV (1 << 14) +#define SDMA_GCR_GL2_DISCARD (1 << 13) +#define SDMA_GCR_GL2_RANGE(x) (((x) & 0x3) << 11) +#define SDMA_GCR_GL2_US (1 << 10) +#define SDMA_GCR_GL1_INV (1 << 9) +#define SDMA_GCR_GLV_INV (1 << 8) +#define SDMA_GCR_GLK_INV (1 << 7) +#define SDMA_GCR_GLK_WB (1 << 6) +#define SDMA_GCR_GLM_INV (1 << 5) +#define SDMA_GCR_GLM_WB (1 << 4) +#define SDMA_GCR_GL1_RANGE(x) (((x) & 0x3) << 2) +#define SDMA_GCR_GLI_INV(x) (((x) & 0x3) << 0) + /*define for op field*/ #define SDMA_PKT_HEADER_op_offset 0 #define SDMA_PKT_HEADER_op_mask 0x000000FF diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v2_3.c b/drivers/gpu/drm/amd/amdgpu/nbio_v2_3.c index f3a3fe746222..cbcf04578b99 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v2_3.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v2_3.c @@ -290,23 +290,6 @@ const struct nbio_hdp_flush_reg nbio_v2_3_hdp_flush_reg = { .ref_and_mask_sdma1 = BIF_BX_PF_GPU_HDP_FLUSH_DONE__SDMA1_MASK, }; -static void nbio_v2_3_detect_hw_virt(struct amdgpu_device *adev) -{ - uint32_t reg; - - reg = RREG32_SOC15(NBIO, 0, mmRCC_DEV0_EPF0_RCC_IOV_FUNC_IDENTIFIER); - if (reg & 1) - adev->virt.caps |= AMDGPU_SRIOV_CAPS_IS_VF; - - if (reg & 0x80000000) - adev->virt.caps |= AMDGPU_SRIOV_CAPS_ENABLE_IOV; - - if (!reg) { - if (is_virtual_machine()) /* passthrough mode exclus sriov mod */ - adev->virt.caps |= AMDGPU_PASSTHROUGH_MODE; - } -} - static void nbio_v2_3_init_registers(struct amdgpu_device *adev) { uint32_t def, data; @@ -338,6 +321,5 @@ const struct amdgpu_nbio_funcs nbio_v2_3_funcs = { .get_clockgating_state = nbio_v2_3_get_clockgating_state, .ih_control = nbio_v2_3_ih_control, .init_registers = nbio_v2_3_init_registers, - .detect_hw_virt = nbio_v2_3_detect_hw_virt, .remap_hdp_registers = nbio_v2_3_remap_hdp_registers, }; diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c b/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c index 635d9e1fc0a3..7b2fb050407d 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c @@ -241,23 +241,6 @@ const struct nbio_hdp_flush_reg nbio_v6_1_hdp_flush_reg = { .ref_and_mask_sdma1 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__SDMA1_MASK }; -static void nbio_v6_1_detect_hw_virt(struct amdgpu_device *adev) -{ - uint32_t reg; - - reg = RREG32_SOC15(NBIO, 0, mmRCC_PF_0_0_RCC_IOV_FUNC_IDENTIFIER); - if (reg & 1) - adev->virt.caps |= AMDGPU_SRIOV_CAPS_IS_VF; - - if (reg & 0x80000000) - adev->virt.caps |= AMDGPU_SRIOV_CAPS_ENABLE_IOV; - - if (!reg) { - if (is_virtual_machine()) /* passthrough mode exclus sriov mod */ - adev->virt.caps |= AMDGPU_PASSTHROUGH_MODE; - } -} - static void nbio_v6_1_init_registers(struct amdgpu_device *adev) { uint32_t def, data; @@ -294,5 +277,4 @@ const struct amdgpu_nbio_funcs nbio_v6_1_funcs = { .get_clockgating_state = nbio_v6_1_get_clockgating_state, .ih_control = nbio_v6_1_ih_control, .init_registers = nbio_v6_1_init_registers, - .detect_hw_virt = nbio_v6_1_detect_hw_virt, }; diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_0.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_0.c index d6cbf26074bc..d34628e113fc 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_0.c @@ -280,12 +280,6 @@ const struct nbio_hdp_flush_reg nbio_v7_0_hdp_flush_reg = { .ref_and_mask_sdma1 = GPU_HDP_FLUSH_DONE__SDMA1_MASK, }; -static void nbio_v7_0_detect_hw_virt(struct amdgpu_device *adev) -{ - if (is_virtual_machine()) /* passthrough mode exclus sriov mod */ - adev->virt.caps |= AMDGPU_PASSTHROUGH_MODE; -} - static void nbio_v7_0_init_registers(struct amdgpu_device *adev) { @@ -310,6 +304,5 @@ const struct amdgpu_nbio_funcs nbio_v7_0_funcs = { .get_clockgating_state = nbio_v7_0_get_clockgating_state, .ih_control = nbio_v7_0_ih_control, .init_registers = nbio_v7_0_init_registers, - .detect_hw_virt = nbio_v7_0_detect_hw_virt, .remap_hdp_registers = nbio_v7_0_remap_hdp_registers, }; diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c index 149d386590df..e629156173d3 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c @@ -185,7 +185,7 @@ static void nbio_v7_4_ih_doorbell_range(struct amdgpu_device *adev, if (use_doorbell) { ih_doorbell_range = REG_SET_FIELD(ih_doorbell_range, BIF_IH_DOORBELL_RANGE, OFFSET, doorbell_index); - ih_doorbell_range = REG_SET_FIELD(ih_doorbell_range, BIF_IH_DOORBELL_RANGE, SIZE, 2); + ih_doorbell_range = REG_SET_FIELD(ih_doorbell_range, BIF_IH_DOORBELL_RANGE, SIZE, 4); } else ih_doorbell_range = REG_SET_FIELD(ih_doorbell_range, BIF_IH_DOORBELL_RANGE, SIZE, 0); @@ -292,23 +292,6 @@ const struct nbio_hdp_flush_reg nbio_v7_4_hdp_flush_reg = { .ref_and_mask_sdma7 = GPU_HDP_FLUSH_DONE__RSVD_ENG5_MASK, }; -static void nbio_v7_4_detect_hw_virt(struct amdgpu_device *adev) -{ - uint32_t reg; - - reg = RREG32_SOC15(NBIO, 0, mmRCC_IOV_FUNC_IDENTIFIER); - if (reg & 1) - adev->virt.caps |= AMDGPU_SRIOV_CAPS_IS_VF; - - if (reg & 0x80000000) - adev->virt.caps |= AMDGPU_SRIOV_CAPS_ENABLE_IOV; - - if (!reg) { - if (is_virtual_machine()) /* passthrough mode exclus sriov mod */ - adev->virt.caps |= AMDGPU_PASSTHROUGH_MODE; - } -} - static void nbio_v7_4_init_registers(struct amdgpu_device *adev) { @@ -340,14 +323,20 @@ static void nbio_v7_4_handle_ras_controller_intr_no_bifring(struct amdgpu_device obj->err_data.ce_count += err_data.ce_count; if (err_data.ce_count) - DRM_INFO("%ld correctable errors detected in %s block\n", - obj->err_data.ce_count, adev->nbio.ras_if->name); + dev_info(adev->dev, "%ld correctable hardware " + "errors detected in %s block, " + "no user action is needed.\n", + obj->err_data.ce_count, + adev->nbio.ras_if->name); if (err_data.ue_count) - DRM_INFO("%ld uncorrectable errors detected in %s block\n", - obj->err_data.ue_count, adev->nbio.ras_if->name); + dev_info(adev->dev, "%ld uncorrectable hardware " + "errors detected in %s block\n", + obj->err_data.ue_count, + adev->nbio.ras_if->name); - DRM_WARN("RAS controller interrupt triggered by NBIF error\n"); + dev_info(adev->dev, "RAS controller interrupt triggered " + "by NBIF error\n"); /* ras_controller_int is dedicated for nbif ras error, * not the global interrupt for sync flood @@ -561,7 +550,6 @@ const struct amdgpu_nbio_funcs nbio_v7_4_funcs = { .get_clockgating_state = nbio_v7_4_get_clockgating_state, .ih_control = nbio_v7_4_ih_control, .init_registers = nbio_v7_4_init_registers, - .detect_hw_virt = nbio_v7_4_detect_hw_virt, .remap_hdp_registers = nbio_v7_4_remap_hdp_registers, .handle_ras_controller_intr_no_bifring = nbio_v7_4_handle_ras_controller_intr_no_bifring, .handle_ras_err_event_athub_intr_no_bifring = nbio_v7_4_handle_ras_err_event_athub_intr_no_bifring, diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c index 033cbbca2072..6655dd2009b6 100644 --- a/drivers/gpu/drm/amd/amdgpu/nv.c +++ b/drivers/gpu/drm/amd/amdgpu/nv.c @@ -351,8 +351,6 @@ static int nv_asic_reset(struct amdgpu_device *adev) struct smu_context *smu = &adev->smu; if (nv_asic_reset_method(adev) == AMD_RESET_METHOD_BACO) { - if (!adev->in_suspend) - amdgpu_inc_vram_lost(adev); ret = smu_baco_enter(smu); if (ret) return ret; @@ -360,8 +358,6 @@ static int nv_asic_reset(struct amdgpu_device *adev) if (ret) return ret; } else { - if (!adev->in_suspend) - amdgpu_inc_vram_lost(adev); ret = nv_asic_mode1_reset(adev); } @@ -457,18 +453,19 @@ int nv_set_ip_blocks(struct amdgpu_device *adev) { int r; - /* Set IP register base before any HW register access */ - r = nv_reg_base_init(adev); - if (r) - return r; - adev->nbio.funcs = &nbio_v2_3_funcs; adev->nbio.hdp_flush_reg = &nbio_v2_3_hdp_flush_reg; - adev->nbio.funcs->detect_hw_virt(adev); - - if (amdgpu_sriov_vf(adev)) + if (amdgpu_sriov_vf(adev)) { adev->virt.ops = &xgpu_nv_virt_ops; + /* try send GPU_INIT_DATA request to host */ + amdgpu_virt_request_init_data(adev); + } + + /* Set IP register base before any HW register access */ + r = nv_reg_base_init(adev); + if (r) + return r; switch (adev->asic_type) { case CHIP_NAVI10: @@ -501,8 +498,7 @@ int nv_set_ip_blocks(struct amdgpu_device *adev) amdgpu_device_ip_block_add(adev, &gmc_v10_0_ip_block); amdgpu_device_ip_block_add(adev, &navi10_ih_ip_block); amdgpu_device_ip_block_add(adev, &psp_v11_0_ip_block); - if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP && - !amdgpu_sriov_vf(adev)) + if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) amdgpu_device_ip_block_add(adev, &smu_v11_0_ip_block); if (adev->enable_virtual_display || amdgpu_sriov_vf(adev)) amdgpu_device_ip_block_add(adev, &dce_virtual_ip_block); @@ -552,13 +548,6 @@ static bool nv_need_full_reset(struct amdgpu_device *adev) return true; } -static void nv_get_pcie_usage(struct amdgpu_device *adev, - uint64_t *count0, - uint64_t *count1) -{ - /*TODO*/ -} - static bool nv_need_reset_on_init(struct amdgpu_device *adev) { #if 0 @@ -633,7 +622,6 @@ static const struct amdgpu_asic_funcs nv_asic_funcs = .invalidate_hdp = &nv_invalidate_hdp, .init_doorbell_index = &nv_init_doorbell_index, .need_full_reset = &nv_need_full_reset, - .get_pcie_usage = &nv_get_pcie_usage, .need_reset_on_init = &nv_need_reset_on_init, .get_pcie_replay_count = &nv_get_pcie_replay_count, .supports_baco = &nv_asic_supports_baco, diff --git a/drivers/gpu/drm/amd/amdgpu/nvd.h b/drivers/gpu/drm/amd/amdgpu/nvd.h index 1de984647dbb..fd6b58243b03 100644 --- a/drivers/gpu/drm/amd/amdgpu/nvd.h +++ b/drivers/gpu/drm/amd/amdgpu/nvd.h @@ -256,6 +256,54 @@ #define PACKET3_BLK_CNTX_UPDATE 0x53 #define PACKET3_INCR_UPDT_STATE 0x55 #define PACKET3_ACQUIRE_MEM 0x58 +/* 1. HEADER + * 2. COHER_CNTL [30:0] + * 2.1 ENGINE_SEL [31:31] + * 2. COHER_SIZE [31:0] + * 3. COHER_SIZE_HI [7:0] + * 4. COHER_BASE_LO [31:0] + * 5. COHER_BASE_HI [23:0] + * 7. POLL_INTERVAL [15:0] + * 8. GCR_CNTL [18:0] + */ +#define PACKET3_ACQUIRE_MEM_GCR_CNTL_GLI_INV(x) ((x) << 0) + /* + * 0:NOP + * 1:ALL + * 2:RANGE + * 3:FIRST_LAST + */ +#define PACKET3_ACQUIRE_MEM_GCR_CNTL_GL1_RANGE(x) ((x) << 2) + /* + * 0:ALL + * 1:reserved + * 2:RANGE + * 3:FIRST_LAST + */ +#define PACKET3_ACQUIRE_MEM_GCR_CNTL_GLM_WB(x) ((x) << 4) +#define PACKET3_ACQUIRE_MEM_GCR_CNTL_GLM_INV(x) ((x) << 5) +#define PACKET3_ACQUIRE_MEM_GCR_CNTL_GLK_WB(x) ((x) << 6) +#define PACKET3_ACQUIRE_MEM_GCR_CNTL_GLK_INV(x) ((x) << 7) +#define PACKET3_ACQUIRE_MEM_GCR_CNTL_GLV_INV(x) ((x) << 8) +#define PACKET3_ACQUIRE_MEM_GCR_CNTL_GL1_INV(x) ((x) << 9) +#define PACKET3_ACQUIRE_MEM_GCR_CNTL_GL2_US(x) ((x) << 10) +#define PACKET3_ACQUIRE_MEM_GCR_CNTL_GL2_RANGE(x) ((x) << 11) + /* + * 0:ALL + * 1:VOL + * 2:RANGE + * 3:FIRST_LAST + */ +#define PACKET3_ACQUIRE_MEM_GCR_CNTL_GL2_DISCARD(x) ((x) << 13) +#define PACKET3_ACQUIRE_MEM_GCR_CNTL_GL2_INV(x) ((x) << 14) +#define PACKET3_ACQUIRE_MEM_GCR_CNTL_GL2_WB(x) ((x) << 15) +#define PACKET3_ACQUIRE_MEM_GCR_CNTL_SEQ(x) ((x) << 16) + /* + * 0: PARALLEL + * 1: FORWARD + * 2: REVERSE + */ +#define PACKET3_ACQUIRE_MEM_GCR_RANGE_IS_PA (1 << 18) #define PACKET3_REWIND 0x59 #define PACKET3_INTERRUPT 0x5A #define PACKET3_GEN_PDEPTE 0x5B @@ -306,6 +354,7 @@ #define PACKET3_GET_LOD_STATS 0x8E #define PACKET3_DRAW_MULTI_PREAMBLE 0x8F #define PACKET3_FRAME_CONTROL 0x90 +# define FRAME_TMZ (1 << 0) # define FRAME_CMD(x) ((x) << 28) /* * x=0: tmz_begin diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c index 7539104175e8..d7f92634eba2 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c @@ -50,15 +50,14 @@ static int psp_v10_0_init_microcode(struct psp_context *psp) const char *chip_name; char fw_name[30]; int err = 0; - const struct psp_firmware_header_v1_0 *hdr; const struct ta_firmware_header_v1_0 *ta_hdr; DRM_DEBUG("\n"); switch (adev->asic_type) { case CHIP_RAVEN: - if (adev->rev_id >= 0x8) + if (adev->apu_flags & AMD_APU_IS_RAVEN2) chip_name = "raven2"; - else if (adev->pdev->device == 0x15d8) + else if (adev->apu_flags & AMD_APU_IS_PICASSO) chip_name = "picasso"; else chip_name = "raven"; @@ -66,22 +65,10 @@ static int psp_v10_0_init_microcode(struct psp_context *psp) default: BUG(); } - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_asd.bin", chip_name); - err = request_firmware(&adev->psp.asd_fw, fw_name, adev->dev); + err = psp_init_asd_microcode(psp, chip_name); if (err) goto out; - err = amdgpu_ucode_validate(adev->psp.asd_fw); - if (err) - goto out; - - hdr = (const struct psp_firmware_header_v1_0 *)adev->psp.asd_fw->data; - adev->psp.asd_fw_version = le32_to_cpu(hdr->header.ucode_version); - adev->psp.asd_feature_version = le32_to_cpu(hdr->ucode_feature_version); - adev->psp.asd_ucode_size = le32_to_cpu(hdr->header.ucode_size_bytes); - adev->psp.asd_start_addr = (uint8_t *)hdr + - le32_to_cpu(hdr->header.ucode_array_offset_bytes); - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_ta.bin", chip_name); err = request_firmware(&adev->psp.ta_fw, fw_name, adev->dev); if (err) { @@ -126,8 +113,6 @@ out: dev_err(adev->dev, "psp v10.0: Failed to load firmware \"%s\"\n", fw_name); - release_firmware(adev->psp.asd_fw); - adev->psp.asd_fw = NULL; } return err; @@ -230,129 +215,6 @@ static int psp_v10_0_ring_destroy(struct psp_context *psp, return ret; } -static int -psp_v10_0_sram_map(struct amdgpu_device *adev, - unsigned int *sram_offset, unsigned int *sram_addr_reg_offset, - unsigned int *sram_data_reg_offset, - enum AMDGPU_UCODE_ID ucode_id) -{ - int ret = 0; - - switch(ucode_id) { -/* TODO: needs to confirm */ -#if 0 - case AMDGPU_UCODE_ID_SMC: - *sram_offset = 0; - *sram_addr_reg_offset = 0; - *sram_data_reg_offset = 0; - break; -#endif - - case AMDGPU_UCODE_ID_CP_CE: - *sram_offset = 0x0; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_CE_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_CE_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_CP_PFP: - *sram_offset = 0x0; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_PFP_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_PFP_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_CP_ME: - *sram_offset = 0x0; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_ME_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_ME_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_CP_MEC1: - *sram_offset = 0x10000; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_MEC_ME1_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_MEC_ME1_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_CP_MEC2: - *sram_offset = 0x10000; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_MEC2_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_MEC2_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_RLC_G: - *sram_offset = 0x2000; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmRLC_GPM_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmRLC_GPM_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_SDMA0: - *sram_offset = 0x0; - *sram_addr_reg_offset = SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_UCODE_DATA); - break; - -/* TODO: needs to confirm */ -#if 0 - case AMDGPU_UCODE_ID_SDMA1: - *sram_offset = ; - *sram_addr_reg_offset = ; - break; - - case AMDGPU_UCODE_ID_UVD: - *sram_offset = ; - *sram_addr_reg_offset = ; - break; - - case AMDGPU_UCODE_ID_VCE: - *sram_offset = ; - *sram_addr_reg_offset = ; - break; -#endif - - case AMDGPU_UCODE_ID_MAXIMUM: - default: - ret = -EINVAL; - break; - } - - return ret; -} - -static bool psp_v10_0_compare_sram_data(struct psp_context *psp, - struct amdgpu_firmware_info *ucode, - enum AMDGPU_UCODE_ID ucode_type) -{ - int err = 0; - unsigned int fw_sram_reg_val = 0; - unsigned int fw_sram_addr_reg_offset = 0; - unsigned int fw_sram_data_reg_offset = 0; - unsigned int ucode_size; - uint32_t *ucode_mem = NULL; - struct amdgpu_device *adev = psp->adev; - - err = psp_v10_0_sram_map(adev, &fw_sram_reg_val, &fw_sram_addr_reg_offset, - &fw_sram_data_reg_offset, ucode_type); - if (err) - return false; - - WREG32(fw_sram_addr_reg_offset, fw_sram_reg_val); - - ucode_size = ucode->ucode_size; - ucode_mem = (uint32_t *)ucode->kaddr; - while (!ucode_size) { - fw_sram_reg_val = RREG32(fw_sram_data_reg_offset); - - if (*ucode_mem != fw_sram_reg_val) - return false; - - ucode_mem++; - /* 4 bytes */ - ucode_size -= 4; - } - - return true; -} - - static int psp_v10_0_mode1_reset(struct psp_context *psp) { DRM_INFO("psp mode 1 reset not supported now! \n"); @@ -379,7 +241,6 @@ static const struct psp_funcs psp_v10_0_funcs = { .ring_create = psp_v10_0_ring_create, .ring_stop = psp_v10_0_ring_stop, .ring_destroy = psp_v10_0_ring_destroy, - .compare_sram_data = psp_v10_0_compare_sram_data, .mode1_reset = psp_v10_0_mode1_reset, .ring_get_wptr = psp_v10_0_ring_get_wptr, .ring_set_wptr = psp_v10_0_ring_set_wptr, diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c index 0afd610a1263..1de89cc3c355 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c @@ -75,10 +75,6 @@ static int psp_v11_0_init_microcode(struct psp_context *psp) const char *chip_name; char fw_name[30]; int err = 0; - const struct psp_firmware_header_v1_0 *sos_hdr; - const struct psp_firmware_header_v1_1 *sos_hdr_v1_1; - const struct psp_firmware_header_v1_2 *sos_hdr_v1_2; - const struct psp_firmware_header_v1_0 *asd_hdr; const struct ta_firmware_header_v1_0 *ta_hdr; DRM_DEBUG("\n"); @@ -103,66 +99,13 @@ static int psp_v11_0_init_microcode(struct psp_context *psp) BUG(); } - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sos.bin", chip_name); - err = request_firmware(&adev->psp.sos_fw, fw_name, adev->dev); + err = psp_init_sos_microcode(psp, chip_name); if (err) - goto out; + return err; - err = amdgpu_ucode_validate(adev->psp.sos_fw); + err = psp_init_asd_microcode(psp, chip_name); if (err) - goto out; - - sos_hdr = (const struct psp_firmware_header_v1_0 *)adev->psp.sos_fw->data; - amdgpu_ucode_print_psp_hdr(&sos_hdr->header); - - switch (sos_hdr->header.header_version_major) { - case 1: - adev->psp.sos_fw_version = le32_to_cpu(sos_hdr->header.ucode_version); - adev->psp.sos_feature_version = le32_to_cpu(sos_hdr->ucode_feature_version); - adev->psp.sos_bin_size = le32_to_cpu(sos_hdr->sos_size_bytes); - adev->psp.sys_bin_size = le32_to_cpu(sos_hdr->sos_offset_bytes); - adev->psp.sys_start_addr = (uint8_t *)sos_hdr + - le32_to_cpu(sos_hdr->header.ucode_array_offset_bytes); - adev->psp.sos_start_addr = (uint8_t *)adev->psp.sys_start_addr + - le32_to_cpu(sos_hdr->sos_offset_bytes); - if (sos_hdr->header.header_version_minor == 1) { - sos_hdr_v1_1 = (const struct psp_firmware_header_v1_1 *)adev->psp.sos_fw->data; - adev->psp.toc_bin_size = le32_to_cpu(sos_hdr_v1_1->toc_size_bytes); - adev->psp.toc_start_addr = (uint8_t *)adev->psp.sys_start_addr + - le32_to_cpu(sos_hdr_v1_1->toc_offset_bytes); - adev->psp.kdb_bin_size = le32_to_cpu(sos_hdr_v1_1->kdb_size_bytes); - adev->psp.kdb_start_addr = (uint8_t *)adev->psp.sys_start_addr + - le32_to_cpu(sos_hdr_v1_1->kdb_offset_bytes); - } - if (sos_hdr->header.header_version_minor == 2) { - sos_hdr_v1_2 = (const struct psp_firmware_header_v1_2 *)adev->psp.sos_fw->data; - adev->psp.kdb_bin_size = le32_to_cpu(sos_hdr_v1_2->kdb_size_bytes); - adev->psp.kdb_start_addr = (uint8_t *)adev->psp.sys_start_addr + - le32_to_cpu(sos_hdr_v1_2->kdb_offset_bytes); - } - break; - default: - dev_err(adev->dev, - "Unsupported psp sos firmware\n"); - err = -EINVAL; - goto out; - } - - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_asd.bin", chip_name); - err = request_firmware(&adev->psp.asd_fw, fw_name, adev->dev); - if (err) - goto out1; - - err = amdgpu_ucode_validate(adev->psp.asd_fw); - if (err) - goto out1; - - asd_hdr = (const struct psp_firmware_header_v1_0 *)adev->psp.asd_fw->data; - adev->psp.asd_fw_version = le32_to_cpu(asd_hdr->header.ucode_version); - adev->psp.asd_feature_version = le32_to_cpu(asd_hdr->ucode_feature_version); - adev->psp.asd_ucode_size = le32_to_cpu(asd_hdr->header.ucode_size_bytes); - adev->psp.asd_start_addr = (uint8_t *)asd_hdr + - le32_to_cpu(asd_hdr->header.ucode_array_offset_bytes); + return err; switch (adev->asic_type) { case CHIP_VEGA20: @@ -194,6 +137,8 @@ static int psp_v11_0_init_microcode(struct psp_context *psp) case CHIP_NAVI10: case CHIP_NAVI14: case CHIP_NAVI12: + if (amdgpu_sriov_vf(adev)) + break; snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_ta.bin", chip_name); err = request_firmware(&adev->psp.ta_fw, fw_name, adev->dev); if (err) { @@ -229,15 +174,6 @@ static int psp_v11_0_init_microcode(struct psp_context *psp) out2: release_firmware(adev->psp.ta_fw); adev->psp.ta_fw = NULL; -out1: - release_firmware(adev->psp.asd_fw); - adev->psp.asd_fw = NULL; -out: - dev_err(adev->dev, - "psp v11.0: Failed to load firmware \"%s\"\n", fw_name); - release_firmware(adev->psp.sos_fw); - adev->psp.sos_fw = NULL; - return err; } @@ -283,11 +219,8 @@ static int psp_v11_0_bootloader_load_kdb(struct psp_context *psp) /* Check tOS sign of life register to confirm sys driver and sOS * are already been loaded. */ - if (psp_v11_0_is_sos_alive(psp)) { - psp->sos_fw_version = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_58); - dev_info(adev->dev, "sos fw version = 0x%x.\n", psp->sos_fw_version); + if (psp_v11_0_is_sos_alive(psp)) return 0; - } ret = psp_v11_0_wait_for_bootloader(psp); if (ret) @@ -319,11 +252,8 @@ static int psp_v11_0_bootloader_load_sysdrv(struct psp_context *psp) /* Check sOS sign of life register to confirm sys driver and sOS * are already been loaded. */ - if (psp_v11_0_is_sos_alive(psp)) { - psp->sos_fw_version = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_58); - dev_info(adev->dev, "sos fw version = 0x%x.\n", psp->sos_fw_version); + if (psp_v11_0_is_sos_alive(psp)) return 0; - } ret = psp_v11_0_wait_for_bootloader(psp); if (ret) @@ -446,13 +376,6 @@ static int psp_v11_0_ring_init(struct psp_context *psp, return 0; } -static bool psp_v11_0_support_vmr_ring(struct psp_context *psp) -{ - if (amdgpu_sriov_vf(psp->adev) && psp->sos_fw_version > 0x80045) - return true; - return false; -} - static int psp_v11_0_ring_stop(struct psp_context *psp, enum psp_ring_type ring_type) { @@ -460,7 +383,7 @@ static int psp_v11_0_ring_stop(struct psp_context *psp, struct amdgpu_device *adev = psp->adev; /* Write the ring destroy command*/ - if (psp_v11_0_support_vmr_ring(psp)) + if (amdgpu_sriov_vf(adev)) WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_101, GFX_CTRL_CMD_ID_DESTROY_GPCOM_RING); else @@ -471,7 +394,7 @@ static int psp_v11_0_ring_stop(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) */ - if (psp_v11_0_support_vmr_ring(psp)) + if (amdgpu_sriov_vf(adev)) ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), 0x80000000, 0x80000000, false); else @@ -489,7 +412,7 @@ static int psp_v11_0_ring_create(struct psp_context *psp, struct psp_ring *ring = &psp->km_ring; struct amdgpu_device *adev = psp->adev; - if (psp_v11_0_support_vmr_ring(psp)) { + if (amdgpu_sriov_vf(adev)) { ret = psp_v11_0_ring_stop(psp, ring_type); if (ret) { DRM_ERROR("psp_v11_0_ring_stop_sriov failed!\n"); @@ -567,138 +490,6 @@ static int psp_v11_0_ring_destroy(struct psp_context *psp, return ret; } -static int -psp_v11_0_sram_map(struct amdgpu_device *adev, - unsigned int *sram_offset, unsigned int *sram_addr_reg_offset, - unsigned int *sram_data_reg_offset, - enum AMDGPU_UCODE_ID ucode_id) -{ - int ret = 0; - - switch (ucode_id) { -/* TODO: needs to confirm */ -#if 0 - case AMDGPU_UCODE_ID_SMC: - *sram_offset = 0; - *sram_addr_reg_offset = 0; - *sram_data_reg_offset = 0; - break; -#endif - - case AMDGPU_UCODE_ID_CP_CE: - *sram_offset = 0x0; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_CE_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_CE_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_CP_PFP: - *sram_offset = 0x0; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_PFP_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_PFP_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_CP_ME: - *sram_offset = 0x0; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_ME_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_ME_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_CP_MEC1: - *sram_offset = 0x10000; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_MEC_ME1_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_MEC_ME1_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_CP_MEC2: - *sram_offset = 0x10000; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_MEC2_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_MEC2_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_RLC_G: - *sram_offset = 0x2000; - if (adev->asic_type < CHIP_NAVI10) { - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmRLC_GPM_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmRLC_GPM_UCODE_DATA); - } else { - *sram_addr_reg_offset = adev->reg_offset[GC_HWIP][0][1] + mmRLC_GPM_UCODE_ADDR_NV10; - *sram_data_reg_offset = adev->reg_offset[GC_HWIP][0][1] + mmRLC_GPM_UCODE_DATA_NV10; - } - break; - - case AMDGPU_UCODE_ID_SDMA0: - *sram_offset = 0x0; - if (adev->asic_type < CHIP_NAVI10) { - *sram_addr_reg_offset = SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_UCODE_DATA); - } else { - *sram_addr_reg_offset = adev->reg_offset[GC_HWIP][0][1] + mmSDMA0_UCODE_ADDR_NV10; - *sram_data_reg_offset = adev->reg_offset[GC_HWIP][0][1] + mmSDMA0_UCODE_DATA_NV10; - } - break; - -/* TODO: needs to confirm */ -#if 0 - case AMDGPU_UCODE_ID_SDMA1: - *sram_offset = ; - *sram_addr_reg_offset = ; - break; - - case AMDGPU_UCODE_ID_UVD: - *sram_offset = ; - *sram_addr_reg_offset = ; - break; - - case AMDGPU_UCODE_ID_VCE: - *sram_offset = ; - *sram_addr_reg_offset = ; - break; -#endif - - case AMDGPU_UCODE_ID_MAXIMUM: - default: - ret = -EINVAL; - break; - } - - return ret; -} - -static bool psp_v11_0_compare_sram_data(struct psp_context *psp, - struct amdgpu_firmware_info *ucode, - enum AMDGPU_UCODE_ID ucode_type) -{ - int err = 0; - unsigned int fw_sram_reg_val = 0; - unsigned int fw_sram_addr_reg_offset = 0; - unsigned int fw_sram_data_reg_offset = 0; - unsigned int ucode_size; - uint32_t *ucode_mem = NULL; - struct amdgpu_device *adev = psp->adev; - - err = psp_v11_0_sram_map(adev, &fw_sram_reg_val, &fw_sram_addr_reg_offset, - &fw_sram_data_reg_offset, ucode_type); - if (err) - return false; - - WREG32(fw_sram_addr_reg_offset, fw_sram_reg_val); - - ucode_size = ucode->ucode_size; - ucode_mem = (uint32_t *)ucode->kaddr; - while (ucode_size) { - fw_sram_reg_val = RREG32(fw_sram_data_reg_offset); - - if (*ucode_mem != fw_sram_reg_val) - return false; - - ucode_mem++; - /* 4 bytes */ - ucode_size -= 4; - } - - return true; -} - static int psp_v11_0_mode1_reset(struct psp_context *psp) { int ret; @@ -733,181 +524,6 @@ static int psp_v11_0_mode1_reset(struct psp_context *psp) return 0; } -/* TODO: Fill in follow functions once PSP firmware interface for XGMI is ready. - * For now, return success and hack the hive_id so high level code can - * start testing - */ -static int psp_v11_0_xgmi_get_topology_info(struct psp_context *psp, - int number_devices, struct psp_xgmi_topology_info *topology) -{ - struct ta_xgmi_shared_memory *xgmi_cmd; - struct ta_xgmi_cmd_get_topology_info_input *topology_info_input; - struct ta_xgmi_cmd_get_topology_info_output *topology_info_output; - int i; - int ret; - - if (!topology || topology->num_nodes > TA_XGMI__MAX_CONNECTED_NODES) - return -EINVAL; - - xgmi_cmd = (struct ta_xgmi_shared_memory*)psp->xgmi_context.xgmi_shared_buf; - memset(xgmi_cmd, 0, sizeof(struct ta_xgmi_shared_memory)); - - /* Fill in the shared memory with topology information as input */ - topology_info_input = &xgmi_cmd->xgmi_in_message.get_topology_info; - xgmi_cmd->cmd_id = TA_COMMAND_XGMI__GET_GET_TOPOLOGY_INFO; - topology_info_input->num_nodes = number_devices; - - for (i = 0; i < topology_info_input->num_nodes; i++) { - topology_info_input->nodes[i].node_id = topology->nodes[i].node_id; - topology_info_input->nodes[i].num_hops = topology->nodes[i].num_hops; - topology_info_input->nodes[i].is_sharing_enabled = topology->nodes[i].is_sharing_enabled; - topology_info_input->nodes[i].sdma_engine = topology->nodes[i].sdma_engine; - } - - /* Invoke xgmi ta to get the topology information */ - ret = psp_xgmi_invoke(psp, TA_COMMAND_XGMI__GET_GET_TOPOLOGY_INFO); - if (ret) - return ret; - - /* Read the output topology information from the shared memory */ - topology_info_output = &xgmi_cmd->xgmi_out_message.get_topology_info; - topology->num_nodes = xgmi_cmd->xgmi_out_message.get_topology_info.num_nodes; - for (i = 0; i < topology->num_nodes; i++) { - topology->nodes[i].node_id = topology_info_output->nodes[i].node_id; - topology->nodes[i].num_hops = topology_info_output->nodes[i].num_hops; - topology->nodes[i].is_sharing_enabled = topology_info_output->nodes[i].is_sharing_enabled; - topology->nodes[i].sdma_engine = topology_info_output->nodes[i].sdma_engine; - } - - return 0; -} - -static int psp_v11_0_xgmi_set_topology_info(struct psp_context *psp, - int number_devices, struct psp_xgmi_topology_info *topology) -{ - struct ta_xgmi_shared_memory *xgmi_cmd; - struct ta_xgmi_cmd_get_topology_info_input *topology_info_input; - int i; - - if (!topology || topology->num_nodes > TA_XGMI__MAX_CONNECTED_NODES) - return -EINVAL; - - xgmi_cmd = (struct ta_xgmi_shared_memory*)psp->xgmi_context.xgmi_shared_buf; - memset(xgmi_cmd, 0, sizeof(struct ta_xgmi_shared_memory)); - - topology_info_input = &xgmi_cmd->xgmi_in_message.get_topology_info; - xgmi_cmd->cmd_id = TA_COMMAND_XGMI__SET_TOPOLOGY_INFO; - topology_info_input->num_nodes = number_devices; - - for (i = 0; i < topology_info_input->num_nodes; i++) { - topology_info_input->nodes[i].node_id = topology->nodes[i].node_id; - topology_info_input->nodes[i].num_hops = topology->nodes[i].num_hops; - topology_info_input->nodes[i].is_sharing_enabled = 1; - topology_info_input->nodes[i].sdma_engine = topology->nodes[i].sdma_engine; - } - - /* Invoke xgmi ta to set topology information */ - return psp_xgmi_invoke(psp, TA_COMMAND_XGMI__SET_TOPOLOGY_INFO); -} - -static int psp_v11_0_xgmi_get_hive_id(struct psp_context *psp, uint64_t *hive_id) -{ - struct ta_xgmi_shared_memory *xgmi_cmd; - int ret; - - xgmi_cmd = (struct ta_xgmi_shared_memory*)psp->xgmi_context.xgmi_shared_buf; - memset(xgmi_cmd, 0, sizeof(struct ta_xgmi_shared_memory)); - - xgmi_cmd->cmd_id = TA_COMMAND_XGMI__GET_HIVE_ID; - - /* Invoke xgmi ta to get hive id */ - ret = psp_xgmi_invoke(psp, xgmi_cmd->cmd_id); - if (ret) - return ret; - - *hive_id = xgmi_cmd->xgmi_out_message.get_hive_id.hive_id; - - return 0; -} - -static int psp_v11_0_xgmi_get_node_id(struct psp_context *psp, uint64_t *node_id) -{ - struct ta_xgmi_shared_memory *xgmi_cmd; - int ret; - - xgmi_cmd = (struct ta_xgmi_shared_memory*)psp->xgmi_context.xgmi_shared_buf; - memset(xgmi_cmd, 0, sizeof(struct ta_xgmi_shared_memory)); - - xgmi_cmd->cmd_id = TA_COMMAND_XGMI__GET_NODE_ID; - - /* Invoke xgmi ta to get the node id */ - ret = psp_xgmi_invoke(psp, xgmi_cmd->cmd_id); - if (ret) - return ret; - - *node_id = xgmi_cmd->xgmi_out_message.get_node_id.node_id; - - return 0; -} - -static int psp_v11_0_ras_trigger_error(struct psp_context *psp, - struct ta_ras_trigger_error_input *info) -{ - struct ta_ras_shared_memory *ras_cmd; - int ret; - - if (!psp->ras.ras_initialized) - return -EINVAL; - - ras_cmd = (struct ta_ras_shared_memory *)psp->ras.ras_shared_buf; - memset(ras_cmd, 0, sizeof(struct ta_ras_shared_memory)); - - ras_cmd->cmd_id = TA_RAS_COMMAND__TRIGGER_ERROR; - ras_cmd->ras_in_message.trigger_error = *info; - - ret = psp_ras_invoke(psp, ras_cmd->cmd_id); - if (ret) - return -EINVAL; - - /* If err_event_athub occurs error inject was successful, however - return status from TA is no long reliable */ - if (amdgpu_ras_intr_triggered()) - return 0; - - return ras_cmd->ras_status; -} - -static int psp_v11_0_ras_cure_posion(struct psp_context *psp, uint64_t *mode_ptr) -{ -#if 0 - // not support yet. - struct ta_ras_shared_memory *ras_cmd; - int ret; - - if (!psp->ras.ras_initialized) - return -EINVAL; - - ras_cmd = (struct ta_ras_shared_memory *)psp->ras.ras_shared_buf; - memset(ras_cmd, 0, sizeof(struct ta_ras_shared_memory)); - - ras_cmd->cmd_id = TA_RAS_COMMAND__CURE_POISON; - ras_cmd->ras_in_message.cure_poison.mode_ptr = mode_ptr; - - ret = psp_ras_invoke(psp, ras_cmd->cmd_id); - if (ret) - return -EINVAL; - - return ras_cmd->ras_status; -#else - return -EINVAL; -#endif -} - -static int psp_v11_0_rlc_autoload_start(struct psp_context *psp) -{ - return psp_rlc_autoload_start(psp); -} - static int psp_v11_0_memory_training_send_msg(struct psp_context *psp, int msg) { int ret; @@ -1099,7 +715,7 @@ static uint32_t psp_v11_0_ring_get_wptr(struct psp_context *psp) uint32_t data; struct amdgpu_device *adev = psp->adev; - if (psp_v11_0_support_vmr_ring(psp)) + if (amdgpu_sriov_vf(adev)) data = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_102); else data = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_67); @@ -1111,7 +727,7 @@ static void psp_v11_0_ring_set_wptr(struct psp_context *psp, uint32_t value) { struct amdgpu_device *adev = psp->adev; - if (psp_v11_0_support_vmr_ring(psp)) { + if (amdgpu_sriov_vf(adev)) { WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_102, value); WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_101, GFX_CTRL_CMD_ID_CONSUME_CMD); } else @@ -1203,16 +819,7 @@ static const struct psp_funcs psp_v11_0_funcs = { .ring_create = psp_v11_0_ring_create, .ring_stop = psp_v11_0_ring_stop, .ring_destroy = psp_v11_0_ring_destroy, - .compare_sram_data = psp_v11_0_compare_sram_data, .mode1_reset = psp_v11_0_mode1_reset, - .xgmi_get_topology_info = psp_v11_0_xgmi_get_topology_info, - .xgmi_set_topology_info = psp_v11_0_xgmi_set_topology_info, - .xgmi_get_hive_id = psp_v11_0_xgmi_get_hive_id, - .xgmi_get_node_id = psp_v11_0_xgmi_get_node_id, - .support_vmr_ring = psp_v11_0_support_vmr_ring, - .ras_trigger_error = psp_v11_0_ras_trigger_error, - .ras_cure_posion = psp_v11_0_ras_cure_posion, - .rlc_autoload_start = psp_v11_0_rlc_autoload_start, .mem_training_init = psp_v11_0_memory_training_init, .mem_training_fini = psp_v11_0_memory_training_fini, .mem_training = psp_v11_0_memory_training, diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c index 58d8b6d732e8..6c9614f77d33 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c @@ -45,11 +45,7 @@ static int psp_v12_0_init_microcode(struct psp_context *psp) { struct amdgpu_device *adev = psp->adev; const char *chip_name; - char fw_name[30]; int err = 0; - const struct psp_firmware_header_v1_0 *asd_hdr; - - DRM_DEBUG("\n"); switch (adev->asic_type) { case CHIP_RENOIR: @@ -59,28 +55,7 @@ static int psp_v12_0_init_microcode(struct psp_context *psp) BUG(); } - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_asd.bin", chip_name); - err = request_firmware(&adev->psp.asd_fw, fw_name, adev->dev); - if (err) - goto out1; - - err = amdgpu_ucode_validate(adev->psp.asd_fw); - if (err) - goto out1; - - asd_hdr = (const struct psp_firmware_header_v1_0 *)adev->psp.asd_fw->data; - adev->psp.asd_fw_version = le32_to_cpu(asd_hdr->header.ucode_version); - adev->psp.asd_feature_version = le32_to_cpu(asd_hdr->ucode_feature_version); - adev->psp.asd_ucode_size = le32_to_cpu(asd_hdr->header.ucode_size_bytes); - adev->psp.asd_start_addr = (uint8_t *)asd_hdr + - le32_to_cpu(asd_hdr->header.ucode_array_offset_bytes); - - return 0; - -out1: - release_firmware(adev->psp.asd_fw); - adev->psp.asd_fw = NULL; - + err = psp_init_asd_microcode(psp, chip_name); return err; } @@ -95,11 +70,8 @@ static int psp_v12_0_bootloader_load_sysdrv(struct psp_context *psp) * are already been loaded. */ sol_reg = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81); - if (sol_reg) { - psp->sos_fw_version = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_58); - printk("sos fw version = 0x%x.\n", psp->sos_fw_version); + if (sol_reg) return 0; - } /* Wait for bootloader to signify that is ready having bit 31 of C2PMSG_35 set to 1 */ ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), @@ -228,13 +200,6 @@ static int psp_v12_0_ring_init(struct psp_context *psp, return 0; } -static bool psp_v12_0_support_vmr_ring(struct psp_context *psp) -{ - if (amdgpu_sriov_vf(psp->adev) && psp->sos_fw_version > 0x80045) - return true; - return false; -} - static int psp_v12_0_ring_create(struct psp_context *psp, enum psp_ring_type ring_type) { @@ -243,7 +208,7 @@ static int psp_v12_0_ring_create(struct psp_context *psp, struct psp_ring *ring = &psp->km_ring; struct amdgpu_device *adev = psp->adev; - if (psp_v12_0_support_vmr_ring(psp)) { + if (amdgpu_sriov_vf(psp->adev)) { /* Write low address of the ring to C2PMSG_102 */ psp_ring_reg = lower_32_bits(ring->ring_mem_mc_addr); WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_102, psp_ring_reg); @@ -295,7 +260,7 @@ static int psp_v12_0_ring_stop(struct psp_context *psp, struct amdgpu_device *adev = psp->adev; /* Write the ring destroy command*/ - if (psp_v12_0_support_vmr_ring(psp)) + if (amdgpu_sriov_vf(adev)) WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_101, GFX_CTRL_CMD_ID_DESTROY_GPCOM_RING); else @@ -306,7 +271,7 @@ static int psp_v12_0_ring_stop(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) */ - if (psp_v12_0_support_vmr_ring(psp)) + if (amdgpu_sriov_vf(adev)) ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), 0x80000000, 0x80000000, false); else @@ -334,128 +299,6 @@ static int psp_v12_0_ring_destroy(struct psp_context *psp, return ret; } -static int -psp_v12_0_sram_map(struct amdgpu_device *adev, - unsigned int *sram_offset, unsigned int *sram_addr_reg_offset, - unsigned int *sram_data_reg_offset, - enum AMDGPU_UCODE_ID ucode_id) -{ - int ret = 0; - - switch (ucode_id) { -/* TODO: needs to confirm */ -#if 0 - case AMDGPU_UCODE_ID_SMC: - *sram_offset = 0; - *sram_addr_reg_offset = 0; - *sram_data_reg_offset = 0; - break; -#endif - - case AMDGPU_UCODE_ID_CP_CE: - *sram_offset = 0x0; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_CE_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_CE_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_CP_PFP: - *sram_offset = 0x0; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_PFP_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_PFP_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_CP_ME: - *sram_offset = 0x0; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_ME_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_ME_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_CP_MEC1: - *sram_offset = 0x10000; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_MEC_ME1_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_MEC_ME1_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_CP_MEC2: - *sram_offset = 0x10000; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_MEC2_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_MEC2_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_RLC_G: - *sram_offset = 0x2000; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmRLC_GPM_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmRLC_GPM_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_SDMA0: - *sram_offset = 0x0; - *sram_addr_reg_offset = SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_UCODE_DATA); - break; - -/* TODO: needs to confirm */ -#if 0 - case AMDGPU_UCODE_ID_SDMA1: - *sram_offset = ; - *sram_addr_reg_offset = ; - break; - - case AMDGPU_UCODE_ID_UVD: - *sram_offset = ; - *sram_addr_reg_offset = ; - break; - - case AMDGPU_UCODE_ID_VCE: - *sram_offset = ; - *sram_addr_reg_offset = ; - break; -#endif - - case AMDGPU_UCODE_ID_MAXIMUM: - default: - ret = -EINVAL; - break; - } - - return ret; -} - -static bool psp_v12_0_compare_sram_data(struct psp_context *psp, - struct amdgpu_firmware_info *ucode, - enum AMDGPU_UCODE_ID ucode_type) -{ - int err = 0; - unsigned int fw_sram_reg_val = 0; - unsigned int fw_sram_addr_reg_offset = 0; - unsigned int fw_sram_data_reg_offset = 0; - unsigned int ucode_size; - uint32_t *ucode_mem = NULL; - struct amdgpu_device *adev = psp->adev; - - err = psp_v12_0_sram_map(adev, &fw_sram_reg_val, &fw_sram_addr_reg_offset, - &fw_sram_data_reg_offset, ucode_type); - if (err) - return false; - - WREG32(fw_sram_addr_reg_offset, fw_sram_reg_val); - - ucode_size = ucode->ucode_size; - ucode_mem = (uint32_t *)ucode->kaddr; - while (ucode_size) { - fw_sram_reg_val = RREG32(fw_sram_data_reg_offset); - - if (*ucode_mem != fw_sram_reg_val) - return false; - - ucode_mem++; - /* 4 bytes */ - ucode_size -= 4; - } - - return true; -} - static int psp_v12_0_mode1_reset(struct psp_context *psp) { int ret; @@ -495,7 +338,7 @@ static uint32_t psp_v12_0_ring_get_wptr(struct psp_context *psp) uint32_t data; struct amdgpu_device *adev = psp->adev; - if (psp_v12_0_support_vmr_ring(psp)) + if (amdgpu_sriov_vf(adev)) data = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_102); else data = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_67); @@ -507,7 +350,7 @@ static void psp_v12_0_ring_set_wptr(struct psp_context *psp, uint32_t value) { struct amdgpu_device *adev = psp->adev; - if (psp_v12_0_support_vmr_ring(psp)) { + if (amdgpu_sriov_vf(adev)) { WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_102, value); WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_101, GFX_CTRL_CMD_ID_CONSUME_CMD); } else @@ -522,7 +365,6 @@ static const struct psp_funcs psp_v12_0_funcs = { .ring_create = psp_v12_0_ring_create, .ring_stop = psp_v12_0_ring_stop, .ring_destroy = psp_v12_0_ring_destroy, - .compare_sram_data = psp_v12_0_compare_sram_data, .mode1_reset = psp_v12_0_mode1_reset, .ring_get_wptr = psp_v12_0_ring_get_wptr, .ring_set_wptr = psp_v12_0_ring_set_wptr, diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c index 735c43c7daab..f2e725f72d2f 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c @@ -50,9 +50,6 @@ MODULE_FIRMWARE("amdgpu/vega12_asd.bin"); #define smnMP1_FIRMWARE_FLAGS 0x3010028 -static uint32_t sos_old_versions[] = {1517616, 1510592, 1448594, 1446554}; - -static bool psp_v3_1_support_vmr_ring(struct psp_context *psp); static int psp_v3_1_ring_stop(struct psp_context *psp, enum psp_ring_type ring_type); @@ -60,9 +57,7 @@ static int psp_v3_1_init_microcode(struct psp_context *psp) { struct amdgpu_device *adev = psp->adev; const char *chip_name; - char fw_name[30]; int err = 0; - const struct psp_firmware_header_v1_0 *hdr; DRM_DEBUG("\n"); @@ -76,55 +71,15 @@ static int psp_v3_1_init_microcode(struct psp_context *psp) default: BUG(); } - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sos.bin", chip_name); - err = request_firmware(&adev->psp.sos_fw, fw_name, adev->dev); - if (err) - goto out; - - err = amdgpu_ucode_validate(adev->psp.sos_fw); + err = psp_init_sos_microcode(psp, chip_name); if (err) - goto out; - - hdr = (const struct psp_firmware_header_v1_0 *)adev->psp.sos_fw->data; - adev->psp.sos_fw_version = le32_to_cpu(hdr->header.ucode_version); - adev->psp.sos_feature_version = le32_to_cpu(hdr->ucode_feature_version); - adev->psp.sos_bin_size = le32_to_cpu(hdr->sos_size_bytes); - adev->psp.sys_bin_size = le32_to_cpu(hdr->header.ucode_size_bytes) - - le32_to_cpu(hdr->sos_size_bytes); - adev->psp.sys_start_addr = (uint8_t *)hdr + - le32_to_cpu(hdr->header.ucode_array_offset_bytes); - adev->psp.sos_start_addr = (uint8_t *)adev->psp.sys_start_addr + - le32_to_cpu(hdr->sos_offset_bytes); - - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_asd.bin", chip_name); - err = request_firmware(&adev->psp.asd_fw, fw_name, adev->dev); - if (err) - goto out; + return err; - err = amdgpu_ucode_validate(adev->psp.asd_fw); + err = psp_init_asd_microcode(psp, chip_name); if (err) - goto out; - - hdr = (const struct psp_firmware_header_v1_0 *)adev->psp.asd_fw->data; - adev->psp.asd_fw_version = le32_to_cpu(hdr->header.ucode_version); - adev->psp.asd_feature_version = le32_to_cpu(hdr->ucode_feature_version); - adev->psp.asd_ucode_size = le32_to_cpu(hdr->header.ucode_size_bytes); - adev->psp.asd_start_addr = (uint8_t *)hdr + - le32_to_cpu(hdr->header.ucode_array_offset_bytes); + return err; return 0; -out: - if (err) { - dev_err(adev->dev, - "psp v3.1: Failed to load firmware \"%s\"\n", - fw_name); - release_firmware(adev->psp.sos_fw); - adev->psp.sos_fw = NULL; - release_firmware(adev->psp.asd_fw); - adev->psp.asd_fw = NULL; - } - - return err; } static int psp_v3_1_bootloader_load_sysdrv(struct psp_context *psp) @@ -168,41 +123,19 @@ static int psp_v3_1_bootloader_load_sysdrv(struct psp_context *psp) return ret; } -static bool psp_v3_1_match_version(struct amdgpu_device *adev, uint32_t ver) -{ - int i; - - if (ver == adev->psp.sos_fw_version) - return true; - - /* - * Double check if the latest four legacy versions. - * If yes, it is still the right version. - */ - for (i = 0; i < ARRAY_SIZE(sos_old_versions); i++) { - if (sos_old_versions[i] == adev->psp.sos_fw_version) - return true; - } - - return false; -} - static int psp_v3_1_bootloader_load_sos(struct psp_context *psp) { int ret; unsigned int psp_gfxdrv_command_reg = 0; struct amdgpu_device *adev = psp->adev; - uint32_t sol_reg, ver; + uint32_t sol_reg; /* Check sOS sign of life register to confirm sys driver and sOS * are already been loaded. */ sol_reg = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81); - if (sol_reg) { - psp->sos_fw_version = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_58); - printk("sos fw version = 0x%x.\n", psp->sos_fw_version); + if (sol_reg) return 0; - } /* Wait for bootloader to signify that is ready having bit 31 of C2PMSG_35 set to 1 */ ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), @@ -227,11 +160,6 @@ static int psp_v3_1_bootloader_load_sos(struct psp_context *psp) ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_81), RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81), 0, true); - - ver = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_58); - if (!psp_v3_1_match_version(adev, ver)) - DRM_WARN("SOS version doesn't match\n"); - return ret; } @@ -302,7 +230,7 @@ static int psp_v3_1_ring_create(struct psp_context *psp, psp_v3_1_reroute_ih(psp); - if (psp_v3_1_support_vmr_ring(psp)) { + if (amdgpu_sriov_vf(adev)) { ret = psp_v3_1_ring_stop(psp, ring_type); if (ret) { DRM_ERROR("psp_v3_1_ring_stop_sriov failed!\n"); @@ -360,34 +288,26 @@ static int psp_v3_1_ring_stop(struct psp_context *psp, enum psp_ring_type ring_type) { int ret = 0; - unsigned int psp_ring_reg = 0; struct amdgpu_device *adev = psp->adev; - if (psp_v3_1_support_vmr_ring(psp)) { - /* Write the Destroy GPCOM ring command to C2PMSG_101 */ - psp_ring_reg = GFX_CTRL_CMD_ID_DESTROY_GPCOM_RING; - WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_101, psp_ring_reg); - - /* there might be handshake issue which needs delay */ - mdelay(20); - - /* Wait for response flag (bit 31) in C2PMSG_101 */ - ret = psp_wait_for(psp, - SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), - 0x80000000, 0x80000000, false); - } else { - /* Write the ring destroy command to C2PMSG_64 */ - psp_ring_reg = 3 << 16; - WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, psp_ring_reg); + /* Write the ring destroy command*/ + if (amdgpu_sriov_vf(adev)) + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_101, + GFX_CTRL_CMD_ID_DESTROY_GPCOM_RING); + else + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, + GFX_CTRL_CMD_ID_DESTROY_RINGS); - /* there might be handshake issue which needs delay */ - mdelay(20); + /* there might be handshake issue with hardware which needs delay */ + mdelay(20); - /* Wait for response flag (bit 31) in C2PMSG_64 */ - ret = psp_wait_for(psp, - SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); - } + /* Wait for response flag (bit 31) */ + if (amdgpu_sriov_vf(adev)) + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), + 0x80000000, 0x80000000, false); + else + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + 0x80000000, 0x80000000, false); return ret; } @@ -410,128 +330,6 @@ static int psp_v3_1_ring_destroy(struct psp_context *psp, return ret; } -static int -psp_v3_1_sram_map(struct amdgpu_device *adev, - unsigned int *sram_offset, unsigned int *sram_addr_reg_offset, - unsigned int *sram_data_reg_offset, - enum AMDGPU_UCODE_ID ucode_id) -{ - int ret = 0; - - switch(ucode_id) { -/* TODO: needs to confirm */ -#if 0 - case AMDGPU_UCODE_ID_SMC: - *sram_offset = 0; - *sram_addr_reg_offset = 0; - *sram_data_reg_offset = 0; - break; -#endif - - case AMDGPU_UCODE_ID_CP_CE: - *sram_offset = 0x0; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_CE_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_CE_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_CP_PFP: - *sram_offset = 0x0; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_PFP_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_PFP_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_CP_ME: - *sram_offset = 0x0; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_ME_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_ME_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_CP_MEC1: - *sram_offset = 0x10000; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_MEC_ME1_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_MEC_ME1_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_CP_MEC2: - *sram_offset = 0x10000; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_MEC2_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_HYP_MEC2_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_RLC_G: - *sram_offset = 0x2000; - *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmRLC_GPM_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmRLC_GPM_UCODE_DATA); - break; - - case AMDGPU_UCODE_ID_SDMA0: - *sram_offset = 0x0; - *sram_addr_reg_offset = SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_UCODE_ADDR); - *sram_data_reg_offset = SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_UCODE_DATA); - break; - -/* TODO: needs to confirm */ -#if 0 - case AMDGPU_UCODE_ID_SDMA1: - *sram_offset = ; - *sram_addr_reg_offset = ; - break; - - case AMDGPU_UCODE_ID_UVD: - *sram_offset = ; - *sram_addr_reg_offset = ; - break; - - case AMDGPU_UCODE_ID_VCE: - *sram_offset = ; - *sram_addr_reg_offset = ; - break; -#endif - - case AMDGPU_UCODE_ID_MAXIMUM: - default: - ret = -EINVAL; - break; - } - - return ret; -} - -static bool psp_v3_1_compare_sram_data(struct psp_context *psp, - struct amdgpu_firmware_info *ucode, - enum AMDGPU_UCODE_ID ucode_type) -{ - int err = 0; - unsigned int fw_sram_reg_val = 0; - unsigned int fw_sram_addr_reg_offset = 0; - unsigned int fw_sram_data_reg_offset = 0; - unsigned int ucode_size; - uint32_t *ucode_mem = NULL; - struct amdgpu_device *adev = psp->adev; - - err = psp_v3_1_sram_map(adev, &fw_sram_reg_val, &fw_sram_addr_reg_offset, - &fw_sram_data_reg_offset, ucode_type); - if (err) - return false; - - WREG32(fw_sram_addr_reg_offset, fw_sram_reg_val); - - ucode_size = ucode->ucode_size; - ucode_mem = (uint32_t *)ucode->kaddr; - while (ucode_size) { - fw_sram_reg_val = RREG32(fw_sram_data_reg_offset); - - if (*ucode_mem != fw_sram_reg_val) - return false; - - ucode_mem++; - /* 4 bytes */ - ucode_size -= 4; - } - - return true; -} - static bool psp_v3_1_smu_reload_quirk(struct psp_context *psp) { struct amdgpu_device *adev = psp->adev; @@ -575,20 +373,12 @@ static int psp_v3_1_mode1_reset(struct psp_context *psp) return 0; } -static bool psp_v3_1_support_vmr_ring(struct psp_context *psp) -{ - if (amdgpu_sriov_vf(psp->adev)) - return true; - - return false; -} - static uint32_t psp_v3_1_ring_get_wptr(struct psp_context *psp) { uint32_t data; struct amdgpu_device *adev = psp->adev; - if (psp_v3_1_support_vmr_ring(psp)) + if (amdgpu_sriov_vf(adev)) data = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_102); else data = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_67); @@ -599,7 +389,7 @@ static void psp_v3_1_ring_set_wptr(struct psp_context *psp, uint32_t value) { struct amdgpu_device *adev = psp->adev; - if (psp_v3_1_support_vmr_ring(psp)) { + if (amdgpu_sriov_vf(adev)) { WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_102, value); /* send interrupt to PSP for SRIOV ring write pointer update */ WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_101, @@ -616,10 +406,8 @@ static const struct psp_funcs psp_v3_1_funcs = { .ring_create = psp_v3_1_ring_create, .ring_stop = psp_v3_1_ring_stop, .ring_destroy = psp_v3_1_ring_destroy, - .compare_sram_data = psp_v3_1_compare_sram_data, .smu_reload_quirk = psp_v3_1_smu_reload_quirk, .mode1_reset = psp_v3_1_mode1_reset, - .support_vmr_ring = psp_v3_1_support_vmr_ring, .ring_get_wptr = psp_v3_1_ring_get_wptr, .ring_set_wptr = psp_v3_1_ring_set_wptr, }; diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c index 7d509a40076f..5f304d61999e 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c @@ -355,8 +355,6 @@ static void sdma_v2_4_gfx_stop(struct amdgpu_device *adev) ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 0); WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl); } - sdma0->sched.ready = false; - sdma1->sched.ready = false; } /** @@ -614,7 +612,8 @@ static int sdma_v2_4_ring_test_ib(struct amdgpu_ring *ring, long timeout) tmp = 0xCAFEDEAD; adev->wb.wb[index] = cpu_to_le32(tmp); memset(&ib, 0, sizeof(ib)); - r = amdgpu_ib_get(adev, NULL, 256, &ib); + r = amdgpu_ib_get(adev, NULL, 256, + AMDGPU_IB_POOL_DIRECT, &ib); if (r) goto err0; @@ -874,7 +873,8 @@ static int sdma_v2_4_sw_init(void *handle) &adev->sdma.trap_irq, (i == 0) ? AMDGPU_SDMA_IRQ_INSTANCE0 : - AMDGPU_SDMA_IRQ_INSTANCE1); + AMDGPU_SDMA_IRQ_INSTANCE1, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } @@ -1200,7 +1200,8 @@ static void sdma_v2_4_set_irq_funcs(struct amdgpu_device *adev) static void sdma_v2_4_emit_copy_buffer(struct amdgpu_ib *ib, uint64_t src_offset, uint64_t dst_offset, - uint32_t byte_count) + uint32_t byte_count, + bool tmz) { ib->ptr[ib->length_dw++] = SDMA_PKT_HEADER_OP(SDMA_OP_COPY) | SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_COPY_LINEAR); diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c index b6109a99fc43..c59f6f6f4c09 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c @@ -529,8 +529,6 @@ static void sdma_v3_0_gfx_stop(struct amdgpu_device *adev) ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 0); WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl); } - sdma0->sched.ready = false; - sdma1->sched.ready = false; } /** @@ -886,7 +884,8 @@ static int sdma_v3_0_ring_test_ib(struct amdgpu_ring *ring, long timeout) tmp = 0xCAFEDEAD; adev->wb.wb[index] = cpu_to_le32(tmp); memset(&ib, 0, sizeof(ib)); - r = amdgpu_ib_get(adev, NULL, 256, &ib); + r = amdgpu_ib_get(adev, NULL, 256, + AMDGPU_IB_POOL_DIRECT, &ib); if (r) goto err0; @@ -1158,7 +1157,8 @@ static int sdma_v3_0_sw_init(void *handle) &adev->sdma.trap_irq, (i == 0) ? AMDGPU_SDMA_IRQ_INSTANCE0 : - AMDGPU_SDMA_IRQ_INSTANCE1); + AMDGPU_SDMA_IRQ_INSTANCE1, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } @@ -1638,7 +1638,8 @@ static void sdma_v3_0_set_irq_funcs(struct amdgpu_device *adev) static void sdma_v3_0_emit_copy_buffer(struct amdgpu_ib *ib, uint64_t src_offset, uint64_t dst_offset, - uint32_t byte_count) + uint32_t byte_count, + bool tmz) { ib->ptr[ib->length_dw++] = SDMA_PKT_HEADER_OP(SDMA_OP_COPY) | SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_COPY_LINEAR); diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c index 5f3a5ee2a3f4..33501c6c7189 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c @@ -115,17 +115,21 @@ static const struct soc15_reg_golden golden_settings_sdma_4[] = { static const struct soc15_reg_golden golden_settings_sdma_vg10[] = { SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_GB_ADDR_CONFIG, 0x0018773f, 0x00104002), SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_GB_ADDR_CONFIG_READ, 0x0018773f, 0x00104002), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_UTCL1_TIMEOUT, 0xffffffff, 0x00010001), SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_CHICKEN_BITS, 0xfe931f07, 0x02831d07), SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_GB_ADDR_CONFIG, 0x0018773f, 0x00104002), - SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_GB_ADDR_CONFIG_READ, 0x0018773f, 0x00104002) + SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_GB_ADDR_CONFIG_READ, 0x0018773f, 0x00104002), + SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_UTCL1_TIMEOUT, 0xffffffff, 0x00010001), }; static const struct soc15_reg_golden golden_settings_sdma_vg12[] = { SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_GB_ADDR_CONFIG, 0x0018773f, 0x00104001), SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_GB_ADDR_CONFIG_READ, 0x0018773f, 0x00104001), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_UTCL1_TIMEOUT, 0xffffffff, 0x00010001), SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_CHICKEN_BITS, 0xfe931f07, 0x02831d07), SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_GB_ADDR_CONFIG, 0x0018773f, 0x00104001), - SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_GB_ADDR_CONFIG_READ, 0x0018773f, 0x00104001) + SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_GB_ADDR_CONFIG_READ, 0x0018773f, 0x00104001), + SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_UTCL1_TIMEOUT, 0xffffffff, 0x00010001), }; static const struct soc15_reg_golden golden_settings_sdma_4_1[] = { @@ -174,6 +178,7 @@ static const struct soc15_reg_golden golden_settings_sdma0_4_2[] = SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_RLC7_RB_RPTR_ADDR_LO, 0xfffffffd, 0x00000001), SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_RLC7_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_UTCL1_PAGE, 0x000003ff, 0x000003c0), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_UTCL1_TIMEOUT, 0xffffffff, 0x00010001), }; static const struct soc15_reg_golden golden_settings_sdma1_4_2[] = { @@ -203,6 +208,7 @@ static const struct soc15_reg_golden golden_settings_sdma1_4_2[] = { SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_RLC7_RB_RPTR_ADDR_LO, 0xfffffffd, 0x00000001), SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_RLC7_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_UTCL1_PAGE, 0x000003ff, 0x000003c0), + SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_UTCL1_TIMEOUT, 0xffffffff, 0x00010001), }; static const struct soc15_reg_golden golden_settings_sdma_rv1[] = @@ -222,27 +228,35 @@ static const struct soc15_reg_golden golden_settings_sdma_arct[] = SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_CHICKEN_BITS, 0xfe931f07, 0x02831f07), SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_GB_ADDR_CONFIG, 0x0000773f, 0x00004002), SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_GB_ADDR_CONFIG_READ, 0x0000773f, 0x00004002), + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_UTCL1_TIMEOUT, 0xffffffff, 0x00010001), SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_CHICKEN_BITS, 0xfe931f07, 0x02831f07), SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_GB_ADDR_CONFIG, 0x0000773f, 0x00004002), SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_GB_ADDR_CONFIG_READ, 0x0000773f, 0x00004002), + SOC15_REG_GOLDEN_VALUE(SDMA1, 0, mmSDMA1_UTCL1_TIMEOUT, 0xffffffff, 0x00010001), SOC15_REG_GOLDEN_VALUE(SDMA2, 0, mmSDMA2_CHICKEN_BITS, 0xfe931f07, 0x02831f07), SOC15_REG_GOLDEN_VALUE(SDMA2, 0, mmSDMA2_GB_ADDR_CONFIG, 0x0000773f, 0x00004002), SOC15_REG_GOLDEN_VALUE(SDMA2, 0, mmSDMA2_GB_ADDR_CONFIG_READ, 0x0000773f, 0x00004002), + SOC15_REG_GOLDEN_VALUE(SDMA2, 0, mmSDMA2_UTCL1_TIMEOUT, 0xffffffff, 0x00010001), SOC15_REG_GOLDEN_VALUE(SDMA3, 0, mmSDMA3_CHICKEN_BITS, 0xfe931f07, 0x02831f07), SOC15_REG_GOLDEN_VALUE(SDMA3, 0, mmSDMA3_GB_ADDR_CONFIG, 0x0000773f, 0x00004002), SOC15_REG_GOLDEN_VALUE(SDMA3, 0, mmSDMA3_GB_ADDR_CONFIG_READ, 0x0000773f, 0x00004002), + SOC15_REG_GOLDEN_VALUE(SDMA3, 0, mmSDMA3_UTCL1_TIMEOUT, 0xffffffff, 0x00010001), SOC15_REG_GOLDEN_VALUE(SDMA4, 0, mmSDMA4_CHICKEN_BITS, 0xfe931f07, 0x02831f07), SOC15_REG_GOLDEN_VALUE(SDMA4, 0, mmSDMA4_GB_ADDR_CONFIG, 0x0000773f, 0x00004002), SOC15_REG_GOLDEN_VALUE(SDMA4, 0, mmSDMA4_GB_ADDR_CONFIG_READ, 0x0000773f, 0x00004002), + SOC15_REG_GOLDEN_VALUE(SDMA4, 0, mmSDMA4_UTCL1_TIMEOUT, 0xffffffff, 0x00010001), SOC15_REG_GOLDEN_VALUE(SDMA5, 0, mmSDMA5_CHICKEN_BITS, 0xfe931f07, 0x02831f07), SOC15_REG_GOLDEN_VALUE(SDMA5, 0, mmSDMA5_GB_ADDR_CONFIG, 0x0000773f, 0x00004002), SOC15_REG_GOLDEN_VALUE(SDMA5, 0, mmSDMA5_GB_ADDR_CONFIG_READ, 0x0000773f, 0x00004002), + SOC15_REG_GOLDEN_VALUE(SDMA5, 0, mmSDMA5_UTCL1_TIMEOUT, 0xffffffff, 0x00010001), SOC15_REG_GOLDEN_VALUE(SDMA6, 0, mmSDMA6_CHICKEN_BITS, 0xfe931f07, 0x02831f07), SOC15_REG_GOLDEN_VALUE(SDMA6, 0, mmSDMA6_GB_ADDR_CONFIG, 0x0000773f, 0x00004002), SOC15_REG_GOLDEN_VALUE(SDMA6, 0, mmSDMA6_GB_ADDR_CONFIG_READ, 0x0000773f, 0x00004002), + SOC15_REG_GOLDEN_VALUE(SDMA6, 0, mmSDMA6_UTCL1_TIMEOUT, 0xffffffff, 0x00010001), SOC15_REG_GOLDEN_VALUE(SDMA7, 0, mmSDMA7_CHICKEN_BITS, 0xfe931f07, 0x02831f07), SOC15_REG_GOLDEN_VALUE(SDMA7, 0, mmSDMA7_GB_ADDR_CONFIG, 0x0000773f, 0x00004002), - SOC15_REG_GOLDEN_VALUE(SDMA7, 0, mmSDMA7_GB_ADDR_CONFIG_READ, 0x0000773f, 0x00004002) + SOC15_REG_GOLDEN_VALUE(SDMA7, 0, mmSDMA7_GB_ADDR_CONFIG_READ, 0x0000773f, 0x00004002), + SOC15_REG_GOLDEN_VALUE(SDMA7, 0, mmSDMA7_UTCL1_TIMEOUT, 0xffffffff, 0x00010001) }; static const struct soc15_reg_golden golden_settings_sdma_4_3[] = { @@ -472,7 +486,7 @@ static void sdma_v4_0_init_golden_registers(struct amdgpu_device *adev) soc15_program_register_sequence(adev, golden_settings_sdma_4_1, ARRAY_SIZE(golden_settings_sdma_4_1)); - if (adev->rev_id >= 8) + if (adev->apu_flags & AMD_APU_IS_RAVEN2) soc15_program_register_sequence(adev, golden_settings_sdma_rv2, ARRAY_SIZE(golden_settings_sdma_rv2)); @@ -561,9 +575,9 @@ static int sdma_v4_0_init_microcode(struct amdgpu_device *adev) chip_name = "vega20"; break; case CHIP_RAVEN: - if (adev->rev_id >= 8) + if (adev->apu_flags & AMD_APU_IS_RAVEN2) chip_name = "raven2"; - else if (adev->pdev->device == 0x15d8) + else if (adev->apu_flags & AMD_APU_IS_PICASSO) chip_name = "picasso"; else chip_name = "raven"; @@ -923,8 +937,6 @@ static void sdma_v4_0_gfx_stop(struct amdgpu_device *adev) ib_cntl = RREG32_SDMA(i, mmSDMA0_GFX_IB_CNTL); ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 0); WREG32_SDMA(i, mmSDMA0_GFX_IB_CNTL, ib_cntl); - - sdma[i]->sched.ready = false; } } @@ -971,8 +983,6 @@ static void sdma_v4_0_page_stop(struct amdgpu_device *adev) ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_PAGE_IB_CNTL, IB_ENABLE, 0); WREG32_SDMA(i, mmSDMA0_PAGE_IB_CNTL, ib_cntl); - - sdma[i]->sched.ready = false; } } @@ -1539,7 +1549,8 @@ static int sdma_v4_0_ring_test_ib(struct amdgpu_ring *ring, long timeout) tmp = 0xCAFEDEAD; adev->wb.wb[index] = cpu_to_le32(tmp); memset(&ib, 0, sizeof(ib)); - r = amdgpu_ib_get(adev, NULL, 256, &ib); + r = amdgpu_ib_get(adev, NULL, 256, + AMDGPU_IB_POOL_DIRECT, &ib); if (r) goto err0; @@ -1840,7 +1851,7 @@ static int sdma_v4_0_sw_init(void *handle) ring->ring_obj = NULL; ring->use_doorbell = true; - DRM_INFO("use_doorbell being set to: [%s]\n", + DRM_DEBUG("SDMA %d use_doorbell being set to: [%s]\n", i, ring->use_doorbell?"true":"false"); /* doorbell size is 2 dwords, get DWORD offset */ @@ -1848,7 +1859,8 @@ static int sdma_v4_0_sw_init(void *handle) sprintf(ring->name, "sdma%d", i); r = amdgpu_ring_init(adev, ring, 1024, &adev->sdma.trap_irq, - AMDGPU_SDMA_IRQ_INSTANCE0 + i); + AMDGPU_SDMA_IRQ_INSTANCE0 + i, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; @@ -1866,7 +1878,8 @@ static int sdma_v4_0_sw_init(void *handle) sprintf(ring->name, "page%d", i); r = amdgpu_ring_init(adev, ring, 1024, &adev->sdma.trap_irq, - AMDGPU_SDMA_IRQ_INSTANCE0 + i); + AMDGPU_SDMA_IRQ_INSTANCE0 + i, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } @@ -2445,10 +2458,12 @@ static void sdma_v4_0_set_irq_funcs(struct amdgpu_device *adev) static void sdma_v4_0_emit_copy_buffer(struct amdgpu_ib *ib, uint64_t src_offset, uint64_t dst_offset, - uint32_t byte_count) + uint32_t byte_count, + bool tmz) { ib->ptr[ib->length_dw++] = SDMA_PKT_HEADER_OP(SDMA_OP_COPY) | - SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_COPY_LINEAR); + SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_COPY_LINEAR) | + SDMA_PKT_COPY_LINEAR_HEADER_TMZ(tmz ? 1 : 0); ib->ptr[ib->length_dw++] = byte_count - 1; ib->ptr[ib->length_dw++] = 0; /* src/dst endian swap */ ib->ptr[ib->length_dw++] = lower_32_bits(src_offset); diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c index ebfd2cdf4e65..b544baf306f2 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c @@ -88,6 +88,29 @@ static const struct soc15_reg_golden golden_settings_sdma_5[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_UTCL1_PAGE, 0x00ffffff, 0x000c5c00) }; +static const struct soc15_reg_golden golden_settings_sdma_5_sriov[] = { + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_GFX_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_PAGE_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC0_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC1_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC2_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC3_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC4_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC5_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC6_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC7_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_GFX_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_PAGE_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC0_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC1_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC2_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC3_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC4_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC5_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC6_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC7_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), +}; + static const struct soc15_reg_golden golden_settings_sdma_nv10[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA0_RLC3_RB_WPTR_POLL_CNTL, 0x0000fff0, 0x00403000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSDMA1_RLC3_RB_WPTR_POLL_CNTL, 0x0000fff0, 0x00403000), @@ -141,9 +164,14 @@ static void sdma_v5_0_init_golden_registers(struct amdgpu_device *adev) (const u32)ARRAY_SIZE(golden_settings_sdma_nv14)); break; case CHIP_NAVI12: - soc15_program_register_sequence(adev, - golden_settings_sdma_5, - (const u32)ARRAY_SIZE(golden_settings_sdma_5)); + if (amdgpu_sriov_vf(adev)) + soc15_program_register_sequence(adev, + golden_settings_sdma_5_sriov, + (const u32)ARRAY_SIZE(golden_settings_sdma_5_sriov)); + else + soc15_program_register_sequence(adev, + golden_settings_sdma_5, + (const u32)ARRAY_SIZE(golden_settings_sdma_5)); soc15_program_register_sequence(adev, golden_settings_sdma_nv12, (const u32)ARRAY_SIZE(golden_settings_sdma_nv12)); @@ -382,6 +410,18 @@ static void sdma_v5_0_ring_emit_ib(struct amdgpu_ring *ring, unsigned vmid = AMDGPU_JOB_GET_VMID(job); uint64_t csa_mc_addr = amdgpu_sdma_get_csa_mc_addr(ring, vmid); + /* Invalidate L2, because if we don't do it, we might get stale cache + * lines from previous IBs. + */ + amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_GCR_REQ)); + amdgpu_ring_write(ring, 0); + amdgpu_ring_write(ring, (SDMA_GCR_GL2_INV | + SDMA_GCR_GL2_WB | + SDMA_GCR_GLM_INV | + SDMA_GCR_GLM_WB) << 16); + amdgpu_ring_write(ring, 0xffffff80); + amdgpu_ring_write(ring, 0xffff); + /* An IB packet must end on a 8 DW boundary--the next dword * must be on a 8-dword boundary. Our IB packet below is 6 * dwords long, thus add x number of NOPs, such that, in @@ -502,9 +542,6 @@ static void sdma_v5_0_gfx_stop(struct amdgpu_device *adev) ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 0); WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl); } - - sdma0->sched.ready = false; - sdma1->sched.ready = false; } /** @@ -529,7 +566,7 @@ static void sdma_v5_0_rlc_stop(struct amdgpu_device *adev) */ static void sdma_v5_0_ctx_switch_enable(struct amdgpu_device *adev, bool enable) { - u32 f32_cntl, phase_quantum = 0; + u32 f32_cntl = 0, phase_quantum = 0; int i; if (amdgpu_sdma_phase_quantum) { @@ -557,9 +594,12 @@ static void sdma_v5_0_ctx_switch_enable(struct amdgpu_device *adev, bool enable) } for (i = 0; i < adev->sdma.num_instances; i++) { - f32_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL)); - f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_CNTL, - AUTO_CTXSW_ENABLE, enable ? 1 : 0); + if (!amdgpu_sriov_vf(adev)) { + f32_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL)); + f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_CNTL, + AUTO_CTXSW_ENABLE, enable ? 1 : 0); + } + if (enable && amdgpu_sdma_phase_quantum) { WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_PHASE0_QUANTUM), phase_quantum); @@ -568,7 +608,8 @@ static void sdma_v5_0_ctx_switch_enable(struct amdgpu_device *adev, bool enable) WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_PHASE2_QUANTUM), phase_quantum); } - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL), f32_cntl); + if (!amdgpu_sriov_vf(adev)) + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL), f32_cntl); } } @@ -591,6 +632,9 @@ static void sdma_v5_0_enable(struct amdgpu_device *adev, bool enable) sdma_v5_0_rlc_stop(adev); } + if (amdgpu_sriov_vf(adev)) + return; + for (i = 0; i < adev->sdma.num_instances; i++) { f32_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_F32_CNTL)); f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_F32_CNTL, HALT, enable ? 0 : 1); @@ -623,7 +667,8 @@ static int sdma_v5_0_gfx_resume(struct amdgpu_device *adev) ring = &adev->sdma.instance[i].ring; wb_offset = (ring->rptr_offs * 4); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_SEM_WAIT_FAIL_TIMER_CNTL), 0); + if (!amdgpu_sriov_vf(adev)) + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_SEM_WAIT_FAIL_TIMER_CNTL), 0); /* Set ring buffer size in dwords */ rb_bufsz = order_base_2(ring->ring_size / 4); @@ -699,26 +744,28 @@ static int sdma_v5_0_gfx_resume(struct amdgpu_device *adev) /* set minor_ptr_update to 0 after wptr programed */ WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_MINOR_PTR_UPDATE), 0); - /* set utc l1 enable flag always to 1 */ - temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL)); - temp = REG_SET_FIELD(temp, SDMA0_CNTL, UTC_L1_ENABLE, 1); - - /* enable MCBP */ - temp = REG_SET_FIELD(temp, SDMA0_CNTL, MIDCMD_PREEMPT_ENABLE, 1); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL), temp); - - /* Set up RESP_MODE to non-copy addresses */ - temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL)); - temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, RESP_MODE, 3); - temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, REDO_DELAY, 9); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL), temp); - - /* program default cache read and write policy */ - temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE)); - /* clean read policy and write policy bits */ - temp &= 0xFF0FFF; - temp |= ((CACHE_READ_POLICY_L2__DEFAULT << 12) | (CACHE_WRITE_POLICY_L2__DEFAULT << 14)); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE), temp); + if (!amdgpu_sriov_vf(adev)) { + /* set utc l1 enable flag always to 1 */ + temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL)); + temp = REG_SET_FIELD(temp, SDMA0_CNTL, UTC_L1_ENABLE, 1); + + /* enable MCBP */ + temp = REG_SET_FIELD(temp, SDMA0_CNTL, MIDCMD_PREEMPT_ENABLE, 1); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL), temp); + + /* Set up RESP_MODE to non-copy addresses */ + temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL)); + temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, RESP_MODE, 3); + temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, REDO_DELAY, 9); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL), temp); + + /* program default cache read and write policy */ + temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE)); + /* clean read policy and write policy bits */ + temp &= 0xFF0FFF; + temp |= ((CACHE_READ_POLICY_L2__DEFAULT << 12) | (CACHE_WRITE_POLICY_L2__DEFAULT << 14)); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE), temp); + } if (!amdgpu_sriov_vf(adev)) { /* unhalt engine */ @@ -948,7 +995,8 @@ static int sdma_v5_0_ring_test_ib(struct amdgpu_ring *ring, long timeout) tmp = 0xCAFEDEAD; adev->wb.wb[index] = cpu_to_le32(tmp); memset(&ib, 0, sizeof(ib)); - r = amdgpu_ib_get(adev, NULL, 256, &ib); + r = amdgpu_ib_get(adev, NULL, 256, + AMDGPU_IB_POOL_DIRECT, &ib); if (r) { DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r); goto err0; @@ -1224,7 +1272,7 @@ static int sdma_v5_0_sw_init(void *handle) ring->ring_obj = NULL; ring->use_doorbell = true; - DRM_INFO("use_doorbell being set to: [%s]\n", + DRM_DEBUG("SDMA %d use_doorbell being set to: [%s]\n", i, ring->use_doorbell?"true":"false"); ring->doorbell_index = (i == 0) ? @@ -1236,7 +1284,8 @@ static int sdma_v5_0_sw_init(void *handle) &adev->sdma.trap_irq, (i == 0) ? AMDGPU_SDMA_IRQ_INSTANCE0 : - AMDGPU_SDMA_IRQ_INSTANCE1); + AMDGPU_SDMA_IRQ_INSTANCE1, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } @@ -1387,14 +1436,16 @@ static int sdma_v5_0_set_trap_irq_state(struct amdgpu_device *adev, { u32 sdma_cntl; - u32 reg_offset = (type == AMDGPU_SDMA_IRQ_INSTANCE0) ? - sdma_v5_0_get_reg_offset(adev, 0, mmSDMA0_CNTL) : - sdma_v5_0_get_reg_offset(adev, 1, mmSDMA0_CNTL); + if (!amdgpu_sriov_vf(adev)) { + u32 reg_offset = (type == AMDGPU_SDMA_IRQ_INSTANCE0) ? + sdma_v5_0_get_reg_offset(adev, 0, mmSDMA0_CNTL) : + sdma_v5_0_get_reg_offset(adev, 1, mmSDMA0_CNTL); - sdma_cntl = RREG32(reg_offset); - sdma_cntl = REG_SET_FIELD(sdma_cntl, SDMA0_CNTL, TRAP_ENABLE, - state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0); - WREG32(reg_offset, sdma_cntl); + sdma_cntl = RREG32(reg_offset); + sdma_cntl = REG_SET_FIELD(sdma_cntl, SDMA0_CNTL, TRAP_ENABLE, + state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0); + WREG32(reg_offset, sdma_cntl); + } return 0; } @@ -1595,7 +1646,7 @@ static const struct amdgpu_ring_funcs sdma_v5_0_ring_funcs = { SOC15_FLUSH_GPU_TLB_NUM_WREG * 3 + SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 6 * 2 + 10 + 10 + 10, /* sdma_v5_0_ring_emit_fence x3 for user fence, vm fence */ - .emit_ib_size = 7 + 6, /* sdma_v5_0_ring_emit_ib */ + .emit_ib_size = 5 + 7 + 6, /* sdma_v5_0_ring_emit_ib */ .emit_ib = sdma_v5_0_ring_emit_ib, .emit_fence = sdma_v5_0_ring_emit_fence, .emit_pipeline_sync = sdma_v5_0_ring_emit_pipeline_sync, @@ -1655,10 +1706,12 @@ static void sdma_v5_0_set_irq_funcs(struct amdgpu_device *adev) static void sdma_v5_0_emit_copy_buffer(struct amdgpu_ib *ib, uint64_t src_offset, uint64_t dst_offset, - uint32_t byte_count) + uint32_t byte_count, + bool tmz) { ib->ptr[ib->length_dw++] = SDMA_PKT_HEADER_OP(SDMA_OP_COPY) | - SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_COPY_LINEAR); + SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_COPY_LINEAR) | + SDMA_PKT_COPY_LINEAR_HEADER_TMZ(tmz ? 1 : 0); ib->ptr[ib->length_dw++] = byte_count - 1; ib->ptr[ib->length_dw++] = 0; /* src/dst endian swap */ ib->ptr[ib->length_dw++] = lower_32_bits(src_offset); diff --git a/drivers/gpu/drm/amd/amdgpu/si.c b/drivers/gpu/drm/amd/amdgpu/si.c index 4d415bfdb42f..153db3f763bc 100644 --- a/drivers/gpu/drm/amd/amdgpu/si.c +++ b/drivers/gpu/drm/amd/amdgpu/si.c @@ -1249,12 +1249,6 @@ static int si_set_uvd_clocks(struct amdgpu_device *adev, u32 vclk, u32 dclk) return 0; } -static void si_detect_hw_virtualization(struct amdgpu_device *adev) -{ - if (is_virtual_machine()) /* passthrough mode */ - adev->virt.caps |= AMDGPU_PASSTHROUGH_MODE; -} - static void si_flush_hdp(struct amdgpu_device *adev, struct amdgpu_ring *ring) { if (!ring || !ring->funcs->emit_wreg) { @@ -2165,8 +2159,6 @@ static const struct amdgpu_ip_block_version si_common_ip_block = int si_set_ip_blocks(struct amdgpu_device *adev) { - si_detect_hw_virtualization(adev); - switch (adev->asic_type) { case CHIP_VERDE: case CHIP_TAHITI: diff --git a/drivers/gpu/drm/amd/amdgpu/si_dma.c b/drivers/gpu/drm/amd/amdgpu/si_dma.c index 42d5601b6bf3..7d2bbcbe547b 100644 --- a/drivers/gpu/drm/amd/amdgpu/si_dma.c +++ b/drivers/gpu/drm/amd/amdgpu/si_dma.c @@ -124,7 +124,6 @@ static void si_dma_stop(struct amdgpu_device *adev) if (adev->mman.buffer_funcs_ring == ring) amdgpu_ttm_set_buffer_funcs_status(adev, false); - ring->sched.ready = false; } } @@ -267,7 +266,8 @@ static int si_dma_ring_test_ib(struct amdgpu_ring *ring, long timeout) tmp = 0xCAFEDEAD; adev->wb.wb[index] = cpu_to_le32(tmp); memset(&ib, 0, sizeof(ib)); - r = amdgpu_ib_get(adev, NULL, 256, &ib); + r = amdgpu_ib_get(adev, NULL, 256, + AMDGPU_IB_POOL_DIRECT, &ib); if (r) goto err0; @@ -504,7 +504,8 @@ static int si_dma_sw_init(void *handle) &adev->sdma.trap_irq, (i == 0) ? AMDGPU_SDMA_IRQ_INSTANCE0 : - AMDGPU_SDMA_IRQ_INSTANCE1); + AMDGPU_SDMA_IRQ_INSTANCE1, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } @@ -775,7 +776,8 @@ static void si_dma_set_irq_funcs(struct amdgpu_device *adev) static void si_dma_emit_copy_buffer(struct amdgpu_ib *ib, uint64_t src_offset, uint64_t dst_offset, - uint32_t byte_count) + uint32_t byte_count, + bool tmz) { ib->ptr[ib->length_dw++] = DMA_PACKET(DMA_PACKET_COPY, 1, 0, 0, byte_count); diff --git a/drivers/gpu/drm/amd/amdgpu/si_dpm.c b/drivers/gpu/drm/amd/amdgpu/si_dpm.c index 0860e85a2d35..c00ba4b23c9a 100644 --- a/drivers/gpu/drm/amd/amdgpu/si_dpm.c +++ b/drivers/gpu/drm/amd/amdgpu/si_dpm.c @@ -345,26 +345,6 @@ static const struct si_dte_data dte_data_tahiti = false }; -#if 0 -static const struct si_dte_data dte_data_tahiti_le = -{ - { 0x1E8480, 0x7A1200, 0x2160EC0, 0x3938700, 0 }, - { 0x7D, 0x7D, 0x4E4, 0xB00, 0 }, - 0x5, - 0xAFC8, - 0x64, - 0x32, - 1, - 0, - 0x10, - { 0x78, 0x7C, 0x82, 0x88, 0x8E, 0x94, 0x9A, 0xA0, 0xA6, 0xAC, 0xB0, 0xB4, 0xB8, 0xBC, 0xC0, 0xC4 }, - { 0x3938700, 0x3938700, 0x3938700, 0x3938700, 0x3938700, 0x3938700, 0x3938700, 0x3938700, 0x3938700, 0x3938700, 0x3938700, 0x3938700, 0x3938700, 0x3938700, 0x3938700, 0x3938700 }, - { 0x2AF8, 0x2AF8, 0x29BB, 0x27F9, 0x2637, 0x2475, 0x22B3, 0x20F1, 0x1F2F, 0x1D6D, 0x1734, 0x1414, 0x10F4, 0xDD4, 0xAB4, 0x794 }, - 85, - true -}; -#endif - static const struct si_dte_data dte_data_tahiti_pro = { { 0x1E8480, 0x3D0900, 0x989680, 0x2625A00, 0x0 }, diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c index a40499d51c93..c7c9e07962b9 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.c +++ b/drivers/gpu/drm/amd/amdgpu/soc15.c @@ -564,19 +564,16 @@ soc15_asic_reset_method(struct amdgpu_device *adev) static int soc15_asic_reset(struct amdgpu_device *adev) { /* original raven doesn't have full asic reset */ - if (adev->pdev->device == 0x15dd && adev->rev_id < 0x8) + if ((adev->apu_flags & AMD_APU_IS_RAVEN) && + !(adev->apu_flags & AMD_APU_IS_RAVEN2)) return 0; switch (soc15_asic_reset_method(adev)) { case AMD_RESET_METHOD_BACO: - if (!adev->in_suspend) - amdgpu_inc_vram_lost(adev); return soc15_asic_baco_reset(adev); case AMD_RESET_METHOD_MODE2: return amdgpu_dpm_mode2_reset(adev); default: - if (!adev->in_suspend) - amdgpu_inc_vram_lost(adev); return soc15_asic_mode1_reset(adev); } } @@ -712,7 +709,6 @@ int soc15_set_ip_blocks(struct amdgpu_device *adev) adev->df.funcs = &df_v1_7_funcs; adev->rev_id = soc15_get_rev_id(adev); - adev->nbio.funcs->detect_hw_virt(adev); if (amdgpu_sriov_vf(adev)) adev->virt.ops = &xgpu_ai_virt_ops; @@ -1134,16 +1130,23 @@ static int soc15_common_early_init(void *handle) break; case CHIP_RAVEN: adev->asic_funcs = &soc15_asic_funcs; + if (adev->pdev->device == 0x15dd) + adev->apu_flags |= AMD_APU_IS_RAVEN; + if (adev->pdev->device == 0x15d8) + adev->apu_flags |= AMD_APU_IS_PICASSO; if (adev->rev_id >= 0x8) + adev->apu_flags |= AMD_APU_IS_RAVEN2; + + if (adev->apu_flags & AMD_APU_IS_RAVEN2) adev->external_rev_id = adev->rev_id + 0x79; - else if (adev->pdev->device == 0x15d8) + else if (adev->apu_flags & AMD_APU_IS_PICASSO) adev->external_rev_id = adev->rev_id + 0x41; else if (adev->rev_id == 1) adev->external_rev_id = adev->rev_id + 0x20; else adev->external_rev_id = adev->rev_id + 0x01; - if (adev->rev_id >= 0x8) { + if (adev->apu_flags & AMD_APU_IS_RAVEN2) { adev->cg_flags = AMD_CG_SUPPORT_GFX_MGCG | AMD_CG_SUPPORT_GFX_MGLS | AMD_CG_SUPPORT_GFX_CP_LS | @@ -1161,7 +1164,7 @@ static int soc15_common_early_init(void *handle) AMD_CG_SUPPORT_VCN_MGCG; adev->pg_flags = AMD_PG_SUPPORT_SDMA | AMD_PG_SUPPORT_VCN; - } else if (adev->pdev->device == 0x15d8) { + } else if (adev->apu_flags & AMD_APU_IS_PICASSO) { adev->cg_flags = AMD_CG_SUPPORT_GFX_MGCG | AMD_CG_SUPPORT_GFX_MGLS | AMD_CG_SUPPORT_GFX_CP_LS | @@ -1222,11 +1225,12 @@ static int soc15_common_early_init(void *handle) AMD_CG_SUPPORT_IH_CG | AMD_CG_SUPPORT_VCN_MGCG | AMD_CG_SUPPORT_JPEG_MGCG; - adev->pg_flags = 0; + adev->pg_flags = AMD_PG_SUPPORT_VCN | AMD_PG_SUPPORT_VCN_DPG; adev->external_rev_id = adev->rev_id + 0x32; break; case CHIP_RENOIR: adev->asic_funcs = &soc15_asic_funcs; + adev->apu_flags |= AMD_APU_IS_RENOIR; adev->cg_flags = AMD_CG_SUPPORT_GFX_MGCG | AMD_CG_SUPPORT_GFX_MGLS | AMD_CG_SUPPORT_GFX_3D_CGCG | diff --git a/drivers/gpu/drm/amd/amdgpu/soc15_common.h b/drivers/gpu/drm/amd/amdgpu/soc15_common.h index c893c645a4b2..56d02aa690a7 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15_common.h +++ b/drivers/gpu/drm/amd/amdgpu/soc15_common.h @@ -35,6 +35,9 @@ #define RREG32_SOC15(ip, inst, reg) \ RREG32(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) +#define RREG32_SOC15_NO_KIQ(ip, inst, reg) \ + RREG32_NO_KIQ(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) + #define RREG32_SOC15_OFFSET(ip, inst, reg, offset) \ RREG32((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) + offset) diff --git a/drivers/gpu/drm/amd/amdgpu/soc15d.h b/drivers/gpu/drm/amd/amdgpu/soc15d.h index edfe50821cd9..799925d22fc8 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15d.h +++ b/drivers/gpu/drm/amd/amdgpu/soc15d.h @@ -253,7 +253,30 @@ # define PACKET3_DMA_DATA_CMD_SAIC (1 << 28) # define PACKET3_DMA_DATA_CMD_DAIC (1 << 29) # define PACKET3_DMA_DATA_CMD_RAW_WAIT (1 << 30) -#define PACKET3_AQUIRE_MEM 0x58 +#define PACKET3_ACQUIRE_MEM 0x58 +/* 1. HEADER + * 2. COHER_CNTL [30:0] + * 2.1 ENGINE_SEL [31:31] + * 3. COHER_SIZE [31:0] + * 4. COHER_SIZE_HI [7:0] + * 5. COHER_BASE_LO [31:0] + * 6. COHER_BASE_HI [23:0] + * 7. POLL_INTERVAL [15:0] + */ +/* COHER_CNTL fields for CP_COHER_CNTL */ +#define PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_TC_NC_ACTION_ENA(x) ((x) << 3) +#define PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_TC_WC_ACTION_ENA(x) ((x) << 4) +#define PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_TC_INV_METADATA_ACTION_ENA(x) ((x) << 5) +#define PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_TCL1_VOL_ACTION_ENA(x) ((x) << 15) +#define PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_TC_WB_ACTION_ENA(x) ((x) << 18) +#define PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_TCL1_ACTION_ENA(x) ((x) << 22) +#define PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_TC_ACTION_ENA(x) ((x) << 23) +#define PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_CB_ACTION_ENA(x) ((x) << 25) +#define PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_DB_ACTION_ENA(x) ((x) << 26) +#define PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_SH_KCACHE_ACTION_ENA(x) ((x) << 27) +#define PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_SH_KCACHE_VOL_ACTION_ENA(x) ((x) << 28) +#define PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_SH_ICACHE_ACTION_ENA(x) ((x) << 29) +#define PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_SH_KCACHE_WB_ACTION_ENA(x) ((x) << 30) #define PACKET3_REWIND 0x59 #define PACKET3_LOAD_UCONFIG_REG 0x5E #define PACKET3_LOAD_SH_REG 0x5F @@ -286,6 +309,7 @@ #define PACKET3_WAIT_ON_DE_COUNTER_DIFF 0x88 #define PACKET3_SWITCH_BUFFER 0x8B #define PACKET3_FRAME_CONTROL 0x90 +# define FRAME_TMZ (1 << 0) # define FRAME_CMD(x) ((x) << 28) /* * x=0: tmz_begin diff --git a/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h b/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h index ca7d05993ca2..745ed0fba1ed 100644 --- a/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h +++ b/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h @@ -24,6 +24,8 @@ #ifndef _TA_RAS_IF_H #define _TA_RAS_IF_H +#define RAS_TA_HOST_IF_VER 0 + /* Responses have bit 31 set */ #define RSP_ID_MASK (1U << 31) #define RSP_ID(cmdId) (((uint32_t)(cmdId)) | RSP_ID_MASK) @@ -36,18 +38,24 @@ enum ras_command { TA_RAS_COMMAND__TRIGGER_ERROR, }; -enum ta_ras_status { - TA_RAS_STATUS__SUCCESS = 0x00, - TA_RAS_STATUS__RESET_NEEDED = 0x01, - TA_RAS_STATUS__ERROR_INVALID_PARAMETER = 0x02, - TA_RAS_STATUS__ERROR_RAS_NOT_AVAILABLE = 0x03, - TA_RAS_STATUS__ERROR_RAS_DUPLICATE_CMD = 0x04, - TA_RAS_STATUS__ERROR_INJECTION_FAILED = 0x05, - TA_RAS_STATUS__ERROR_ASD_READ_WRITE = 0x06, - TA_RAS_STATUS__ERROR_TOGGLE_DF_CSTATE = 0x07, - TA_RAS_STATUS__ERROR_TIMEOUT = 0x08, - TA_RAS_STATUS__ERROR_BLOCK_DISABLED = 0x09, - TA_RAS_STATUS__ERROR_GENERIC = 0x10, +enum ta_ras_status +{ + TA_RAS_STATUS__SUCCESS = 0x00, + TA_RAS_STATUS__RESET_NEEDED = 0xA001, + TA_RAS_STATUS__ERROR_INVALID_PARAMETER = 0xA002, + TA_RAS_STATUS__ERROR_RAS_NOT_AVAILABLE = 0xA003, + TA_RAS_STATUS__ERROR_RAS_DUPLICATE_CMD = 0xA004, + TA_RAS_STATUS__ERROR_INJECTION_FAILED = 0xA005, + TA_RAS_STATUS__ERROR_ASD_READ_WRITE = 0xA006, + TA_RAS_STATUS__ERROR_TOGGLE_DF_CSTATE = 0xA007, + TA_RAS_STATUS__ERROR_TIMEOUT = 0xA008, + TA_RAS_STATUS__ERROR_BLOCK_DISABLED = 0XA009, + TA_RAS_STATUS__ERROR_GENERIC = 0xA00A, + TA_RAS_STATUS__ERROR_RAS_MMHUB_INIT = 0xA00B, + TA_RAS_STATUS__ERROR_GET_DEV_INFO = 0xA00C, + TA_RAS_STATUS__ERROR_UNSUPPORTED_DEV = 0xA00D, + TA_RAS_STATUS__ERROR_NOT_INITIALIZED = 0xA00E, + TA_RAS_STATUS__ERROR_TEE_INTERNAL = 0xA00F }; enum ta_ras_block { @@ -97,22 +105,39 @@ struct ta_ras_trigger_error_input { uint64_t value; // method if error injection. i.e persistent, coherent etc. }; +struct ta_ras_output_flags +{ + uint8_t ras_init_success_flag; + uint8_t err_inject_switch_disable_flag; + uint8_t reg_access_failure_flag; +}; + /* Common input structure for RAS callbacks */ /**********************************************************/ union ta_ras_cmd_input { struct ta_ras_enable_features_input enable_features; struct ta_ras_disable_features_input disable_features; struct ta_ras_trigger_error_input trigger_error; + + uint32_t reserve_pad[256]; +}; + +union ta_ras_cmd_output +{ + struct ta_ras_output_flags flags; + + uint32_t reserve_pad[256]; }; /* Shared Memory structures */ /**********************************************************/ struct ta_ras_shared_memory { - uint32_t cmd_id; - uint32_t resp_id; - enum ta_ras_status ras_status; - uint32_t reserved; - union ta_ras_cmd_input ras_in_message; + uint32_t cmd_id; + uint32_t resp_id; + uint32_t ras_status; + uint32_t if_version; + union ta_ras_cmd_input ras_in_message; + union ta_ras_cmd_output ras_out_message; }; #endif // TL_RAS_IF_H_ diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c b/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c index 14d346321a5f..418cf097c918 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c +++ b/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c @@ -56,24 +56,43 @@ const uint32_t static void umc_v6_1_enable_umc_index_mode(struct amdgpu_device *adev) { - WREG32_FIELD15(RSMU, 0, RSMU_UMC_INDEX_REGISTER_NBIF_VG20_GPU, + uint32_t rsmu_umc_addr, rsmu_umc_val; + + rsmu_umc_addr = SOC15_REG_OFFSET(RSMU, 0, + mmRSMU_UMC_INDEX_REGISTER_NBIF_VG20_GPU); + rsmu_umc_val = RREG32_PCIE(rsmu_umc_addr * 4); + + rsmu_umc_val = REG_SET_FIELD(rsmu_umc_val, + RSMU_UMC_INDEX_REGISTER_NBIF_VG20_GPU, RSMU_UMC_INDEX_MODE_EN, 1); + + WREG32_PCIE(rsmu_umc_addr * 4, rsmu_umc_val); } static void umc_v6_1_disable_umc_index_mode(struct amdgpu_device *adev) { - WREG32_FIELD15(RSMU, 0, RSMU_UMC_INDEX_REGISTER_NBIF_VG20_GPU, + uint32_t rsmu_umc_addr, rsmu_umc_val; + + rsmu_umc_addr = SOC15_REG_OFFSET(RSMU, 0, + mmRSMU_UMC_INDEX_REGISTER_NBIF_VG20_GPU); + rsmu_umc_val = RREG32_PCIE(rsmu_umc_addr * 4); + + rsmu_umc_val = REG_SET_FIELD(rsmu_umc_val, + RSMU_UMC_INDEX_REGISTER_NBIF_VG20_GPU, RSMU_UMC_INDEX_MODE_EN, 0); + + WREG32_PCIE(rsmu_umc_addr * 4, rsmu_umc_val); } static uint32_t umc_v6_1_get_umc_index_mode_state(struct amdgpu_device *adev) { - uint32_t rsmu_umc_index; + uint32_t rsmu_umc_addr, rsmu_umc_val; - rsmu_umc_index = RREG32_SOC15(RSMU, 0, + rsmu_umc_addr = SOC15_REG_OFFSET(RSMU, 0, mmRSMU_UMC_INDEX_REGISTER_NBIF_VG20_GPU); + rsmu_umc_val = RREG32_PCIE(rsmu_umc_addr * 4); - return REG_GET_FIELD(rsmu_umc_index, + return REG_GET_FIELD(rsmu_umc_val, RSMU_UMC_INDEX_REGISTER_NBIF_VG20_GPU, RSMU_UMC_INDEX_MODE_EN); } @@ -85,6 +104,81 @@ static inline uint32_t get_umc_6_reg_offset(struct amdgpu_device *adev, return adev->umc.channel_offs*ch_inst + UMC_6_INST_DIST*umc_inst; } +static void umc_v6_1_clear_error_count_per_channel(struct amdgpu_device *adev, + uint32_t umc_reg_offset) +{ + uint32_t ecc_err_cnt_addr; + uint32_t ecc_err_cnt_sel, ecc_err_cnt_sel_addr; + + if (adev->asic_type == CHIP_ARCTURUS) { + /* UMC 6_1_2 registers */ + ecc_err_cnt_sel_addr = + SOC15_REG_OFFSET(UMC, 0, + mmUMCCH0_0_EccErrCntSel_ARCT); + ecc_err_cnt_addr = + SOC15_REG_OFFSET(UMC, 0, + mmUMCCH0_0_EccErrCnt_ARCT); + } else { + /* UMC 6_1_1 registers */ + ecc_err_cnt_sel_addr = + SOC15_REG_OFFSET(UMC, 0, + mmUMCCH0_0_EccErrCntSel); + ecc_err_cnt_addr = + SOC15_REG_OFFSET(UMC, 0, + mmUMCCH0_0_EccErrCnt); + } + + /* select the lower chip */ + ecc_err_cnt_sel = RREG32_PCIE((ecc_err_cnt_sel_addr + + umc_reg_offset) * 4); + ecc_err_cnt_sel = REG_SET_FIELD(ecc_err_cnt_sel, + UMCCH0_0_EccErrCntSel, + EccErrCntCsSel, 0); + WREG32_PCIE((ecc_err_cnt_sel_addr + umc_reg_offset) * 4, + ecc_err_cnt_sel); + + /* clear lower chip error count */ + WREG32_PCIE((ecc_err_cnt_addr + umc_reg_offset) * 4, + UMC_V6_1_CE_CNT_INIT); + + /* select the higher chip */ + ecc_err_cnt_sel = RREG32_PCIE((ecc_err_cnt_sel_addr + + umc_reg_offset) * 4); + ecc_err_cnt_sel = REG_SET_FIELD(ecc_err_cnt_sel, + UMCCH0_0_EccErrCntSel, + EccErrCntCsSel, 1); + WREG32_PCIE((ecc_err_cnt_sel_addr + umc_reg_offset) * 4, + ecc_err_cnt_sel); + + /* clear higher chip error count */ + WREG32_PCIE((ecc_err_cnt_addr + umc_reg_offset) * 4, + UMC_V6_1_CE_CNT_INIT); +} + +static void umc_v6_1_clear_error_count(struct amdgpu_device *adev) +{ + uint32_t umc_inst = 0; + uint32_t ch_inst = 0; + uint32_t umc_reg_offset = 0; + uint32_t rsmu_umc_index_state = + umc_v6_1_get_umc_index_mode_state(adev); + + if (rsmu_umc_index_state) + umc_v6_1_disable_umc_index_mode(adev); + + LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) { + umc_reg_offset = get_umc_6_reg_offset(adev, + umc_inst, + ch_inst); + + umc_v6_1_clear_error_count_per_channel(adev, + umc_reg_offset); + } + + if (rsmu_umc_index_state) + umc_v6_1_enable_umc_index_mode(adev); +} + static void umc_v6_1_query_correctable_error_count(struct amdgpu_device *adev, uint32_t umc_reg_offset, unsigned long *error_count) @@ -117,23 +211,21 @@ static void umc_v6_1_query_correctable_error_count(struct amdgpu_device *adev, ecc_err_cnt_sel = REG_SET_FIELD(ecc_err_cnt_sel, UMCCH0_0_EccErrCntSel, EccErrCntCsSel, 0); WREG32_PCIE((ecc_err_cnt_sel_addr + umc_reg_offset) * 4, ecc_err_cnt_sel); + ecc_err_cnt = RREG32_PCIE((ecc_err_cnt_addr + umc_reg_offset) * 4); *error_count += (REG_GET_FIELD(ecc_err_cnt, UMCCH0_0_EccErrCnt, EccErrCnt) - UMC_V6_1_CE_CNT_INIT); - /* clear the lower chip err count */ - WREG32_PCIE((ecc_err_cnt_addr + umc_reg_offset) * 4, UMC_V6_1_CE_CNT_INIT); /* select the higher chip and check the err counter */ ecc_err_cnt_sel = REG_SET_FIELD(ecc_err_cnt_sel, UMCCH0_0_EccErrCntSel, EccErrCntCsSel, 1); WREG32_PCIE((ecc_err_cnt_sel_addr + umc_reg_offset) * 4, ecc_err_cnt_sel); + ecc_err_cnt = RREG32_PCIE((ecc_err_cnt_addr + umc_reg_offset) * 4); *error_count += (REG_GET_FIELD(ecc_err_cnt, UMCCH0_0_EccErrCnt, EccErrCnt) - UMC_V6_1_CE_CNT_INIT); - /* clear the higher chip err count */ - WREG32_PCIE((ecc_err_cnt_addr + umc_reg_offset) * 4, UMC_V6_1_CE_CNT_INIT); /* check for SRAM correctable error MCUMC_STATUS is a 64 bit register */ @@ -209,6 +301,8 @@ static void umc_v6_1_query_ras_error_count(struct amdgpu_device *adev, if (rsmu_umc_index_state) umc_v6_1_enable_umc_index_mode(adev); + + umc_v6_1_clear_error_count(adev); } static void umc_v6_1_query_error_address(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c index 82abd8e728ab..3cafba726587 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c @@ -118,7 +118,8 @@ static int uvd_v4_2_sw_init(void *handle) ring = &adev->uvd.inst->ring; sprintf(ring->name, "uvd"); - r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst->irq, 0); + r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst->irq, 0, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; @@ -210,13 +211,10 @@ done: static int uvd_v4_2_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_ring *ring = &adev->uvd.inst->ring; if (RREG32(mmUVD_STATUS) != 0) uvd_v4_2_stop(adev); - ring->sched.ready = false; - return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c index 0fa8aae2d78e..a566ff926e90 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c @@ -116,7 +116,8 @@ static int uvd_v5_0_sw_init(void *handle) ring = &adev->uvd.inst->ring; sprintf(ring->name, "uvd"); - r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst->irq, 0); + r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst->irq, 0, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; @@ -208,13 +209,10 @@ done: static int uvd_v5_0_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_ring *ring = &adev->uvd.inst->ring; if (RREG32(mmUVD_STATUS) != 0) uvd_v5_0_stop(adev); - ring->sched.ready = false; - return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c index e0aadcaf6c8b..0a880bc101b8 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c @@ -216,7 +216,8 @@ static int uvd_v6_0_enc_get_create_msg(struct amdgpu_ring *ring, uint32_t handle uint64_t addr; int i, r; - r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, &job); + r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, + AMDGPU_IB_POOL_DIRECT, &job); if (r) return r; @@ -279,7 +280,8 @@ static int uvd_v6_0_enc_get_destroy_msg(struct amdgpu_ring *ring, uint64_t addr; int i, r; - r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, &job); + r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, + AMDGPU_IB_POOL_DIRECT, &job); if (r) return r; @@ -416,7 +418,8 @@ static int uvd_v6_0_sw_init(void *handle) ring = &adev->uvd.inst->ring; sprintf(ring->name, "uvd"); - r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst->irq, 0); + r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst->irq, 0, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; @@ -428,7 +431,9 @@ static int uvd_v6_0_sw_init(void *handle) for (i = 0; i < adev->uvd.num_enc_rings; ++i) { ring = &adev->uvd.inst->ring_enc[i]; sprintf(ring->name, "uvd_enc%d", i); - r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst->irq, 0); + r = amdgpu_ring_init(adev, ring, 512, + &adev->uvd.inst->irq, 0, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } @@ -535,13 +540,10 @@ done: static int uvd_v6_0_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_ring *ring = &adev->uvd.inst->ring; if (RREG32(mmUVD_STATUS) != 0) uvd_v6_0_stop(adev); - ring->sched.ready = false; - return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c index 0995378d8263..7a55457e6f9e 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c @@ -224,7 +224,8 @@ static int uvd_v7_0_enc_get_create_msg(struct amdgpu_ring *ring, uint32_t handle uint64_t addr; int i, r; - r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, &job); + r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, + AMDGPU_IB_POOL_DIRECT, &job); if (r) return r; @@ -286,7 +287,8 @@ static int uvd_v7_0_enc_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handl uint64_t addr; int i, r; - r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, &job); + r = amdgpu_job_alloc_with_ib(ring->adev, ib_size_dw * 4, + AMDGPU_IB_POOL_DIRECT, &job); if (r) return r; @@ -450,7 +452,9 @@ static int uvd_v7_0_sw_init(void *handle) if (!amdgpu_sriov_vf(adev)) { ring = &adev->uvd.inst[j].ring; sprintf(ring->name, "uvd_%d", ring->me); - r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst[j].irq, 0); + r = amdgpu_ring_init(adev, ring, 512, + &adev->uvd.inst[j].irq, 0, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } @@ -469,7 +473,9 @@ static int uvd_v7_0_sw_init(void *handle) else ring->doorbell_index = adev->doorbell_index.uvd_vce.uvd_ring2_3 * 2 + 1; } - r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst[j].irq, 0); + r = amdgpu_ring_init(adev, ring, 512, + &adev->uvd.inst[j].irq, 0, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } @@ -598,7 +604,6 @@ done: static int uvd_v7_0_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - int i; if (!amdgpu_sriov_vf(adev)) uvd_v7_0_stop(adev); @@ -607,12 +612,6 @@ static int uvd_v7_0_hw_fini(void *handle) DRM_DEBUG("For SRIOV client, shouldn't do anything.\n"); } - for (i = 0; i < adev->uvd.num_uvd_inst; ++i) { - if (adev->uvd.harvest_config & (1 << i)) - continue; - adev->uvd.inst[i].ring.sched.ready = false; - } - return 0; } @@ -1694,7 +1693,7 @@ static int uvd_v7_0_set_clockgating_state(void *handle, enum amd_clockgating_state state) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - bool enable = (state == AMD_CG_STATE_GATE) ? true : false; + bool enable = (state == AMD_CG_STATE_GATE); uvd_v7_0_set_bypass_mode(adev, enable); diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c index b6837fcfdba7..0e2945baf0f1 100644 --- a/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c @@ -434,7 +434,8 @@ static int vce_v2_0_sw_init(void *handle) ring = &adev->vce.ring[i]; sprintf(ring->name, "vce%d", i); r = amdgpu_ring_init(adev, ring, 512, - &adev->vce.irq, 0); + &adev->vce.irq, 0, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c index 217db187207c..6d9108fa22e0 100644 --- a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c @@ -442,7 +442,8 @@ static int vce_v3_0_sw_init(void *handle) for (i = 0; i < adev->vce.num_rings; i++) { ring = &adev->vce.ring[i]; sprintf(ring->name, "vce%d", i); - r = amdgpu_ring_init(adev, ring, 512, &adev->vce.irq, 0); + r = amdgpu_ring_init(adev, ring, 512, &adev->vce.irq, 0, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c index 3fd102efb7af..a0fb119240f4 100644 --- a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c @@ -476,7 +476,8 @@ static int vce_v4_0_sw_init(void *handle) else ring->doorbell_index = adev->doorbell_index.uvd_vce.vce_ring2_3 * 2 + 1; } - r = amdgpu_ring_init(adev, ring, 512, &adev->vce.irq, 0); + r = amdgpu_ring_init(adev, ring, 512, &adev->vce.irq, 0, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } @@ -539,7 +540,6 @@ static int vce_v4_0_hw_init(void *handle) static int vce_v4_0_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - int i; if (!amdgpu_sriov_vf(adev)) { /* vce_v4_0_wait_for_idle(handle); */ @@ -549,9 +549,6 @@ static int vce_v4_0_hw_fini(void *handle) DRM_DEBUG("For SRIOV client, shouldn't do anything.\n"); } - for (i = 0; i < adev->vce.num_rings; i++) - adev->vce.ring[i].sched.ready = false; - return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c index 09b0572b838d..1ad79155ed00 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c @@ -127,7 +127,8 @@ static int vcn_v1_0_sw_init(void *handle) ring = &adev->vcn.inst->ring_dec; sprintf(ring->name, "vcn_dec"); - r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0); + r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; @@ -145,7 +146,8 @@ static int vcn_v1_0_sw_init(void *handle) for (i = 0; i < adev->vcn.num_enc_rings; ++i) { ring = &adev->vcn.inst->ring_enc[i]; sprintf(ring->name, "vcn_enc%d", i); - r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0); + r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } @@ -227,14 +229,11 @@ done: static int vcn_v1_0_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_ring *ring = &adev->vcn.inst->ring_dec; if ((adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) || RREG32_SOC15(VCN, 0, mmUVD_STATUS)) vcn_v1_0_set_powergating_state(adev, AMD_PG_STATE_GATE); - ring->sched.ready = false; - return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c index ec8091a661df..90ed773695ea 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c @@ -92,6 +92,7 @@ static int vcn_v2_0_sw_init(void *handle) struct amdgpu_ring *ring; int i, r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; + volatile struct amdgpu_fw_shared *fw_shared; /* VCN DEC TRAP */ r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN, @@ -133,7 +134,8 @@ static int vcn_v2_0_sw_init(void *handle) ring->doorbell_index = adev->doorbell_index.vcn.vcn_ring0_1 << 1; sprintf(ring->name, "vcn_dec"); - r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0); + r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; @@ -163,7 +165,8 @@ static int vcn_v2_0_sw_init(void *handle) else ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1 + i; sprintf(ring->name, "vcn_enc%d", i); - r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0); + r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } @@ -174,6 +177,8 @@ static int vcn_v2_0_sw_init(void *handle) if (r) return r; + fw_shared = adev->vcn.inst->fw_shared_cpu_addr; + fw_shared->present_flag_0 = cpu_to_le32(AMDGPU_VCN_MULTI_QUEUE_FLAG); return 0; } @@ -188,6 +193,9 @@ static int vcn_v2_0_sw_fini(void *handle) { int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; + volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared_cpu_addr; + + fw_shared->present_flag_0 = 0; amdgpu_virt_free_mm_table(adev); @@ -223,6 +231,10 @@ static int vcn_v2_0_hw_init(void *handle) if (r) goto done; + //Disable vcn decode for sriov + if (amdgpu_sriov_vf(adev)) + ring->sched.ready = false; + for (i = 0; i < adev->vcn.num_enc_rings; ++i) { ring = &adev->vcn.inst->ring_enc[i]; r = amdgpu_ring_test_helper(ring); @@ -248,21 +260,12 @@ done: static int vcn_v2_0_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_ring *ring = &adev->vcn.inst->ring_dec; - int i; if ((adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) || (adev->vcn.cur_state != AMD_PG_STATE_GATE && RREG32_SOC15(VCN, 0, mmUVD_STATUS))) vcn_v2_0_set_powergating_state(adev, AMD_PG_STATE_GATE); - ring->sched.ready = false; - - for (i = 0; i < adev->vcn.num_enc_rings; ++i) { - ring = &adev->vcn.inst->ring_enc[i]; - ring->sched.ready = false; - } - return 0; } @@ -359,6 +362,15 @@ static void vcn_v2_0_mc_resume(struct amdgpu_device *adev) WREG32_SOC15(UVD, 0, mmUVD_VCPU_CACHE_OFFSET2, 0); WREG32_SOC15(UVD, 0, mmUVD_VCPU_CACHE_SIZE2, AMDGPU_VCN_CONTEXT_SIZE); + /* non-cache window */ + WREG32_SOC15(UVD, 0, mmUVD_LMI_VCPU_NC0_64BIT_BAR_LOW, + lower_32_bits(adev->vcn.inst->fw_shared_gpu_addr)); + WREG32_SOC15(UVD, 0, mmUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH, + upper_32_bits(adev->vcn.inst->fw_shared_gpu_addr)); + WREG32_SOC15(UVD, 0, mmUVD_VCPU_NONCACHE_OFFSET0, 0); + WREG32_SOC15(UVD, 0, mmUVD_VCPU_NONCACHE_SIZE0, + AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_fw_shared))); + WREG32_SOC15(UVD, 0, mmUVD_GFX10_ADDR_CONFIG, adev->gfx.config.gb_addr_config); } @@ -442,13 +454,16 @@ static void vcn_v2_0_mc_resume_dpg_mode(struct amdgpu_device *adev, bool indirec /* non-cache window */ WREG32_SOC15_DPG_MODE_2_0(0, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_LMI_VCPU_NC0_64BIT_BAR_LOW), 0, 0, indirect); + UVD, 0, mmUVD_LMI_VCPU_NC0_64BIT_BAR_LOW), + lower_32_bits(adev->vcn.inst->fw_shared_gpu_addr), 0, indirect); WREG32_SOC15_DPG_MODE_2_0(0, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH), 0, 0, indirect); + UVD, 0, mmUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH), + upper_32_bits(adev->vcn.inst->fw_shared_gpu_addr), 0, indirect); WREG32_SOC15_DPG_MODE_2_0(0, SOC15_DPG_MODE_OFFSET_2_0( UVD, 0, mmUVD_VCPU_NONCACHE_OFFSET0), 0, 0, indirect); WREG32_SOC15_DPG_MODE_2_0(0, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_VCPU_NONCACHE_SIZE0), 0, 0, indirect); + UVD, 0, mmUVD_VCPU_NONCACHE_SIZE0), + AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_fw_shared)), 0, indirect); /* VCN global tiling registers */ WREG32_SOC15_DPG_MODE_2_0(0, SOC15_DPG_MODE_OFFSET_2_0( @@ -773,6 +788,7 @@ static void vcn_v2_0_enable_static_power_gating(struct amdgpu_device *adev) static int vcn_v2_0_start_dpg_mode(struct amdgpu_device *adev, bool indirect) { + volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared_cpu_addr; struct amdgpu_ring *ring = &adev->vcn.inst->ring_dec; uint32_t rb_bufsz, tmp; @@ -872,6 +888,12 @@ static int vcn_v2_0_start_dpg_mode(struct amdgpu_device *adev, bool indirect) tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_RPTR_WR_EN, 1); WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_CNTL, tmp); + /* Stall DPG before WPTR/RPTR reset */ + WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_POWER_STATUS), + UVD_POWER_STATUS__STALL_DPG_POWER_UP_MASK, + ~UVD_POWER_STATUS__STALL_DPG_POWER_UP_MASK); + fw_shared->multi_queue.decode_queue_mode |= FW_QUEUE_RING_RESET; + /* set the write pointer delay */ WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_WPTR_CNTL, 0); @@ -894,11 +916,16 @@ static int vcn_v2_0_start_dpg_mode(struct amdgpu_device *adev, bool indirect) WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_WPTR, lower_32_bits(ring->wptr)); + fw_shared->multi_queue.decode_queue_mode &= ~FW_QUEUE_RING_RESET; + /* Unstall DPG */ + WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_POWER_STATUS), + 0, ~UVD_POWER_STATUS__STALL_DPG_POWER_UP_MASK); return 0; } static int vcn_v2_0_start(struct amdgpu_device *adev) { + volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared_cpu_addr; struct amdgpu_ring *ring = &adev->vcn.inst->ring_dec; uint32_t rb_bufsz, tmp; uint32_t lmi_swap_cntl; @@ -1033,6 +1060,7 @@ static int vcn_v2_0_start(struct amdgpu_device *adev) tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_RPTR_WR_EN, 1); WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_CNTL, tmp); + fw_shared->multi_queue.decode_queue_mode |= FW_QUEUE_RING_RESET; /* programm the RB_BASE for ring buffer */ WREG32_SOC15(UVD, 0, mmUVD_LMI_RBC_RB_64BIT_BAR_LOW, lower_32_bits(ring->gpu_addr)); @@ -1045,20 +1073,25 @@ static int vcn_v2_0_start(struct amdgpu_device *adev) ring->wptr = RREG32_SOC15(UVD, 0, mmUVD_RBC_RB_RPTR); WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_WPTR, lower_32_bits(ring->wptr)); + fw_shared->multi_queue.decode_queue_mode &= ~FW_QUEUE_RING_RESET; + fw_shared->multi_queue.encode_generalpurpose_queue_mode |= FW_QUEUE_RING_RESET; ring = &adev->vcn.inst->ring_enc[0]; WREG32_SOC15(UVD, 0, mmUVD_RB_RPTR, lower_32_bits(ring->wptr)); WREG32_SOC15(UVD, 0, mmUVD_RB_WPTR, lower_32_bits(ring->wptr)); WREG32_SOC15(UVD, 0, mmUVD_RB_BASE_LO, ring->gpu_addr); WREG32_SOC15(UVD, 0, mmUVD_RB_BASE_HI, upper_32_bits(ring->gpu_addr)); WREG32_SOC15(UVD, 0, mmUVD_RB_SIZE, ring->ring_size / 4); + fw_shared->multi_queue.encode_generalpurpose_queue_mode &= ~FW_QUEUE_RING_RESET; + fw_shared->multi_queue.encode_lowlatency_queue_mode |= FW_QUEUE_RING_RESET; ring = &adev->vcn.inst->ring_enc[1]; WREG32_SOC15(UVD, 0, mmUVD_RB_RPTR2, lower_32_bits(ring->wptr)); WREG32_SOC15(UVD, 0, mmUVD_RB_WPTR2, lower_32_bits(ring->wptr)); WREG32_SOC15(UVD, 0, mmUVD_RB_BASE_LO2, ring->gpu_addr); WREG32_SOC15(UVD, 0, mmUVD_RB_BASE_HI2, upper_32_bits(ring->gpu_addr)); WREG32_SOC15(UVD, 0, mmUVD_RB_SIZE2, ring->ring_size / 4); + fw_shared->multi_queue.encode_lowlatency_queue_mode &= ~FW_QUEUE_RING_RESET; return 0; } @@ -1180,6 +1213,7 @@ static int vcn_v2_0_pause_dpg_mode(struct amdgpu_device *adev, UVD_POWER_STATUS__UVD_POWER_STATUS_MASK, ret_code); if (!ret_code) { + volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared_cpu_addr; /* pause DPG */ reg_data |= UVD_DPG_PAUSE__NJ_PAUSE_DPG_REQ_MASK; WREG32_SOC15(UVD, 0, mmUVD_DPG_PAUSE, reg_data); @@ -1189,23 +1223,38 @@ static int vcn_v2_0_pause_dpg_mode(struct amdgpu_device *adev, UVD_DPG_PAUSE__NJ_PAUSE_DPG_ACK_MASK, UVD_DPG_PAUSE__NJ_PAUSE_DPG_ACK_MASK, ret_code); + /* Stall DPG before WPTR/RPTR reset */ + WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_POWER_STATUS), + UVD_POWER_STATUS__STALL_DPG_POWER_UP_MASK, + ~UVD_POWER_STATUS__STALL_DPG_POWER_UP_MASK); /* Restore */ + fw_shared->multi_queue.encode_generalpurpose_queue_mode |= FW_QUEUE_RING_RESET; ring = &adev->vcn.inst->ring_enc[0]; + ring->wptr = 0; WREG32_SOC15(UVD, 0, mmUVD_RB_BASE_LO, ring->gpu_addr); WREG32_SOC15(UVD, 0, mmUVD_RB_BASE_HI, upper_32_bits(ring->gpu_addr)); WREG32_SOC15(UVD, 0, mmUVD_RB_SIZE, ring->ring_size / 4); WREG32_SOC15(UVD, 0, mmUVD_RB_RPTR, lower_32_bits(ring->wptr)); WREG32_SOC15(UVD, 0, mmUVD_RB_WPTR, lower_32_bits(ring->wptr)); + fw_shared->multi_queue.encode_generalpurpose_queue_mode &= ~FW_QUEUE_RING_RESET; + fw_shared->multi_queue.encode_lowlatency_queue_mode |= FW_QUEUE_RING_RESET; ring = &adev->vcn.inst->ring_enc[1]; + ring->wptr = 0; WREG32_SOC15(UVD, 0, mmUVD_RB_BASE_LO2, ring->gpu_addr); WREG32_SOC15(UVD, 0, mmUVD_RB_BASE_HI2, upper_32_bits(ring->gpu_addr)); WREG32_SOC15(UVD, 0, mmUVD_RB_SIZE2, ring->ring_size / 4); WREG32_SOC15(UVD, 0, mmUVD_RB_RPTR2, lower_32_bits(ring->wptr)); WREG32_SOC15(UVD, 0, mmUVD_RB_WPTR2, lower_32_bits(ring->wptr)); + fw_shared->multi_queue.encode_lowlatency_queue_mode &= ~FW_QUEUE_RING_RESET; + fw_shared->multi_queue.decode_queue_mode |= FW_QUEUE_RING_RESET; WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_WPTR, RREG32_SOC15(UVD, 0, mmUVD_SCRATCH2) & 0x7FFFFFFF); + fw_shared->multi_queue.decode_queue_mode &= ~FW_QUEUE_RING_RESET; + /* Unstall DPG */ + WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_POWER_STATUS), + 0, ~UVD_POWER_STATUS__STALL_DPG_POWER_UP_MASK); SOC15_WAIT_ON_RREG(UVD, 0, mmUVD_POWER_STATUS, UVD_PGFSM_CONFIG__UVDM_UVDU_PWR_ON, @@ -1796,7 +1845,6 @@ static int vcn_v2_0_start_sriov(struct amdgpu_device *adev) uint32_t table_size = 0; struct mmsch_v2_0_cmd_direct_write direct_wt = { {0} }; struct mmsch_v2_0_cmd_direct_read_modify_write direct_rd_mod_wt = { {0} }; - struct mmsch_v2_0_cmd_direct_polling direct_poll = { {0} }; struct mmsch_v2_0_cmd_end end = { {0} }; struct mmsch_v2_0_init_header *header; uint32_t *init_table = adev->virt.mm_table.cpu_addr; @@ -1806,8 +1854,6 @@ static int vcn_v2_0_start_sriov(struct amdgpu_device *adev) direct_wt.cmd_header.command_type = MMSCH_COMMAND__DIRECT_REG_WRITE; direct_rd_mod_wt.cmd_header.command_type = MMSCH_COMMAND__DIRECT_REG_READ_MODIFY_WRITE; - direct_poll.cmd_header.command_type = - MMSCH_COMMAND__DIRECT_REG_POLLING; end.cmd_header.command_type = MMSCH_COMMAND__END; if (header->vcn_table_offset == 0 && header->vcn_table_size == 0) { diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c index c6363f5ad564..3c6eafb62ee6 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c @@ -86,7 +86,7 @@ static int vcn_v2_5_early_init(void *handle) adev->vcn.num_vcn_inst = VCN25_MAX_HW_INSTANCES_ARCTURUS; for (i = 0; i < adev->vcn.num_vcn_inst; i++) { - harvest = RREG32_SOC15(UVD, i, mmCC_UVD_HARVESTING); + harvest = RREG32_SOC15(VCN, i, mmCC_UVD_HARVESTING); if (harvest & CC_UVD_HARVESTING__UVD_DISABLE_MASK) adev->vcn.harvest_config |= 1 << i; } @@ -165,6 +165,8 @@ static int vcn_v2_5_sw_init(void *handle) return r; for (j = 0; j < adev->vcn.num_vcn_inst; j++) { + volatile struct amdgpu_fw_shared *fw_shared; + if (adev->vcn.harvest_config & (1 << j)) continue; adev->vcn.internal.context_id = mmUVD_CONTEXT_ID_INTERNAL_OFFSET; @@ -175,15 +177,15 @@ static int vcn_v2_5_sw_init(void *handle) adev->vcn.internal.gp_scratch8 = mmUVD_GP_SCRATCH8_INTERNAL_OFFSET; adev->vcn.internal.scratch9 = mmUVD_SCRATCH9_INTERNAL_OFFSET; - adev->vcn.inst[j].external.scratch9 = SOC15_REG_OFFSET(UVD, j, mmUVD_SCRATCH9); + adev->vcn.inst[j].external.scratch9 = SOC15_REG_OFFSET(VCN, j, mmUVD_SCRATCH9); adev->vcn.internal.data0 = mmUVD_GPCOM_VCPU_DATA0_INTERNAL_OFFSET; - adev->vcn.inst[j].external.data0 = SOC15_REG_OFFSET(UVD, j, mmUVD_GPCOM_VCPU_DATA0); + adev->vcn.inst[j].external.data0 = SOC15_REG_OFFSET(VCN, j, mmUVD_GPCOM_VCPU_DATA0); adev->vcn.internal.data1 = mmUVD_GPCOM_VCPU_DATA1_INTERNAL_OFFSET; - adev->vcn.inst[j].external.data1 = SOC15_REG_OFFSET(UVD, j, mmUVD_GPCOM_VCPU_DATA1); + adev->vcn.inst[j].external.data1 = SOC15_REG_OFFSET(VCN, j, mmUVD_GPCOM_VCPU_DATA1); adev->vcn.internal.cmd = mmUVD_GPCOM_VCPU_CMD_INTERNAL_OFFSET; - adev->vcn.inst[j].external.cmd = SOC15_REG_OFFSET(UVD, j, mmUVD_GPCOM_VCPU_CMD); + adev->vcn.inst[j].external.cmd = SOC15_REG_OFFSET(VCN, j, mmUVD_GPCOM_VCPU_CMD); adev->vcn.internal.nop = mmUVD_NO_OP_INTERNAL_OFFSET; - adev->vcn.inst[j].external.nop = SOC15_REG_OFFSET(UVD, j, mmUVD_NO_OP); + adev->vcn.inst[j].external.nop = SOC15_REG_OFFSET(VCN, j, mmUVD_NO_OP); ring = &adev->vcn.inst[j].ring_dec; ring->use_doorbell = true; @@ -191,7 +193,8 @@ static int vcn_v2_5_sw_init(void *handle) ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + (amdgpu_sriov_vf(adev) ? 2*j : 8*j); sprintf(ring->name, "vcn_dec_%d", j); - r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst[j].irq, 0); + r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst[j].irq, + 0, AMDGPU_RING_PRIO_DEFAULT); if (r) return r; @@ -203,10 +206,15 @@ static int vcn_v2_5_sw_init(void *handle) (amdgpu_sriov_vf(adev) ? (1 + i + 2*j) : (2 + i + 8*j)); sprintf(ring->name, "vcn_enc_%d.%d", j, i); - r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst[j].irq, 0); + r = amdgpu_ring_init(adev, ring, 512, + &adev->vcn.inst[j].irq, 0, + AMDGPU_RING_PRIO_DEFAULT); if (r) return r; } + + fw_shared = adev->vcn.inst[j].fw_shared_cpu_addr; + fw_shared->present_flag_0 = cpu_to_le32(AMDGPU_VCN_MULTI_QUEUE_FLAG); } if (amdgpu_sriov_vf(adev)) { @@ -230,8 +238,16 @@ static int vcn_v2_5_sw_init(void *handle) */ static int vcn_v2_5_sw_fini(void *handle) { - int r; + int i, r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; + volatile struct amdgpu_fw_shared *fw_shared; + + for (i = 0; i < adev->vcn.num_vcn_inst; i++) { + if (adev->vcn.harvest_config & (1 << i)) + continue; + fw_shared = adev->vcn.inst[i].fw_shared_cpu_addr; + fw_shared->present_flag_0 = 0; + } if (amdgpu_sriov_vf(adev)) amdgpu_virt_free_mm_table(adev); @@ -308,25 +324,16 @@ done: static int vcn_v2_5_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_ring *ring; - int i, j; + int i; for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { if (adev->vcn.harvest_config & (1 << i)) continue; - ring = &adev->vcn.inst[i].ring_dec; if ((adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) || (adev->vcn.cur_state != AMD_PG_STATE_GATE && RREG32_SOC15(VCN, i, mmUVD_STATUS))) vcn_v2_5_set_powergating_state(adev, AMD_PG_STATE_GATE); - - ring->sched.ready = false; - - for (j = 0; j < adev->vcn.num_enc_rings; ++j) { - ring = &adev->vcn.inst[i].ring_enc[j]; - ring->sched.ready = false; - } } return 0; @@ -392,38 +399,47 @@ static void vcn_v2_5_mc_resume(struct amdgpu_device *adev) continue; /* cache window 0: fw */ if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { - WREG32_SOC15(UVD, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW, + WREG32_SOC15(VCN, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW, (adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + i].tmr_mc_addr_lo)); - WREG32_SOC15(UVD, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH, + WREG32_SOC15(VCN, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH, (adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + i].tmr_mc_addr_hi)); - WREG32_SOC15(UVD, i, mmUVD_VCPU_CACHE_OFFSET0, 0); + WREG32_SOC15(VCN, i, mmUVD_VCPU_CACHE_OFFSET0, 0); offset = 0; } else { - WREG32_SOC15(UVD, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW, + WREG32_SOC15(VCN, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW, lower_32_bits(adev->vcn.inst[i].gpu_addr)); - WREG32_SOC15(UVD, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH, + WREG32_SOC15(VCN, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH, upper_32_bits(adev->vcn.inst[i].gpu_addr)); offset = size; - WREG32_SOC15(UVD, i, mmUVD_VCPU_CACHE_OFFSET0, + WREG32_SOC15(VCN, i, mmUVD_VCPU_CACHE_OFFSET0, AMDGPU_UVD_FIRMWARE_OFFSET >> 3); } - WREG32_SOC15(UVD, i, mmUVD_VCPU_CACHE_SIZE0, size); + WREG32_SOC15(VCN, i, mmUVD_VCPU_CACHE_SIZE0, size); /* cache window 1: stack */ - WREG32_SOC15(UVD, i, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW, + WREG32_SOC15(VCN, i, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW, lower_32_bits(adev->vcn.inst[i].gpu_addr + offset)); - WREG32_SOC15(UVD, i, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH, + WREG32_SOC15(VCN, i, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH, upper_32_bits(adev->vcn.inst[i].gpu_addr + offset)); - WREG32_SOC15(UVD, i, mmUVD_VCPU_CACHE_OFFSET1, 0); - WREG32_SOC15(UVD, i, mmUVD_VCPU_CACHE_SIZE1, AMDGPU_VCN_STACK_SIZE); + WREG32_SOC15(VCN, i, mmUVD_VCPU_CACHE_OFFSET1, 0); + WREG32_SOC15(VCN, i, mmUVD_VCPU_CACHE_SIZE1, AMDGPU_VCN_STACK_SIZE); /* cache window 2: context */ - WREG32_SOC15(UVD, i, mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW, + WREG32_SOC15(VCN, i, mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW, lower_32_bits(adev->vcn.inst[i].gpu_addr + offset + AMDGPU_VCN_STACK_SIZE)); - WREG32_SOC15(UVD, i, mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH, + WREG32_SOC15(VCN, i, mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH, upper_32_bits(adev->vcn.inst[i].gpu_addr + offset + AMDGPU_VCN_STACK_SIZE)); - WREG32_SOC15(UVD, i, mmUVD_VCPU_CACHE_OFFSET2, 0); - WREG32_SOC15(UVD, i, mmUVD_VCPU_CACHE_SIZE2, AMDGPU_VCN_CONTEXT_SIZE); + WREG32_SOC15(VCN, i, mmUVD_VCPU_CACHE_OFFSET2, 0); + WREG32_SOC15(VCN, i, mmUVD_VCPU_CACHE_SIZE2, AMDGPU_VCN_CONTEXT_SIZE); + + /* non-cache window */ + WREG32_SOC15(VCN, i, mmUVD_LMI_VCPU_NC0_64BIT_BAR_LOW, + lower_32_bits(adev->vcn.inst[i].fw_shared_gpu_addr)); + WREG32_SOC15(VCN, i, mmUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH, + upper_32_bits(adev->vcn.inst[i].fw_shared_gpu_addr)); + WREG32_SOC15(VCN, i, mmUVD_VCPU_NONCACHE_OFFSET0, 0); + WREG32_SOC15(VCN, i, mmUVD_VCPU_NONCACHE_SIZE0, + AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_fw_shared))); } } @@ -436,88 +452,91 @@ static void vcn_v2_5_mc_resume_dpg_mode(struct amdgpu_device *adev, int inst_idx if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { if (!indirect) { WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), + VCN, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), (adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + inst_idx].tmr_mc_addr_lo), 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), + VCN, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), (adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + inst_idx].tmr_mc_addr_hi), 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_VCPU_CACHE_OFFSET0), 0, 0, indirect); + VCN, 0, mmUVD_VCPU_CACHE_OFFSET0), 0, 0, indirect); } else { WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), 0, 0, indirect); + VCN, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), 0, 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), 0, 0, indirect); + VCN, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), 0, 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_VCPU_CACHE_OFFSET0), 0, 0, indirect); + VCN, 0, mmUVD_VCPU_CACHE_OFFSET0), 0, 0, indirect); } offset = 0; } else { WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), + VCN, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr), 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), + VCN, 0, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr), 0, indirect); offset = size; WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_VCPU_CACHE_OFFSET0), + VCN, 0, mmUVD_VCPU_CACHE_OFFSET0), AMDGPU_UVD_FIRMWARE_OFFSET >> 3, 0, indirect); } if (!indirect) WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_VCPU_CACHE_SIZE0), size, 0, indirect); + VCN, 0, mmUVD_VCPU_CACHE_SIZE0), size, 0, indirect); else WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_VCPU_CACHE_SIZE0), 0, 0, indirect); + VCN, 0, mmUVD_VCPU_CACHE_SIZE0), 0, 0, indirect); /* cache window 1: stack */ if (!indirect) { WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW), + VCN, 0, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW), lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset), 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH), + VCN, 0, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH), upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset), 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_VCPU_CACHE_OFFSET1), 0, 0, indirect); + VCN, 0, mmUVD_VCPU_CACHE_OFFSET1), 0, 0, indirect); } else { WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW), 0, 0, indirect); + VCN, 0, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW), 0, 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH), 0, 0, indirect); + VCN, 0, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH), 0, 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_VCPU_CACHE_OFFSET1), 0, 0, indirect); + VCN, 0, mmUVD_VCPU_CACHE_OFFSET1), 0, 0, indirect); } WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_VCPU_CACHE_SIZE1), AMDGPU_VCN_STACK_SIZE, 0, indirect); + VCN, 0, mmUVD_VCPU_CACHE_SIZE1), AMDGPU_VCN_STACK_SIZE, 0, indirect); /* cache window 2: context */ WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW), + VCN, 0, mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW), lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset + AMDGPU_VCN_STACK_SIZE), 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH), + VCN, 0, mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH), upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset + AMDGPU_VCN_STACK_SIZE), 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_VCPU_CACHE_OFFSET2), 0, 0, indirect); + VCN, 0, mmUVD_VCPU_CACHE_OFFSET2), 0, 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_VCPU_CACHE_SIZE2), AMDGPU_VCN_CONTEXT_SIZE, 0, indirect); + VCN, 0, mmUVD_VCPU_CACHE_SIZE2), AMDGPU_VCN_CONTEXT_SIZE, 0, indirect); /* non-cache window */ WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_LMI_VCPU_NC0_64BIT_BAR_LOW), 0, 0, indirect); + VCN, 0, mmUVD_LMI_VCPU_NC0_64BIT_BAR_LOW), + lower_32_bits(adev->vcn.inst[inst_idx].fw_shared_gpu_addr), 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH), 0, 0, indirect); + VCN, 0, mmUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH), + upper_32_bits(adev->vcn.inst[inst_idx].fw_shared_gpu_addr), 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_VCPU_NONCACHE_OFFSET0), 0, 0, indirect); + VCN, 0, mmUVD_VCPU_NONCACHE_OFFSET0), 0, 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_VCPU_NONCACHE_SIZE0), 0, 0, indirect); + VCN, 0, mmUVD_VCPU_NONCACHE_SIZE0), + AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_fw_shared)), 0, indirect); /* VCN global tiling registers */ WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_GFX8_ADDR_CONFIG), adev->gfx.config.gb_addr_config, 0, indirect); + VCN, 0, mmUVD_GFX8_ADDR_CONFIG), adev->gfx.config.gb_addr_config, 0, indirect); } /** @@ -671,19 +690,19 @@ static void vcn_v2_5_clock_gating_dpg_mode(struct amdgpu_device *adev, UVD_CGC_CTRL__VCPU_MODE_MASK | UVD_CGC_CTRL__MMSCH_MODE_MASK); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_CGC_CTRL), reg_data, sram_sel, indirect); + VCN, 0, mmUVD_CGC_CTRL), reg_data, sram_sel, indirect); /* turn off clock gating */ WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_CGC_GATE), 0, sram_sel, indirect); + VCN, 0, mmUVD_CGC_GATE), 0, sram_sel, indirect); /* turn on SUVD clock gating */ WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_SUVD_CGC_GATE), 1, sram_sel, indirect); + VCN, 0, mmUVD_SUVD_CGC_GATE), 1, sram_sel, indirect); /* turn on sw mode in UVD_SUVD_CGC_CTRL */ WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_SUVD_CGC_CTRL), 0, sram_sel, indirect); + VCN, 0, mmUVD_SUVD_CGC_CTRL), 0, sram_sel, indirect); } /** @@ -750,17 +769,18 @@ static void vcn_v2_5_enable_clock_gating(struct amdgpu_device *adev) static int vcn_v2_5_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, bool indirect) { + volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared_cpu_addr; struct amdgpu_ring *ring; uint32_t rb_bufsz, tmp; /* disable register anti-hang mechanism */ - WREG32_P(SOC15_REG_OFFSET(UVD, inst_idx, mmUVD_POWER_STATUS), 1, + WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, mmUVD_POWER_STATUS), 1, ~UVD_POWER_STATUS__UVD_POWER_STATUS_MASK); /* enable dynamic power gating mode */ - tmp = RREG32_SOC15(UVD, inst_idx, mmUVD_POWER_STATUS); + tmp = RREG32_SOC15(VCN, inst_idx, mmUVD_POWER_STATUS); tmp |= UVD_POWER_STATUS__UVD_PG_MODE_MASK; tmp |= UVD_POWER_STATUS__UVD_PG_EN_MASK; - WREG32_SOC15(UVD, inst_idx, mmUVD_POWER_STATUS, tmp); + WREG32_SOC15(VCN, inst_idx, mmUVD_POWER_STATUS, tmp); if (indirect) adev->vcn.inst[inst_idx].dpg_sram_curr_addr = (uint32_t*)adev->vcn.inst[inst_idx].dpg_sram_cpu_addr; @@ -773,11 +793,11 @@ static int vcn_v2_5_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, boo tmp |= UVD_VCPU_CNTL__CLK_EN_MASK; tmp |= UVD_VCPU_CNTL__BLK_RST_MASK; WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_VCPU_CNTL), tmp, 0, indirect); + VCN, 0, mmUVD_VCPU_CNTL), tmp, 0, indirect); /* disable master interupt */ WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_MASTINT_EN), 0, 0, indirect); + VCN, 0, mmUVD_MASTINT_EN), 0, 0, indirect); /* setup mmUVD_LMI_CTRL */ tmp = (0x8 | UVD_LMI_CTRL__WRITE_CLEAN_TIMER_EN_MASK | @@ -789,28 +809,28 @@ static int vcn_v2_5_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, boo (8 << UVD_LMI_CTRL__WRITE_CLEAN_TIMER__SHIFT) | 0x00100000L); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_LMI_CTRL), tmp, 0, indirect); + VCN, 0, mmUVD_LMI_CTRL), tmp, 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_MPC_CNTL), + VCN, 0, mmUVD_MPC_CNTL), 0x2 << UVD_MPC_CNTL__REPLACEMENT_MODE__SHIFT, 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_MPC_SET_MUXA0), + VCN, 0, mmUVD_MPC_SET_MUXA0), ((0x1 << UVD_MPC_SET_MUXA0__VARA_1__SHIFT) | (0x2 << UVD_MPC_SET_MUXA0__VARA_2__SHIFT) | (0x3 << UVD_MPC_SET_MUXA0__VARA_3__SHIFT) | (0x4 << UVD_MPC_SET_MUXA0__VARA_4__SHIFT)), 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_MPC_SET_MUXB0), + VCN, 0, mmUVD_MPC_SET_MUXB0), ((0x1 << UVD_MPC_SET_MUXB0__VARB_1__SHIFT) | (0x2 << UVD_MPC_SET_MUXB0__VARB_2__SHIFT) | (0x3 << UVD_MPC_SET_MUXB0__VARB_3__SHIFT) | (0x4 << UVD_MPC_SET_MUXB0__VARB_4__SHIFT)), 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_MPC_SET_MUX), + VCN, 0, mmUVD_MPC_SET_MUX), ((0x0 << UVD_MPC_SET_MUX__SET_0__SHIFT) | (0x1 << UVD_MPC_SET_MUX__SET_1__SHIFT) | (0x2 << UVD_MPC_SET_MUX__SET_2__SHIFT)), 0, indirect); @@ -818,26 +838,26 @@ static int vcn_v2_5_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, boo vcn_v2_5_mc_resume_dpg_mode(adev, inst_idx, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_REG_XX_MASK), 0x10, 0, indirect); + VCN, 0, mmUVD_REG_XX_MASK), 0x10, 0, indirect); WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_RBC_XX_IB_REG_CHECK), 0x3, 0, indirect); + VCN, 0, mmUVD_RBC_XX_IB_REG_CHECK), 0x3, 0, indirect); /* enable LMI MC and UMC channels */ WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_LMI_CTRL2), 0, 0, indirect); + VCN, 0, mmUVD_LMI_CTRL2), 0, 0, indirect); /* unblock VCPU register access */ WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_RB_ARB_CTRL), 0, 0, indirect); + VCN, 0, mmUVD_RB_ARB_CTRL), 0, 0, indirect); tmp = (0xFF << UVD_VCPU_CNTL__PRB_TIMEOUT_VAL__SHIFT); tmp |= UVD_VCPU_CNTL__CLK_EN_MASK; WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_VCPU_CNTL), tmp, 0, indirect); + VCN, 0, mmUVD_VCPU_CNTL), tmp, 0, indirect); /* enable master interrupt */ WREG32_SOC15_DPG_MODE_2_0(inst_idx, SOC15_DPG_MODE_OFFSET_2_0( - UVD, 0, mmUVD_MASTINT_EN), + VCN, 0, mmUVD_MASTINT_EN), UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect); if (indirect) @@ -853,30 +873,41 @@ static int vcn_v2_5_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, boo tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_NO_FETCH, 1); tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_NO_UPDATE, 1); tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_RPTR_WR_EN, 1); - WREG32_SOC15(UVD, inst_idx, mmUVD_RBC_RB_CNTL, tmp); + WREG32_SOC15(VCN, inst_idx, mmUVD_RBC_RB_CNTL, tmp); + + /* Stall DPG before WPTR/RPTR reset */ + WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, mmUVD_POWER_STATUS), + UVD_POWER_STATUS__STALL_DPG_POWER_UP_MASK, + ~UVD_POWER_STATUS__STALL_DPG_POWER_UP_MASK); + fw_shared->multi_queue.decode_queue_mode |= FW_QUEUE_RING_RESET; /* set the write pointer delay */ - WREG32_SOC15(UVD, inst_idx, mmUVD_RBC_RB_WPTR_CNTL, 0); + WREG32_SOC15(VCN, inst_idx, mmUVD_RBC_RB_WPTR_CNTL, 0); /* set the wb address */ - WREG32_SOC15(UVD, inst_idx, mmUVD_RBC_RB_RPTR_ADDR, + WREG32_SOC15(VCN, inst_idx, mmUVD_RBC_RB_RPTR_ADDR, (upper_32_bits(ring->gpu_addr) >> 2)); /* programm the RB_BASE for ring buffer */ - WREG32_SOC15(UVD, inst_idx, mmUVD_LMI_RBC_RB_64BIT_BAR_LOW, + WREG32_SOC15(VCN, inst_idx, mmUVD_LMI_RBC_RB_64BIT_BAR_LOW, lower_32_bits(ring->gpu_addr)); - WREG32_SOC15(UVD, inst_idx, mmUVD_LMI_RBC_RB_64BIT_BAR_HIGH, + WREG32_SOC15(VCN, inst_idx, mmUVD_LMI_RBC_RB_64BIT_BAR_HIGH, upper_32_bits(ring->gpu_addr)); /* Initialize the ring buffer's read and write pointers */ - WREG32_SOC15(UVD, inst_idx, mmUVD_RBC_RB_RPTR, 0); + WREG32_SOC15(VCN, inst_idx, mmUVD_RBC_RB_RPTR, 0); - WREG32_SOC15(UVD, inst_idx, mmUVD_SCRATCH2, 0); + WREG32_SOC15(VCN, inst_idx, mmUVD_SCRATCH2, 0); - ring->wptr = RREG32_SOC15(UVD, inst_idx, mmUVD_RBC_RB_RPTR); - WREG32_SOC15(UVD, inst_idx, mmUVD_RBC_RB_WPTR, + ring->wptr = RREG32_SOC15(VCN, inst_idx, mmUVD_RBC_RB_RPTR); + WREG32_SOC15(VCN, inst_idx, mmUVD_RBC_RB_WPTR, lower_32_bits(ring->wptr)); + fw_shared->multi_queue.decode_queue_mode &= ~FW_QUEUE_RING_RESET; + /* Unstall DPG */ + WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, mmUVD_POWER_STATUS), + 0, ~UVD_POWER_STATUS__STALL_DPG_POWER_UP_MASK); + return 0; } @@ -898,12 +929,12 @@ static int vcn_v2_5_start(struct amdgpu_device *adev) } /* disable register anti-hang mechanism */ - WREG32_P(SOC15_REG_OFFSET(UVD, i, mmUVD_POWER_STATUS), 0, + WREG32_P(SOC15_REG_OFFSET(VCN, i, mmUVD_POWER_STATUS), 0, ~UVD_POWER_STATUS__UVD_POWER_STATUS_MASK); /* set uvd status busy */ - tmp = RREG32_SOC15(UVD, i, mmUVD_STATUS) | UVD_STATUS__UVD_BUSY; - WREG32_SOC15(UVD, i, mmUVD_STATUS, tmp); + tmp = RREG32_SOC15(VCN, i, mmUVD_STATUS) | UVD_STATUS__UVD_BUSY; + WREG32_SOC15(VCN, i, mmUVD_STATUS, tmp); } if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) @@ -916,44 +947,44 @@ static int vcn_v2_5_start(struct amdgpu_device *adev) if (adev->vcn.harvest_config & (1 << i)) continue; /* enable VCPU clock */ - WREG32_P(SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CNTL), + WREG32_P(SOC15_REG_OFFSET(VCN, i, mmUVD_VCPU_CNTL), UVD_VCPU_CNTL__CLK_EN_MASK, ~UVD_VCPU_CNTL__CLK_EN_MASK); /* disable master interrupt */ - WREG32_P(SOC15_REG_OFFSET(UVD, i, mmUVD_MASTINT_EN), 0, + WREG32_P(SOC15_REG_OFFSET(VCN, i, mmUVD_MASTINT_EN), 0, ~UVD_MASTINT_EN__VCPU_EN_MASK); /* setup mmUVD_LMI_CTRL */ - tmp = RREG32_SOC15(UVD, i, mmUVD_LMI_CTRL); + tmp = RREG32_SOC15(VCN, i, mmUVD_LMI_CTRL); tmp &= ~0xff; - WREG32_SOC15(UVD, i, mmUVD_LMI_CTRL, tmp | 0x8| + WREG32_SOC15(VCN, i, mmUVD_LMI_CTRL, tmp | 0x8| UVD_LMI_CTRL__WRITE_CLEAN_TIMER_EN_MASK | UVD_LMI_CTRL__MASK_MC_URGENT_MASK | UVD_LMI_CTRL__DATA_COHERENCY_EN_MASK | UVD_LMI_CTRL__VCPU_DATA_COHERENCY_EN_MASK); /* setup mmUVD_MPC_CNTL */ - tmp = RREG32_SOC15(UVD, i, mmUVD_MPC_CNTL); + tmp = RREG32_SOC15(VCN, i, mmUVD_MPC_CNTL); tmp &= ~UVD_MPC_CNTL__REPLACEMENT_MODE_MASK; tmp |= 0x2 << UVD_MPC_CNTL__REPLACEMENT_MODE__SHIFT; WREG32_SOC15(VCN, i, mmUVD_MPC_CNTL, tmp); /* setup UVD_MPC_SET_MUXA0 */ - WREG32_SOC15(UVD, i, mmUVD_MPC_SET_MUXA0, + WREG32_SOC15(VCN, i, mmUVD_MPC_SET_MUXA0, ((0x1 << UVD_MPC_SET_MUXA0__VARA_1__SHIFT) | (0x2 << UVD_MPC_SET_MUXA0__VARA_2__SHIFT) | (0x3 << UVD_MPC_SET_MUXA0__VARA_3__SHIFT) | (0x4 << UVD_MPC_SET_MUXA0__VARA_4__SHIFT))); /* setup UVD_MPC_SET_MUXB0 */ - WREG32_SOC15(UVD, i, mmUVD_MPC_SET_MUXB0, + WREG32_SOC15(VCN, i, mmUVD_MPC_SET_MUXB0, ((0x1 << UVD_MPC_SET_MUXB0__VARB_1__SHIFT) | (0x2 << UVD_MPC_SET_MUXB0__VARB_2__SHIFT) | (0x3 << UVD_MPC_SET_MUXB0__VARB_3__SHIFT) | (0x4 << UVD_MPC_SET_MUXB0__VARB_4__SHIFT))); /* setup mmUVD_MPC_SET_MUX */ - WREG32_SOC15(UVD, i, mmUVD_MPC_SET_MUX, + WREG32_SOC15(VCN, i, mmUVD_MPC_SET_MUX, ((0x0 << UVD_MPC_SET_MUX__SET_0__SHIFT) | (0x1 << UVD_MPC_SET_MUX__SET_1__SHIFT) | (0x2 << UVD_MPC_SET_MUX__SET_2__SHIFT))); @@ -962,30 +993,31 @@ static int vcn_v2_5_start(struct amdgpu_device *adev) vcn_v2_5_mc_resume(adev); for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { + volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[i].fw_shared_cpu_addr; if (adev->vcn.harvest_config & (1 << i)) continue; /* VCN global tiling registers */ - WREG32_SOC15(UVD, i, mmUVD_GFX8_ADDR_CONFIG, + WREG32_SOC15(VCN, i, mmUVD_GFX8_ADDR_CONFIG, adev->gfx.config.gb_addr_config); - WREG32_SOC15(UVD, i, mmUVD_GFX8_ADDR_CONFIG, + WREG32_SOC15(VCN, i, mmUVD_GFX8_ADDR_CONFIG, adev->gfx.config.gb_addr_config); /* enable LMI MC and UMC channels */ - WREG32_P(SOC15_REG_OFFSET(UVD, i, mmUVD_LMI_CTRL2), 0, + WREG32_P(SOC15_REG_OFFSET(VCN, i, mmUVD_LMI_CTRL2), 0, ~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK); /* unblock VCPU register access */ - WREG32_P(SOC15_REG_OFFSET(UVD, i, mmUVD_RB_ARB_CTRL), 0, + WREG32_P(SOC15_REG_OFFSET(VCN, i, mmUVD_RB_ARB_CTRL), 0, ~UVD_RB_ARB_CTRL__VCPU_DIS_MASK); - WREG32_P(SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CNTL), 0, + WREG32_P(SOC15_REG_OFFSET(VCN, i, mmUVD_VCPU_CNTL), 0, ~UVD_VCPU_CNTL__BLK_RST_MASK); for (k = 0; k < 10; ++k) { uint32_t status; for (j = 0; j < 100; ++j) { - status = RREG32_SOC15(UVD, i, mmUVD_STATUS); + status = RREG32_SOC15(VCN, i, mmUVD_STATUS); if (status & 2) break; if (amdgpu_emu_mode == 1) @@ -998,11 +1030,11 @@ static int vcn_v2_5_start(struct amdgpu_device *adev) break; DRM_ERROR("VCN decode not responding, trying to reset the VCPU!!!\n"); - WREG32_P(SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CNTL), + WREG32_P(SOC15_REG_OFFSET(VCN, i, mmUVD_VCPU_CNTL), UVD_VCPU_CNTL__BLK_RST_MASK, ~UVD_VCPU_CNTL__BLK_RST_MASK); mdelay(10); - WREG32_P(SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CNTL), 0, + WREG32_P(SOC15_REG_OFFSET(VCN, i, mmUVD_VCPU_CNTL), 0, ~UVD_VCPU_CNTL__BLK_RST_MASK); mdelay(10); @@ -1015,15 +1047,15 @@ static int vcn_v2_5_start(struct amdgpu_device *adev) } /* enable master interrupt */ - WREG32_P(SOC15_REG_OFFSET(UVD, i, mmUVD_MASTINT_EN), + WREG32_P(SOC15_REG_OFFSET(VCN, i, mmUVD_MASTINT_EN), UVD_MASTINT_EN__VCPU_EN_MASK, ~UVD_MASTINT_EN__VCPU_EN_MASK); /* clear the busy bit of VCN_STATUS */ - WREG32_P(SOC15_REG_OFFSET(UVD, i, mmUVD_STATUS), 0, + WREG32_P(SOC15_REG_OFFSET(VCN, i, mmUVD_STATUS), 0, ~(2 << UVD_STATUS__VCPU_REPORT__SHIFT)); - WREG32_SOC15(UVD, i, mmUVD_LMI_RBC_RB_VMID, 0); + WREG32_SOC15(VCN, i, mmUVD_LMI_RBC_RB_VMID, 0); ring = &adev->vcn.inst[i].ring_dec; /* force RBC into idle state */ @@ -1033,33 +1065,40 @@ static int vcn_v2_5_start(struct amdgpu_device *adev) tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_NO_FETCH, 1); tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_NO_UPDATE, 1); tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_RPTR_WR_EN, 1); - WREG32_SOC15(UVD, i, mmUVD_RBC_RB_CNTL, tmp); + WREG32_SOC15(VCN, i, mmUVD_RBC_RB_CNTL, tmp); + fw_shared->multi_queue.decode_queue_mode |= FW_QUEUE_RING_RESET; /* programm the RB_BASE for ring buffer */ - WREG32_SOC15(UVD, i, mmUVD_LMI_RBC_RB_64BIT_BAR_LOW, + WREG32_SOC15(VCN, i, mmUVD_LMI_RBC_RB_64BIT_BAR_LOW, lower_32_bits(ring->gpu_addr)); - WREG32_SOC15(UVD, i, mmUVD_LMI_RBC_RB_64BIT_BAR_HIGH, + WREG32_SOC15(VCN, i, mmUVD_LMI_RBC_RB_64BIT_BAR_HIGH, upper_32_bits(ring->gpu_addr)); /* Initialize the ring buffer's read and write pointers */ - WREG32_SOC15(UVD, i, mmUVD_RBC_RB_RPTR, 0); + WREG32_SOC15(VCN, i, mmUVD_RBC_RB_RPTR, 0); - ring->wptr = RREG32_SOC15(UVD, i, mmUVD_RBC_RB_RPTR); - WREG32_SOC15(UVD, i, mmUVD_RBC_RB_WPTR, + ring->wptr = RREG32_SOC15(VCN, i, mmUVD_RBC_RB_RPTR); + WREG32_SOC15(VCN, i, mmUVD_RBC_RB_WPTR, lower_32_bits(ring->wptr)); - ring = &adev->vcn.inst[i].ring_enc[0]; - WREG32_SOC15(UVD, i, mmUVD_RB_RPTR, lower_32_bits(ring->wptr)); - WREG32_SOC15(UVD, i, mmUVD_RB_WPTR, lower_32_bits(ring->wptr)); - WREG32_SOC15(UVD, i, mmUVD_RB_BASE_LO, ring->gpu_addr); - WREG32_SOC15(UVD, i, mmUVD_RB_BASE_HI, upper_32_bits(ring->gpu_addr)); - WREG32_SOC15(UVD, i, mmUVD_RB_SIZE, ring->ring_size / 4); + fw_shared->multi_queue.decode_queue_mode &= ~FW_QUEUE_RING_RESET; + fw_shared->multi_queue.encode_generalpurpose_queue_mode |= FW_QUEUE_RING_RESET; + ring = &adev->vcn.inst[i].ring_enc[0]; + WREG32_SOC15(VCN, i, mmUVD_RB_RPTR, lower_32_bits(ring->wptr)); + WREG32_SOC15(VCN, i, mmUVD_RB_WPTR, lower_32_bits(ring->wptr)); + WREG32_SOC15(VCN, i, mmUVD_RB_BASE_LO, ring->gpu_addr); + WREG32_SOC15(VCN, i, mmUVD_RB_BASE_HI, upper_32_bits(ring->gpu_addr)); + WREG32_SOC15(VCN, i, mmUVD_RB_SIZE, ring->ring_size / 4); + fw_shared->multi_queue.encode_generalpurpose_queue_mode &= ~FW_QUEUE_RING_RESET; + + fw_shared->multi_queue.encode_lowlatency_queue_mode |= FW_QUEUE_RING_RESET; ring = &adev->vcn.inst[i].ring_enc[1]; - WREG32_SOC15(UVD, i, mmUVD_RB_RPTR2, lower_32_bits(ring->wptr)); - WREG32_SOC15(UVD, i, mmUVD_RB_WPTR2, lower_32_bits(ring->wptr)); - WREG32_SOC15(UVD, i, mmUVD_RB_BASE_LO2, ring->gpu_addr); - WREG32_SOC15(UVD, i, mmUVD_RB_BASE_HI2, upper_32_bits(ring->gpu_addr)); - WREG32_SOC15(UVD, i, mmUVD_RB_SIZE2, ring->ring_size / 4); + WREG32_SOC15(VCN, i, mmUVD_RB_RPTR2, lower_32_bits(ring->wptr)); + WREG32_SOC15(VCN, i, mmUVD_RB_WPTR2, lower_32_bits(ring->wptr)); + WREG32_SOC15(VCN, i, mmUVD_RB_BASE_LO2, ring->gpu_addr); + WREG32_SOC15(VCN, i, mmUVD_RB_BASE_HI2, upper_32_bits(ring->gpu_addr)); + WREG32_SOC15(VCN, i, mmUVD_RB_SIZE2, ring->ring_size / 4); + fw_shared->multi_queue.encode_lowlatency_queue_mode &= ~FW_QUEUE_RING_RESET; } return 0; @@ -1079,33 +1118,33 @@ static int vcn_v2_5_mmsch_start(struct amdgpu_device *adev, * 1, write to vce_mmsch_vf_ctx_addr_lo/hi register with GPU mc addr of * memory descriptor location */ - WREG32_SOC15(UVD, 0, mmMMSCH_VF_CTX_ADDR_LO, lower_32_bits(addr)); - WREG32_SOC15(UVD, 0, mmMMSCH_VF_CTX_ADDR_HI, upper_32_bits(addr)); + WREG32_SOC15(VCN, 0, mmMMSCH_VF_CTX_ADDR_LO, lower_32_bits(addr)); + WREG32_SOC15(VCN, 0, mmMMSCH_VF_CTX_ADDR_HI, upper_32_bits(addr)); /* 2, update vmid of descriptor */ - data = RREG32_SOC15(UVD, 0, mmMMSCH_VF_VMID); + data = RREG32_SOC15(VCN, 0, mmMMSCH_VF_VMID); data &= ~MMSCH_VF_VMID__VF_CTX_VMID_MASK; /* use domain0 for MM scheduler */ data |= (0 << MMSCH_VF_VMID__VF_CTX_VMID__SHIFT); - WREG32_SOC15(UVD, 0, mmMMSCH_VF_VMID, data); + WREG32_SOC15(VCN, 0, mmMMSCH_VF_VMID, data); /* 3, notify mmsch about the size of this descriptor */ - WREG32_SOC15(UVD, 0, mmMMSCH_VF_CTX_SIZE, size); + WREG32_SOC15(VCN, 0, mmMMSCH_VF_CTX_SIZE, size); /* 4, set resp to zero */ - WREG32_SOC15(UVD, 0, mmMMSCH_VF_MAILBOX_RESP, 0); + WREG32_SOC15(VCN, 0, mmMMSCH_VF_MAILBOX_RESP, 0); /* * 5, kick off the initialization and wait until * VCE_MMSCH_VF_MAILBOX_RESP becomes non-zero */ - WREG32_SOC15(UVD, 0, mmMMSCH_VF_MAILBOX_HOST, 0x10000001); + WREG32_SOC15(VCN, 0, mmMMSCH_VF_MAILBOX_HOST, 0x10000001); - data = RREG32_SOC15(UVD, 0, mmMMSCH_VF_MAILBOX_RESP); + data = RREG32_SOC15(VCN, 0, mmMMSCH_VF_MAILBOX_RESP); loop = 10; while ((data & 0x10000002) != 0x10000002) { udelay(100); - data = RREG32_SOC15(UVD, 0, mmMMSCH_VF_MAILBOX_RESP); + data = RREG32_SOC15(VCN, 0, mmMMSCH_VF_MAILBOX_RESP); loop--; if (!loop) break; @@ -1128,14 +1167,12 @@ static int vcn_v2_5_sriov_start(struct amdgpu_device *adev) uint32_t table_size = 0; struct mmsch_v1_0_cmd_direct_write direct_wt = { { 0 } }; struct mmsch_v1_0_cmd_direct_read_modify_write direct_rd_mod_wt = { { 0 } }; - struct mmsch_v1_0_cmd_direct_polling direct_poll = { { 0 } }; struct mmsch_v1_0_cmd_end end = { { 0 } }; uint32_t *init_table = adev->virt.mm_table.cpu_addr; struct mmsch_v1_1_init_header *header = (struct mmsch_v1_1_init_header *)init_table; direct_wt.cmd_header.command_type = MMSCH_COMMAND__DIRECT_REG_WRITE; direct_rd_mod_wt.cmd_header.command_type = MMSCH_COMMAND__DIRECT_REG_READ_MODIFY_WRITE; - direct_poll.cmd_header.command_type = MMSCH_COMMAND__DIRECT_REG_POLLING; end.cmd_header.command_type = MMSCH_COMMAND__END; header->version = MMSCH_VERSION; @@ -1150,93 +1187,93 @@ static int vcn_v2_5_sriov_start(struct amdgpu_device *adev) table_size = 0; MMSCH_V1_0_INSERT_DIRECT_RD_MOD_WT( - SOC15_REG_OFFSET(UVD, i, mmUVD_STATUS), + SOC15_REG_OFFSET(VCN, i, mmUVD_STATUS), ~UVD_STATUS__UVD_BUSY, UVD_STATUS__UVD_BUSY); size = AMDGPU_GPU_PAGE_ALIGN(adev->vcn.fw->size + 4); /* mc resume*/ if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, + SOC15_REG_OFFSET(VCN, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + i].tmr_mc_addr_lo); MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, + SOC15_REG_OFFSET(VCN, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + i].tmr_mc_addr_hi); offset = 0; MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_OFFSET0), 0); + SOC15_REG_OFFSET(VCN, i, mmUVD_VCPU_CACHE_OFFSET0), 0); } else { MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, + SOC15_REG_OFFSET(VCN, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), lower_32_bits(adev->vcn.inst[i].gpu_addr)); MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, + SOC15_REG_OFFSET(VCN, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), upper_32_bits(adev->vcn.inst[i].gpu_addr)); offset = size; MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_OFFSET0), + SOC15_REG_OFFSET(VCN, i, mmUVD_VCPU_CACHE_OFFSET0), AMDGPU_UVD_FIRMWARE_OFFSET >> 3); } MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_SIZE0), + SOC15_REG_OFFSET(VCN, i, mmUVD_VCPU_CACHE_SIZE0), size); MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, + SOC15_REG_OFFSET(VCN, i, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW), lower_32_bits(adev->vcn.inst[i].gpu_addr + offset)); MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, + SOC15_REG_OFFSET(VCN, i, mmUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH), upper_32_bits(adev->vcn.inst[i].gpu_addr + offset)); MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_OFFSET1), + SOC15_REG_OFFSET(VCN, i, mmUVD_VCPU_CACHE_OFFSET1), 0); MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_SIZE1), + SOC15_REG_OFFSET(VCN, i, mmUVD_VCPU_CACHE_SIZE1), AMDGPU_VCN_STACK_SIZE); MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, + SOC15_REG_OFFSET(VCN, i, mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW), lower_32_bits(adev->vcn.inst[i].gpu_addr + offset + AMDGPU_VCN_STACK_SIZE)); MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, + SOC15_REG_OFFSET(VCN, i, mmUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH), upper_32_bits(adev->vcn.inst[i].gpu_addr + offset + AMDGPU_VCN_STACK_SIZE)); MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_OFFSET2), + SOC15_REG_OFFSET(VCN, i, mmUVD_VCPU_CACHE_OFFSET2), 0); MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CACHE_SIZE2), + SOC15_REG_OFFSET(VCN, i, mmUVD_VCPU_CACHE_SIZE2), AMDGPU_VCN_CONTEXT_SIZE); ring = &adev->vcn.inst[i].ring_enc[0]; ring->wptr = 0; MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, mmUVD_RB_BASE_LO), + SOC15_REG_OFFSET(VCN, i, mmUVD_RB_BASE_LO), lower_32_bits(ring->gpu_addr)); MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, mmUVD_RB_BASE_HI), + SOC15_REG_OFFSET(VCN, i, mmUVD_RB_BASE_HI), upper_32_bits(ring->gpu_addr)); MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, mmUVD_RB_SIZE), + SOC15_REG_OFFSET(VCN, i, mmUVD_RB_SIZE), ring->ring_size / 4); ring = &adev->vcn.inst[i].ring_dec; ring->wptr = 0; MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, + SOC15_REG_OFFSET(VCN, i, mmUVD_LMI_RBC_RB_64BIT_BAR_LOW), lower_32_bits(ring->gpu_addr)); MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, + SOC15_REG_OFFSET(VCN, i, mmUVD_LMI_RBC_RB_64BIT_BAR_HIGH), upper_32_bits(ring->gpu_addr)); @@ -1248,7 +1285,7 @@ static int vcn_v2_5_sriov_start(struct amdgpu_device *adev) tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_NO_UPDATE, 1); tmp = REG_SET_FIELD(tmp, UVD_RBC_RB_CNTL, RB_RPTR_WR_EN, 1); MMSCH_V1_0_INSERT_DIRECT_WT( - SOC15_REG_OFFSET(UVD, i, mmUVD_RBC_RB_CNTL), tmp); + SOC15_REG_OFFSET(VCN, i, mmUVD_RBC_RB_CNTL), tmp); /* add end packet */ memcpy((void *)init_table, &end, sizeof(struct mmsch_v1_0_cmd_end)); @@ -1269,24 +1306,24 @@ static int vcn_v2_5_stop_dpg_mode(struct amdgpu_device *adev, int inst_idx) uint32_t tmp; /* Wait for power status to be 1 */ - SOC15_WAIT_ON_RREG(UVD, inst_idx, mmUVD_POWER_STATUS, 1, + SOC15_WAIT_ON_RREG(VCN, inst_idx, mmUVD_POWER_STATUS, 1, UVD_POWER_STATUS__UVD_POWER_STATUS_MASK, ret_code); /* wait for read ptr to be equal to write ptr */ - tmp = RREG32_SOC15(UVD, inst_idx, mmUVD_RB_WPTR); - SOC15_WAIT_ON_RREG(UVD, inst_idx, mmUVD_RB_RPTR, tmp, 0xFFFFFFFF, ret_code); + tmp = RREG32_SOC15(VCN, inst_idx, mmUVD_RB_WPTR); + SOC15_WAIT_ON_RREG(VCN, inst_idx, mmUVD_RB_RPTR, tmp, 0xFFFFFFFF, ret_code); - tmp = RREG32_SOC15(UVD, inst_idx, mmUVD_RB_WPTR2); - SOC15_WAIT_ON_RREG(UVD, inst_idx, mmUVD_RB_RPTR2, tmp, 0xFFFFFFFF, ret_code); + tmp = RREG32_SOC15(VCN, inst_idx, mmUVD_RB_WPTR2); + SOC15_WAIT_ON_RREG(VCN, inst_idx, mmUVD_RB_RPTR2, tmp, 0xFFFFFFFF, ret_code); - tmp = RREG32_SOC15(UVD, inst_idx, mmUVD_RBC_RB_WPTR) & 0x7FFFFFFF; - SOC15_WAIT_ON_RREG(UVD, inst_idx, mmUVD_RBC_RB_RPTR, tmp, 0xFFFFFFFF, ret_code); + tmp = RREG32_SOC15(VCN, inst_idx, mmUVD_RBC_RB_WPTR) & 0x7FFFFFFF; + SOC15_WAIT_ON_RREG(VCN, inst_idx, mmUVD_RBC_RB_RPTR, tmp, 0xFFFFFFFF, ret_code); - SOC15_WAIT_ON_RREG(UVD, inst_idx, mmUVD_POWER_STATUS, 1, + SOC15_WAIT_ON_RREG(VCN, inst_idx, mmUVD_POWER_STATUS, 1, UVD_POWER_STATUS__UVD_POWER_STATUS_MASK, ret_code); /* disable dynamic power gating mode */ - WREG32_P(SOC15_REG_OFFSET(UVD, inst_idx, mmUVD_POWER_STATUS), 0, + WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, mmUVD_POWER_STATUS), 0, ~UVD_POWER_STATUS__UVD_PG_MODE_MASK); return 0; @@ -1330,17 +1367,17 @@ static int vcn_v2_5_stop(struct amdgpu_device *adev) return r; /* block VCPU register access */ - WREG32_P(SOC15_REG_OFFSET(UVD, i, mmUVD_RB_ARB_CTRL), + WREG32_P(SOC15_REG_OFFSET(VCN, i, mmUVD_RB_ARB_CTRL), UVD_RB_ARB_CTRL__VCPU_DIS_MASK, ~UVD_RB_ARB_CTRL__VCPU_DIS_MASK); /* reset VCPU */ - WREG32_P(SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CNTL), + WREG32_P(SOC15_REG_OFFSET(VCN, i, mmUVD_VCPU_CNTL), UVD_VCPU_CNTL__BLK_RST_MASK, ~UVD_VCPU_CNTL__BLK_RST_MASK); /* disable VCPU clock */ - WREG32_P(SOC15_REG_OFFSET(UVD, i, mmUVD_VCPU_CNTL), 0, + WREG32_P(SOC15_REG_OFFSET(VCN, i, mmUVD_VCPU_CNTL), 0, ~(UVD_VCPU_CNTL__CLK_EN_MASK)); /* clear status */ @@ -1349,7 +1386,7 @@ static int vcn_v2_5_stop(struct amdgpu_device *adev) vcn_v2_5_enable_clock_gating(adev); /* enable register anti-hang mechanism */ - WREG32_P(SOC15_REG_OFFSET(UVD, i, mmUVD_POWER_STATUS), + WREG32_P(SOC15_REG_OFFSET(VCN, i, mmUVD_POWER_STATUS), UVD_POWER_STATUS__UVD_POWER_STATUS_MASK, ~UVD_POWER_STATUS__UVD_POWER_STATUS_MASK); } @@ -1365,55 +1402,69 @@ static int vcn_v2_5_pause_dpg_mode(struct amdgpu_device *adev, { struct amdgpu_ring *ring; uint32_t reg_data = 0; - int ret_code; + int ret_code = 0; /* pause/unpause if state is changed */ if (adev->vcn.inst[inst_idx].pause_state.fw_based != new_state->fw_based) { DRM_DEBUG("dpg pause state changed %d -> %d", adev->vcn.inst[inst_idx].pause_state.fw_based, new_state->fw_based); - reg_data = RREG32_SOC15(UVD, inst_idx, mmUVD_DPG_PAUSE) & + reg_data = RREG32_SOC15(VCN, inst_idx, mmUVD_DPG_PAUSE) & (~UVD_DPG_PAUSE__NJ_PAUSE_DPG_ACK_MASK); if (new_state->fw_based == VCN_DPG_STATE__PAUSE) { - ret_code = 0; - SOC15_WAIT_ON_RREG(UVD, inst_idx, mmUVD_POWER_STATUS, 0x1, + SOC15_WAIT_ON_RREG(VCN, inst_idx, mmUVD_POWER_STATUS, 0x1, UVD_POWER_STATUS__UVD_POWER_STATUS_MASK, ret_code); if (!ret_code) { + volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared_cpu_addr; + /* pause DPG */ reg_data |= UVD_DPG_PAUSE__NJ_PAUSE_DPG_REQ_MASK; - WREG32_SOC15(UVD, inst_idx, mmUVD_DPG_PAUSE, reg_data); + WREG32_SOC15(VCN, inst_idx, mmUVD_DPG_PAUSE, reg_data); /* wait for ACK */ - SOC15_WAIT_ON_RREG(UVD, inst_idx, mmUVD_DPG_PAUSE, + SOC15_WAIT_ON_RREG(VCN, inst_idx, mmUVD_DPG_PAUSE, UVD_DPG_PAUSE__NJ_PAUSE_DPG_ACK_MASK, UVD_DPG_PAUSE__NJ_PAUSE_DPG_ACK_MASK, ret_code); + /* Stall DPG before WPTR/RPTR reset */ + WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, mmUVD_POWER_STATUS), + UVD_POWER_STATUS__STALL_DPG_POWER_UP_MASK, + ~UVD_POWER_STATUS__STALL_DPG_POWER_UP_MASK); + /* Restore */ + fw_shared->multi_queue.encode_generalpurpose_queue_mode |= FW_QUEUE_RING_RESET; ring = &adev->vcn.inst[inst_idx].ring_enc[0]; - WREG32_SOC15(UVD, inst_idx, mmUVD_RB_BASE_LO, ring->gpu_addr); - WREG32_SOC15(UVD, inst_idx, mmUVD_RB_BASE_HI, upper_32_bits(ring->gpu_addr)); - WREG32_SOC15(UVD, inst_idx, mmUVD_RB_SIZE, ring->ring_size / 4); - WREG32_SOC15(UVD, inst_idx, mmUVD_RB_RPTR, lower_32_bits(ring->wptr)); - WREG32_SOC15(UVD, inst_idx, mmUVD_RB_WPTR, lower_32_bits(ring->wptr)); - + ring->wptr = 0; + WREG32_SOC15(VCN, inst_idx, mmUVD_RB_BASE_LO, ring->gpu_addr); + WREG32_SOC15(VCN, inst_idx, mmUVD_RB_BASE_HI, upper_32_bits(ring->gpu_addr)); + WREG32_SOC15(VCN, inst_idx, mmUVD_RB_SIZE, ring->ring_size / 4); + WREG32_SOC15(VCN, inst_idx, mmUVD_RB_RPTR, lower_32_bits(ring->wptr)); + WREG32_SOC15(VCN, inst_idx, mmUVD_RB_WPTR, lower_32_bits(ring->wptr)); + fw_shared->multi_queue.encode_generalpurpose_queue_mode &= ~FW_QUEUE_RING_RESET; + + fw_shared->multi_queue.encode_lowlatency_queue_mode |= FW_QUEUE_RING_RESET; ring = &adev->vcn.inst[inst_idx].ring_enc[1]; - WREG32_SOC15(UVD, inst_idx, mmUVD_RB_BASE_LO2, ring->gpu_addr); - WREG32_SOC15(UVD, inst_idx, mmUVD_RB_BASE_HI2, upper_32_bits(ring->gpu_addr)); - WREG32_SOC15(UVD, inst_idx, mmUVD_RB_SIZE2, ring->ring_size / 4); - WREG32_SOC15(UVD, inst_idx, mmUVD_RB_RPTR2, lower_32_bits(ring->wptr)); - WREG32_SOC15(UVD, inst_idx, mmUVD_RB_WPTR2, lower_32_bits(ring->wptr)); - - WREG32_SOC15(UVD, inst_idx, mmUVD_RBC_RB_WPTR, - RREG32_SOC15(UVD, inst_idx, mmUVD_SCRATCH2) & 0x7FFFFFFF); - - SOC15_WAIT_ON_RREG(UVD, inst_idx, mmUVD_POWER_STATUS, + ring->wptr = 0; + WREG32_SOC15(VCN, inst_idx, mmUVD_RB_BASE_LO2, ring->gpu_addr); + WREG32_SOC15(VCN, inst_idx, mmUVD_RB_BASE_HI2, upper_32_bits(ring->gpu_addr)); + WREG32_SOC15(VCN, inst_idx, mmUVD_RB_SIZE2, ring->ring_size / 4); + WREG32_SOC15(VCN, inst_idx, mmUVD_RB_RPTR2, lower_32_bits(ring->wptr)); + WREG32_SOC15(VCN, inst_idx, mmUVD_RB_WPTR2, lower_32_bits(ring->wptr)); + fw_shared->multi_queue.encode_lowlatency_queue_mode &= ~FW_QUEUE_RING_RESET; + + /* Unstall DPG */ + WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, mmUVD_POWER_STATUS), + 0, ~UVD_POWER_STATUS__STALL_DPG_POWER_UP_MASK); + + SOC15_WAIT_ON_RREG(VCN, inst_idx, mmUVD_POWER_STATUS, UVD_PGFSM_CONFIG__UVDM_UVDU_PWR_ON, UVD_POWER_STATUS__UVD_POWER_STATUS_MASK, ret_code); } } else { - /* unpause dpg, no need to wait */ reg_data &= ~UVD_DPG_PAUSE__NJ_PAUSE_DPG_REQ_MASK; - WREG32_SOC15(UVD, inst_idx, mmUVD_DPG_PAUSE, reg_data); + WREG32_SOC15(VCN, inst_idx, mmUVD_DPG_PAUSE, reg_data); + SOC15_WAIT_ON_RREG(VCN, inst_idx, mmUVD_POWER_STATUS, 0x1, + UVD_POWER_STATUS__UVD_POWER_STATUS_MASK, ret_code); } adev->vcn.inst[inst_idx].pause_state.fw_based = new_state->fw_based; } @@ -1432,7 +1483,7 @@ static uint64_t vcn_v2_5_dec_ring_get_rptr(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; - return RREG32_SOC15(UVD, ring->me, mmUVD_RBC_RB_RPTR); + return RREG32_SOC15(VCN, ring->me, mmUVD_RBC_RB_RPTR); } /** @@ -1449,7 +1500,7 @@ static uint64_t vcn_v2_5_dec_ring_get_wptr(struct amdgpu_ring *ring) if (ring->use_doorbell) return adev->wb.wb[ring->wptr_offs]; else - return RREG32_SOC15(UVD, ring->me, mmUVD_RBC_RB_WPTR); + return RREG32_SOC15(VCN, ring->me, mmUVD_RBC_RB_WPTR); } /** @@ -1463,15 +1514,11 @@ static void vcn_v2_5_dec_ring_set_wptr(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; - if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) - WREG32_SOC15(UVD, ring->me, mmUVD_SCRATCH2, - lower_32_bits(ring->wptr) | 0x80000000); - if (ring->use_doorbell) { adev->wb.wb[ring->wptr_offs] = lower_32_bits(ring->wptr); WDOORBELL32(ring->doorbell_index, lower_32_bits(ring->wptr)); } else { - WREG32_SOC15(UVD, ring->me, mmUVD_RBC_RB_WPTR, lower_32_bits(ring->wptr)); + WREG32_SOC15(VCN, ring->me, mmUVD_RBC_RB_WPTR, lower_32_bits(ring->wptr)); } } @@ -1517,9 +1564,9 @@ static uint64_t vcn_v2_5_enc_ring_get_rptr(struct amdgpu_ring *ring) struct amdgpu_device *adev = ring->adev; if (ring == &adev->vcn.inst[ring->me].ring_enc[0]) - return RREG32_SOC15(UVD, ring->me, mmUVD_RB_RPTR); + return RREG32_SOC15(VCN, ring->me, mmUVD_RB_RPTR); else - return RREG32_SOC15(UVD, ring->me, mmUVD_RB_RPTR2); + return RREG32_SOC15(VCN, ring->me, mmUVD_RB_RPTR2); } /** @@ -1537,12 +1584,12 @@ static uint64_t vcn_v2_5_enc_ring_get_wptr(struct amdgpu_ring *ring) if (ring->use_doorbell) return adev->wb.wb[ring->wptr_offs]; else - return RREG32_SOC15(UVD, ring->me, mmUVD_RB_WPTR); + return RREG32_SOC15(VCN, ring->me, mmUVD_RB_WPTR); } else { if (ring->use_doorbell) return adev->wb.wb[ring->wptr_offs]; else - return RREG32_SOC15(UVD, ring->me, mmUVD_RB_WPTR2); + return RREG32_SOC15(VCN, ring->me, mmUVD_RB_WPTR2); } } @@ -1562,14 +1609,14 @@ static void vcn_v2_5_enc_ring_set_wptr(struct amdgpu_ring *ring) adev->wb.wb[ring->wptr_offs] = lower_32_bits(ring->wptr); WDOORBELL32(ring->doorbell_index, lower_32_bits(ring->wptr)); } else { - WREG32_SOC15(UVD, ring->me, mmUVD_RB_WPTR, lower_32_bits(ring->wptr)); + WREG32_SOC15(VCN, ring->me, mmUVD_RB_WPTR, lower_32_bits(ring->wptr)); } } else { if (ring->use_doorbell) { adev->wb.wb[ring->wptr_offs] = lower_32_bits(ring->wptr); WDOORBELL32(ring->doorbell_index, lower_32_bits(ring->wptr)); } else { - WREG32_SOC15(UVD, ring->me, mmUVD_RB_WPTR2, lower_32_bits(ring->wptr)); + WREG32_SOC15(VCN, ring->me, mmUVD_RB_WPTR2, lower_32_bits(ring->wptr)); } } } diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c index 78b35901643b..af8986a55354 100644 --- a/drivers/gpu/drm/amd/amdgpu/vi.c +++ b/drivers/gpu/drm/amd/amdgpu/vi.c @@ -448,27 +448,6 @@ static bool vi_read_bios_from_rom(struct amdgpu_device *adev, return true; } -static void vi_detect_hw_virtualization(struct amdgpu_device *adev) -{ - uint32_t reg = 0; - - if (adev->asic_type == CHIP_TONGA || - adev->asic_type == CHIP_FIJI) { - reg = RREG32(mmBIF_IOV_FUNC_IDENTIFIER); - /* bit0: 0 means pf and 1 means vf */ - if (REG_GET_FIELD(reg, BIF_IOV_FUNC_IDENTIFIER, FUNC_IDENTIFIER)) - adev->virt.caps |= AMDGPU_SRIOV_CAPS_IS_VF; - /* bit31: 0 means disable IOV and 1 means enable */ - if (REG_GET_FIELD(reg, BIF_IOV_FUNC_IDENTIFIER, IOV_ENABLE)) - adev->virt.caps |= AMDGPU_SRIOV_CAPS_ENABLE_IOV; - } - - if (reg == 0) { - if (is_virtual_machine()) /* passthrough mode exclus sr-iov mode */ - adev->virt.caps |= AMDGPU_PASSTHROUGH_MODE; - } -} - static const struct amdgpu_allowed_register_entry vi_allowed_read_registers[] = { {mmGRBM_STATUS}, {mmGRBM_STATUS2}, @@ -765,8 +744,6 @@ static int vi_asic_reset(struct amdgpu_device *adev) int r; if (vi_asic_reset_method(adev) == AMD_RESET_METHOD_BACO) { - if (!adev->in_suspend) - amdgpu_inc_vram_lost(adev); r = amdgpu_dpm_baco_reset(adev); } else { r = vi_asic_pci_config_reset(adev); @@ -1730,9 +1707,6 @@ static const struct amdgpu_ip_block_version vi_common_ip_block = int vi_set_ip_blocks(struct amdgpu_device *adev) { - /* in early init stage, vbios code won't work */ - vi_detect_hw_virtualization(adev); - if (amdgpu_sriov_vf(adev)) adev->virt.ops = &xgpu_vi_virt_ops; diff --git a/drivers/gpu/drm/amd/amdgpu/vid.h b/drivers/gpu/drm/amd/amdgpu/vid.h index 19ddd2312e00..7a01e6133798 100644 --- a/drivers/gpu/drm/amd/amdgpu/vid.h +++ b/drivers/gpu/drm/amd/amdgpu/vid.h @@ -332,7 +332,7 @@ # define PACKET3_DMA_DATA_CMD_SAIC (1 << 28) # define PACKET3_DMA_DATA_CMD_DAIC (1 << 29) # define PACKET3_DMA_DATA_CMD_RAW_WAIT (1 << 30) -#define PACKET3_AQUIRE_MEM 0x58 +#define PACKET3_ACQUIRE_MEM 0x58 #define PACKET3_REWIND 0x59 #define PACKET3_LOAD_UCONFIG_REG 0x5E #define PACKET3_LOAD_SH_REG 0x5F diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c index 0ec5f25adf56..cf0017f4d9d5 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c @@ -215,6 +215,7 @@ static int set_queue_properties_from_user(struct queue_properties *q_properties, } q_properties->is_interop = false; + q_properties->is_gws = false; q_properties->queue_percent = args->queue_percentage; q_properties->priority = args->queue_priority; q_properties->queue_address = args->ring_base_address; @@ -1322,6 +1323,10 @@ static int kfd_ioctl_alloc_memory_of_gpu(struct file *filep, goto err_free; } + /* Update the VRAM usage count */ + if (flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) + WRITE_ONCE(pdd->vram_usage, pdd->vram_usage + args->size); + mutex_unlock(&p->mutex); args->handle = MAKE_HANDLE(args->gpu_id, idr_handle); @@ -1337,7 +1342,7 @@ static int kfd_ioctl_alloc_memory_of_gpu(struct file *filep, return 0; err_free: - amdgpu_amdkfd_gpuvm_free_memory_of_gpu(dev->kgd, (struct kgd_mem *)mem); + amdgpu_amdkfd_gpuvm_free_memory_of_gpu(dev->kgd, (struct kgd_mem *)mem, NULL); err_unlock: mutex_unlock(&p->mutex); return err; @@ -1351,6 +1356,7 @@ static int kfd_ioctl_free_memory_of_gpu(struct file *filep, void *mem; struct kfd_dev *dev; int ret; + uint64_t size = 0; dev = kfd_device_by_id(GET_GPU_ID(args->handle)); if (!dev) @@ -1373,7 +1379,7 @@ static int kfd_ioctl_free_memory_of_gpu(struct file *filep, } ret = amdgpu_amdkfd_gpuvm_free_memory_of_gpu(dev->kgd, - (struct kgd_mem *)mem); + (struct kgd_mem *)mem, &size); /* If freeing the buffer failed, leave the handle in place for * clean-up during process tear-down. @@ -1382,6 +1388,8 @@ static int kfd_ioctl_free_memory_of_gpu(struct file *filep, kfd_process_device_remove_obj_handle( pdd, GET_IDR_HANDLE(args->handle)); + WRITE_ONCE(pdd->vram_usage, pdd->vram_usage - size); + err_unlock: mutex_unlock(&p->mutex); return ret; @@ -1584,6 +1592,45 @@ copy_from_user_failed: return err; } +static int kfd_ioctl_alloc_queue_gws(struct file *filep, + struct kfd_process *p, void *data) +{ + int retval; + struct kfd_ioctl_alloc_queue_gws_args *args = data; + struct queue *q; + struct kfd_dev *dev; + + mutex_lock(&p->mutex); + q = pqm_get_user_queue(&p->pqm, args->queue_id); + + if (q) { + dev = q->device; + } else { + retval = -EINVAL; + goto out_unlock; + } + + if (!dev->gws) { + retval = -ENODEV; + goto out_unlock; + } + + if (dev->dqm->sched_policy == KFD_SCHED_POLICY_NO_HWS) { + retval = -ENODEV; + goto out_unlock; + } + + retval = pqm_set_gws(&p->pqm, args->queue_id, args->num_gws ? dev->gws : NULL); + mutex_unlock(&p->mutex); + + args->first_gws = 0; + return retval; + +out_unlock: + mutex_unlock(&p->mutex); + return retval; +} + static int kfd_ioctl_get_dmabuf_info(struct file *filep, struct kfd_process *p, void *data) { @@ -1687,7 +1734,7 @@ static int kfd_ioctl_import_dmabuf(struct file *filep, return 0; err_free: - amdgpu_amdkfd_gpuvm_free_memory_of_gpu(dev->kgd, (struct kgd_mem *)mem); + amdgpu_amdkfd_gpuvm_free_memory_of_gpu(dev->kgd, (struct kgd_mem *)mem, NULL); err_unlock: mutex_unlock(&p->mutex); return r; @@ -1786,6 +1833,8 @@ static const struct amdkfd_ioctl_desc amdkfd_ioctls[] = { AMDKFD_IOCTL_DEF(AMDKFD_IOC_IMPORT_DMABUF, kfd_ioctl_import_dmabuf, 0), + AMDKFD_IOCTL_DEF(AMDKFD_IOC_ALLOC_QUEUE_GWS, + kfd_ioctl_alloc_queue_gws, 0), }; #define AMDKFD_CORE_IOCTL_COUNT ARRAY_SIZE(amdkfd_ioctls) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c index de9f68d5c312..1009a3b8dcc2 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c @@ -502,7 +502,7 @@ int kfd_parse_crat_table(void *crat_image, struct list_head *device_list, num_nodes = crat_table->num_domains; image_len = crat_table->length; - pr_info("Parsing CRAT table with %d nodes\n", num_nodes); + pr_debug("Parsing CRAT table with %d nodes\n", num_nodes); for (node_id = 0; node_id < num_nodes; node_id++) { top_dev = kfd_create_topology_device(device_list); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c index 05bc6d96ec52..0491ab2b4a9b 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c @@ -569,6 +569,23 @@ static void kfd_cwsr_init(struct kfd_dev *kfd) } } +static int kfd_gws_init(struct kfd_dev *kfd) +{ + int ret = 0; + + if (kfd->dqm->sched_policy == KFD_SCHED_POLICY_NO_HWS) + return 0; + + if (hws_gws_support + || (kfd->device_info->asic_family >= CHIP_VEGA10 + && kfd->device_info->asic_family <= CHIP_RAVEN + && kfd->mec2_fw_version >= 0x1b3)) + ret = amdgpu_amdkfd_alloc_gws(kfd->kgd, + amdgpu_amdkfd_get_num_gws(kfd->kgd), &kfd->gws); + + return ret; +} + bool kgd2kfd_device_init(struct kfd_dev *kfd, struct drm_device *ddev, const struct kgd2kfd_shared_resources *gpu_resources) @@ -578,6 +595,8 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, kfd->ddev = ddev; kfd->mec_fw_version = amdgpu_amdkfd_get_fw_version(kfd->kgd, KGD_ENGINE_MEC1); + kfd->mec2_fw_version = amdgpu_amdkfd_get_fw_version(kfd->kgd, + KGD_ENGINE_MEC2); kfd->sdma_fw_version = amdgpu_amdkfd_get_fw_version(kfd->kgd, KGD_ENGINE_SDMA1); kfd->shared_resources = *gpu_resources; @@ -598,13 +617,6 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, } else kfd->max_proc_per_quantum = hws_max_conc_proc; - /* Allocate global GWS that is shared by all KFD processes */ - if (hws_gws_support && amdgpu_amdkfd_alloc_gws(kfd->kgd, - amdgpu_amdkfd_get_num_gws(kfd->kgd), &kfd->gws)) { - dev_err(kfd_device, "Could not allocate %d gws\n", - amdgpu_amdkfd_get_num_gws(kfd->kgd)); - goto out; - } /* calculate max size of mqds needed for queues */ size = max_num_of_queues_per_device * kfd->device_info->mqd_size_aligned; @@ -662,6 +674,15 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, goto device_queue_manager_error; } + /* If supported on this device, allocate global GWS that is shared + * by all KFD processes + */ + if (kfd_gws_init(kfd)) { + dev_err(kfd_device, "Could not allocate %d gws\n", + amdgpu_amdkfd_get_num_gws(kfd->kgd)); + goto gws_error; + } + if (kfd_iommu_device_init(kfd)) { dev_err(kfd_device, "Error initializing iommuv2\n"); goto device_iommu_error; @@ -691,6 +712,7 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, kfd_topology_add_device_error: kfd_resume_error: device_iommu_error: +gws_error: device_queue_manager_uninit(kfd->dqm); device_queue_manager_error: kfd_interrupt_exit(kfd); @@ -701,7 +723,7 @@ kfd_doorbell_error: kfd_gtt_sa_init_error: amdgpu_amdkfd_free_gtt_mem(kfd->kgd, kfd->gtt_mem); alloc_gtt_mem_failure: - if (hws_gws_support) + if (kfd->gws) amdgpu_amdkfd_free_gws(kfd->kgd, kfd->gws); dev_err(kfd_device, "device %x:%x NOT added due to errors\n", @@ -720,7 +742,7 @@ void kgd2kfd_device_exit(struct kfd_dev *kfd) kfd_doorbell_fini(kfd); kfd_gtt_sa_fini(kfd); amdgpu_amdkfd_free_gtt_mem(kfd->kgd, kfd->gtt_mem); - if (hws_gws_support) + if (kfd->gws) amdgpu_amdkfd_free_gws(kfd->kgd, kfd->gws); } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index 77ea0f0cb163..e9c4867abeff 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -505,8 +505,13 @@ static int destroy_queue_nocpsch_locked(struct device_queue_manager *dqm, deallocate_vmid(dqm, qpd, q); } qpd->queue_count--; - if (q->properties.is_active) + if (q->properties.is_active) { decrement_queue_count(dqm, q->properties.type); + if (q->properties.is_gws) { + dqm->gws_queue_count--; + qpd->mapped_gws_queue = false; + } + } return retval; } @@ -583,6 +588,20 @@ static int update_queue(struct device_queue_manager *dqm, struct queue *q) else if (!q->properties.is_active && prev_active) decrement_queue_count(dqm, q->properties.type); + if (q->gws && !q->properties.is_gws) { + if (q->properties.is_active) { + dqm->gws_queue_count++; + pdd->qpd.mapped_gws_queue = true; + } + q->properties.is_gws = true; + } else if (!q->gws && q->properties.is_gws) { + if (q->properties.is_active) { + dqm->gws_queue_count--; + pdd->qpd.mapped_gws_queue = false; + } + q->properties.is_gws = false; + } + if (dqm->sched_policy != KFD_SCHED_POLICY_NO_HWS) retval = map_queues_cpsch(dqm); else if (q->properties.is_active && @@ -631,6 +650,10 @@ static int evict_process_queues_nocpsch(struct device_queue_manager *dqm, q->properties.type)]; q->properties.is_active = false; decrement_queue_count(dqm, q->properties.type); + if (q->properties.is_gws) { + dqm->gws_queue_count--; + qpd->mapped_gws_queue = false; + } if (WARN_ONCE(!dqm->sched_running, "Evict when stopped\n")) continue; @@ -744,6 +767,10 @@ static int restore_process_queues_nocpsch(struct device_queue_manager *dqm, q->properties.type)]; q->properties.is_active = true; increment_queue_count(dqm, q->properties.type); + if (q->properties.is_gws) { + dqm->gws_queue_count++; + qpd->mapped_gws_queue = true; + } if (WARN_ONCE(!dqm->sched_running, "Restore when stopped\n")) continue; @@ -913,6 +940,7 @@ static int initialize_nocpsch(struct device_queue_manager *dqm) INIT_LIST_HEAD(&dqm->queues); dqm->active_queue_count = dqm->next_pipe_to_allocate = 0; dqm->active_cp_queue_count = 0; + dqm->gws_queue_count = 0; for (pipe = 0; pipe < get_pipes_per_mec(dqm); pipe++) { int pipe_offset = pipe * get_queues_per_pipe(dqm); @@ -1061,7 +1089,9 @@ static int set_sched_resources(struct device_queue_manager *dqm) break; } - res.queue_mask |= (1ull << i); + res.queue_mask |= 1ull + << amdgpu_queue_mask_bit_to_set_resource_bit( + (struct amdgpu_device *)dqm->dev->kgd, i); } res.gws_mask = ~0ull; res.oac_mask = res.gds_heap_base = res.gds_heap_size = 0; @@ -1082,7 +1112,7 @@ static int initialize_cpsch(struct device_queue_manager *dqm) INIT_LIST_HEAD(&dqm->queues); dqm->active_queue_count = dqm->processes_count = 0; dqm->active_cp_queue_count = 0; - + dqm->gws_queue_count = 0; dqm->active_runlist = false; dqm->sdma_bitmap = ~0ULL >> (64 - get_num_sdma_queues(dqm)); dqm->xgmi_sdma_bitmap = ~0ULL >> (64 - get_num_xgmi_sdma_queues(dqm)); @@ -1432,6 +1462,10 @@ static int destroy_queue_cpsch(struct device_queue_manager *dqm, KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0); if (retval == -ETIME) qpd->reset_wavefronts = true; + if (q->properties.is_gws) { + dqm->gws_queue_count--; + qpd->mapped_gws_queue = false; + } } /* @@ -1650,8 +1684,13 @@ static int process_termination_cpsch(struct device_queue_manager *dqm, else if (q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) deallocate_sdma_queue(dqm, q); - if (q->properties.is_active) + if (q->properties.is_active) { decrement_queue_count(dqm, q->properties.type); + if (q->properties.is_gws) { + dqm->gws_queue_count--; + qpd->mapped_gws_queue = false; + } + } dqm->total_queue_count--; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h index 50d919f814e9..4afa015c69b1 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h @@ -182,6 +182,7 @@ struct device_queue_manager { unsigned int processes_count; unsigned int active_queue_count; unsigned int active_cp_queue_count; + unsigned int gws_queue_count; unsigned int total_queue_count; unsigned int next_pipe_to_allocate; unsigned int *allocated_queues; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c index 15476fca8fa6..a9583b95fcc1 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c @@ -901,7 +901,7 @@ void kfd_signal_iommu_event(struct kfd_dev *dev, unsigned int pasid, memset(&memory_exception_data, 0, sizeof(memory_exception_data)); - down_read(&mm->mmap_sem); + mmap_read_lock(mm); vma = find_vma(mm, address); memory_exception_data.gpu_id = dev->id; @@ -924,7 +924,7 @@ void kfd_signal_iommu_event(struct kfd_dev *dev, unsigned int pasid, memory_exception_data.failure.NoExecute = 0; } - up_read(&mm->mmap_sem); + mmap_read_unlock(mm); mmput(mm); pr_debug("notpresent %d, noexecute %d, readonly %d\n", diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c index e05d75ecda21..fce6ccabe38b 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c @@ -37,7 +37,7 @@ static bool event_interrupt_isr_v9(struct kfd_dev *dev, vmid = SOC15_VMID_FROM_IH_ENTRY(ih_ring_entry); if (vmid < dev->vm_info.first_vmid_kfd || vmid > dev->vm_info.last_vmid_kfd) - return 0; + return false; source_id = SOC15_SOURCE_ID_FROM_IH_ENTRY(ih_ring_entry); client_id = SOC15_CLIENT_ID_FROM_IH_ENTRY(ih_ring_entry); @@ -69,7 +69,7 @@ static bool event_interrupt_isr_v9(struct kfd_dev *dev, /* If there is no valid PASID, it's likely a bug */ if (WARN_ONCE(pasid == 0, "Bug: No PASID in KFD interrupt")) - return 0; + return false; /* Interrupt types we care about: various signals and faults. * They will be forwarded to a work queue (see below). diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c b/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c index 8d871514671e..7c8786b9eb0a 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c @@ -192,7 +192,7 @@ static int iommu_invalid_ppr_cb(struct pci_dev *pdev, int pasid, dev_warn_ratelimited(kfd_device, "Invalid PPR device %x:%x.%x pasid 0x%x address 0x%lX flags 0x%X", - PCI_BUS_NUM(pdev->devfn), + pdev->bus->number, PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn), pasid, diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c index bae706462f96..a2b77d1df854 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c @@ -126,6 +126,7 @@ static bool kq_initialize(struct kernel_queue *kq, struct kfd_dev *dev, prop.queue_size = queue_size; prop.is_interop = false; + prop.is_gws = false; prop.priority = 1; prop.queue_percent = 100; prop.type = type; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c index efdb75e7677b..685ca82d42fe 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c @@ -41,7 +41,7 @@ static void pm_calc_rlib_size(struct packet_manager *pm, unsigned int *rlib_size, bool *over_subscription) { - unsigned int process_count, queue_count, compute_queue_count; + unsigned int process_count, queue_count, compute_queue_count, gws_queue_count; unsigned int map_queue_size; unsigned int max_proc_per_quantum = 1; struct kfd_dev *dev = pm->dqm->dev; @@ -49,6 +49,7 @@ static void pm_calc_rlib_size(struct packet_manager *pm, process_count = pm->dqm->processes_count; queue_count = pm->dqm->active_queue_count; compute_queue_count = pm->dqm->active_cp_queue_count; + gws_queue_count = pm->dqm->gws_queue_count; /* check if there is over subscription * Note: the arbitration between the number of VMIDs and @@ -61,7 +62,8 @@ static void pm_calc_rlib_size(struct packet_manager *pm, max_proc_per_quantum = dev->max_proc_per_quantum; if ((process_count > max_proc_per_quantum) || - compute_queue_count > get_cp_queues_num(pm->dqm)) { + compute_queue_count > get_cp_queues_num(pm->dqm) || + gws_queue_count > 1) { *over_subscription = true; pr_debug("Over subscribed runlist\n"); } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c index 2de01009f1b6..bdca9dc5f118 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c @@ -43,7 +43,7 @@ static int pm_map_process_v9(struct packet_manager *pm, packet->bitfields2.pasid = qpd->pqm->process->pasid; packet->bitfields14.gds_size = qpd->gds_size & 0x3F; packet->bitfields14.gds_size_hi = (qpd->gds_size >> 6) & 0xF; - packet->bitfields14.num_gws = qpd->num_gws; + packet->bitfields14.num_gws = (qpd->mapped_gws_queue) ? qpd->num_gws : 0; packet->bitfields14.num_oac = qpd->num_oac; packet->bitfields14.sdma_enable = 1; packet->bitfields14.num_queues = (qpd->is_debug) ? 0 : qpd->queue_count; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h index 4a3049841086..fee60921fccf 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h @@ -40,6 +40,7 @@ #include <drm/drm_file.h> #include <drm/drm_drv.h> #include <drm/drm_device.h> +#include <drm/drm_ioctl.h> #include <kgd_kfd_interface.h> #include <linux/swap.h> @@ -282,6 +283,7 @@ struct kfd_dev { /* Firmware versions */ uint16_t mec_fw_version; + uint16_t mec2_fw_version; uint16_t sdma_fw_version; /* Maximum process number mapped to HW scheduler */ @@ -410,6 +412,10 @@ enum KFD_QUEUE_PRIORITY { * @is_active: Defines if the queue is active or not. @is_active and * @is_evicted are protected by the DQM lock. * + * @is_gws: Defines if the queue has been updated to be GWS-capable or not. + * @is_gws should be protected by the DQM lock, since changing it can yield the + * possibility of updating DQM state on number of GWS queues. + * * @vmid: If the scheduling mode is no cp scheduling the field defines the vmid * of the queue. * @@ -432,6 +438,7 @@ struct queue_properties { bool is_interop; bool is_evicted; bool is_active; + bool is_gws; /* Not relevant for user mode queues in cp scheduling */ unsigned int vmid; /* Relevant only for sdma queues*/ @@ -563,6 +570,14 @@ struct qcm_process_device { */ bool reset_wavefronts; + /* This flag tells us if this process has a GWS-capable + * queue that will be mapped into the runlist. It's + * possible to request a GWS BO, but not have the queue + * currently mapped, and this changes how the MAP_PROCESS + * PM4 packet is configured. + */ + bool mapped_gws_queue; + /* * All the memory management data should be here too */ @@ -615,6 +630,8 @@ enum kfd_pdd_bound { PDD_BOUND_SUSPENDED, }; +#define MAX_VRAM_FILENAME_LEN 11 + /* Data that is per-process-per device. */ struct kfd_process_device { /* @@ -657,6 +674,11 @@ struct kfd_process_device { /* Is this process/pasid bound to this device? (amd_iommu_bind_pasid) */ enum kfd_pdd_bound bound; + + /* VRAM usage */ + uint64_t vram_usage; + struct attribute attr_vram; + char vram_filename[MAX_VRAM_FILENAME_LEN]; }; #define qpd_to_pdd(x) container_of(x, struct kfd_process_device, qpd) @@ -923,6 +945,8 @@ int pqm_set_gws(struct process_queue_manager *pqm, unsigned int qid, void *gws); struct kernel_queue *pqm_get_kernel_queue(struct process_queue_manager *pqm, unsigned int qid); +struct queue *pqm_get_user_queue(struct process_queue_manager *pqm, + unsigned int qid); int pqm_get_wave_state(struct process_queue_manager *pqm, unsigned int qid, void __user *ctl_stack, @@ -1050,10 +1074,10 @@ void kfd_dec_compute_active(struct kfd_dev *dev); /* Check with device cgroup if @kfd device is accessible */ static inline int kfd_devcgroup_check_permission(struct kfd_dev *kfd) { -#if defined(CONFIG_CGROUP_DEVICE) +#if defined(CONFIG_CGROUP_DEVICE) || defined(CONFIG_CGROUP_BPF) struct drm_device *ddev = kfd->ddev; - return devcgroup_check_permission(DEVCG_DEV_CHAR, ddev->driver->major, + return devcgroup_check_permission(DEVCG_DEV_CHAR, DRM_MAJOR, ddev->render->index, DEVCG_ACC_WRITE | DEVCG_ACC_READ); #else diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index fe0cd49d4ea7..d27221ddcdeb 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -79,18 +79,22 @@ static struct kfd_procfs_tree procfs; static ssize_t kfd_procfs_show(struct kobject *kobj, struct attribute *attr, char *buffer) { - int val = 0; - if (strcmp(attr->name, "pasid") == 0) { struct kfd_process *p = container_of(attr, struct kfd_process, attr_pasid); - val = p->pasid; + + return snprintf(buffer, PAGE_SIZE, "%d\n", p->pasid); + } else if (strncmp(attr->name, "vram_", 5) == 0) { + struct kfd_process_device *pdd = container_of(attr, struct kfd_process_device, + attr_vram); + if (pdd) + return snprintf(buffer, PAGE_SIZE, "%llu\n", READ_ONCE(pdd->vram_usage)); } else { pr_err("Invalid attribute"); return -EINVAL; } - return snprintf(buffer, PAGE_SIZE, "%d\n", val); + return 0; } static void kfd_procfs_kobj_release(struct kobject *kobj) @@ -206,6 +210,34 @@ int kfd_procfs_add_queue(struct queue *q) return 0; } +int kfd_procfs_add_vram_usage(struct kfd_process *p) +{ + int ret = 0; + struct kfd_process_device *pdd; + + if (!p) + return -EINVAL; + + if (!p->kobj) + return -EFAULT; + + /* Create proc/<pid>/vram_<gpuid> file for each GPU */ + list_for_each_entry(pdd, &p->per_device_data, per_device_list) { + snprintf(pdd->vram_filename, MAX_VRAM_FILENAME_LEN, "vram_%u", + pdd->dev->id); + pdd->attr_vram.name = pdd->vram_filename; + pdd->attr_vram.mode = KFD_SYSFS_FILE_MODE; + sysfs_attr_init(&pdd->attr_vram); + ret = sysfs_create_file(p->kobj, &pdd->attr_vram); + if (ret) + pr_warn("Creating vram usage for gpu id %d failed", + (int)pdd->dev->id); + } + + return ret; +} + + void kfd_procfs_del_queue(struct queue *q) { if (!q) @@ -248,7 +280,7 @@ static void kfd_process_free_gpuvm(struct kgd_mem *mem, struct kfd_dev *dev = pdd->dev; amdgpu_amdkfd_gpuvm_unmap_memory_from_gpu(dev->kgd, mem, pdd->vm); - amdgpu_amdkfd_gpuvm_free_memory_of_gpu(dev->kgd, mem); + amdgpu_amdkfd_gpuvm_free_memory_of_gpu(dev->kgd, mem, NULL); } /* kfd_process_alloc_gpuvm - Allocate GPU VM for the KFD process @@ -312,7 +344,7 @@ sync_memory_failed: return err; err_map_mem: - amdgpu_amdkfd_gpuvm_free_memory_of_gpu(kdev->kgd, mem); + amdgpu_amdkfd_gpuvm_free_memory_of_gpu(kdev->kgd, mem, NULL); err_alloc_mem: *kptr = NULL; return err; @@ -411,6 +443,11 @@ struct kfd_process *kfd_create_process(struct file *filep) process->kobj); if (!process->kobj_queues) pr_warn("Creating KFD proc/queues folder failed"); + + ret = kfd_procfs_add_vram_usage(process); + if (ret) + pr_warn("Creating vram usage file for pid %d failed", + (int)process->lead_thread->pid); } out: if (!IS_ERR(process)) @@ -488,7 +525,7 @@ static void kfd_process_device_free_bos(struct kfd_process_device *pdd) peer_pdd->dev->kgd, mem, peer_pdd->vm); } - amdgpu_amdkfd_gpuvm_free_memory_of_gpu(pdd->dev->kgd, mem); + amdgpu_amdkfd_gpuvm_free_memory_of_gpu(pdd->dev->kgd, mem, NULL); kfd_process_device_remove_obj_handle(pdd, id); } } @@ -551,6 +588,7 @@ static void kfd_process_wq_release(struct work_struct *work) { struct kfd_process *p = container_of(work, struct kfd_process, release_work); + struct kfd_process_device *pdd; /* Remove the procfs files */ if (p->kobj) { @@ -558,6 +596,10 @@ static void kfd_process_wq_release(struct work_struct *work) kobject_del(p->kobj_queues); kobject_put(p->kobj_queues); p->kobj_queues = NULL; + + list_for_each_entry(pdd, &p->per_device_data, per_device_list) + sysfs_remove_file(p->kobj, &pdd->attr_vram); + kobject_del(p->kobj); kobject_put(p->kobj); p->kobj = NULL; @@ -858,10 +900,12 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_dev *dev, pdd->qpd.dqm = dev->dqm; pdd->qpd.pqm = &p->pqm; pdd->qpd.evicted = 0; + pdd->qpd.mapped_gws_queue = false; pdd->process = p; pdd->bound = PDD_UNBOUND; pdd->already_dequeued = false; pdd->runtime_inuse = false; + pdd->vram_usage = 0; list_add(&pdd->per_device_list, &p->per_device_data); /* Init idr used for memory handle translation */ @@ -1078,7 +1122,7 @@ struct kfd_process *kfd_lookup_process_by_mm(const struct mm_struct *mm) return p; } -/* process_evict_queues - Evict all user queues of a process +/* kfd_process_evict_queues - Evict all user queues of a process * * Eviction is reference-counted per process-device. This means multiple * evictions from different sources can be nested safely. @@ -1118,7 +1162,7 @@ fail: return r; } -/* process_restore_queues - Restore all user queues of a process */ +/* kfd_process_restore_queues - Restore all user queues of a process */ int kfd_process_restore_queues(struct kfd_process *p) { struct kfd_process_device *pdd; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c index 084c35f55d59..eb1635ac8988 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c @@ -476,6 +476,15 @@ struct kernel_queue *pqm_get_kernel_queue( return NULL; } +struct queue *pqm_get_user_queue(struct process_queue_manager *pqm, + unsigned int qid) +{ + struct process_queue_node *pqn; + + pqn = get_queue_by_qid(pqm, qid); + return pqn ? pqn->q : NULL; +} + int pqm_get_wave_state(struct process_queue_manager *pqm, unsigned int qid, void __user *ctl_stack, diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index aa0bfa78a667..bb77f7af2b6d 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -478,6 +478,8 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr, dev->node_props.device_id); sysfs_show_32bit_prop(buffer, "location_id", dev->node_props.location_id); + sysfs_show_32bit_prop(buffer, "domain", + dev->node_props.domain); sysfs_show_32bit_prop(buffer, "drm_render_minor", dev->node_props.drm_render_minor); sysfs_show_64bit_prop(buffer, "hive_id", @@ -787,7 +789,6 @@ static int kfd_topology_update_sysfs(void) { int ret; - pr_info("Creating topology SYSFS entries\n"); if (!sys_props.kobj_topology) { sys_props.kobj_topology = kfd_alloc_struct(sys_props.kobj_topology); @@ -1048,7 +1049,6 @@ int kfd_topology_init(void) sys_props.generation_count++; kfd_update_system_properties(); kfd_debug_print_topology(); - pr_info("Finished initializing topology\n"); } else pr_err("Failed to update topology in sysfs ret=%d\n", ret); @@ -1303,7 +1303,12 @@ int kfd_topology_add_device(struct kfd_dev *gpu) dev->node_props.vendor_id = gpu->pdev->vendor; dev->node_props.device_id = gpu->pdev->device; + dev->node_props.capability |= + ((amdgpu_amdkfd_get_asic_rev_id(dev->gpu->kgd) << + HSA_CAP_ASIC_REVISION_SHIFT) & + HSA_CAP_ASIC_REVISION_MASK); dev->node_props.location_id = pci_dev_id(gpu->pdev); + dev->node_props.domain = pci_domain_nr(gpu->pdev->bus); dev->node_props.max_engine_clk_fcompute = amdgpu_amdkfd_get_max_engine_clock_in_mhz(dev->gpu->kgd); dev->node_props.max_engine_clk_ccompute = @@ -1317,7 +1322,7 @@ int kfd_topology_add_device(struct kfd_dev *gpu) gpu->device_info->num_xgmi_sdma_engines; dev->node_props.num_sdma_queues_per_engine = gpu->device_info->num_sdma_queues_per_engine; - dev->node_props.num_gws = (hws_gws_support && + dev->node_props.num_gws = (dev->gpu->gws && dev->gpu->dqm->sched_policy != KFD_SCHED_POLICY_NO_HWS) ? amdgpu_amdkfd_get_num_gws(dev->gpu->kgd) : 0; dev->node_props.num_cp_queues = get_cp_queues_num(dev->gpu->dqm); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h index 46eeecaf1b68..326d9b26b7aa 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h @@ -41,7 +41,6 @@ #define HSA_CAP_WATCH_POINTS_TOTALBITS_SHIFT 8 #define HSA_CAP_DOORBELL_TYPE_TOTALBITS_MASK 0x00003000 #define HSA_CAP_DOORBELL_TYPE_TOTALBITS_SHIFT 12 -#define HSA_CAP_RESERVED 0xffffc000 #define HSA_CAP_DOORBELL_TYPE_PRE_1_0 0x0 #define HSA_CAP_DOORBELL_TYPE_1_0 0x1 @@ -51,6 +50,10 @@ #define HSA_CAP_SRAM_EDCSUPPORTED 0x00080000 #define HSA_CAP_MEM_EDCSUPPORTED 0x00100000 #define HSA_CAP_RASEVENTNOTIFY 0x00200000 +#define HSA_CAP_ASIC_REVISION_MASK 0x03c00000 +#define HSA_CAP_ASIC_REVISION_SHIFT 22 + +#define HSA_CAP_RESERVED 0xfc078000 struct kfd_node_properties { uint64_t hive_id; @@ -77,6 +80,7 @@ struct kfd_node_properties { uint32_t vendor_id; uint32_t device_id; uint32_t location_id; + uint32_t domain; uint32_t max_engine_clk_fcompute; uint32_t max_engine_clk_ccompute; int32_t drm_render_minor; diff --git a/drivers/gpu/drm/amd/display/Kconfig b/drivers/gpu/drm/amd/display/Kconfig index 87858bc57e64..1911a34cc060 100644 --- a/drivers/gpu/drm/amd/display/Kconfig +++ b/drivers/gpu/drm/amd/display/Kconfig @@ -21,16 +21,12 @@ config DRM_AMD_DC_HDCP bool "Enable HDCP support in DC" depends on DRM_AMD_DC help - Choose this option - if you want to support - HDCP authentication + Choose this option if you want to support HDCP authentication. config DEBUG_KERNEL_DC bool "Enable kgdb break in DC" depends on DRM_AMD_DC help - Choose this option - if you want to hit - kdgb_break in assert. + Choose this option if you want to hit kdgb_break in assert. endmenu diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index f7c5cdc10a70..7ced9f87be97 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -30,7 +30,7 @@ #include "dc.h" #include "dc/inc/core_types.h" #include "dal_asic_id.h" -#include "dmub/inc/dmub_srv.h" +#include "dmub/dmub_srv.h" #include "dc/inc/hw/dmcu.h" #include "dc/inc/hw/abm.h" #include "dc/dc_dmub_srv.h" @@ -441,7 +441,7 @@ static void dm_vupdate_high_irq(void *interrupt_params) /** * dm_crtc_high_irq() - Handles CRTC interrupt - * @interrupt_params: ignored + * @interrupt_params: used for determining the CRTC instance * * Handles the CRTC/VSYNC interrupt by notfying DRM's VBLANK * event handler. @@ -455,70 +455,6 @@ static void dm_crtc_high_irq(void *interrupt_params) unsigned long flags; acrtc = get_crtc_by_otg_inst(adev, irq_params->irq_src - IRQ_TYPE_VBLANK); - - if (acrtc) { - acrtc_state = to_dm_crtc_state(acrtc->base.state); - - DRM_DEBUG_VBL("crtc:%d, vupdate-vrr:%d\n", - acrtc->crtc_id, - amdgpu_dm_vrr_active(acrtc_state)); - - /* Core vblank handling at start of front-porch is only possible - * in non-vrr mode, as only there vblank timestamping will give - * valid results while done in front-porch. Otherwise defer it - * to dm_vupdate_high_irq after end of front-porch. - */ - if (!amdgpu_dm_vrr_active(acrtc_state)) - drm_crtc_handle_vblank(&acrtc->base); - - /* Following stuff must happen at start of vblank, for crc - * computation and below-the-range btr support in vrr mode. - */ - amdgpu_dm_crtc_handle_crc_irq(&acrtc->base); - - if (acrtc_state->stream && adev->family >= AMDGPU_FAMILY_AI && - acrtc_state->vrr_params.supported && - acrtc_state->freesync_config.state == VRR_STATE_ACTIVE_VARIABLE) { - spin_lock_irqsave(&adev->ddev->event_lock, flags); - mod_freesync_handle_v_update( - adev->dm.freesync_module, - acrtc_state->stream, - &acrtc_state->vrr_params); - - dc_stream_adjust_vmin_vmax( - adev->dm.dc, - acrtc_state->stream, - &acrtc_state->vrr_params.adjust); - spin_unlock_irqrestore(&adev->ddev->event_lock, flags); - } - } -} - -#if defined(CONFIG_DRM_AMD_DC_DCN) -/** - * dm_dcn_crtc_high_irq() - Handles VStartup interrupt for DCN generation ASICs - * @interrupt params - interrupt parameters - * - * Notify DRM's vblank event handler at VSTARTUP - * - * Unlike DCE hardware, we trigger the handler at VSTARTUP. at which: - * * We are close enough to VUPDATE - the point of no return for hw - * * We are in the fixed portion of variable front porch when vrr is enabled - * * We are before VUPDATE, where double-buffered vrr registers are swapped - * - * It is therefore the correct place to signal vblank, send user flip events, - * and update VRR. - */ -static void dm_dcn_crtc_high_irq(void *interrupt_params) -{ - struct common_irq_params *irq_params = interrupt_params; - struct amdgpu_device *adev = irq_params->adev; - struct amdgpu_crtc *acrtc; - struct dm_crtc_state *acrtc_state; - unsigned long flags; - - acrtc = get_crtc_by_otg_inst(adev, irq_params->irq_src - IRQ_TYPE_VBLANK); - if (!acrtc) return; @@ -528,22 +464,35 @@ static void dm_dcn_crtc_high_irq(void *interrupt_params) amdgpu_dm_vrr_active(acrtc_state), acrtc_state->active_planes); + /** + * Core vblank handling at start of front-porch is only possible + * in non-vrr mode, as only there vblank timestamping will give + * valid results while done in front-porch. Otherwise defer it + * to dm_vupdate_high_irq after end of front-porch. + */ + if (!amdgpu_dm_vrr_active(acrtc_state)) + drm_crtc_handle_vblank(&acrtc->base); + + /** + * Following stuff must happen at start of vblank, for crc + * computation and below-the-range btr support in vrr mode. + */ amdgpu_dm_crtc_handle_crc_irq(&acrtc->base); - drm_crtc_handle_vblank(&acrtc->base); + + /* BTR updates need to happen before VUPDATE on Vega and above. */ + if (adev->family < AMDGPU_FAMILY_AI) + return; spin_lock_irqsave(&adev->ddev->event_lock, flags); - if (acrtc_state->vrr_params.supported && + if (acrtc_state->stream && acrtc_state->vrr_params.supported && acrtc_state->freesync_config.state == VRR_STATE_ACTIVE_VARIABLE) { - mod_freesync_handle_v_update( - adev->dm.freesync_module, - acrtc_state->stream, - &acrtc_state->vrr_params); + mod_freesync_handle_v_update(adev->dm.freesync_module, + acrtc_state->stream, + &acrtc_state->vrr_params); - dc_stream_adjust_vmin_vmax( - adev->dm.dc, - acrtc_state->stream, - &acrtc_state->vrr_params.adjust); + dc_stream_adjust_vmin_vmax(adev->dm.dc, acrtc_state->stream, + &acrtc_state->vrr_params.adjust); } /* @@ -556,7 +505,8 @@ static void dm_dcn_crtc_high_irq(void *interrupt_params) * avoid race conditions between flip programming and completion, * which could cause too early flip completion events. */ - if (acrtc->pflip_status == AMDGPU_FLIP_SUBMITTED && + if (adev->family >= AMDGPU_FAMILY_RV && + acrtc->pflip_status == AMDGPU_FLIP_SUBMITTED && acrtc_state->active_planes == 0) { if (acrtc->event) { drm_crtc_send_vblank_event(&acrtc->base, acrtc->event); @@ -568,7 +518,6 @@ static void dm_dcn_crtc_high_irq(void *interrupt_params) spin_unlock_irqrestore(&adev->ddev->event_lock, flags); } -#endif static int dm_set_clockgating_state(void *handle, enum amd_clockgating_state state) @@ -825,8 +774,9 @@ static int dm_dmub_hw_init(struct amdgpu_device *adev) fw_inst_const_size); } - memcpy(fb_info->fb[DMUB_WINDOW_2_BSS_DATA].cpu_addr, fw_bss_data, - fw_bss_data_size); + if (fw_bss_data_size) + memcpy(fb_info->fb[DMUB_WINDOW_2_BSS_DATA].cpu_addr, + fw_bss_data, fw_bss_data_size); /* Copy firmware bios info into FB memory. */ memcpy(fb_info->fb[DMUB_WINDOW_3_VBIOS].cpu_addr, adev->bios, @@ -968,6 +918,23 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) goto error; } + if (amdgpu_dc_debug_mask & DC_DISABLE_PIPE_SPLIT) { + adev->dm.dc->debug.force_single_disp_pipe_split = false; + adev->dm.dc->debug.pipe_split_policy = MPC_SPLIT_AVOID; + } + + if (adev->asic_type != CHIP_CARRIZO && adev->asic_type != CHIP_STONEY) + adev->dm.dc->debug.disable_stutter = amdgpu_pp_feature_mask & PP_STUTTER_MODE ? false : true; + + if (amdgpu_dc_debug_mask & DC_DISABLE_STUTTER) + adev->dm.dc->debug.disable_stutter = true; + + if (amdgpu_dc_debug_mask & DC_DISABLE_DSC) + adev->dm.dc->debug.disable_dsc = true; + + if (amdgpu_dc_debug_mask & DC_DISABLE_CLOCK_GATING) + adev->dm.dc->debug.disable_clock_gate = true; + r = dm_dmub_hw_init(adev); if (r) { DRM_ERROR("DMUB interface failed to initialize: status=%d\n", r); @@ -1265,6 +1232,10 @@ static int dm_dmub_sw_init(struct amdgpu_device *adev) adev->dm.dmub_fw->data + le32_to_cpu(hdr->header.ucode_array_offset_bytes) + le32_to_cpu(hdr->inst_const_bytes); + region_params.fw_inst_const = + adev->dm.dmub_fw->data + + le32_to_cpu(hdr->header.ucode_array_offset_bytes) + + PSP_HEADER_BYTES; status = dmub_srv_calc_region_info(dmub_srv, ®ion_params, ®ion_info); @@ -1384,8 +1355,13 @@ static int dm_late_init(void *handle) struct dmcu_iram_parameters params; unsigned int linear_lut[16]; int i; - struct dmcu *dmcu = adev->dm.dc->res_pool->dmcu; - bool ret = false; + struct dmcu *dmcu = NULL; + bool ret; + + if (!adev->dm.fw_dmcu) + return detect_mst_link_for_all_connectors(adev->ddev); + + dmcu = adev->dm.dc->res_pool->dmcu; for (i = 0; i < 16; i++) linear_lut[i] = 0xFFFF * i / 15; @@ -1401,13 +1377,10 @@ static int dm_late_init(void *handle) */ params.min_abm_backlight = 0x28F; - /* todo will enable for navi10 */ - if (adev->asic_type <= CHIP_RAVEN) { - ret = dmcu_load_iram(dmcu, params); + ret = dmcu_load_iram(dmcu, params); - if (!ret) - return -EINVAL; - } + if (!ret) + return -EINVAL; return detect_mst_link_for_all_connectors(adev->ddev); } @@ -1562,12 +1535,115 @@ static int dm_hw_fini(void *handle) return 0; } + +static int dm_enable_vblank(struct drm_crtc *crtc); +static void dm_disable_vblank(struct drm_crtc *crtc); + +static void dm_gpureset_toggle_interrupts(struct amdgpu_device *adev, + struct dc_state *state, bool enable) +{ + enum dc_irq_source irq_source; + struct amdgpu_crtc *acrtc; + int rc = -EBUSY; + int i = 0; + + for (i = 0; i < state->stream_count; i++) { + acrtc = get_crtc_by_otg_inst( + adev, state->stream_status[i].primary_otg_inst); + + if (acrtc && state->stream_status[i].plane_count != 0) { + irq_source = IRQ_TYPE_PFLIP + acrtc->otg_inst; + rc = dc_interrupt_set(adev->dm.dc, irq_source, enable) ? 0 : -EBUSY; + DRM_DEBUG("crtc %d - vupdate irq %sabling: r=%d\n", + acrtc->crtc_id, enable ? "en" : "dis", rc); + if (rc) + DRM_WARN("Failed to %s pflip interrupts\n", + enable ? "enable" : "disable"); + + if (enable) { + rc = dm_enable_vblank(&acrtc->base); + if (rc) + DRM_WARN("Failed to enable vblank interrupts\n"); + } else { + dm_disable_vblank(&acrtc->base); + } + + } + } + +} + +enum dc_status amdgpu_dm_commit_zero_streams(struct dc *dc) +{ + struct dc_state *context = NULL; + enum dc_status res = DC_ERROR_UNEXPECTED; + int i; + struct dc_stream_state *del_streams[MAX_PIPES]; + int del_streams_count = 0; + + memset(del_streams, 0, sizeof(del_streams)); + + context = dc_create_state(dc); + if (context == NULL) + goto context_alloc_fail; + + dc_resource_state_copy_construct_current(dc, context); + + /* First remove from context all streams */ + for (i = 0; i < context->stream_count; i++) { + struct dc_stream_state *stream = context->streams[i]; + + del_streams[del_streams_count++] = stream; + } + + /* Remove all planes for removed streams and then remove the streams */ + for (i = 0; i < del_streams_count; i++) { + if (!dc_rem_all_planes_for_stream(dc, del_streams[i], context)) { + res = DC_FAIL_DETACH_SURFACES; + goto fail; + } + + res = dc_remove_stream_from_ctx(dc, context, del_streams[i]); + if (res != DC_OK) + goto fail; + } + + + res = dc_validate_global_state(dc, context, false); + + if (res != DC_OK) { + DRM_ERROR("%s:resource validation failed, dc_status:%d\n", __func__, res); + goto fail; + } + + res = dc_commit_state(dc, context); + +fail: + dc_release_state(context); + +context_alloc_fail: + return res; +} + static int dm_suspend(void *handle) { struct amdgpu_device *adev = handle; struct amdgpu_display_manager *dm = &adev->dm; int ret = 0; + if (adev->in_gpu_reset) { + mutex_lock(&dm->dc_lock); + dm->cached_dc_state = dc_copy_state(dm->dc->current_state); + + dm_gpureset_toggle_interrupts(adev, dm->cached_dc_state, false); + + amdgpu_dm_commit_zero_streams(dm->dc); + + amdgpu_dm_irq_suspend(adev); + + return ret; + } + WARN_ON(adev->dm.cached_state); adev->dm.cached_state = drm_atomic_helper_suspend(adev->ddev); @@ -1578,7 +1654,7 @@ static int dm_suspend(void *handle) dc_set_power_state(dm->dc, DC_ACPI_CM_POWER_STATE_D3); - return ret; + return 0; } static struct amdgpu_dm_connector * @@ -1682,6 +1758,46 @@ static void emulated_link_detect(struct dc_link *link) } +static void dm_gpureset_commit_state(struct dc_state *dc_state, + struct amdgpu_display_manager *dm) +{ + struct { + struct dc_surface_update surface_updates[MAX_SURFACES]; + struct dc_plane_info plane_infos[MAX_SURFACES]; + struct dc_scaling_info scaling_infos[MAX_SURFACES]; + struct dc_flip_addrs flip_addrs[MAX_SURFACES]; + struct dc_stream_update stream_update; + } * bundle; + int k, m; + + bundle = kzalloc(sizeof(*bundle), GFP_KERNEL); + + if (!bundle) { + dm_error("Failed to allocate update bundle\n"); + goto cleanup; + } + + for (k = 0; k < dc_state->stream_count; k++) { + bundle->stream_update.stream = dc_state->streams[k]; + + for (m = 0; m < dc_state->stream_status->plane_count; m++) { + bundle->surface_updates[m].surface = + dc_state->stream_status->plane_states[m]; + bundle->surface_updates[m].surface->force_full_update = + true; + } + dc_commit_updates_for_stream( + dm->dc, bundle->surface_updates, + dc_state->stream_status->plane_count, + dc_state->streams[k], &bundle->stream_update, dc_state); + } + +cleanup: + kfree(bundle); + + return; +} + static int dm_resume(void *handle) { struct amdgpu_device *adev = handle; @@ -1698,8 +1814,44 @@ static int dm_resume(void *handle) struct dm_plane_state *dm_new_plane_state; struct dm_atomic_state *dm_state = to_dm_atomic_state(dm->atomic_obj.state); enum dc_connection_type new_connection_type = dc_connection_none; - int i, r; + struct dc_state *dc_state; + int i, r, j; + + if (adev->in_gpu_reset) { + dc_state = dm->cached_dc_state; + + r = dm_dmub_hw_init(adev); + if (r) + DRM_ERROR("DMUB interface failed to initialize: status=%d\n", r); + dc_set_power_state(dm->dc, DC_ACPI_CM_POWER_STATE_D0); + dc_resume(dm->dc); + + amdgpu_dm_irq_resume_early(adev); + + for (i = 0; i < dc_state->stream_count; i++) { + dc_state->streams[i]->mode_changed = true; + for (j = 0; j < dc_state->stream_status->plane_count; j++) { + dc_state->stream_status->plane_states[j]->update_flags.raw + = 0xffffffff; + } + } + + WARN_ON(!dc_commit_state(dm->dc, dc_state)); + + dm_gpureset_commit_state(dm->cached_dc_state, dm); + + dm_gpureset_toggle_interrupts(adev, dm->cached_dc_state, true); + + dc_release_state(dm->cached_dc_state); + dm->cached_dc_state = NULL; + + amdgpu_dm_irq_resume_late(adev); + + mutex_unlock(&dm->dc_lock); + + return 0; + } /* Recreate dc_state - DC invalidates it when setting power state to S3. */ dc_release_state(dm_state->context); dm_state->context = dc_create_state(dm->dc); @@ -2008,17 +2160,22 @@ void amdgpu_dm_update_connector_after_detect( dc_sink_retain(aconnector->dc_sink); if (sink->dc_edid.length == 0) { aconnector->edid = NULL; - drm_dp_cec_unset_edid(&aconnector->dm_dp_aux.aux); + if (aconnector->dc_link->aux_mode) { + drm_dp_cec_unset_edid( + &aconnector->dm_dp_aux.aux); + } } else { aconnector->edid = - (struct edid *) sink->dc_edid.raw_edid; - + (struct edid *)sink->dc_edid.raw_edid; drm_connector_update_edid_property(connector, - aconnector->edid); - drm_dp_cec_set_edid(&aconnector->dm_dp_aux.aux, - aconnector->edid); + aconnector->edid); + + if (aconnector->dc_link->aux_mode) + drm_dp_cec_set_edid(&aconnector->dm_dp_aux.aux, + aconnector->edid); } + amdgpu_dm_update_freesync_caps(connector, aconnector->edid); update_connector_ext_caps(aconnector); } else { @@ -2440,8 +2597,36 @@ static int dcn10_register_irq_handlers(struct amdgpu_device *adev) c_irq_params->adev = adev; c_irq_params->irq_src = int_params.irq_source; + amdgpu_dm_irq_register_interrupt( + adev, &int_params, dm_crtc_high_irq, c_irq_params); + } + + /* Use VUPDATE_NO_LOCK interrupt on DCN, which seems to correspond to + * the regular VUPDATE interrupt on DCE. We want DC_IRQ_SOURCE_VUPDATEx + * to trigger at end of each vblank, regardless of state of the lock, + * matching DCE behaviour. + */ + for (i = DCN_1_0__SRCID__OTG0_IHC_V_UPDATE_NO_LOCK_INTERRUPT; + i <= DCN_1_0__SRCID__OTG0_IHC_V_UPDATE_NO_LOCK_INTERRUPT + adev->mode_info.num_crtc - 1; + i++) { + r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_DCE, i, &adev->vupdate_irq); + + if (r) { + DRM_ERROR("Failed to add vupdate irq id!\n"); + return r; + } + + int_params.int_context = INTERRUPT_HIGH_IRQ_CONTEXT; + int_params.irq_source = + dc_interrupt_to_irq_source(dc, i, 0); + + c_irq_params = &adev->dm.vupdate_params[int_params.irq_source - DC_IRQ_SOURCE_VUPDATE1]; + + c_irq_params->adev = adev; + c_irq_params->irq_src = int_params.irq_source; + amdgpu_dm_irq_register_interrupt(adev, &int_params, - dm_dcn_crtc_high_irq, c_irq_params); + dm_vupdate_high_irq, c_irq_params); } /* Use GRPH_PFLIP interrupt */ @@ -3031,9 +3216,6 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev) goto fail; } - if (adev->asic_type != CHIP_CARRIZO && adev->asic_type != CHIP_STONEY) - dm->dc->debug.disable_stutter = amdgpu_pp_feature_mask & PP_STUTTER_MODE ? false : true; - /* No userspace support. */ dm->dc->debug.disable_tri_buf = true; @@ -3304,7 +3486,7 @@ static int fill_dc_scaling_info(const struct drm_plane_state *state, } static int get_fb_info(const struct amdgpu_framebuffer *amdgpu_fb, - uint64_t *tiling_flags) + uint64_t *tiling_flags, bool *tmz_surface) { struct amdgpu_bo *rbo = gem_to_amdgpu_bo(amdgpu_fb->base.obj[0]); int r = amdgpu_bo_reserve(rbo, false); @@ -3319,6 +3501,9 @@ static int get_fb_info(const struct amdgpu_framebuffer *amdgpu_fb, if (tiling_flags) amdgpu_bo_get_tiling_flags(rbo, tiling_flags); + if (tmz_surface) + *tmz_surface = amdgpu_bo_encrypted(rbo); + amdgpu_bo_unreserve(rbo); return r; @@ -3340,7 +3525,8 @@ fill_plane_dcc_attributes(struct amdgpu_device *adev, const union dc_tiling_info *tiling_info, const uint64_t info, struct dc_plane_dcc_param *dcc, - struct dc_plane_address *address) + struct dc_plane_address *address, + bool force_disable_dcc) { struct dc *dc = adev->dm.dc; struct dc_dcc_surface_param input; @@ -3352,6 +3538,9 @@ fill_plane_dcc_attributes(struct amdgpu_device *adev, memset(&input, 0, sizeof(input)); memset(&output, 0, sizeof(output)); + if (force_disable_dcc) + return 0; + if (!offset) return 0; @@ -3401,7 +3590,9 @@ fill_plane_buffer_attributes(struct amdgpu_device *adev, union dc_tiling_info *tiling_info, struct plane_size *plane_size, struct dc_plane_dcc_param *dcc, - struct dc_plane_address *address) + struct dc_plane_address *address, + bool tmz_surface, + bool force_disable_dcc) { const struct drm_framebuffer *fb = &afb->base; int ret; @@ -3411,6 +3602,8 @@ fill_plane_buffer_attributes(struct amdgpu_device *adev, memset(dcc, 0, sizeof(*dcc)); memset(address, 0, sizeof(*address)); + address->tmz_surface = tmz_surface; + if (format < SURFACE_PIXEL_FORMAT_VIDEO_BEGIN) { plane_size->surface_size.x = 0; plane_size->surface_size.y = 0; @@ -3507,7 +3700,8 @@ fill_plane_buffer_attributes(struct amdgpu_device *adev, ret = fill_plane_dcc_attributes(adev, afb, format, rotation, plane_size, tiling_info, - tiling_flags, dcc, address); + tiling_flags, dcc, address, + force_disable_dcc); if (ret) return ret; } @@ -3599,7 +3793,9 @@ fill_dc_plane_info_and_addr(struct amdgpu_device *adev, const struct drm_plane_state *plane_state, const uint64_t tiling_flags, struct dc_plane_info *plane_info, - struct dc_plane_address *address) + struct dc_plane_address *address, + bool tmz_surface, + bool force_disable_dcc) { const struct drm_framebuffer *fb = plane_state->fb; const struct amdgpu_framebuffer *afb = @@ -3642,6 +3838,14 @@ fill_dc_plane_info_and_addr(struct amdgpu_device *adev, case DRM_FORMAT_P010: plane_info->format = SURFACE_PIXEL_FORMAT_VIDEO_420_10bpc_YCrCb; break; + case DRM_FORMAT_XRGB16161616F: + case DRM_FORMAT_ARGB16161616F: + plane_info->format = SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616F; + break; + case DRM_FORMAT_XBGR16161616F: + case DRM_FORMAT_ABGR16161616F: + plane_info->format = SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616F; + break; default: DRM_ERROR( "Unsupported screen format %s\n", @@ -3681,7 +3885,8 @@ fill_dc_plane_info_and_addr(struct amdgpu_device *adev, plane_info->rotation, tiling_flags, &plane_info->tiling_info, &plane_info->plane_size, - &plane_info->dcc, address); + &plane_info->dcc, address, tmz_surface, + force_disable_dcc); if (ret) return ret; @@ -3704,6 +3909,8 @@ static int fill_dc_plane_attributes(struct amdgpu_device *adev, struct dc_plane_info plane_info; uint64_t tiling_flags; int ret; + bool tmz_surface = false; + bool force_disable_dcc = false; ret = fill_dc_scaling_info(plane_state, &scaling_info); if (ret) @@ -3714,13 +3921,16 @@ static int fill_dc_plane_attributes(struct amdgpu_device *adev, dc_plane_state->clip_rect = scaling_info.clip_rect; dc_plane_state->scaling_quality = scaling_info.scaling_quality; - ret = get_fb_info(amdgpu_fb, &tiling_flags); + ret = get_fb_info(amdgpu_fb, &tiling_flags, &tmz_surface); if (ret) return ret; + force_disable_dcc = adev->asic_type == CHIP_RAVEN && adev->in_suspend; ret = fill_dc_plane_info_and_addr(adev, plane_state, tiling_flags, &plane_info, - &dc_plane_state->address); + &dc_plane_state->address, + tmz_surface, + force_disable_dcc); if (ret) return ret; @@ -3807,8 +4017,7 @@ static void update_stream_scaling_settings(const struct drm_display_mode *mode, static enum dc_color_depth convert_color_depth_from_display_info(const struct drm_connector *connector, - const struct drm_connector_state *state, - bool is_y420) + bool is_y420, int requested_bpc) { uint8_t bpc; @@ -3828,10 +4037,7 @@ convert_color_depth_from_display_info(const struct drm_connector *connector, bpc = bpc ? bpc : 8; } - if (!state) - state = connector->state; - - if (state) { + if (requested_bpc > 0) { /* * Cap display bpc based on the user requested value. * @@ -3840,7 +4046,7 @@ convert_color_depth_from_display_info(const struct drm_connector *connector, * or if this was called outside of atomic check, so it * can't be used directly. */ - bpc = min(bpc, state->max_requested_bpc); + bpc = min_t(u8, bpc, requested_bpc); /* Round down to the nearest even number. */ bpc = bpc - (bpc & 1); @@ -3962,7 +4168,8 @@ static void fill_stream_properties_from_drm_display_mode( const struct drm_display_mode *mode_in, const struct drm_connector *connector, const struct drm_connector_state *connector_state, - const struct dc_stream_state *old_stream) + const struct dc_stream_state *old_stream, + int requested_bpc) { struct dc_crtc_timing *timing_out = &stream->timing; const struct drm_display_info *info = &connector->display_info; @@ -3992,8 +4199,9 @@ static void fill_stream_properties_from_drm_display_mode( timing_out->timing_3d_format = TIMING_3D_FORMAT_NONE; timing_out->display_color_depth = convert_color_depth_from_display_info( - connector, connector_state, - (timing_out->pixel_encoding == PIXEL_ENCODING_YCBCR420)); + connector, + (timing_out->pixel_encoding == PIXEL_ENCODING_YCBCR420), + requested_bpc); timing_out->scan_type = SCANNING_TYPE_NODATA; timing_out->hdmi_vic = 0; @@ -4199,7 +4407,8 @@ static struct dc_stream_state * create_stream_for_sink(struct amdgpu_dm_connector *aconnector, const struct drm_display_mode *drm_mode, const struct dm_connector_state *dm_state, - const struct dc_stream_state *old_stream) + const struct dc_stream_state *old_stream, + int requested_bpc) { struct drm_display_mode *preferred_mode = NULL; struct drm_connector *drm_connector; @@ -4284,10 +4493,10 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, */ if (!scale || mode_refresh != preferred_refresh) fill_stream_properties_from_drm_display_mode(stream, - &mode, &aconnector->base, con_state, NULL); + &mode, &aconnector->base, con_state, NULL, requested_bpc); else fill_stream_properties_from_drm_display_mode(stream, - &mode, &aconnector->base, con_state, old_stream); + &mode, &aconnector->base, con_state, old_stream, requested_bpc); stream->timing.flags.DSC = 0; @@ -4324,14 +4533,10 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, if (stream->signal == SIGNAL_TYPE_HDMI_TYPE_A) mod_build_hf_vsif_infopacket(stream, &stream->vsp_infopacket, false, false); - if (stream->link->psr_feature_enabled) { + if (stream->link->psr_settings.psr_feature_enabled) { struct dc *core_dc = stream->link->ctx->dc; if (dc_is_dmcu_initialized(core_dc)) { - struct dmcu *dmcu = core_dc->res_pool->dmcu; - - stream->psr_version = dmcu->dmcu_version.psr_version; - // // should decide stream support vsc sdp colorimetry capability // before building vsc info packet @@ -4437,10 +4642,6 @@ static inline int dm_set_vupdate_irq(struct drm_crtc *crtc, bool enable) struct amdgpu_device *adev = crtc->dev->dev_private; int rc; - /* Do not set vupdate for DCN hardware */ - if (adev->family > AMDGPU_FAMILY_AI) - return 0; - irq_source = IRQ_TYPE_VUPDATE + acrtc->otg_inst; rc = dc_interrupt_set(adev->dm.dc, irq_source, enable) ? 0 : -EBUSY; @@ -4664,6 +4865,7 @@ static void amdgpu_dm_connector_destroy(struct drm_connector *connector) i2c_del_adapter(&aconnector->i2c->base); kfree(aconnector->i2c); } + kfree(aconnector->dm_dp_aux.aux.name); kfree(connector); } @@ -4723,10 +4925,19 @@ amdgpu_dm_connector_atomic_duplicate_state(struct drm_connector *connector) static int amdgpu_dm_connector_late_register(struct drm_connector *connector) { -#if defined(CONFIG_DEBUG_FS) struct amdgpu_dm_connector *amdgpu_dm_connector = to_amdgpu_dm_connector(connector); + int r; + + if ((connector->connector_type == DRM_MODE_CONNECTOR_DisplayPort) || + (connector->connector_type == DRM_MODE_CONNECTOR_eDP)) { + amdgpu_dm_connector->dm_dp_aux.aux.dev = connector->kdev; + r = drm_dp_aux_register(&amdgpu_dm_connector->dm_dp_aux.aux); + if (r) + return r; + } +#if defined(CONFIG_DEBUG_FS) connector_debugfs_init(amdgpu_dm_connector); #endif @@ -4804,16 +5015,54 @@ static void handle_edid_mgmt(struct amdgpu_dm_connector *aconnector) create_eml_sink(aconnector); } +static struct dc_stream_state * +create_validate_stream_for_sink(struct amdgpu_dm_connector *aconnector, + const struct drm_display_mode *drm_mode, + const struct dm_connector_state *dm_state, + const struct dc_stream_state *old_stream) +{ + struct drm_connector *connector = &aconnector->base; + struct amdgpu_device *adev = connector->dev->dev_private; + struct dc_stream_state *stream; + int requested_bpc = connector->state ? connector->state->max_requested_bpc : 8; + enum dc_status dc_result = DC_OK; + + do { + stream = create_stream_for_sink(aconnector, drm_mode, + dm_state, old_stream, + requested_bpc); + if (stream == NULL) { + DRM_ERROR("Failed to create stream for sink!\n"); + break; + } + + dc_result = dc_validate_stream(adev->dm.dc, stream); + + if (dc_result != DC_OK) { + DRM_DEBUG_KMS("Mode %dx%d (clk %d) failed DC validation with error %d\n", + drm_mode->hdisplay, + drm_mode->vdisplay, + drm_mode->clock, + dc_result); + + dc_stream_release(stream); + stream = NULL; + requested_bpc -= 2; /* lower bpc to retry validation */ + } + + } while (stream == NULL && requested_bpc >= 6); + + return stream; +} + enum drm_mode_status amdgpu_dm_connector_mode_valid(struct drm_connector *connector, struct drm_display_mode *mode) { int result = MODE_ERROR; struct dc_sink *dc_sink; - struct amdgpu_device *adev = connector->dev->dev_private; /* TODO: Unhardcode stream count */ struct dc_stream_state *stream; struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); - enum dc_status dc_result = DC_OK; if ((mode->flags & DRM_MODE_FLAG_INTERLACE) || (mode->flags & DRM_MODE_FLAG_DBLSCAN)) @@ -4834,24 +5083,11 @@ enum drm_mode_status amdgpu_dm_connector_mode_valid(struct drm_connector *connec goto fail; } - stream = create_stream_for_sink(aconnector, mode, NULL, NULL); - if (stream == NULL) { - DRM_ERROR("Failed to create stream for sink!\n"); - goto fail; - } - - dc_result = dc_validate_stream(adev->dm.dc, stream); - - if (dc_result == DC_OK) + stream = create_validate_stream_for_sink(aconnector, mode, NULL, NULL); + if (stream) { + dc_stream_release(stream); result = MODE_OK; - else - DRM_DEBUG_KMS("Mode %dx%d (clk %d) failed DC validation with error %d\n", - mode->hdisplay, - mode->vdisplay, - mode->clock, - dc_result); - - dc_stream_release(stream); + } fail: /* TODO: error handling*/ @@ -5174,10 +5410,12 @@ static int dm_encoder_helper_atomic_check(struct drm_encoder *encoder, return 0; if (!state->duplicated) { + int max_bpc = conn_state->max_requested_bpc; is_y420 = drm_mode_is_420_also(&connector->display_info, adjusted_mode) && aconnector->force_yuv420_output; - color_depth = convert_color_depth_from_display_info(connector, conn_state, - is_y420); + color_depth = convert_color_depth_from_display_info(connector, + is_y420, + max_bpc); bpp = convert_dc_color_depth_into_bpc(color_depth) * 3; clock = adjusted_mode->clock; dm_new_connector_state->pbn = drm_dp_calc_pbn_mode(clock, bpp, false); @@ -5332,6 +5570,8 @@ static int dm_plane_helper_prepare_fb(struct drm_plane *plane, uint64_t tiling_flags; uint32_t domain; int r; + bool tmz_surface = false; + bool force_disable_dcc = false; dm_plane_state_old = to_dm_plane_state(plane->state); dm_plane_state_new = to_dm_plane_state(new_state); @@ -5380,6 +5620,8 @@ static int dm_plane_helper_prepare_fb(struct drm_plane *plane, amdgpu_bo_get_tiling_flags(rbo, &tiling_flags); + tmz_surface = amdgpu_bo_encrypted(rbo); + ttm_eu_backoff_reservation(&ticket, &list); afb->address = amdgpu_bo_gpu_offset(rbo); @@ -5390,11 +5632,13 @@ static int dm_plane_helper_prepare_fb(struct drm_plane *plane, dm_plane_state_old->dc_state != dm_plane_state_new->dc_state) { struct dc_plane_state *plane_state = dm_plane_state_new->dc_state; + force_disable_dcc = adev->asic_type == CHIP_RAVEN && adev->in_suspend; fill_plane_buffer_attributes( adev, afb, plane_state->format, plane_state->rotation, tiling_flags, &plane_state->tiling_info, &plane_state->plane_size, &plane_state->dcc, - &plane_state->address); + &plane_state->address, tmz_surface, + force_disable_dcc); } return 0; @@ -5540,6 +5784,12 @@ static int get_plane_formats(const struct drm_plane *plane, formats[num_formats++] = DRM_FORMAT_NV12; if (plane_cap && plane_cap->pixel_format_support.p010) formats[num_formats++] = DRM_FORMAT_P010; + if (plane_cap && plane_cap->pixel_format_support.fp16) { + formats[num_formats++] = DRM_FORMAT_XRGB16161616F; + formats[num_formats++] = DRM_FORMAT_ARGB16161616F; + formats[num_formats++] = DRM_FORMAT_XBGR16161616F; + formats[num_formats++] = DRM_FORMAT_ABGR16161616F; + } break; case DRM_PLANE_TYPE_OVERLAY: @@ -6092,7 +6342,7 @@ static int amdgpu_dm_connector_init(struct amdgpu_display_manager *dm, if (connector_type == DRM_MODE_CONNECTOR_DisplayPort || connector_type == DRM_MODE_CONNECTOR_eDP) - amdgpu_dm_initialize_dp_connector(dm, aconnector); + amdgpu_dm_initialize_dp_connector(dm, aconnector, link->link_index); out_free: if (res) { @@ -6567,6 +6817,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, unsigned long flags; struct amdgpu_bo *abo; uint64_t tiling_flags; + bool tmz_surface = false; uint32_t target_vblank, last_flip_vblank; bool vrr_active = amdgpu_dm_vrr_active(acrtc_state); bool pflip_present = false; @@ -6619,6 +6870,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, if (new_pcrtc_state->color_mgmt_changed) { bundle->surface_updates[planes_count].gamma = dc_plane->gamma_correction; bundle->surface_updates[planes_count].in_transfer_func = dc_plane->in_transfer_func; + bundle->surface_updates[planes_count].gamut_remap_matrix = &dc_plane->gamut_remap_matrix; } fill_dc_scaling_info(new_plane_state, @@ -6661,12 +6913,20 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, amdgpu_bo_get_tiling_flags(abo, &tiling_flags); + tmz_surface = amdgpu_bo_encrypted(abo); + amdgpu_bo_unreserve(abo); fill_dc_plane_info_and_addr( dm->adev, new_plane_state, tiling_flags, &bundle->plane_infos[planes_count], - &bundle->flip_addrs[planes_count].address); + &bundle->flip_addrs[planes_count].address, + tmz_surface, + false); + + DRM_DEBUG_DRIVER("plane: id=%d dcc_en=%d\n", + new_plane_state->plane->index, + bundle->plane_infos[planes_count].dcc.enable); bundle->surface_updates[planes_count].plane_info = &bundle->plane_infos[planes_count]; @@ -6807,7 +7067,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, } mutex_lock(&dm->dc_lock); if ((acrtc_state->update_type > UPDATE_TYPE_FAST) && - acrtc_state->stream->link->psr_allow_active) + acrtc_state->stream->link->psr_settings.psr_allow_active) amdgpu_dm_psr_disable(acrtc_state->stream); dc_commit_updates_for_stream(dm->dc, @@ -6818,12 +7078,12 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, dc_state); if ((acrtc_state->update_type > UPDATE_TYPE_FAST) && - acrtc_state->stream->psr_version && - !acrtc_state->stream->link->psr_feature_enabled) + acrtc_state->stream->link->psr_settings.psr_version != DC_PSR_VERSION_UNSUPPORTED && + !acrtc_state->stream->link->psr_settings.psr_feature_enabled) amdgpu_dm_link_setup_psr(acrtc_state->stream); else if ((acrtc_state->update_type == UPDATE_TYPE_FAST) && - acrtc_state->stream->link->psr_feature_enabled && - !acrtc_state->stream->link->psr_allow_active) { + acrtc_state->stream->link->psr_settings.psr_feature_enabled && + !acrtc_state->stream->link->psr_settings.psr_allow_active) { amdgpu_dm_psr_enable(acrtc_state->stream); } @@ -7137,7 +7397,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) DRM_DEBUG_DRIVER("Atomic commit: RESET. crtc id %d:[%p]\n", acrtc->crtc_id, acrtc); /* i.e. reset mode */ if (dm_old_crtc_state->stream) { - if (dm_old_crtc_state->stream->link->psr_allow_active) + if (dm_old_crtc_state->stream->link->psr_settings.psr_allow_active) amdgpu_dm_psr_disable(dm_old_crtc_state->stream); remove_stream(adev, acrtc, dm_old_crtc_state->stream); @@ -7585,10 +7845,10 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm, if (!drm_atomic_crtc_needs_modeset(new_crtc_state)) goto skip_modeset; - new_stream = create_stream_for_sink(aconnector, - &new_crtc_state->mode, - dm_new_conn_state, - dm_old_crtc_state->stream); + new_stream = create_validate_stream_for_sink(aconnector, + &new_crtc_state->mode, + dm_new_conn_state, + dm_old_crtc_state->stream); /* * we can have no stream on ACTION_SET if a display @@ -7848,6 +8108,7 @@ static int dm_update_plane_state(struct dc *dc, struct drm_crtc_state *old_crtc_state, *new_crtc_state; struct dm_crtc_state *dm_new_crtc_state, *dm_old_crtc_state; struct dm_plane_state *dm_new_plane_state, *dm_old_plane_state; + struct amdgpu_crtc *new_acrtc; bool needs_reset; int ret = 0; @@ -7857,9 +8118,23 @@ static int dm_update_plane_state(struct dc *dc, dm_new_plane_state = to_dm_plane_state(new_plane_state); dm_old_plane_state = to_dm_plane_state(old_plane_state); - /*TODO Implement atomic check for cursor plane */ - if (plane->type == DRM_PLANE_TYPE_CURSOR) + /*TODO Implement better atomic check for cursor plane */ + if (plane->type == DRM_PLANE_TYPE_CURSOR) { + if (!enable || !new_plane_crtc || + drm_atomic_plane_disabling(plane->state, new_plane_state)) + return 0; + + new_acrtc = to_amdgpu_crtc(new_plane_crtc); + + if ((new_plane_state->crtc_w > new_acrtc->max_cursor_width) || + (new_plane_state->crtc_h > new_acrtc->max_cursor_height)) { + DRM_DEBUG_ATOMIC("Bad cursor size %d x %d\n", + new_plane_state->crtc_w, new_plane_state->crtc_h); + return -EINVAL; + } + return 0; + } needs_reset = should_reset_plane(state, plane, old_plane_state, new_plane_state); @@ -8034,6 +8309,7 @@ dm_determine_update_type_for_commit(struct amdgpu_display_manager *dm, struct dc_flip_addrs *flip_addr = &bundle->flip_addrs[num_plane]; struct dc_scaling_info *scaling_info = &bundle->scaling_infos[num_plane]; uint64_t tiling_flags; + bool tmz_surface = false; new_plane_crtc = new_plane_state->crtc; new_dm_plane_state = to_dm_plane_state(new_plane_state); @@ -8063,6 +8339,8 @@ dm_determine_update_type_for_commit(struct amdgpu_display_manager *dm, new_dm_plane_state->dc_state->gamma_correction; bundle->surface_updates[num_plane].in_transfer_func = new_dm_plane_state->dc_state->in_transfer_func; + bundle->surface_updates[num_plane].gamut_remap_matrix = + &new_dm_plane_state->dc_state->gamut_remap_matrix; bundle->stream_update.gamut_remap = &new_dm_crtc_state->stream->gamut_remap_matrix; bundle->stream_update.output_csc_transform = @@ -8079,14 +8357,15 @@ dm_determine_update_type_for_commit(struct amdgpu_display_manager *dm, bundle->surface_updates[num_plane].scaling_info = scaling_info; if (amdgpu_fb) { - ret = get_fb_info(amdgpu_fb, &tiling_flags); + ret = get_fb_info(amdgpu_fb, &tiling_flags, &tmz_surface); if (ret) goto cleanup; ret = fill_dc_plane_info_and_addr( dm->adev, new_plane_state, tiling_flags, plane_info, - &flip_addr->address); + &flip_addr->address, tmz_surface, + false); if (ret) goto cleanup; @@ -8586,8 +8865,17 @@ static void amdgpu_dm_set_psr_caps(struct dc_link *link) return; if (dm_helpers_dp_read_dpcd(NULL, link, DP_PSR_SUPPORT, dpcd_data, sizeof(dpcd_data))) { - link->psr_feature_enabled = dpcd_data[0] ? true:false; - DRM_INFO("PSR support:%d\n", link->psr_feature_enabled); + link->dpcd_caps.psr_caps.psr_version = dpcd_data[0]; + + if (dpcd_data[0] == 0) { + link->psr_settings.psr_version = DC_PSR_VERSION_UNSUPPORTED; + link->psr_settings.psr_feature_enabled = false; + } else { + link->psr_settings.psr_version = DC_PSR_VERSION_1; + link->psr_settings.psr_feature_enabled = true; + } + + DRM_INFO("PSR support:%d\n", link->psr_settings.psr_feature_enabled); } } @@ -8602,16 +8890,14 @@ static bool amdgpu_dm_link_setup_psr(struct dc_stream_state *stream) struct dc_link *link = NULL; struct psr_config psr_config = {0}; struct psr_context psr_context = {0}; - struct dc *dc = NULL; bool ret = false; if (stream == NULL) return false; link = stream->link; - dc = link->ctx->dc; - psr_config.psr_version = dc->res_pool->dmcu->dmcu_version.psr_version; + psr_config.psr_version = link->dpcd_caps.psr_caps.psr_version; if (psr_config.psr_version > 0) { psr_config.psr_exit_link_training_required = 0x1; @@ -8623,7 +8909,7 @@ static bool amdgpu_dm_link_setup_psr(struct dc_stream_state *stream) ret = dc_link_setup_psr(link, stream, &psr_config, &psr_context); } - DRM_DEBUG_DRIVER("PSR link: %d\n", link->psr_feature_enabled); + DRM_DEBUG_DRIVER("PSR link: %d\n", link->psr_settings.psr_feature_enabled); return ret; } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h index 5cab3e65d992..d61186ff411d 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -315,6 +315,7 @@ struct amdgpu_display_manager { #endif struct drm_atomic_state *cached_state; + struct dc_state *cached_dc_state; struct dm_comressor_info compressor; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c index 2233d293a707..4dfb6b55bb2e 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c @@ -239,7 +239,8 @@ static int __set_output_tf(struct dc_transfer_func *func, * instead to simulate this. */ gamma->type = GAMMA_CUSTOM; - res = mod_color_calculate_degamma_params(func, gamma, true); + res = mod_color_calculate_degamma_params(NULL, func, + gamma, true); } else { /* * Assume sRGB. The actual mapping will depend on whether the @@ -271,7 +272,7 @@ static int __set_input_tf(struct dc_transfer_func *func, __drm_lut_to_dc_gamma(lut, gamma, false); - res = mod_color_calculate_degamma_params(func, gamma, true); + res = mod_color_calculate_degamma_params(NULL, func, gamma, true); dc_gamma_release(&gamma); return res ? 0 : -ENOMEM; @@ -419,9 +420,21 @@ int amdgpu_dm_update_plane_color_mgmt(struct dm_crtc_state *crtc, struct dc_plane_state *dc_plane_state) { const struct drm_color_lut *degamma_lut; + enum dc_transfer_func_predefined tf = TRANSFER_FUNCTION_SRGB; uint32_t degamma_size; int r; + /* Get the correct base transfer function for implicit degamma. */ + switch (dc_plane_state->format) { + case SURFACE_PIXEL_FORMAT_VIDEO_420_YCbCr: + case SURFACE_PIXEL_FORMAT_VIDEO_420_YCrCb: + /* DC doesn't have a transfer function for BT601 specifically. */ + tf = TRANSFER_FUNCTION_BT709; + break; + default: + break; + } + if (crtc->cm_has_degamma) { degamma_lut = __extract_blob_lut(crtc->base.degamma_lut, °amma_size); @@ -455,8 +468,7 @@ int amdgpu_dm_update_plane_color_mgmt(struct dm_crtc_state *crtc, * map these to the atomic one instead. */ if (crtc->cm_is_degamma_srgb) - dc_plane_state->in_transfer_func->tf = - TRANSFER_FUNCTION_SRGB; + dc_plane_state->in_transfer_func->tf = tf; else dc_plane_state->in_transfer_func->tf = TRANSFER_FUNCTION_LINEAR; @@ -471,7 +483,12 @@ int amdgpu_dm_update_plane_color_mgmt(struct dm_crtc_state *crtc, * in linear space. Assume that the input is sRGB. */ dc_plane_state->in_transfer_func->type = TF_TYPE_PREDEFINED; - dc_plane_state->in_transfer_func->tf = TRANSFER_FUNCTION_SRGB; + dc_plane_state->in_transfer_func->tf = tf; + + if (tf != TRANSFER_FUNCTION_SRGB && + !mod_color_calculate_degamma_params(NULL, + dc_plane_state->in_transfer_func, NULL, false)) + return -ENOMEM; } else { /* ...Otherwise we can just bypass the DGM block. */ dc_plane_state->in_transfer_func->type = TF_TYPE_BYPASS; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c index 0461fecd68db..076af267b488 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c @@ -32,7 +32,7 @@ #include "amdgpu_dm.h" #include "amdgpu_dm_debugfs.h" #include "dm_helpers.h" -#include "dmub/inc/dmub_srv.h" +#include "dmub/dmub_srv.h" struct dmub_debugfs_trace_header { uint32_t entry_count; @@ -838,6 +838,44 @@ static int vrr_range_show(struct seq_file *m, void *data) return 0; } +#ifdef CONFIG_DRM_AMD_DC_HDCP +/* + * Returns the HDCP capability of the Display (1.4 for now). + * + * NOTE* Not all HDMI displays report their HDCP caps even when they are capable. + * Since its rare for a display to not be HDCP 1.4 capable, we set HDMI as always capable. + * + * Example usage: cat /sys/kernel/debug/dri/0/DP-1/hdcp_sink_capability + * or cat /sys/kernel/debug/dri/0/HDMI-A-1/hdcp_sink_capability + */ +static int hdcp_sink_capability_show(struct seq_file *m, void *data) +{ + struct drm_connector *connector = m->private; + struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); + bool hdcp_cap, hdcp2_cap; + + if (connector->status != connector_status_connected) + return -ENODEV; + + seq_printf(m, "%s:%d HDCP version: ", connector->name, connector->base.id); + + hdcp_cap = dc_link_is_hdcp14(aconnector->dc_link); + hdcp2_cap = dc_link_is_hdcp22(aconnector->dc_link); + + + if (hdcp_cap) + seq_printf(m, "%s ", "HDCP1.4"); + if (hdcp2_cap) + seq_printf(m, "%s ", "HDCP2.2"); + + if (!hdcp_cap && !hdcp2_cap) + seq_printf(m, "%s ", "None"); + + seq_puts(m, "\n"); + + return 0; +} +#endif /* function description * * generic SDP message access for testing @@ -964,6 +1002,9 @@ DEFINE_SHOW_ATTRIBUTE(dmub_fw_state); DEFINE_SHOW_ATTRIBUTE(dmub_tracebuffer); DEFINE_SHOW_ATTRIBUTE(output_bpc); DEFINE_SHOW_ATTRIBUTE(vrr_range); +#ifdef CONFIG_DRM_AMD_DC_HDCP +DEFINE_SHOW_ATTRIBUTE(hdcp_sink_capability); +#endif static const struct file_operations dp_link_settings_debugfs_fops = { .owner = THIS_MODULE, @@ -1019,12 +1060,23 @@ static const struct { {"test_pattern", &dp_phy_test_pattern_fops}, {"output_bpc", &output_bpc_fops}, {"vrr_range", &vrr_range_fops}, +#ifdef CONFIG_DRM_AMD_DC_HDCP + {"hdcp_sink_capability", &hdcp_sink_capability_fops}, +#endif {"sdp_message", &sdp_message_fops}, {"aux_dpcd_address", &dp_dpcd_address_debugfs_fops}, {"aux_dpcd_size", &dp_dpcd_size_debugfs_fops}, {"aux_dpcd_data", &dp_dpcd_data_debugfs_fops} }; +#ifdef CONFIG_DRM_AMD_DC_HDCP +static const struct { + char *name; + const struct file_operations *fops; +} hdmi_debugfs_entries[] = { + {"hdcp_sink_capability", &hdcp_sink_capability_fops} +}; +#endif /* * Force YUV420 output if available from the given mode */ @@ -1093,6 +1145,15 @@ void connector_debugfs_init(struct amdgpu_dm_connector *connector) connector->debugfs_dpcd_address = 0; connector->debugfs_dpcd_size = 0; +#ifdef CONFIG_DRM_AMD_DC_HDCP + if (connector->base.connector_type == DRM_MODE_CONNECTOR_HDMIA) { + for (i = 0; i < ARRAY_SIZE(hdmi_debugfs_entries); i++) { + debugfs_create_file(hdmi_debugfs_entries[i].name, + 0644, dir, connector, + hdmi_debugfs_entries[i].fops); + } + } +#endif } /* @@ -1167,8 +1228,9 @@ static int current_backlight_read(struct seq_file *m, void *data) struct drm_info_node *node = (struct drm_info_node *)m->private; struct drm_device *dev = node->minor->dev; struct amdgpu_device *adev = dev->dev_private; - struct dc *dc = adev->dm.dc; - unsigned int backlight = dc_get_current_backlight_pwm(dc); + struct amdgpu_display_manager *dm = &adev->dm; + + unsigned int backlight = dc_link_get_backlight_level(dm->backlight_link); seq_printf(m, "0x%x\n", backlight); return 0; @@ -1184,8 +1246,9 @@ static int target_backlight_read(struct seq_file *m, void *data) struct drm_info_node *node = (struct drm_info_node *)m->private; struct drm_device *dev = node->minor->dev; struct amdgpu_device *adev = dev->dev_private; - struct dc *dc = adev->dm.dc; - unsigned int backlight = dc_get_target_backlight_pwm(dc); + struct amdgpu_display_manager *dm = &adev->dm; + + unsigned int backlight = dc_link_get_target_backlight_pwm(dm->backlight_link); seq_printf(m, "0x%x\n", backlight); return 0; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c index 78e1c11d4ae5..dcf84a61de37 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c @@ -398,15 +398,15 @@ static void update_config(void *handle, struct cp_psp_stream_config *config) struct mod_hdcp_display *display = &hdcp_work[link_index].display; struct mod_hdcp_link *link = &hdcp_work[link_index].link; - memset(display, 0, sizeof(*display)); - memset(link, 0, sizeof(*link)); - - display->index = aconnector->base.index; - if (config->dpms_off) { hdcp_remove_display(hdcp_work, link_index, aconnector); return; } + + memset(display, 0, sizeof(*display)); + memset(link, 0, sizeof(*link)); + + display->index = aconnector->base.index; display->state = MOD_HDCP_DISPLAY_ACTIVE; if (aconnector->dc_sink != NULL) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c index c20fb08c450b..b086d5c906e0 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c @@ -445,7 +445,7 @@ bool dm_helpers_dp_read_dpcd( struct amdgpu_dm_connector *aconnector = link->priv; if (!aconnector) { - DRM_ERROR("Failed to find connector for link!"); + DC_LOG_DC("Failed to find connector for link!\n"); return false; } @@ -554,6 +554,7 @@ enum dc_edid_status dm_helpers_read_local_edid( struct dc_sink *sink) { struct amdgpu_dm_connector *aconnector = link->priv; + struct drm_connector *connector = &aconnector->base; struct i2c_adapter *ddc; int retry = 3; enum dc_edid_status edid_status; @@ -571,6 +572,15 @@ enum dc_edid_status dm_helpers_read_local_edid( edid = drm_get_edid(&aconnector->base, ddc); + /* DP Compliance Test 4.2.2.6 */ + if (link->aux_mode && connector->edid_corrupt) + drm_dp_send_real_edid_checksum(&aconnector->dm_dp_aux.aux, connector->real_edid_checksum); + + if (!edid && connector->edid_corrupt) { + connector->edid_corrupt = false; + return EDID_BAD_CHECKSUM; + } + if (!edid) return EDID_NO_RESPONSE; @@ -605,34 +615,10 @@ enum dc_edid_status dm_helpers_read_local_edid( DRM_ERROR("EDID err: %d, on connector: %s", edid_status, aconnector->base.name); - if (link->aux_mode) { - union test_request test_request = { {0} }; - union test_response test_response = { {0} }; - - dm_helpers_dp_read_dpcd(ctx, - link, - DP_TEST_REQUEST, - &test_request.raw, - sizeof(union test_request)); - - if (!test_request.bits.EDID_READ) - return edid_status; - test_response.bits.EDID_CHECKSUM_WRITE = 1; - - dm_helpers_dp_write_dpcd(ctx, - link, - DP_TEST_EDID_CHECKSUM, - &sink->dc_edid.raw_edid[sink->dc_edid.length-1], - 1); - - dm_helpers_dp_write_dpcd(ctx, - link, - DP_TEST_RESPONSE, - &test_response.raw, - sizeof(test_response)); - - } + /* DP Compliance Test 4.2.2.3 */ + if (link->aux_mode) + drm_dp_send_real_edid_checksum(&aconnector->dm_dp_aux.aux, sink->dc_edid.raw_edid[sink->dc_edid.length-1]); return edid_status; } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index 3db1ec35d2b4..ae0a7ef1d595 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -41,53 +41,10 @@ #include "amdgpu_dm_debugfs.h" #endif - #if defined(CONFIG_DRM_AMD_DC_DCN) #include "dc/dcn20/dcn20_resource.h" #endif -/* #define TRACE_DPCD */ - -#ifdef TRACE_DPCD -#define SIDE_BAND_MSG(address) (address >= DP_SIDEBAND_MSG_DOWN_REQ_BASE && address < DP_SINK_COUNT_ESI) - -static inline char *side_band_msg_type_to_str(uint32_t address) -{ - static char str[10] = {0}; - - if (address < DP_SIDEBAND_MSG_UP_REP_BASE) - strcpy(str, "DOWN_REQ"); - else if (address < DP_SIDEBAND_MSG_DOWN_REP_BASE) - strcpy(str, "UP_REP"); - else if (address < DP_SIDEBAND_MSG_UP_REQ_BASE) - strcpy(str, "DOWN_REP"); - else - strcpy(str, "UP_REQ"); - - return str; -} - -static void log_dpcd(uint8_t type, - uint32_t address, - uint8_t *data, - uint32_t size, - bool res) -{ - DRM_DEBUG_KMS("Op: %s, addr: %04x, SideBand Msg: %s, Op res: %s\n", - (type == DP_AUX_NATIVE_READ) || - (type == DP_AUX_I2C_READ) ? - "Read" : "Write", - address, - SIDE_BAND_MSG(address) ? - side_band_msg_type_to_str(address) : "Nop", - res ? "OK" : "Fail"); - - if (res) { - print_hex_dump(KERN_INFO, "Body: ", DUMP_PREFIX_NONE, 16, 1, data, size, false); - } -} -#endif - static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg) { @@ -162,16 +119,16 @@ amdgpu_dm_mst_connector_late_register(struct drm_connector *connector) to_amdgpu_dm_connector(connector); int r; - amdgpu_dm_connector->dm_dp_aux.aux.dev = connector->kdev; - r = drm_dp_aux_register(&amdgpu_dm_connector->dm_dp_aux.aux); - if (r) + r = drm_dp_mst_connector_late_register(connector, + amdgpu_dm_connector->port); + if (r < 0) return r; #if defined(CONFIG_DEBUG_FS) connector_debugfs_init(amdgpu_dm_connector); #endif - return r; + return 0; } static void @@ -451,9 +408,12 @@ static const struct drm_dp_mst_topology_cbs dm_mst_cbs = { }; void amdgpu_dm_initialize_dp_connector(struct amdgpu_display_manager *dm, - struct amdgpu_dm_connector *aconnector) + struct amdgpu_dm_connector *aconnector, + int link_index) { - aconnector->dm_dp_aux.aux.name = "dmdc"; + aconnector->dm_dp_aux.aux.name = + kasprintf(GFP_KERNEL, "AMDGPU DM aux hw bus %d", + link_index); aconnector->dm_dp_aux.aux.transfer = dm_dp_aux_transfer; aconnector->dm_dp_aux.ddc_service = aconnector->dc_link->ddc; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h index d6813ce67bbd..d2c56579a2cc 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h @@ -32,7 +32,8 @@ struct amdgpu_dm_connector; int dm_mst_get_pbn_divider(struct dc_link *link); void amdgpu_dm_initialize_dp_connector(struct amdgpu_display_manager *dm, - struct amdgpu_dm_connector *aconnector); + struct amdgpu_dm_connector *aconnector, + int link_index); #if defined(CONFIG_DRM_AMD_DC_DCN) bool compute_mst_dsc_configs_for_state(struct drm_atomic_state *state, diff --git a/drivers/gpu/drm/amd/display/dc/basics/Makefile b/drivers/gpu/drm/amd/display/dc/basics/Makefile index 7ad0cad0f4ef..01b99e0d788e 100644 --- a/drivers/gpu/drm/amd/display/dc/basics/Makefile +++ b/drivers/gpu/drm/amd/display/dc/basics/Makefile @@ -24,8 +24,7 @@ # It provides the general basic services required by other DAL # subcomponents. -BASICS = conversion.o fixpt31_32.o \ - log_helpers.o vector.o dc_common.o +BASICS = conversion.o fixpt31_32.o vector.o dc_common.o AMD_DAL_BASICS = $(addprefix $(AMDDALPATH)/dc/basics/,$(BASICS)) diff --git a/drivers/gpu/drm/amd/display/dc/basics/log_helpers.c b/drivers/gpu/drm/amd/display/dc/basics/log_helpers.c deleted file mode 100644 index 26583f346c39..000000000000 --- a/drivers/gpu/drm/amd/display/dc/basics/log_helpers.c +++ /dev/null @@ -1,39 +0,0 @@ -/* - * Copyright 2012-16 Advanced Micro Devices, Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: AMD - * - */ - -#include "core_types.h" -#include "logger.h" -#include "include/logger_interface.h" -#include "dm_helpers.h" - -void dc_conn_log_hex_linux(const uint8_t *hex_data, int hex_data_count) -{ - int i; - - if (hex_data) - for (i = 0; i < hex_data_count; i++) - DC_LOG_DEBUG("%2.2X ", hex_data[i]); -} - diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c index 8edc2506d49e..bed91572f82a 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c @@ -113,13 +113,19 @@ static void encoder_control_dmcub( struct dc_dmub_srv *dmcub, struct dig_encoder_stream_setup_parameters_v1_5 *dig) { - struct dmub_rb_cmd_digx_encoder_control encoder_control = { 0 }; + union dmub_rb_cmd cmd; - encoder_control.header.type = DMUB_CMD__VBIOS; - encoder_control.header.sub_type = DMUB_CMD__VBIOS_DIGX_ENCODER_CONTROL; - encoder_control.encoder_control.dig.stream_param = *dig; + memset(&cmd, 0, sizeof(cmd)); - dc_dmub_srv_cmd_queue(dmcub, &encoder_control.header); + cmd.digx_encoder_control.header.type = DMUB_CMD__VBIOS; + cmd.digx_encoder_control.header.sub_type = + DMUB_CMD__VBIOS_DIGX_ENCODER_CONTROL; + cmd.digx_encoder_control.header.payload_bytes = + sizeof(cmd.digx_encoder_control) - + sizeof(cmd.digx_encoder_control.header); + cmd.digx_encoder_control.encoder_control.dig.stream_param = *dig; + + dc_dmub_srv_cmd_queue(dmcub, &cmd); dc_dmub_srv_cmd_execute(dmcub); dc_dmub_srv_wait_idle(dmcub); } @@ -238,14 +244,19 @@ static void transmitter_control_dmcub( struct dc_dmub_srv *dmcub, struct dig_transmitter_control_parameters_v1_6 *dig) { - struct dmub_rb_cmd_dig1_transmitter_control transmitter_control; + union dmub_rb_cmd cmd; + + memset(&cmd, 0, sizeof(cmd)); - transmitter_control.header.type = DMUB_CMD__VBIOS; - transmitter_control.header.sub_type = + cmd.dig1_transmitter_control.header.type = DMUB_CMD__VBIOS; + cmd.dig1_transmitter_control.header.sub_type = DMUB_CMD__VBIOS_DIG1_TRANSMITTER_CONTROL; - transmitter_control.transmitter_control.dig = *dig; + cmd.dig1_transmitter_control.header.payload_bytes = + sizeof(cmd.dig1_transmitter_control) - + sizeof(cmd.dig1_transmitter_control.header); + cmd.dig1_transmitter_control.transmitter_control.dig = *dig; - dc_dmub_srv_cmd_queue(dmcub, &transmitter_control.header); + dc_dmub_srv_cmd_queue(dmcub, &cmd); dc_dmub_srv_cmd_execute(dmcub); dc_dmub_srv_wait_idle(dmcub); } @@ -339,13 +350,18 @@ static void set_pixel_clock_dmcub( struct dc_dmub_srv *dmcub, struct set_pixel_clock_parameter_v1_7 *clk) { - struct dmub_rb_cmd_set_pixel_clock pixel_clock = { 0 }; + union dmub_rb_cmd cmd; - pixel_clock.header.type = DMUB_CMD__VBIOS; - pixel_clock.header.sub_type = DMUB_CMD__VBIOS_SET_PIXEL_CLOCK; - pixel_clock.pixel_clock.clk = *clk; + memset(&cmd, 0, sizeof(cmd)); - dc_dmub_srv_cmd_queue(dmcub, &pixel_clock.header); + cmd.set_pixel_clock.header.type = DMUB_CMD__VBIOS; + cmd.set_pixel_clock.header.sub_type = DMUB_CMD__VBIOS_SET_PIXEL_CLOCK; + cmd.set_pixel_clock.header.payload_bytes = + sizeof(cmd.set_pixel_clock) - + sizeof(cmd.set_pixel_clock.header); + cmd.set_pixel_clock.pixel_clock.clk = *clk; + + dc_dmub_srv_cmd_queue(dmcub, &cmd); dc_dmub_srv_cmd_execute(dmcub); dc_dmub_srv_wait_idle(dmcub); } @@ -705,13 +721,19 @@ static void enable_disp_power_gating_dmcub( struct dc_dmub_srv *dmcub, struct enable_disp_power_gating_parameters_v2_1 *pwr) { - struct dmub_rb_cmd_enable_disp_power_gating power_gating; + union dmub_rb_cmd cmd; + + memset(&cmd, 0, sizeof(cmd)); - power_gating.header.type = DMUB_CMD__VBIOS; - power_gating.header.sub_type = DMUB_CMD__VBIOS_ENABLE_DISP_POWER_GATING; - power_gating.power_gating.pwr = *pwr; + cmd.enable_disp_power_gating.header.type = DMUB_CMD__VBIOS; + cmd.enable_disp_power_gating.header.sub_type = + DMUB_CMD__VBIOS_ENABLE_DISP_POWER_GATING; + cmd.enable_disp_power_gating.header.payload_bytes = + sizeof(cmd.enable_disp_power_gating) - + sizeof(cmd.enable_disp_power_gating.header); + cmd.enable_disp_power_gating.power_gating.pwr = *pwr; - dc_dmub_srv_cmd_queue(dmcub, &power_gating.header); + dc_dmub_srv_cmd_queue(dmcub, &cmd); dc_dmub_srv_cmd_execute(dmcub); dc_dmub_srv_wait_idle(dmcub); } diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c index 3960a8db94cb..1e5a92b192a1 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c @@ -690,6 +690,26 @@ static void hack_bounding_box(struct dcn_bw_internal_vars *v, struct dc_debug_options *dbg, struct dc_state *context) { + int i; + + for (i = 0; i < MAX_PIPES; i++) { + struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + + /** + * Workaround for avoiding pipe-split in cases where we'd split + * planes that are too small, resulting in splits that aren't + * valid for the scaler. + */ + if (pipe->plane_state && + (pipe->plane_state->dst_rect.width <= 16 || + pipe->plane_state->dst_rect.height <= 16 || + pipe->plane_state->src_rect.width <= 16 || + pipe->plane_state->src_rect.height <= 16)) { + hack_disable_optional_pipe_split(v); + return; + } + } + if (dbg->pipe_split_policy == MPC_SPLIT_AVOID) hack_disable_optional_pipe_split(v); @@ -702,7 +722,6 @@ static void hack_bounding_box(struct dcn_bw_internal_vars *v, hack_force_pipe_split(v, context->streams[0]->timing.pix_clk_100hz); } - unsigned int get_highest_allowed_voltage_level(uint32_t hw_internal_rev, uint32_t pci_revision_id) { /* for low power RV2 variants, the highest voltage level we want is 0 */ diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c index 8ec2dfe45d40..a5c2114e4292 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c @@ -90,7 +90,7 @@ void clk_mgr_exit_optimized_pwr_state(const struct dc *dc, struct clk_mgr *clk_m dc->hwss.exit_optimized_pwr_state(dc, dc->current_state); if (edp_link) { - clk_mgr->psr_allow_active_cache = edp_link->psr_allow_active; + clk_mgr->psr_allow_active_cache = edp_link->psr_settings.psr_allow_active; dc_link_set_psr_allow_active(edp_link, false, false); } diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c index 26db1c5d4e4d..b210f8e9d592 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c @@ -131,7 +131,7 @@ int dce_get_dp_ref_freq_khz(struct clk_mgr *clk_mgr_base) struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); int dprefclk_wdivider; int dprefclk_src_sel; - int dp_ref_clk_khz = 600000; + int dp_ref_clk_khz; int target_div; /* ASSERT DP Reference Clock source is from DFS*/ diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c index 97b7f32294fd..c320b7af7d34 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c @@ -97,9 +97,6 @@ int rv1_vbios_smu_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_di VBIOSSMC_MSG_SetDispclkFreq, requested_dispclk_khz / 1000); - /* Actual dispclk set is returned in the parameter register */ - actual_dispclk_set_mhz = REG_READ(MP1_SMN_C2PMSG_83) * 1000; - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) { if (clk_mgr->dfs_bypass_disp_clk != actual_dispclk_set_mhz) diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index 8489f1e56892..6f93a6ca4cf0 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -66,6 +66,8 @@ #include "dce/dce_i2c.h" +#include "dmub/dmub_srv.h" + #define CTX \ dc->ctx @@ -348,7 +350,7 @@ bool dc_stream_configure_crc(struct dc *dc, struct dc_stream_state *stream, for (i = 0; i < MAX_PIPES; i++) { pipe = &dc->current_state->res_ctx.pipe_ctx[i]; - if (pipe->stream == stream) + if (pipe->stream == stream && !pipe->top_pipe && !pipe->prev_odm_pipe) break; } /* Stream not found */ @@ -365,6 +367,9 @@ bool dc_stream_configure_crc(struct dc *dc, struct dc_stream_state *stream, param.windowb_x_end = pipe->stream->timing.h_addressable; param.windowb_y_end = pipe->stream->timing.v_addressable; + param.dsc_mode = pipe->stream->timing.flags.DSC ? 1:0; + param.odm_mode = pipe->next_odm_pipe ? 1:0; + /* Default to the union of both windows */ param.selection = UNION_WINDOW_A_B; param.continuous_mode = continuous; @@ -834,11 +839,10 @@ static void disable_dangling_plane(struct dc *dc, struct dc_state *context) static void wait_for_no_pipes_pending(struct dc *dc, struct dc_state *context) { int i; - int count = 0; - struct pipe_ctx *pipe; PERF_TRACE(); for (i = 0; i < MAX_PIPES; i++) { - pipe = &context->res_ctx.pipe_ctx[i]; + int count = 0; + struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; if (!pipe->plane_state) continue; @@ -1012,9 +1016,17 @@ static void program_timing_sync( } } - /* set first pipe with plane as master */ + /* set first unblanked pipe as master */ for (j = 0; j < group_size; j++) { - if (pipe_set[j]->plane_state) { + bool is_blanked; + + if (pipe_set[j]->stream_res.opp->funcs->dpg_is_blanked) + is_blanked = + pipe_set[j]->stream_res.opp->funcs->dpg_is_blanked(pipe_set[j]->stream_res.opp); + else + is_blanked = + pipe_set[j]->stream_res.tg->funcs->is_blanked(pipe_set[j]->stream_res.tg); + if (!is_blanked) { if (j == 0) break; @@ -1035,9 +1047,17 @@ static void program_timing_sync( status->timing_sync_info.master = false; } - /* remove any other pipes with plane as they have already been synced */ + /* remove any other unblanked pipes as they have already been synced */ for (j = j + 1; j < group_size; j++) { - if (pipe_set[j]->plane_state) { + bool is_blanked; + + if (pipe_set[j]->stream_res.opp->funcs->dpg_is_blanked) + is_blanked = + pipe_set[j]->stream_res.opp->funcs->dpg_is_blanked(pipe_set[j]->stream_res.opp); + else + is_blanked = + pipe_set[j]->stream_res.tg->funcs->is_blanked(pipe_set[j]->stream_res.tg); + if (!is_blanked) { group_size--; pipe_set[j] = pipe_set[group_size]; j--; @@ -2205,7 +2225,7 @@ static void commit_planes_do_stream_update(struct dc *dc, if (should_program_abm) { if (*stream_update->abm_level == ABM_LEVEL_IMMEDIATE_DISABLE) { - pipe_ctx->stream_res.abm->funcs->set_abm_immediate_disable(pipe_ctx->stream_res.abm); + dc->hwss.set_abm_immediate_disable(pipe_ctx); } else { pipe_ctx->stream_res.abm->funcs->set_abm_level( pipe_ctx->stream_res.abm, stream->abm_level); @@ -2518,6 +2538,12 @@ void dc_commit_updates_for_stream(struct dc *dc, copy_stream_update_to_stream(dc, context, stream, stream_update); + if (!dc->res_pool->funcs->validate_bandwidth(dc, context, false)) { + DC_ERROR("Mode validation failed for stream update!\n"); + dc_release_state(context); + return; + } + commit_planes_for_stream( dc, srf_updates, @@ -2641,33 +2667,12 @@ void dc_set_power_state( void dc_resume(struct dc *dc) { - uint32_t i; for (i = 0; i < dc->link_count; i++) core_link_resume(dc->links[i]); } -unsigned int dc_get_current_backlight_pwm(struct dc *dc) -{ - struct abm *abm = dc->res_pool->abm; - - if (abm) - return abm->funcs->get_current_backlight(abm); - - return 0; -} - -unsigned int dc_get_target_backlight_pwm(struct dc *dc) -{ - struct abm *abm = dc->res_pool->abm; - - if (abm) - return abm->funcs->get_target_backlight(abm); - - return 0; -} - bool dc_is_dmcu_initialized(struct dc *dc) { struct dmcu *dmcu = dc->res_pool->dmcu; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link.c b/drivers/gpu/drm/amd/display/dc/core/dc_link.c index 67cfff1586e9..48ab51533d5d 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c @@ -26,7 +26,7 @@ #include <linux/slab.h> #include "dm_services.h" -#include "atom.h" +#include "atomfirmware.h" #include "dm_helpers.h" #include "dc.h" #include "grph_object_id.h" @@ -46,10 +46,11 @@ #include "dmcu.h" #include "hw/clk_mgr.h" #include "dce/dmub_psr.h" +#include "dmub/dmub_srv.h" +#include "inc/hw/panel_cntl.h" #define DC_LOGGER_INIT(logger) - #define LINK_INFO(...) \ DC_LOG_HW_HOTPLUG( \ __VA_ARGS__) @@ -64,11 +65,11 @@ enum { PEAK_FACTOR_X1000 = 1006, /* - * Some receivers fail to train on first try and are good - * on subsequent tries. 2 retries should be plenty. If we - * don't have a successful training then we don't expect to - * ever get one. - */ + * Some receivers fail to train on first try and are good + * on subsequent tries. 2 retries should be plenty. If we + * don't have a successful training then we don't expect to + * ever get one. + */ LINK_TRAINING_MAX_VERIFY_RETRY = 2 }; @@ -79,7 +80,7 @@ static void dc_link_destruct(struct dc_link *link) { int i; - if (link->hpd_gpio != NULL) { + if (link->hpd_gpio) { dal_gpio_destroy_irq(&link->hpd_gpio); link->hpd_gpio = NULL; } @@ -87,7 +88,10 @@ static void dc_link_destruct(struct dc_link *link) if (link->ddc) dal_ddc_service_destroy(&link->ddc); - if(link->link_enc) + if (link->panel_cntl) + link->panel_cntl->funcs->destroy(&link->panel_cntl); + + if (link->link_enc) link->link_enc->funcs->destroy(&link->link_enc); if (link->local_sink) @@ -98,8 +102,8 @@ static void dc_link_destruct(struct dc_link *link) } struct gpio *get_hpd_gpio(struct dc_bios *dcb, - struct graphics_object_id link_id, - struct gpio_service *gpio_service) + struct graphics_object_id link_id, + struct gpio_service *gpio_service) { enum bp_result bp_result; struct graphics_object_hpd_info hpd_info; @@ -116,10 +120,9 @@ struct gpio *get_hpd_gpio(struct dc_bios *dcb, return NULL; } - return dal_gpio_service_create_irq( - gpio_service, - pin_info.offset, - pin_info.mask); + return dal_gpio_service_create_irq(gpio_service, + pin_info.offset, + pin_info.mask); } /* @@ -134,13 +137,10 @@ struct gpio *get_hpd_gpio(struct dc_bios *dcb, * @return * true on success, false otherwise */ -static bool program_hpd_filter( - const struct dc_link *link) +static bool program_hpd_filter(const struct dc_link *link) { bool result = false; - struct gpio *hpd; - int delay_on_connect_in_ms = 0; int delay_on_disconnect_in_ms = 0; @@ -159,10 +159,10 @@ static bool program_hpd_filter( case SIGNAL_TYPE_DISPLAY_PORT_MST: /* Program hpd filter to allow DP signal to settle */ /* 500: not able to detect MST <-> SST switch as HPD is low for - * only 100ms on DELL U2413 - * 0: some passive dongle still show aux mode instead of i2c - * 20-50:not enough to hide bouncing HPD with passive dongle. - * also see intermittent i2c read issues. + * only 100ms on DELL U2413 + * 0: some passive dongle still show aux mode instead of i2c + * 20-50: not enough to hide bouncing HPD with passive dongle. + * also see intermittent i2c read issues. */ delay_on_connect_in_ms = 80; delay_on_disconnect_in_ms = 0; @@ -175,7 +175,8 @@ static bool program_hpd_filter( } /* Obtain HPD handle */ - hpd = get_hpd_gpio(link->ctx->dc_bios, link->link_id, link->ctx->gpio_service); + hpd = get_hpd_gpio(link->ctx->dc_bios, link->link_id, + link->ctx->gpio_service); if (!hpd) return result; @@ -226,8 +227,9 @@ bool dc_link_detect_sink(struct dc_link *link, enum dc_connection_type *type) } /* todo: may need to lock gpio access */ - hpd_pin = get_hpd_gpio(link->ctx->dc_bios, link->link_id, link->ctx->gpio_service); - if (hpd_pin == NULL) + hpd_pin = get_hpd_gpio(link->ctx->dc_bios, link->link_id, + link->ctx->gpio_service); + if (!hpd_pin) goto hpd_gpio_failure; dal_gpio_open(hpd_pin, GPIO_MODE_INTERRUPT); @@ -248,8 +250,7 @@ hpd_gpio_failure: return false; } -static enum ddc_transaction_type get_ddc_transaction_type( - enum signal_type sink_signal) +static enum ddc_transaction_type get_ddc_transaction_type(enum signal_type sink_signal) { enum ddc_transaction_type transaction_type = DDC_TRANSACTION_TYPE_NONE; @@ -270,7 +271,8 @@ static enum ddc_transaction_type get_ddc_transaction_type( case SIGNAL_TYPE_DISPLAY_PORT_MST: /* MST does not use I2COverAux, but there is the * SPECIAL use case for "immediate dwnstrm device - * access" (EPR#370830). */ + * access" (EPR#370830). + */ transaction_type = DDC_TRANSACTION_TYPE_I2C_OVER_AUX; break; @@ -281,9 +283,8 @@ static enum ddc_transaction_type get_ddc_transaction_type( return transaction_type; } -static enum signal_type get_basic_signal_type( - struct graphics_object_id encoder, - struct graphics_object_id downstream) +static enum signal_type get_basic_signal_type(struct graphics_object_id encoder, + struct graphics_object_id downstream) { if (downstream.type == OBJECT_TYPE_CONNECTOR) { switch (downstream.id) { @@ -369,10 +370,11 @@ bool dc_link_is_dp_sink_present(struct dc_link *link) /* Open GPIO and set it to I2C mode */ /* Note: this GpioMode_Input will be converted * to GpioConfigType_I2cAuxDualMode in GPIO component, - * which indicates we need additional delay */ + * which indicates we need additional delay + */ - if (GPIO_RESULT_OK != dal_ddc_open( - ddc, GPIO_MODE_INPUT, GPIO_DDC_CONFIG_TYPE_MODE_I2C)) { + if (dal_ddc_open(ddc, GPIO_MODE_INPUT, + GPIO_DDC_CONFIG_TYPE_MODE_I2C) != GPIO_RESULT_OK) { dal_ddc_close(ddc); return present; @@ -406,25 +408,25 @@ bool dc_link_is_dp_sink_present(struct dc_link *link) * @brief * Detect output sink type */ -static enum signal_type link_detect_sink( - struct dc_link *link, - enum dc_detect_reason reason) +static enum signal_type link_detect_sink(struct dc_link *link, + enum dc_detect_reason reason) { - enum signal_type result = get_basic_signal_type( - link->link_enc->id, link->link_id); + enum signal_type result = get_basic_signal_type(link->link_enc->id, + link->link_id); /* Internal digital encoder will detect only dongles - * that require digital signal */ + * that require digital signal + */ /* Detection mechanism is different * for different native connectors. * LVDS connector supports only LVDS signal; * PCIE is a bus slot, the actual connector needs to be detected first; * eDP connector supports only eDP signal; - * HDMI should check straps for audio */ + * HDMI should check straps for audio + */ /* PCIE detects the actual connector on add-on board */ - if (link->link_id.id == CONNECTOR_ID_PCIE) { /* ZAZTODO implement PCIE add-on card detection */ } @@ -432,8 +434,10 @@ static enum signal_type link_detect_sink( switch (link->link_id.id) { case CONNECTOR_ID_HDMI_TYPE_A: { /* check audio support: - * if native HDMI is not supported, switch to DVI */ - struct audio_support *aud_support = &link->dc->res_pool->audio_support; + * if native HDMI is not supported, switch to DVI + */ + struct audio_support *aud_support = + &link->dc->res_pool->audio_support; if (!aud_support->hdmi_audio_native) if (link->link_id.id == CONNECTOR_ID_HDMI_TYPE_A) @@ -461,16 +465,15 @@ static enum signal_type link_detect_sink( return result; } -static enum signal_type decide_signal_from_strap_and_dongle_type( - enum display_dongle_type dongle_type, - struct audio_support *audio_support) +static enum signal_type decide_signal_from_strap_and_dongle_type(enum display_dongle_type dongle_type, + struct audio_support *audio_support) { enum signal_type signal = SIGNAL_TYPE_NONE; switch (dongle_type) { case DISPLAY_DONGLE_DP_HDMI_DONGLE: if (audio_support->hdmi_audio_on_dongle) - signal = SIGNAL_TYPE_HDMI_TYPE_A; + signal = SIGNAL_TYPE_HDMI_TYPE_A; else signal = SIGNAL_TYPE_DVI_SINGLE_LINK; break; @@ -491,16 +494,14 @@ static enum signal_type decide_signal_from_strap_and_dongle_type( return signal; } -static enum signal_type dp_passive_dongle_detection( - struct ddc_service *ddc, - struct display_sink_capability *sink_cap, - struct audio_support *audio_support) +static enum signal_type dp_passive_dongle_detection(struct ddc_service *ddc, + struct display_sink_capability *sink_cap, + struct audio_support *audio_support) { - dal_ddc_service_i2c_query_dp_dual_mode_adaptor( - ddc, sink_cap); - return decide_signal_from_strap_and_dongle_type( - sink_cap->dongle_type, - audio_support); + dal_ddc_service_i2c_query_dp_dual_mode_adaptor(ddc, sink_cap); + + return decide_signal_from_strap_and_dongle_type(sink_cap->dongle_type, + audio_support); } static void link_disconnect_sink(struct dc_link *link) @@ -519,6 +520,96 @@ static void link_disconnect_remap(struct dc_sink *prev_sink, struct dc_link *lin link->local_sink = prev_sink; } +#if defined(CONFIG_DRM_AMD_DC_HDCP) +bool dc_link_is_hdcp14(struct dc_link *link) +{ + bool ret = false; + + switch (link->connector_signal) { + case SIGNAL_TYPE_DISPLAY_PORT: + case SIGNAL_TYPE_DISPLAY_PORT_MST: + ret = link->hdcp_caps.bcaps.bits.HDCP_CAPABLE; + break; + case SIGNAL_TYPE_DVI_SINGLE_LINK: + case SIGNAL_TYPE_DVI_DUAL_LINK: + case SIGNAL_TYPE_HDMI_TYPE_A: + /* HDMI doesn't tell us its HDCP(1.4) capability, so assume to always be capable, + * we can poll for bksv but some displays have an issue with this. Since its so rare + * for a display to not be 1.4 capable, this assumtion is ok + */ + ret = true; + break; + default: + break; + } + return ret; +} + +bool dc_link_is_hdcp22(struct dc_link *link) +{ + bool ret = false; + + switch (link->connector_signal) { + case SIGNAL_TYPE_DISPLAY_PORT: + case SIGNAL_TYPE_DISPLAY_PORT_MST: + ret = (link->hdcp_caps.bcaps.bits.HDCP_CAPABLE && + link->hdcp_caps.rx_caps.fields.byte0.hdcp_capable && + (link->hdcp_caps.rx_caps.fields.version == 0x2)) ? 1 : 0; + break; + case SIGNAL_TYPE_DVI_SINGLE_LINK: + case SIGNAL_TYPE_DVI_DUAL_LINK: + case SIGNAL_TYPE_HDMI_TYPE_A: + ret = (link->hdcp_caps.rx_caps.fields.version == 0x4) ? 1:0; + break; + default: + break; + } + + return ret; +} + +static void query_hdcp_capability(enum signal_type signal, struct dc_link *link) +{ + struct hdcp_protection_message msg22; + struct hdcp_protection_message msg14; + + memset(&msg22, 0, sizeof(struct hdcp_protection_message)); + memset(&msg14, 0, sizeof(struct hdcp_protection_message)); + memset(link->hdcp_caps.rx_caps.raw, 0, + sizeof(link->hdcp_caps.rx_caps.raw)); + + if ((link->connector_signal == SIGNAL_TYPE_DISPLAY_PORT && + link->ddc->transaction_type == + DDC_TRANSACTION_TYPE_I2C_OVER_AUX) || + link->connector_signal == SIGNAL_TYPE_EDP) { + msg22.data = link->hdcp_caps.rx_caps.raw; + msg22.length = sizeof(link->hdcp_caps.rx_caps.raw); + msg22.msg_id = HDCP_MESSAGE_ID_RX_CAPS; + } else { + msg22.data = &link->hdcp_caps.rx_caps.fields.version; + msg22.length = sizeof(link->hdcp_caps.rx_caps.fields.version); + msg22.msg_id = HDCP_MESSAGE_ID_HDCP2VERSION; + } + msg22.version = HDCP_VERSION_22; + msg22.link = HDCP_LINK_PRIMARY; + msg22.max_retries = 5; + dc_process_hdcp_msg(signal, link, &msg22); + + if (signal == SIGNAL_TYPE_DISPLAY_PORT || signal == SIGNAL_TYPE_DISPLAY_PORT_MST) { + enum hdcp_message_status status = HDCP_MESSAGE_UNSUPPORTED; + + msg14.data = &link->hdcp_caps.bcaps.raw; + msg14.length = sizeof(link->hdcp_caps.bcaps.raw); + msg14.msg_id = HDCP_MESSAGE_ID_READ_BCAPS; + msg14.version = HDCP_VERSION_14; + msg14.link = HDCP_LINK_PRIMARY; + msg14.max_retries = 5; + + status = dc_process_hdcp_msg(signal, link, &msg14); + } + +} +#endif static void read_current_link_settings_on_detect(struct dc_link *link) { @@ -532,18 +623,18 @@ static void read_current_link_settings_on_detect(struct dc_link *link) // Read DPCD 00101h to find out the number of lanes currently set for (i = 0; i < read_dpcd_retry_cnt; i++) { - status = core_link_read_dpcd( - link, - DP_LANE_COUNT_SET, - &lane_count_set.raw, - sizeof(lane_count_set)); + status = core_link_read_dpcd(link, + DP_LANE_COUNT_SET, + &lane_count_set.raw, + sizeof(lane_count_set)); /* First DPCD read after VDD ON can fail if the particular board * does not have HPD pin wired correctly. So if DPCD read fails, * which it should never happen, retry a few times. Target worst * case scenario of 80 ms. */ if (status == DC_OK) { - link->cur_link_settings.lane_count = lane_count_set.bits.LANE_COUNT_SET; + link->cur_link_settings.lane_count = + lane_count_set.bits.LANE_COUNT_SET; break; } @@ -552,7 +643,7 @@ static void read_current_link_settings_on_detect(struct dc_link *link) // Read DPCD 00100h to find if standard link rates are set core_link_read_dpcd(link, DP_LINK_BW_SET, - &link_bw_set, sizeof(link_bw_set)); + &link_bw_set, sizeof(link_bw_set)); if (link_bw_set == 0) { if (link->connector_signal == SIGNAL_TYPE_EDP) { @@ -560,12 +651,12 @@ static void read_current_link_settings_on_detect(struct dc_link *link) * Read DPCD 00115h to find the edp link rate set used */ core_link_read_dpcd(link, DP_LINK_RATE_SET, - &link_rate_set, sizeof(link_rate_set)); + &link_rate_set, sizeof(link_rate_set)); // edp_supported_link_rates_count = 0 for DP if (link_rate_set < link->dpcd_caps.edp_supported_link_rates_count) { link->cur_link_settings.link_rate = - link->dpcd_caps.edp_supported_link_rates[link_rate_set]; + link->dpcd_caps.edp_supported_link_rates[link_rate_set]; link->cur_link_settings.link_rate_set = link_rate_set; link->cur_link_settings.use_link_rate_set = true; } @@ -579,7 +670,7 @@ static void read_current_link_settings_on_detect(struct dc_link *link) } // Read DPCD 00003h to find the max down spread. core_link_read_dpcd(link, DP_MAX_DOWNSPREAD, - &max_down_spread.raw, sizeof(max_down_spread)); + &max_down_spread.raw, sizeof(max_down_spread)); link->cur_link_settings.link_spread = max_down_spread.bits.MAX_DOWN_SPREAD ? LINK_SPREAD_05_DOWNSPREAD_30KHZ : LINK_SPREAD_DISABLED; @@ -612,6 +703,12 @@ static bool detect_dp(struct dc_link *link, dal_ddc_service_set_transaction_type(link->ddc, sink_caps->transaction_type); +#if defined(CONFIG_DRM_AMD_DC_HDCP) + /* In case of fallback to SST when topology discovery below fails + * HDCP caps will be querried again later by the upper layer (caller + * of this function). */ + query_hdcp_capability(SIGNAL_TYPE_DISPLAY_PORT_MST, link); +#endif /* * This call will initiate MST topology discovery. Which * will detect MST ports and add new DRM connector DRM @@ -683,12 +780,12 @@ static bool is_same_edid(struct dc_edid *old_edid, struct dc_edid *new_edid) if (new_edid->length == 0) return false; - return (memcmp(old_edid->raw_edid, new_edid->raw_edid, new_edid->length) == 0); + return (memcmp(old_edid->raw_edid, + new_edid->raw_edid, new_edid->length) == 0); } -static bool wait_for_alt_mode(struct dc_link *link) +static bool wait_for_entering_dp_alt_mode(struct dc_link *link) { - /** * something is terribly wrong if time out is > 200ms. (5Hz) * 500 microseconds * 400 tries us 200 ms @@ -703,7 +800,7 @@ static bool wait_for_alt_mode(struct dc_link *link) DC_LOGGER_INIT(link->ctx->logger); - if (link->link_enc->funcs->is_in_alt_mode == NULL) + if (!link->link_enc->funcs->is_in_alt_mode) return true; is_in_alt_mode = link->link_enc->funcs->is_in_alt_mode(link->link_enc); @@ -718,21 +815,21 @@ static bool wait_for_alt_mode(struct dc_link *link) udelay(sleep_time_in_microseconds); /* ask the link if alt mode is enabled, if so return ok */ if (link->link_enc->funcs->is_in_alt_mode(link->link_enc)) { - finish_timestamp = dm_get_timestamp(link->ctx); - time_taken_in_ns = dm_get_elapse_time_in_ns( - link->ctx, finish_timestamp, enter_timestamp); + time_taken_in_ns = + dm_get_elapse_time_in_ns(link->ctx, + finish_timestamp, + enter_timestamp); DC_LOG_WARNING("Alt mode entered finished after %llu ms\n", div_u64(time_taken_in_ns, 1000000)); return true; } - } finish_timestamp = dm_get_timestamp(link->ctx); time_taken_in_ns = dm_get_elapse_time_in_ns(link->ctx, finish_timestamp, enter_timestamp); DC_LOG_WARNING("Alt mode has timed out after %llu ms\n", - div_u64(time_taken_in_ns, 1000000)); + div_u64(time_taken_in_ns, 1000000)); return false; } @@ -768,30 +865,30 @@ static bool dc_link_detect_helper(struct dc_link *link, return false; if ((link->connector_signal == SIGNAL_TYPE_LVDS || - link->connector_signal == SIGNAL_TYPE_EDP) && - link->local_sink) { - + link->connector_signal == SIGNAL_TYPE_EDP) && + link->local_sink) { // need to re-write OUI and brightness in resume case if (link->connector_signal == SIGNAL_TYPE_EDP) { dpcd_set_source_specific_data(link); - dc_link_set_default_brightness_aux(link); //TODO: use cached + dc_link_set_default_brightness_aux(link); + //TODO: use cached } return true; } - if (false == dc_link_detect_sink(link, &new_connection_type)) { + if (!dc_link_detect_sink(link, &new_connection_type)) { BREAK_TO_DEBUGGER(); return false; } prev_sink = link->local_sink; - if (prev_sink != NULL) { + if (prev_sink) { dc_sink_retain(prev_sink); memcpy(&prev_dpcd_caps, &link->dpcd_caps, sizeof(struct dpcd_caps)); } - link_disconnect_sink(link); + link_disconnect_sink(link); if (new_connection_type != dc_connection_none) { link->type = new_connection_type; link->link_state_valid = false; @@ -838,35 +935,31 @@ static bool dc_link_detect_helper(struct dc_link *link, } case SIGNAL_TYPE_DISPLAY_PORT: { - /* wa HPD high coming too early*/ if (link->link_enc->features.flags.bits.DP_IS_USB_C == 1) { - /* if alt mode times out, return false */ - if (wait_for_alt_mode(link) == false) { + if (!wait_for_entering_dp_alt_mode(link)) return false; - } } - if (!detect_dp( - link, - &sink_caps, - &converter_disable_audio, - aud_support, reason)) { - if (prev_sink != NULL) + if (!detect_dp(link, &sink_caps, + &converter_disable_audio, + aud_support, reason)) { + if (prev_sink) dc_sink_release(prev_sink); return false; } // Check if dpcp block is the same - if (prev_sink != NULL) { - if (memcmp(&link->dpcd_caps, &prev_dpcd_caps, sizeof(struct dpcd_caps))) + if (prev_sink) { + if (memcmp(&link->dpcd_caps, &prev_dpcd_caps, + sizeof(struct dpcd_caps))) same_dpcd = false; } /* Active dongle downstream unplug*/ if (link->type == dc_connection_active_dongle && - link->dpcd_caps.sink_count.bits.SINK_COUNT == 0) { - if (prev_sink != NULL) + link->dpcd_caps.sink_count.bits.SINK_COUNT == 0) { + if (prev_sink) /* Downstream unplug */ dc_sink_release(prev_sink); return true; @@ -874,7 +967,7 @@ static bool dc_link_detect_helper(struct dc_link *link, if (link->type == dc_connection_mst_branch) { LINK_INFO("link=%d, mst branch is now Connected\n", - link->link_index); + link->link_index); /* Need to setup mst link_cap struct here * otherwise dc_link_detect() will leave mst link_cap * empty which leads to allocate_mst_payload() has "0" @@ -882,15 +975,15 @@ static bool dc_link_detect_helper(struct dc_link *link, */ dp_verify_mst_link_cap(link); - if (prev_sink != NULL) + if (prev_sink) dc_sink_release(prev_sink); return false; } // For seamless boot, to skip verify link cap, we read UEFI settings and set them as verified. if (reason == DETECT_REASON_BOOT && - dc_ctx->dc->config.power_down_display_on_boot == false && - link->link_status.link_active == true) + !dc_ctx->dc->config.power_down_display_on_boot && + link->link_status.link_active) perform_dp_seamless_boot = true; if (perform_dp_seamless_boot) { @@ -903,24 +996,23 @@ static bool dc_link_detect_helper(struct dc_link *link, default: DC_ERROR("Invalid connector type! signal:%d\n", - link->connector_signal); - if (prev_sink != NULL) + link->connector_signal); + if (prev_sink) dc_sink_release(prev_sink); return false; } /* switch() */ if (link->dpcd_caps.sink_count.bits.SINK_COUNT) - link->dpcd_sink_count = link->dpcd_caps.sink_count. - bits.SINK_COUNT; + link->dpcd_sink_count = + link->dpcd_caps.sink_count.bits.SINK_COUNT; else link->dpcd_sink_count = 1; - dal_ddc_service_set_transaction_type( - link->ddc, - sink_caps.transaction_type); + dal_ddc_service_set_transaction_type(link->ddc, + sink_caps.transaction_type); - link->aux_mode = dal_ddc_service_is_in_aux_transaction_mode( - link->ddc); + link->aux_mode = + dal_ddc_service_is_in_aux_transaction_mode(link->ddc); sink_init_data.link = link; sink_init_data.sink_signal = sink_caps.signal; @@ -928,7 +1020,7 @@ static bool dc_link_detect_helper(struct dc_link *link, sink = dc_sink_create(&sink_init_data); if (!sink) { DC_ERROR("Failed to create sink!\n"); - if (prev_sink != NULL) + if (prev_sink) dc_sink_release(prev_sink); return false; } @@ -939,10 +1031,8 @@ static bool dc_link_detect_helper(struct dc_link *link, /* dc_sink_create returns a new reference */ link->local_sink = sink; - edid_status = dm_helpers_read_local_edid( - link->ctx, - link, - sink); + edid_status = dm_helpers_read_local_edid(link->ctx, + link, sink); switch (edid_status) { case EDID_BAD_CHECKSUM: @@ -950,7 +1040,6 @@ static bool dc_link_detect_helper(struct dc_link *link, break; case EDID_NO_RESPONSE: DC_LOG_ERROR("No EDID read.\n"); - /* * Abort detection for non-DP connectors if we have * no EDID @@ -961,7 +1050,7 @@ static bool dc_link_detect_helper(struct dc_link *link, */ if (dc_is_hdmi_signal(link->connector_signal) || dc_is_dvi_signal(link->connector_signal)) { - if (prev_sink != NULL) + if (prev_sink) dc_sink_release(prev_sink); return false; @@ -974,45 +1063,53 @@ static bool dc_link_detect_helper(struct dc_link *link, link->ctx->dc->debug.disable_fec = true; // Check if edid is the same - if ((prev_sink != NULL) && ((edid_status == EDID_THE_SAME) || (edid_status == EDID_OK))) - same_edid = is_same_edid(&prev_sink->dc_edid, &sink->dc_edid); + if ((prev_sink) && + (edid_status == EDID_THE_SAME || edid_status == EDID_OK)) + same_edid = is_same_edid(&prev_sink->dc_edid, + &sink->dc_edid); if (sink->edid_caps.panel_patch.skip_scdc_overwrite) link->ctx->dc->debug.hdmi20_disable = true; if (link->connector_signal == SIGNAL_TYPE_DISPLAY_PORT && - sink_caps.transaction_type == DDC_TRANSACTION_TYPE_I2C_OVER_AUX) { + sink_caps.transaction_type == + DDC_TRANSACTION_TYPE_I2C_OVER_AUX) { /* * TODO debug why Dell 2413 doesn't like * two link trainings */ +#if defined(CONFIG_DRM_AMD_DC_HDCP) + query_hdcp_capability(sink->sink_signal, link); +#endif // verify link cap for SST non-seamless boot if (!perform_dp_seamless_boot) dp_verify_link_cap_with_retries(link, - &link->reported_link_cap, - LINK_TRAINING_MAX_VERIFY_RETRY); + &link->reported_link_cap, + LINK_TRAINING_MAX_VERIFY_RETRY); } else { // If edid is the same, then discard new sink and revert back to original sink if (same_edid) { link_disconnect_remap(prev_sink, link); sink = prev_sink; prev_sink = NULL; - } +#if defined(CONFIG_DRM_AMD_DC_HDCP) + query_hdcp_capability(sink->sink_signal, link); +#endif } /* HDMI-DVI Dongle */ if (sink->sink_signal == SIGNAL_TYPE_HDMI_TYPE_A && - !sink->edid_caps.edid_hdmi) + !sink->edid_caps.edid_hdmi) sink->sink_signal = SIGNAL_TYPE_DVI_SINGLE_LINK; /* Connectivity log: detection */ for (i = 0; i < sink->dc_edid.length / DC_EDID_BLOCK_SIZE; i++) { CONN_DATA_DETECT(link, - &sink->dc_edid.raw_edid[i * DC_EDID_BLOCK_SIZE], - DC_EDID_BLOCK_SIZE, - "%s: [Block %d] ", sink->edid_caps.display_name, i); + &sink->dc_edid.raw_edid[i * DC_EDID_BLOCK_SIZE], + DC_EDID_BLOCK_SIZE, + "%s: [Block %d] ", sink->edid_caps.display_name, i); } DC_LOG_DETECTION_EDID_PARSER("%s: " @@ -1047,17 +1144,18 @@ static bool dc_link_detect_helper(struct dc_link *link, sink->edid_caps.audio_modes[i].sample_rate, sink->edid_caps.audio_modes[i].sample_size); } - } else { /* From Connected-to-Disconnected. */ if (link->type == dc_connection_mst_branch) { LINK_INFO("link=%d, mst branch is now Disconnected\n", - link->link_index); + link->link_index); dm_helpers_dp_mst_stop_top_mgr(link->ctx, link); link->mst_stream_alloc_table.stream_count = 0; - memset(link->mst_stream_alloc_table.stream_allocations, 0, sizeof(link->mst_stream_alloc_table.stream_allocations)); + memset(link->mst_stream_alloc_table.stream_allocations, + 0, + sizeof(link->mst_stream_alloc_table.stream_allocations)); } link->type = dc_connection_none; @@ -1071,16 +1169,15 @@ static bool dc_link_detect_helper(struct dc_link *link, } LINK_INFO("link=%d, dc_sink_in=%p is now %s prev_sink=%p dpcd same=%d edid same=%d\n", - link->link_index, sink, - (sink_caps.signal == SIGNAL_TYPE_NONE ? - "Disconnected":"Connected"), prev_sink, - same_dpcd, same_edid); + link->link_index, sink, + (sink_caps.signal == + SIGNAL_TYPE_NONE ? "Disconnected" : "Connected"), + prev_sink, same_dpcd, same_edid); - if (prev_sink != NULL) + if (prev_sink) dc_sink_release(prev_sink); return true; - } bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) @@ -1110,13 +1207,13 @@ bool dc_link_get_hpd_state(struct dc_link *dc_link) return state; } -static enum hpd_source_id get_hpd_line( - struct dc_link *link) +static enum hpd_source_id get_hpd_line(struct dc_link *link) { struct gpio *hpd; enum hpd_source_id hpd_id = HPD_SOURCEID_UNKNOWN; - hpd = get_hpd_gpio(link->ctx->dc_bios, link->link_id, link->ctx->gpio_service); + hpd = get_hpd_gpio(link->ctx->dc_bios, link->link_id, + link->ctx->gpio_service); if (hpd) { switch (dal_irq_get_source(hpd)) { @@ -1191,8 +1288,7 @@ static enum channel_id get_ddc_line(struct dc_link *link) return channel; } -static enum transmitter translate_encoder_to_transmitter( - struct graphics_object_id encoder) +static enum transmitter translate_encoder_to_transmitter(struct graphics_object_id encoder) { switch (encoder.id) { case ENCODER_ID_INTERNAL_UNIPHY: @@ -1256,17 +1352,18 @@ static enum transmitter translate_encoder_to_transmitter( } } -static bool dc_link_construct( - struct dc_link *link, - const struct link_init_data *init_params) +static bool dc_link_construct(struct dc_link *link, + const struct link_init_data *init_params) { uint8_t i; struct ddc_service_init_data ddc_service_init_data = { { 0 } }; struct dc_context *dc_ctx = init_params->ctx; struct encoder_init_data enc_init_data = { 0 }; + struct panel_cntl_init_data panel_cntl_init_data = { 0 }; struct integrated_info info = {{{ 0 }}}; struct dc_bios *bios = init_params->dc->ctx->dc_bios; const struct dc_vbios_funcs *bp_funcs = bios->funcs; + DC_LOGGER_INIT(dc_ctx->logger); link->irq_source_hpd = DC_IRQ_SOURCE_INVALID; @@ -1278,23 +1375,27 @@ static bool dc_link_construct( link->ctx = dc_ctx; link->link_index = init_params->link_index; - memset(&link->preferred_training_settings, 0, sizeof(struct dc_link_training_overrides)); - memset(&link->preferred_link_setting, 0, sizeof(struct dc_link_settings)); + memset(&link->preferred_training_settings, 0, + sizeof(struct dc_link_training_overrides)); + memset(&link->preferred_link_setting, 0, + sizeof(struct dc_link_settings)); - link->link_id = bios->funcs->get_connector_id(bios, init_params->connector_index); + link->link_id = + bios->funcs->get_connector_id(bios, init_params->connector_index); if (link->link_id.type != OBJECT_TYPE_CONNECTOR) { dm_output_to_console("%s: Invalid Connector ObjectID from Adapter Service for connector index:%d! type %d expected %d\n", - __func__, init_params->connector_index, - link->link_id.type, OBJECT_TYPE_CONNECTOR); + __func__, init_params->connector_index, + link->link_id.type, OBJECT_TYPE_CONNECTOR); goto create_fail; } if (link->dc->res_pool->funcs->link_init) link->dc->res_pool->funcs->link_init(link); - link->hpd_gpio = get_hpd_gpio(link->ctx->dc_bios, link->link_id, link->ctx->gpio_service); - if (link->hpd_gpio != NULL) { + link->hpd_gpio = get_hpd_gpio(link->ctx->dc_bios, link->link_id, + link->ctx->gpio_service); + if (link->hpd_gpio) { dal_gpio_open(link->hpd_gpio, GPIO_MODE_INTERRUPT); dal_gpio_unlock_pin(link->hpd_gpio); link->irq_source_hpd = dal_irq_get_source(link->hpd_gpio); @@ -1314,9 +1415,9 @@ static bool dc_link_construct( link->connector_signal = SIGNAL_TYPE_DVI_DUAL_LINK; break; case CONNECTOR_ID_DISPLAY_PORT: - link->connector_signal = SIGNAL_TYPE_DISPLAY_PORT; + link->connector_signal = SIGNAL_TYPE_DISPLAY_PORT; - if (link->hpd_gpio != NULL) + if (link->hpd_gpio) link->irq_source_hpd_rx = dal_irq_get_rx_source(link->hpd_gpio); @@ -1324,42 +1425,60 @@ static bool dc_link_construct( case CONNECTOR_ID_EDP: link->connector_signal = SIGNAL_TYPE_EDP; - if (link->hpd_gpio != NULL) { + if (link->hpd_gpio) { link->irq_source_hpd = DC_IRQ_SOURCE_INVALID; link->irq_source_hpd_rx = dal_irq_get_rx_source(link->hpd_gpio); } + break; case CONNECTOR_ID_LVDS: link->connector_signal = SIGNAL_TYPE_LVDS; break; default: - DC_LOG_WARNING("Unsupported Connector type:%d!\n", link->link_id.id); + DC_LOG_WARNING("Unsupported Connector type:%d!\n", + link->link_id.id); goto create_fail; } /* TODO: #DAL3 Implement id to str function.*/ LINK_INFO("Connector[%d] description:" - "signal %d\n", - init_params->connector_index, - link->connector_signal); + "signal %d\n", + init_params->connector_index, + link->connector_signal); ddc_service_init_data.ctx = link->ctx; ddc_service_init_data.id = link->link_id; ddc_service_init_data.link = link; link->ddc = dal_ddc_service_create(&ddc_service_init_data); - if (link->ddc == NULL) { + if (!link->ddc) { DC_ERROR("Failed to create ddc_service!\n"); goto ddc_create_fail; } link->ddc_hw_inst = - dal_ddc_get_line( - dal_ddc_service_get_ddc_pin(link->ddc)); + dal_ddc_get_line(dal_ddc_service_get_ddc_pin(link->ddc)); + + + if (link->dc->res_pool->funcs->panel_cntl_create && + (link->link_id.id == CONNECTOR_ID_EDP || + link->link_id.id == CONNECTOR_ID_LVDS)) { + panel_cntl_init_data.ctx = dc_ctx; + panel_cntl_init_data.inst = 0; + link->panel_cntl = + link->dc->res_pool->funcs->panel_cntl_create( + &panel_cntl_init_data); + + if (link->panel_cntl == NULL) { + DC_ERROR("Failed to create link panel_cntl!\n"); + goto panel_cntl_create_fail; + } + } enc_init_data.ctx = dc_ctx; - bp_funcs->get_src_obj(dc_ctx->dc_bios, link->link_id, 0, &enc_init_data.encoder); + bp_funcs->get_src_obj(dc_ctx->dc_bios, link->link_id, 0, + &enc_init_data.encoder); enc_init_data.connector = link->link_id; enc_init_data.channel = get_ddc_line(link); enc_init_data.hpd_source = get_hpd_line(link); @@ -1367,11 +1486,11 @@ static bool dc_link_construct( link->hpd_src = enc_init_data.hpd_source; enc_init_data.transmitter = - translate_encoder_to_transmitter(enc_init_data.encoder); - link->link_enc = link->dc->res_pool->funcs->link_enc_create( - &enc_init_data); + translate_encoder_to_transmitter(enc_init_data.encoder); + link->link_enc = + link->dc->res_pool->funcs->link_enc_create(&enc_init_data); - if (link->link_enc == NULL) { + if (!link->link_enc) { DC_ERROR("Failed to create link encoder!\n"); goto link_enc_create_fail; } @@ -1379,8 +1498,9 @@ static bool dc_link_construct( link->link_enc_hw_inst = link->link_enc->transmitter; for (i = 0; i < 4; i++) { - if (BP_RESULT_OK != - bp_funcs->get_device_tag(dc_ctx->dc_bios, link->link_id, i, &link->device_tag)) { + if (bp_funcs->get_device_tag(dc_ctx->dc_bios, + link->link_id, i, + &link->device_tag) != BP_RESULT_OK) { DC_ERROR("Failed to find device tag!\n"); goto device_tag_fail; } @@ -1388,13 +1508,14 @@ static bool dc_link_construct( /* Look for device tag that matches connector signal, * CRT for rgb, LCD for other supported signal tyes */ - if (!bp_funcs->is_device_id_supported(dc_ctx->dc_bios, link->device_tag.dev_id)) + if (!bp_funcs->is_device_id_supported(dc_ctx->dc_bios, + link->device_tag.dev_id)) continue; - if (link->device_tag.dev_id.device_type == DEVICE_TYPE_CRT - && link->connector_signal != SIGNAL_TYPE_RGB) + if (link->device_tag.dev_id.device_type == DEVICE_TYPE_CRT && + link->connector_signal != SIGNAL_TYPE_RGB) continue; - if (link->device_tag.dev_id.device_type == DEVICE_TYPE_LCD - && link->connector_signal == SIGNAL_TYPE_RGB) + if (link->device_tag.dev_id.device_type == DEVICE_TYPE_LCD && + link->connector_signal == SIGNAL_TYPE_RGB) continue; break; } @@ -1406,16 +1527,16 @@ static bool dc_link_construct( for (i = 0; i < MAX_NUMBER_OF_EXT_DISPLAY_PATH; i++) { struct external_display_path *path = &info.ext_disp_conn_info.path[i]; - if (path->device_connector_id.enum_id == link->link_id.enum_id - && path->device_connector_id.id == link->link_id.id - && path->device_connector_id.type == link->link_id.type) { - if (link->device_tag.acpi_device != 0 - && path->device_acpi_enum == link->device_tag.acpi_device) { + if (path->device_connector_id.enum_id == link->link_id.enum_id && + path->device_connector_id.id == link->link_id.id && + path->device_connector_id.type == link->link_id.type) { + if (link->device_tag.acpi_device != 0 && + path->device_acpi_enum == link->device_tag.acpi_device) { link->ddi_channel_mapping = path->channel_mapping; link->chip_caps = path->caps; } else if (path->device_tag == - link->device_tag.dev_id.raw_device_tag) { + link->device_tag.dev_id.raw_device_tag) { link->ddi_channel_mapping = path->channel_mapping; link->chip_caps = path->caps; } @@ -1431,15 +1552,20 @@ static bool dc_link_construct( */ program_hpd_filter(link); + link->psr_settings.psr_version = DC_PSR_VERSION_UNSUPPORTED; + return true; device_tag_fail: link->link_enc->funcs->destroy(&link->link_enc); link_enc_create_fail: + if (link->panel_cntl != NULL) + link->panel_cntl->funcs->destroy(&link->panel_cntl); +panel_cntl_create_fail: dal_ddc_service_destroy(&link->ddc); ddc_create_fail: create_fail: - if (link->hpd_gpio != NULL) { + if (link->hpd_gpio) { dal_gpio_destroy_irq(&link->hpd_gpio); link->hpd_gpio = NULL; } @@ -2339,9 +2465,28 @@ enum dc_status dc_link_validate_mode_timing( return DC_OK; } +static struct abm *get_abm_from_stream_res(const struct dc_link *link) +{ + int i; + struct dc *dc = link->ctx->dc; + struct abm *abm = NULL; + + for (i = 0; i < MAX_PIPES; i++) { + struct pipe_ctx pipe_ctx = dc->current_state->res_ctx.pipe_ctx[i]; + struct dc_stream_state *stream = pipe_ctx.stream; + + if (stream && stream->link == link) { + abm = pipe_ctx.stream_res.abm; + break; + } + } + return abm; +} + int dc_link_get_backlight_level(const struct dc_link *link) { - struct abm *abm = link->ctx->dc->res_pool->abm; + + struct abm *abm = get_abm_from_stream_res(link); if (abm == NULL || abm->funcs->get_current_backlight == NULL) return DC_ERROR_UNEXPECTED; @@ -2349,71 +2494,63 @@ int dc_link_get_backlight_level(const struct dc_link *link) return (int) abm->funcs->get_current_backlight(abm); } -bool dc_link_set_backlight_level(const struct dc_link *link, - uint32_t backlight_pwm_u16_16, - uint32_t frame_ramp) +int dc_link_get_target_backlight_pwm(const struct dc_link *link) { - struct dc *dc = link->ctx->dc; - struct abm *abm = dc->res_pool->abm; - struct dmcu *dmcu = dc->res_pool->dmcu; - unsigned int controller_id = 0; - bool use_smooth_brightness = true; - int i; - DC_LOGGER_INIT(link->ctx->logger); + struct abm *abm = get_abm_from_stream_res(link); - if ((dmcu == NULL) || - (abm == NULL) || - (abm->funcs->set_backlight_level_pwm == NULL)) - return false; + if (abm == NULL || abm->funcs->get_target_backlight == NULL) + return DC_ERROR_UNEXPECTED; - use_smooth_brightness = dmcu->funcs->is_dmcu_initialized(dmcu); + return (int) abm->funcs->get_target_backlight(abm); +} - DC_LOG_BACKLIGHT("New Backlight level: %d (0x%X)\n", - backlight_pwm_u16_16, backlight_pwm_u16_16); +static struct pipe_ctx *get_pipe_from_link(const struct dc_link *link) +{ + int i; + struct dc *dc = link->ctx->dc; + struct pipe_ctx *pipe_ctx = NULL; - if (dc_is_embedded_signal(link->connector_signal)) { - for (i = 0; i < MAX_PIPES; i++) { - if (dc->current_state->res_ctx.pipe_ctx[i].stream) { - if (dc->current_state->res_ctx. - pipe_ctx[i].stream->link - == link) { - /* DMCU -1 for all controller id values, - * therefore +1 here - */ - controller_id = - dc->current_state-> - res_ctx.pipe_ctx[i].stream_res.tg->inst + - 1; - - /* Disable brightness ramping when the display is blanked - * as it can hang the DMCU - */ - if (dc->current_state->res_ctx.pipe_ctx[i].plane_state == NULL) - frame_ramp = 0; - } + for (i = 0; i < MAX_PIPES; i++) { + if (dc->current_state->res_ctx.pipe_ctx[i].stream) { + if (dc->current_state->res_ctx.pipe_ctx[i].stream->link == link) { + pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i]; + break; } } - abm->funcs->set_backlight_level_pwm( - abm, - backlight_pwm_u16_16, - frame_ramp, - controller_id, - use_smooth_brightness); } - return true; + return pipe_ctx; } -bool dc_link_set_abm_disable(const struct dc_link *link) +bool dc_link_set_backlight_level(const struct dc_link *link, + uint32_t backlight_pwm_u16_16, + uint32_t frame_ramp) { struct dc *dc = link->ctx->dc; - struct abm *abm = dc->res_pool->abm; - if ((abm == NULL) || (abm->funcs->set_backlight_level_pwm == NULL)) - return false; + DC_LOGGER_INIT(link->ctx->logger); + DC_LOG_BACKLIGHT("New Backlight level: %d (0x%X)\n", + backlight_pwm_u16_16, backlight_pwm_u16_16); + + if (dc_is_embedded_signal(link->connector_signal)) { + struct pipe_ctx *pipe_ctx = get_pipe_from_link(link); - abm->funcs->set_abm_immediate_disable(abm); + if (pipe_ctx) { + /* Disable brightness ramping when the display is blanked + * as it can hang the DMCU + */ + if (pipe_ctx->plane_state == NULL) + frame_ramp = 0; + } else { + ASSERT(false); + return false; + } + dc->hwss.set_backlight_level( + pipe_ctx, + backlight_pwm_u16_16, + frame_ramp); + } return true; } @@ -2423,12 +2560,12 @@ bool dc_link_set_psr_allow_active(struct dc_link *link, bool allow_active, bool struct dmcu *dmcu = dc->res_pool->dmcu; struct dmub_psr *psr = dc->res_pool->psr; - if (psr != NULL && link->psr_feature_enabled) + if (psr != NULL && link->psr_settings.psr_feature_enabled) psr->funcs->psr_enable(psr, allow_active); - else if ((dmcu != NULL && dmcu->funcs->is_dmcu_initialized(dmcu)) && link->psr_feature_enabled) + else if ((dmcu != NULL && dmcu->funcs->is_dmcu_initialized(dmcu)) && link->psr_settings.psr_feature_enabled) dmcu->funcs->set_psr_enable(dmcu, allow_active, wait); - link->psr_allow_active = allow_active; + link->psr_settings.psr_allow_active = allow_active; return true; } @@ -2439,9 +2576,9 @@ bool dc_link_get_psr_state(const struct dc_link *link, uint32_t *psr_state) struct dmcu *dmcu = dc->res_pool->dmcu; struct dmub_psr *psr = dc->res_pool->psr; - if (psr != NULL && link->psr_feature_enabled) + if (psr != NULL && link->psr_settings.psr_feature_enabled) psr->funcs->psr_get_state(psr, psr_state); - else if (dmcu != NULL && link->psr_feature_enabled) + else if (dmcu != NULL && link->psr_settings.psr_feature_enabled) dmcu->funcs->get_psr_state(dmcu, psr_state); return true; @@ -2612,14 +2749,14 @@ bool dc_link_setup_psr(struct dc_link *link, psr_context->frame_delay = 0; if (psr) - link->psr_feature_enabled = psr->funcs->psr_copy_settings(psr, link, psr_context); + link->psr_settings.psr_feature_enabled = psr->funcs->psr_copy_settings(psr, link, psr_context); else - link->psr_feature_enabled = dmcu->funcs->setup_psr(dmcu, link, psr_context); + link->psr_settings.psr_feature_enabled = dmcu->funcs->setup_psr(dmcu, link, psr_context); /* psr_enabled == 0 indicates setup_psr did not succeed, but this * should not happen since firmware should be running at this point */ - if (link->psr_feature_enabled == 0) + if (link->psr_settings.psr_feature_enabled == 0) ASSERT(0); return true; @@ -2966,7 +3103,7 @@ void core_link_enable_stream( enum dc_status status; DC_LOGGER_INIT(pipe_ctx->stream->ctx->logger); - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) && + if (!IS_DIAG_DC(dc->ctx->dce_environment) && dc_is_virtual_signal(pipe_ctx->stream->signal)) return; @@ -3040,6 +3177,18 @@ void core_link_enable_stream( if (pipe_ctx->stream->dpms_off) return; + /* Have to setup DSC before DIG FE and BE are connected (which happens before the + * link training). This is to make sure the bandwidth sent to DIG BE won't be + * bigger than what the link and/or DIG BE can handle. VBID[6]/CompressedStream_flag + * will be automatically set at a later time when the video is enabled + * (DP_VID_STREAM_EN = 1). + */ + if (pipe_ctx->stream->timing.flags.DSC) { + if (dc_is_dp_signal(pipe_ctx->stream->signal) || + dc_is_virtual_signal(pipe_ctx->stream->signal)) + dp_set_dsc_enable(pipe_ctx, true); + } + status = enable_link(state, pipe_ctx); if (status != DC_OK) { @@ -3067,11 +3216,6 @@ void core_link_enable_stream( CONTROLLER_DP_TEST_PATTERN_VIDEOMODE, COLOR_DEPTH_UNDEFINED); - if (pipe_ctx->stream->timing.flags.DSC) { - if (dc_is_dp_signal(pipe_ctx->stream->signal) || - dc_is_virtual_signal(pipe_ctx->stream->signal)) - dp_set_dsc_enable(pipe_ctx, true); - } dc->hwss.enable_stream(pipe_ctx); /* Set DPS PPS SDP (AKA "info frames") */ @@ -3101,6 +3245,10 @@ void core_link_enable_stream( dp_set_dsc_enable(pipe_ctx, true); } + + if (pipe_ctx->stream->signal == SIGNAL_TYPE_HDMI_TYPE_A) { + core_link_set_avmute(pipe_ctx, false); + } } void core_link_disable_stream(struct pipe_ctx *pipe_ctx) @@ -3109,10 +3257,14 @@ void core_link_disable_stream(struct pipe_ctx *pipe_ctx) struct dc_stream_state *stream = pipe_ctx->stream; struct dc_link *link = stream->sink->link; - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) && + if (!IS_DIAG_DC(dc->ctx->dce_environment) && dc_is_virtual_signal(pipe_ctx->stream->signal)) return; + if (pipe_ctx->stream->signal == SIGNAL_TYPE_HDMI_TYPE_A) { + core_link_set_avmute(pipe_ctx, true); + } + #if defined(CONFIG_DRM_AMD_DC_HDCP) update_psp_stream_config(pipe_ctx, true); #endif diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c index 256889eed93e..aefd29a440b5 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c @@ -599,7 +599,7 @@ bool dal_ddc_submit_aux_command(struct ddc_service *ddc, do { struct aux_payload current_payload; bool is_end_of_payload = (retrieved + DEFAULT_AUX_MAX_DATA_SIZE) > - payload->length ? true : false; + payload->length; current_payload.address = payload->address; current_payload.data = &payload->data[retrieved]; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c index aa3c45a69b5e..91cd884d6f25 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c @@ -13,7 +13,6 @@ #include "core_status.h" #include "dpcd_defs.h" -#include "resource.h" #define DC_LOGGER \ link->ctx->logger @@ -220,6 +219,30 @@ static enum dpcd_training_patterns return dpcd_tr_pattern; } +static uint8_t dc_dp_initialize_scrambling_data_symbols( + struct dc_link *link, + enum dc_dp_training_pattern pattern) +{ + uint8_t disable_scrabled_data_symbols = 0; + + switch (pattern) { + case DP_TRAINING_PATTERN_SEQUENCE_1: + case DP_TRAINING_PATTERN_SEQUENCE_2: + case DP_TRAINING_PATTERN_SEQUENCE_3: + disable_scrabled_data_symbols = 1; + break; + case DP_TRAINING_PATTERN_SEQUENCE_4: + disable_scrabled_data_symbols = 0; + break; + default: + ASSERT(0); + DC_LOG_HW_LINK_TRAINING("%s: Invalid HW Training pattern: %d\n", + __func__, pattern); + break; + } + return disable_scrabled_data_symbols; +} + static inline bool is_repeater(struct dc_link *link, uint32_t offset) { return (!link->is_lttpr_mode_transparent && offset != 0); @@ -252,6 +275,9 @@ static void dpcd_set_lt_pattern_and_lane_settings( dpcd_pattern.v1_4.TRAINING_PATTERN_SET = dc_dp_training_pattern_to_dpcd_training_pattern(link, pattern); + dpcd_pattern.v1_4.SCRAMBLING_DISABLE = + dc_dp_initialize_scrambling_data_symbols(link, pattern); + dpcd_lt_buffer[DP_TRAINING_PATTERN_SET - DP_TRAINING_PATTERN_SET] = dpcd_pattern.raw; @@ -1710,19 +1736,10 @@ bool dc_link_dp_sync_lt_end(struct dc_link *link, bool link_down) static struct dc_link_settings get_max_link_cap(struct dc_link *link) { - /* Set Default link settings */ - struct dc_link_settings max_link_cap = {LANE_COUNT_FOUR, LINK_RATE_HIGH, - LINK_SPREAD_05_DOWNSPREAD_30KHZ, false, 0}; - - /* Higher link settings based on feature supported */ - if (link->link_enc->features.flags.bits.IS_HBR2_CAPABLE) - max_link_cap.link_rate = LINK_RATE_HIGH2; - - if (link->link_enc->features.flags.bits.IS_HBR3_CAPABLE) - max_link_cap.link_rate = LINK_RATE_HIGH3; + struct dc_link_settings max_link_cap = {0}; - if (link->link_enc->funcs->get_max_link_cap) - link->link_enc->funcs->get_max_link_cap(link->link_enc, &max_link_cap); + /* get max link encoder capability */ + link->link_enc->funcs->get_max_link_cap(link->link_enc, &max_link_cap); /* Lower link settings based on sink's link cap */ if (link->reported_link_cap.lane_count < max_link_cap.lane_count) @@ -2426,7 +2443,7 @@ static bool handle_hpd_irq_psr_sink(struct dc_link *link) { union dpcd_psr_configuration psr_configuration; - if (!link->psr_feature_enabled) + if (!link->psr_settings.psr_feature_enabled) return false; dm_helpers_dp_read_dpcd( @@ -2911,6 +2928,12 @@ bool dc_link_handle_hpd_rx_irq(struct dc_link *link, union hpd_irq_data *out_hpd for (i = 0; i < MAX_PIPES; i++) { pipe_ctx = &link->dc->current_state->res_ctx.pipe_ctx[i]; if (pipe_ctx && pipe_ctx->stream && pipe_ctx->stream->link == link) + link->dc->hwss.blank_stream(pipe_ctx); + } + + for (i = 0; i < MAX_PIPES; i++) { + pipe_ctx = &link->dc->current_state->res_ctx.pipe_ctx[i]; + if (pipe_ctx && pipe_ctx->stream && pipe_ctx->stream->link == link) break; } @@ -2927,6 +2950,12 @@ bool dc_link_handle_hpd_rx_irq(struct dc_link *link, union hpd_irq_data *out_hpd if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) dc_link_reallocate_mst_payload(link); + for (i = 0; i < MAX_PIPES; i++) { + pipe_ctx = &link->dc->current_state->res_ctx.pipe_ctx[i]; + if (pipe_ctx && pipe_ctx->stream && pipe_ctx->stream->link == link) + link->dc->hwss.unblank_stream(pipe_ctx, &previous_link_settings); + } + status = false; if (out_link_loss) *out_link_loss = true; @@ -4227,6 +4256,21 @@ void dp_set_fec_enable(struct dc_link *link, bool enable) void dpcd_set_source_specific_data(struct dc_link *link) { const uint32_t post_oui_delay = 30; // 30ms + uint8_t dspc = 0; + enum dc_status ret; + + ret = core_link_read_dpcd(link, DP_DOWN_STREAM_PORT_COUNT, &dspc, + sizeof(dspc)); + + if (ret != DC_OK) { + DC_LOG_ERROR("Error in DP aux read transaction," + " not writing source specific data\n"); + return; + } + + /* Return if OUI unsupported */ + if (!(dspc & DP_OUI_SUPPORT)) + return; if (!link->dc->vendor_signature.is_valid) { struct dpcd_amd_signature amd_signature; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c index 51e0ee6e7695..6590f51caefa 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c @@ -400,7 +400,7 @@ static bool dp_set_dsc_on_rx(struct pipe_ctx *pipe_ctx, bool enable) struct dc_stream_state *stream = pipe_ctx->stream; bool result = false; - if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) + if (dc_is_virtual_signal(stream->signal) || IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) result = true; else result = dm_helpers_dp_write_dsc_enable(dc->ctx, stream, enable); diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c index f4bcc71b2920..0c5619364e7d 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -532,6 +532,24 @@ static inline void get_vp_scan_direction( *flip_horz_scan_dir = !*flip_horz_scan_dir; } +int get_num_mpc_splits(struct pipe_ctx *pipe) +{ + int mpc_split_count = 0; + struct pipe_ctx *other_pipe = pipe->bottom_pipe; + + while (other_pipe && other_pipe->plane_state == pipe->plane_state) { + mpc_split_count++; + other_pipe = other_pipe->bottom_pipe; + } + other_pipe = pipe->top_pipe; + while (other_pipe && other_pipe->plane_state == pipe->plane_state) { + mpc_split_count++; + other_pipe = other_pipe->top_pipe; + } + + return mpc_split_count; +} + int get_num_odm_splits(struct pipe_ctx *pipe) { int odm_split_count = 0; @@ -556,16 +574,11 @@ static void calculate_split_count_and_index(struct pipe_ctx *pipe_ctx, int *spli /*Check for mpc split*/ struct pipe_ctx *split_pipe = pipe_ctx->top_pipe; + *split_count = get_num_mpc_splits(pipe_ctx); while (split_pipe && split_pipe->plane_state == pipe_ctx->plane_state) { (*split_idx)++; - (*split_count)++; split_pipe = split_pipe->top_pipe; } - split_pipe = pipe_ctx->bottom_pipe; - while (split_pipe && split_pipe->plane_state == pipe_ctx->plane_state) { - (*split_count)++; - split_pipe = split_pipe->bottom_pipe; - } } else { /*Get odm split index*/ struct pipe_ctx *split_pipe = pipe_ctx->prev_odm_pipe; @@ -692,6 +705,9 @@ static void calculate_viewport(struct pipe_ctx *pipe_ctx) /* Round up, assume original video size always even dimensions */ data->viewport_c.width = (data->viewport.width + vpc_div - 1) / vpc_div; data->viewport_c.height = (data->viewport.height + vpc_div - 1) / vpc_div; + + data->viewport_unadjusted = data->viewport; + data->viewport_c_unadjusted = data->viewport_c; } static void calculate_recout(struct pipe_ctx *pipe_ctx) @@ -1061,8 +1077,8 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) calculate_viewport(pipe_ctx); - if (pipe_ctx->plane_res.scl_data.viewport.height < 16 || - pipe_ctx->plane_res.scl_data.viewport.width < 16) { + if (pipe_ctx->plane_res.scl_data.viewport.height < 12 || + pipe_ctx->plane_res.scl_data.viewport.width < 12) { if (store_h_border_left) { restore_border_left_from_dst(pipe_ctx, store_h_border_left); @@ -1358,9 +1374,6 @@ bool dc_add_plane_to_context( dc_plane_state_retain(plane_state); while (head_pipe) { - tail_pipe = resource_get_tail_pipe(&context->res_ctx, head_pipe); - ASSERT(tail_pipe); - free_pipe = acquire_free_pipe_for_head(context, pool, head_pipe); #if defined(CONFIG_DRM_AMD_DC_DCN) @@ -1378,6 +1391,8 @@ bool dc_add_plane_to_context( free_pipe->plane_state = plane_state; if (head_pipe != free_pipe) { + tail_pipe = resource_get_tail_pipe(&context->res_ctx, head_pipe); + ASSERT(tail_pipe); free_pipe->stream_res.tg = tail_pipe->stream_res.tg; free_pipe->stream_res.abm = tail_pipe->stream_res.abm; free_pipe->stream_res.opp = tail_pipe->stream_res.opp; @@ -1545,35 +1560,6 @@ bool dc_add_all_planes_for_stream( return add_all_planes_for_stream(dc, stream, &set, 1, context); } - -static bool is_hdr_static_meta_changed(struct dc_stream_state *cur_stream, - struct dc_stream_state *new_stream) -{ - if (cur_stream == NULL) - return true; - - if (memcmp(&cur_stream->hdr_static_metadata, - &new_stream->hdr_static_metadata, - sizeof(struct dc_info_packet)) != 0) - return true; - - return false; -} - -static bool is_vsc_info_packet_changed(struct dc_stream_state *cur_stream, - struct dc_stream_state *new_stream) -{ - if (cur_stream == NULL) - return true; - - if (memcmp(&cur_stream->vsc_infopacket, - &new_stream->vsc_infopacket, - sizeof(struct dc_info_packet)) != 0) - return true; - - return false; -} - static bool is_timing_changed(struct dc_stream_state *cur_stream, struct dc_stream_state *new_stream) { @@ -1608,15 +1594,9 @@ static bool are_stream_backends_same( if (is_timing_changed(stream_a, stream_b)) return false; - if (is_hdr_static_meta_changed(stream_a, stream_b)) - return false; - if (stream_a->dpms_off != stream_b->dpms_off) return false; - if (is_vsc_info_packet_changed(stream_a, stream_b)) - return false; - return true; } @@ -1756,21 +1736,6 @@ static struct audio *find_first_free_audio( return 0; } -bool resource_is_stream_unchanged( - struct dc_state *old_context, struct dc_stream_state *stream) -{ - int i; - - for (i = 0; i < old_context->stream_count; i++) { - struct dc_stream_state *old_stream = old_context->streams[i]; - - if (are_stream_backends_same(old_stream, stream)) - return true; - } - - return false; -} - /** * dc_add_stream_to_ctx() - Add a new dc_stream_state to a dc_state. */ @@ -2025,17 +1990,6 @@ enum dc_status resource_map_pool_resources( int pipe_idx = -1; struct dc_bios *dcb = dc->ctx->dc_bios; - /* TODO Check if this is needed */ - /*if (!resource_is_stream_unchanged(old_context, stream)) { - if (stream != NULL && old_context->streams[i] != NULL) { - stream->bit_depth_params = - old_context->streams[i]->bit_depth_params; - stream->clamping = old_context->streams[i]->clamping; - continue; - } - } - */ - calculate_phy_pix_clks(stream); /* TODO: Check Linux */ @@ -2718,19 +2672,16 @@ bool pipe_need_reprogram( if (is_timing_changed(pipe_ctx_old->stream, pipe_ctx->stream)) return true; - if (is_hdr_static_meta_changed(pipe_ctx_old->stream, pipe_ctx->stream)) - return true; - if (pipe_ctx_old->stream->dpms_off != pipe_ctx->stream->dpms_off) return true; - if (is_vsc_info_packet_changed(pipe_ctx_old->stream, pipe_ctx->stream)) - return true; - if (false == pipe_ctx_old->stream->link->link_state_valid && false == pipe_ctx_old->stream->dpms_off) return true; + if (pipe_ctx_old->stream_res.dsc != pipe_ctx->stream_res.dsc) + return true; + return false; } diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_sink.c b/drivers/gpu/drm/amd/display/dc/core/dc_sink.c index a249a0e5edd0..9e16af22e4aa 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_sink.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_sink.c @@ -54,6 +54,7 @@ static bool dc_sink_construct(struct dc_sink *sink, const struct dc_sink_init_da sink->ctx = link->ctx; sink->dongle_max_pix_clk = init_params->dongle_max_pix_clk; sink->converter_disable_audio = init_params->converter_disable_audio; + sink->is_mst_legacy = init_params->sink_is_legacy; sink->dc_container_id = NULL; sink->sink_id = init_params->link->ctx->dc_sink_id_count; // increment dc_sink_id_count because we don't want two sinks with same ID diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c index 6ddbb00ed37a..4f0e7203dba4 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c @@ -231,34 +231,6 @@ struct dc_stream_status *dc_stream_get_status( return dc_stream_get_status_from_state(dc->current_state, stream); } -static void delay_cursor_until_vupdate(struct pipe_ctx *pipe_ctx, struct dc *dc) -{ -#if defined(CONFIG_DRM_AMD_DC_DCN) - unsigned int vupdate_line; - unsigned int lines_to_vupdate, us_to_vupdate, vpos, nvpos; - struct dc_stream_state *stream = pipe_ctx->stream; - unsigned int us_per_line; - - if (stream->ctx->asic_id.chip_family == FAMILY_RV && - ASICREV_IS_RAVEN(stream->ctx->asic_id.hw_internal_rev)) { - - vupdate_line = dc->hwss.get_vupdate_offset_from_vsync(pipe_ctx); - if (!dc_stream_get_crtc_position(dc, &stream, 1, &vpos, &nvpos)) - return; - - if (vpos >= vupdate_line) - return; - - us_per_line = stream->timing.h_total * 10000 / stream->timing.pix_clk_100hz; - lines_to_vupdate = vupdate_line - vpos; - us_to_vupdate = lines_to_vupdate * us_per_line; - - /* 70 us is a conservative estimate of cursor update time*/ - if (us_to_vupdate < 70) - udelay(us_to_vupdate); - } -#endif -} /** * dc_stream_set_cursor_attributes() - Update cursor attributes and set cursor surface address @@ -298,9 +270,7 @@ bool dc_stream_set_cursor_attributes( if (!pipe_to_program) { pipe_to_program = pipe_ctx; - - delay_cursor_until_vupdate(pipe_ctx, dc); - dc->hwss.pipe_control_lock(dc, pipe_to_program, true); + dc->hwss.cursor_lock(dc, pipe_to_program, true); } dc->hwss.set_cursor_attribute(pipe_ctx); @@ -309,7 +279,7 @@ bool dc_stream_set_cursor_attributes( } if (pipe_to_program) - dc->hwss.pipe_control_lock(dc, pipe_to_program, false); + dc->hwss.cursor_lock(dc, pipe_to_program, false); return true; } @@ -349,16 +319,14 @@ bool dc_stream_set_cursor_position( if (!pipe_to_program) { pipe_to_program = pipe_ctx; - - delay_cursor_until_vupdate(pipe_ctx, dc); - dc->hwss.pipe_control_lock(dc, pipe_to_program, true); + dc->hwss.cursor_lock(dc, pipe_to_program, true); } dc->hwss.set_cursor_position(pipe_ctx); } if (pipe_to_program) - dc->hwss.pipe_control_lock(dc, pipe_to_program, false); + dc->hwss.cursor_lock(dc, pipe_to_program, false); return true; } diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index 1935cf6601eb..85908561c741 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -29,6 +29,9 @@ #include "dc_types.h" #include "grph_object_defs.h" #include "logger_types.h" +#if defined(CONFIG_DRM_AMD_DC_HDCP) +#include "hdcp_types.h" +#endif #include "gpio_types.h" #include "link_service_types.h" #include "grph_object_ctrl_defs.h" @@ -39,7 +42,7 @@ #include "inc/hw/dmcu.h" #include "dml/display_mode_lib.h" -#define DC_VER "3.2.76" +#define DC_VER "3.2.84" #define MAX_SURFACES 3 #define MAX_PLANES 6 @@ -95,6 +98,49 @@ struct dc_plane_cap { } max_downscale_factor; }; +// Color management caps (DPP and MPC) +struct rom_curve_caps { + uint16_t srgb : 1; + uint16_t bt2020 : 1; + uint16_t gamma2_2 : 1; + uint16_t pq : 1; + uint16_t hlg : 1; +}; + +struct dpp_color_caps { + uint16_t dcn_arch : 1; // all DCE generations treated the same + // input lut is different than most LUTs, just plain 256-entry lookup + uint16_t input_lut_shared : 1; // shared with DGAM + uint16_t icsc : 1; + uint16_t dgam_ram : 1; + uint16_t post_csc : 1; // before gamut remap + uint16_t gamma_corr : 1; + + // hdr_mult and gamut remap always available in DPP (in that order) + // 3d lut implies shaper LUT, + // it may be shared with MPC - check MPC:shared_3d_lut flag + uint16_t hw_3d_lut : 1; + uint16_t ogam_ram : 1; // blnd gam + uint16_t ocsc : 1; + struct rom_curve_caps dgam_rom_caps; + struct rom_curve_caps ogam_rom_caps; +}; + +struct mpc_color_caps { + uint16_t gamut_remap : 1; + uint16_t ogam_ram : 1; + uint16_t ocsc : 1; + uint16_t num_3dluts : 3; //3d lut always assumes a preceding shaper LUT + uint16_t shared_3d_lut:1; //can be in either DPP or MPC, but single instance + + struct rom_curve_caps ogam_rom_caps; +}; + +struct dc_color_caps { + struct dpp_color_caps dpp; + struct mpc_color_caps mpc; +}; + struct dc_caps { uint32_t max_streams; uint32_t max_links; @@ -117,9 +163,9 @@ struct dc_caps { bool psp_setup_panel_mode; bool extended_aux_timeout_support; bool dmcub_support; - bool hw_3d_lut; enum dp_protocol_version max_dp_protocol_version; struct dc_plane_cap planes[MAX_PLANES]; + struct dc_color_caps color; }; struct dc_bug_wa { @@ -230,7 +276,8 @@ struct dc_config { bool forced_clocks; bool disable_extended_timeout_support; // Used to disable extended timeout and lttpr feature as well bool multi_mon_pp_mclk_switch; - bool psr_on_dmub; + bool disable_dmcu; + bool enable_4to1MPC; }; enum visual_confirm { @@ -238,6 +285,7 @@ enum visual_confirm { VISUAL_CONFIRM_SURFACE = 1, VISUAL_CONFIRM_HDR = 2, VISUAL_CONFIRM_MPCTREE = 4, + VISUAL_CONFIRM_PSR = 5, }; enum dcc_option { @@ -429,6 +477,7 @@ struct dc_debug_options { bool enable_dmcub_surface_flip; bool usbc_combo_phy_reset_wa; bool disable_dsc; + bool enable_dram_clock_change_one_display_vactive; }; struct dc_debug_data { @@ -474,6 +523,7 @@ struct dc_bounding_box_overrides { int urgent_latency_ns; int percent_of_ideal_drambw; int dram_clock_change_latency_ns; + int dummy_clock_change_latency_ns; /* This forces a hard min on the DCFCLK we use * for DML. Unlike the debug option for forcing * DCFCLK, this override affects watermark calculations @@ -987,6 +1037,7 @@ struct dpcd_caps { union dpcd_fec_capability fec_cap; struct dpcd_dsc_capabilities dsc_caps; struct dc_lttpr_caps lttpr_caps; + struct psr_caps psr_caps; }; @@ -1004,6 +1055,35 @@ union dpcd_sink_ext_caps { uint8_t raw; }; +#if defined(CONFIG_DRM_AMD_DC_HDCP) +union hdcp_rx_caps { + struct { + uint8_t version; + uint8_t reserved; + struct { + uint8_t repeater : 1; + uint8_t hdcp_capable : 1; + uint8_t reserved : 6; + } byte0; + } fields; + uint8_t raw[3]; +}; + +union hdcp_bcaps { + struct { + uint8_t HDCP_CAPABLE:1; + uint8_t REPEATER:1; + uint8_t RESERVED:6; + } bits; + uint8_t raw; +}; + +struct hdcp_caps { + union hdcp_rx_caps rx_caps; + union hdcp_bcaps bcaps; +}; +#endif + #include "dc_link.h" /******************************************************************************* @@ -1046,7 +1126,7 @@ struct dc_sink { void *priv; struct stereo_3d_features features_3d[TIMING_3D_FORMAT_MAX]; bool converter_disable_audio; - + bool is_mst_legacy; struct dc_sink_dsc_caps dsc_caps; struct dc_sink_fec_caps fec_caps; @@ -1073,6 +1153,7 @@ struct dc_sink_init_data { struct dc_link *link; uint32_t dongle_max_pix_clk; bool converter_disable_audio; + bool sink_is_legacy; }; struct dc_sink *dc_sink_create(const struct dc_sink_init_data *init_params); @@ -1104,9 +1185,16 @@ void dc_set_power_state( struct dc *dc, enum dc_acpi_cm_power_state power_state); void dc_resume(struct dc *dc); -unsigned int dc_get_current_backlight_pwm(struct dc *dc); -unsigned int dc_get_target_backlight_pwm(struct dc *dc); +#if defined(CONFIG_DRM_AMD_DC_HDCP) +/* + * HDCP Interfaces + */ +enum hdcp_message_status dc_process_hdcp_msg( + enum signal_type signal, + struct dc_link *link, + struct hdcp_protection_message *message_info); +#endif bool dc_is_dmcu_initialized(struct dc *dc); enum dc_status dc_set_clock(struct dc *dc, enum dc_clock_type clock_type, uint32_t clk_khz, uint32_t stepping); diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c index 59c298a6484f..eea2429ac67d 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c @@ -25,7 +25,7 @@ #include "dc.h" #include "dc_dmub_srv.h" -#include "../dmub/inc/dmub_srv.h" +#include "../dmub/dmub_srv.h" static void dc_dmub_srv_construct(struct dc_dmub_srv *dc_srv, struct dc *dc, struct dmub_srv *dmub) @@ -58,7 +58,7 @@ void dc_dmub_srv_destroy(struct dc_dmub_srv **dmub_srv) } void dc_dmub_srv_cmd_queue(struct dc_dmub_srv *dc_dmub_srv, - struct dmub_cmd_header *cmd) + union dmub_rb_cmd *cmd) { struct dmub_srv *dmub = dc_dmub_srv->dmub; struct dc_context *dc_ctx = dc_dmub_srv->ctx; diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h index 754b6077539c..a3a09ccb6d26 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h @@ -27,10 +27,9 @@ #define _DMUB_DC_SRV_H_ #include "os_types.h" -#include "../dmub/inc/dmub_cmd.h" +#include "dmub/dmub_srv.h" struct dmub_srv; -struct dmub_cmd_header; struct dc_reg_helper_state { bool gather_in_progress; @@ -49,7 +48,7 @@ struct dc_dmub_srv { }; void dc_dmub_srv_cmd_queue(struct dc_dmub_srv *dc_dmub_srv, - struct dmub_cmd_header *cmd); + union dmub_rb_cmd *cmd); void dc_dmub_srv_cmd_execute(struct dc_dmub_srv *dc_dmub_srv); diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h index bb2730e9521e..af177c087d3b 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h @@ -740,5 +740,11 @@ struct dpcd_dsc_capabilities { union dpcd_dsc_ext_capabilities dsc_ext_caps; }; +/* These parameters are from PSR capabilities reported by Sink DPCD */ +struct psr_caps { + unsigned char psr_version; + unsigned int psr_rfb_setup_time; + bool psr_exit_link_training_required; +}; #endif /* DC_DP_TYPES_H */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_helper.c b/drivers/gpu/drm/amd/display/dc/dc_helper.c index 737048d8a96c..85a0170be544 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_helper.c +++ b/drivers/gpu/drm/amd/display/dc/dc_helper.c @@ -50,7 +50,7 @@ static inline void submit_dmub_read_modify_write( gather = ctx->dmub_srv->reg_helper_offload.gather_in_progress; ctx->dmub_srv->reg_helper_offload.gather_in_progress = false; - dc_dmub_srv_cmd_queue(ctx->dmub_srv, &cmd_buf->header); + dc_dmub_srv_cmd_queue(ctx->dmub_srv, &offload->cmd_data); ctx->dmub_srv->reg_helper_offload.gather_in_progress = gather; @@ -73,7 +73,7 @@ static inline void submit_dmub_burst_write( gather = ctx->dmub_srv->reg_helper_offload.gather_in_progress; ctx->dmub_srv->reg_helper_offload.gather_in_progress = false; - dc_dmub_srv_cmd_queue(ctx->dmub_srv, &cmd_buf->header); + dc_dmub_srv_cmd_queue(ctx->dmub_srv, &offload->cmd_data); ctx->dmub_srv->reg_helper_offload.gather_in_progress = gather; @@ -92,7 +92,7 @@ static inline void submit_dmub_reg_wait( gather = ctx->dmub_srv->reg_helper_offload.gather_in_progress; ctx->dmub_srv->reg_helper_offload.gather_in_progress = false; - dc_dmub_srv_cmd_queue(ctx->dmub_srv, &cmd_buf->header); + dc_dmub_srv_cmd_queue(ctx->dmub_srv, &offload->cmd_data); memset(cmd_buf, 0, sizeof(*cmd_buf)); offload->reg_seq_count = 0; diff --git a/drivers/gpu/drm/amd/display/dc/dc_link.h b/drivers/gpu/drm/amd/display/dc/dc_link.h index 00ff5e98278c..f63fc25aa6c5 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_link.h +++ b/drivers/gpu/drm/amd/display/dc/dc_link.h @@ -66,6 +66,22 @@ struct time_stamp { struct link_trace { struct time_stamp time_stamp; }; + +/* PSR feature flags */ +struct psr_settings { + bool psr_feature_enabled; // PSR is supported by sink + bool psr_allow_active; // PSR is currently active + enum dc_psr_version psr_version; // Internal PSR version, determined based on DPCD + + /* These parameters are calculated in Driver, + * based on display timing and Sink capabilities. + * If VBLANK region is too small and Sink takes a long time + * to set up RFB, it may take an extra frame to enter PSR state. + */ + bool psr_frame_capture_indication_req; + unsigned int psr_sdp_transmit_line_num_deadline; +}; + /* * A link contains one or more sinks and their connected status. * The currently active signal type (HDMI, DP-SST, DP-MST) is also reported. @@ -118,6 +134,7 @@ struct dc_link { struct dc_context *ctx; + struct panel_cntl *panel_cntl; struct link_encoder *link_enc; struct graphics_object_id link_id; union ddi_channel_mapping ddi_channel_mapping; @@ -126,11 +143,14 @@ struct dc_link { uint32_t dongle_max_pix_clk; unsigned short chip_caps; unsigned int dpcd_sink_count; +#if defined(CONFIG_DRM_AMD_DC_HDCP) + struct hdcp_caps hdcp_caps; +#endif enum edp_revision edp_revision; - bool psr_feature_enabled; - bool psr_allow_active; union dpcd_sink_ext_caps dpcd_sink_ext_caps; + struct psr_settings psr_settings; + /* MST record stream using this link */ struct link_flags { bool dp_keep_receiver_powered; @@ -197,7 +217,7 @@ bool dc_link_set_default_brightness_aux(struct dc_link *link); int dc_link_get_backlight_level(const struct dc_link *dc_link); -bool dc_link_set_abm_disable(const struct dc_link *dc_link); +int dc_link_get_target_backlight_pwm(const struct dc_link *link); bool dc_link_set_psr_allow_active(struct dc_link *dc_link, bool enable, bool wait); @@ -290,6 +310,10 @@ bool dc_link_detect_sink(struct dc_link *link, enum dc_connection_type *type); * DPCD access interfaces */ +#ifdef CONFIG_DRM_AMD_DC_HDCP +bool dc_link_is_hdcp14(struct dc_link *link); +bool dc_link_is_hdcp22(struct dc_link *link); +#endif void dc_link_set_drive_settings(struct dc *dc, struct link_training_settings *lt_settings, const struct dc_link *link); diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h index a5c7ef47b8d3..49aad691e687 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_stream.h +++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h @@ -167,8 +167,6 @@ struct dc_stream_state { /* TODO: custom INFO packets */ /* TODO: ABM info (DMCU) */ - /* PSR info */ - unsigned char psr_version; /* TODO: CEA VIC */ /* DMCU info */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h index 0d210104ba0a..f236da1c1859 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_types.h @@ -862,4 +862,9 @@ struct dsc_dec_dpcd_caps { uint32_t branch_max_line_width; }; +enum dc_psr_version { + DC_PSR_VERSION_1 = 0, + DC_PSR_VERSION_UNSUPPORTED = 0xFFFFFFFF, +}; + #endif /* DC_TYPES_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dce/Makefile b/drivers/gpu/drm/amd/display/dc/dce/Makefile index fbfcff700971..f704a8fd52e8 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dce/Makefile @@ -29,7 +29,7 @@ DCE = dce_audio.o dce_stream_encoder.o dce_link_encoder.o dce_hwseq.o \ dce_mem_input.o dce_clock_source.o dce_scl_filters.o dce_transform.o \ dce_opp.o dce_dmcu.o dce_abm.o dce_ipp.o dce_aux.o \ -dce_i2c.o dce_i2c_hw.o dce_i2c_sw.o dmub_psr.o +dce_i2c.o dce_i2c_hw.o dce_i2c_sw.o dmub_psr.o dmub_abm.o dce_panel_cntl.o AMD_DAL_DCE = $(addprefix $(AMDDALPATH)/dc/dce/,$(DCE)) diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c index b8a3fc505c9b..4e87e70237e3 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c @@ -55,7 +55,7 @@ #define MCP_DISABLE_ABM_IMMEDIATELY 255 -static bool dce_abm_set_pipe(struct abm *abm, uint32_t controller_id) +static bool dce_abm_set_pipe(struct abm *abm, uint32_t controller_id, uint32_t panel_inst) { struct dce_abm *abm_dce = TO_DCE_ABM(abm); uint32_t rampingBoundary = 0xFFFF; @@ -83,125 +83,12 @@ static bool dce_abm_set_pipe(struct abm *abm, uint32_t controller_id) return true; } -static unsigned int calculate_16_bit_backlight_from_pwm(struct dce_abm *abm_dce) -{ - uint64_t current_backlight; - uint32_t round_result; - uint32_t pwm_period_cntl, bl_period, bl_int_count; - uint32_t bl_pwm_cntl, bl_pwm, fractional_duty_cycle_en; - uint32_t bl_period_mask, bl_pwm_mask; - - pwm_period_cntl = REG_READ(BL_PWM_PERIOD_CNTL); - REG_GET(BL_PWM_PERIOD_CNTL, BL_PWM_PERIOD, &bl_period); - REG_GET(BL_PWM_PERIOD_CNTL, BL_PWM_PERIOD_BITCNT, &bl_int_count); - - bl_pwm_cntl = REG_READ(BL_PWM_CNTL); - REG_GET(BL_PWM_CNTL, BL_ACTIVE_INT_FRAC_CNT, (uint32_t *)(&bl_pwm)); - REG_GET(BL_PWM_CNTL, BL_PWM_FRACTIONAL_EN, &fractional_duty_cycle_en); - - if (bl_int_count == 0) - bl_int_count = 16; - - bl_period_mask = (1 << bl_int_count) - 1; - bl_period &= bl_period_mask; - - bl_pwm_mask = bl_period_mask << (16 - bl_int_count); - - if (fractional_duty_cycle_en == 0) - bl_pwm &= bl_pwm_mask; - else - bl_pwm &= 0xFFFF; - - current_backlight = bl_pwm << (1 + bl_int_count); - - if (bl_period == 0) - bl_period = 0xFFFF; - - current_backlight = div_u64(current_backlight, bl_period); - current_backlight = (current_backlight + 1) >> 1; - - current_backlight = (uint64_t)(current_backlight) * bl_period; - - round_result = (uint32_t)(current_backlight & 0xFFFFFFFF); - - round_result = (round_result >> (bl_int_count-1)) & 1; - - current_backlight >>= bl_int_count; - current_backlight += round_result; - - return (uint32_t)(current_backlight); -} - -static void driver_set_backlight_level(struct dce_abm *abm_dce, - uint32_t backlight_pwm_u16_16) -{ - uint32_t backlight_16bit; - uint32_t masked_pwm_period; - uint8_t bit_count; - uint64_t active_duty_cycle; - uint32_t pwm_period_bitcnt; - - /* - * 1. Find 16 bit backlight active duty cycle, where 0 <= backlight - * active duty cycle <= backlight period - */ - - /* 1.1 Apply bitmask for backlight period value based on value of BITCNT - */ - REG_GET_2(BL_PWM_PERIOD_CNTL, - BL_PWM_PERIOD_BITCNT, &pwm_period_bitcnt, - BL_PWM_PERIOD, &masked_pwm_period); - - if (pwm_period_bitcnt == 0) - bit_count = 16; - else - bit_count = pwm_period_bitcnt; - - /* e.g. maskedPwmPeriod = 0x24 when bitCount is 6 */ - masked_pwm_period = masked_pwm_period & ((1 << bit_count) - 1); - - /* 1.2 Calculate integer active duty cycle required upper 16 bits - * contain integer component, lower 16 bits contain fractional component - * of active duty cycle e.g. 0x21BDC0 = 0xEFF0 * 0x24 - */ - active_duty_cycle = backlight_pwm_u16_16 * masked_pwm_period; - - /* 1.3 Calculate 16 bit active duty cycle from integer and fractional - * components shift by bitCount then mask 16 bits and add rounding bit - * from MSB of fraction e.g. 0x86F7 = ((0x21BDC0 >> 6) & 0xFFF) + 0 - */ - backlight_16bit = active_duty_cycle >> bit_count; - backlight_16bit &= 0xFFFF; - backlight_16bit += (active_duty_cycle >> (bit_count - 1)) & 0x1; - - /* - * 2. Program register with updated value - */ - - /* 2.1 Lock group 2 backlight registers */ - - REG_UPDATE_2(BL_PWM_GRP1_REG_LOCK, - BL_PWM_GRP1_IGNORE_MASTER_LOCK_EN, 1, - BL_PWM_GRP1_REG_LOCK, 1); - - // 2.2 Write new active duty cycle - REG_UPDATE(BL_PWM_CNTL, BL_ACTIVE_INT_FRAC_CNT, backlight_16bit); - - /* 2.3 Unlock group 2 backlight registers */ - REG_UPDATE(BL_PWM_GRP1_REG_LOCK, - BL_PWM_GRP1_REG_LOCK, 0); - - /* 3 Wait for pending bit to be cleared */ - REG_WAIT(BL_PWM_GRP1_REG_LOCK, - BL_PWM_GRP1_REG_UPDATE_PENDING, 0, - 1, 10000); -} - static void dmcu_set_backlight_level( struct dce_abm *abm_dce, uint32_t backlight_pwm_u16_16, uint32_t frame_ramp, - uint32_t controller_id) + uint32_t controller_id, + uint32_t panel_id) { unsigned int backlight_8_bit = 0; uint32_t s2; @@ -213,7 +100,7 @@ static void dmcu_set_backlight_level( // Take MSB of fractional part since backlight is not max backlight_8_bit = (backlight_pwm_u16_16 >> 8) & 0xFF; - dce_abm_set_pipe(&abm_dce->base, controller_id); + dce_abm_set_pipe(&abm_dce->base, controller_id, panel_id); /* waitDMCUReadyForCmd */ REG_WAIT(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, @@ -248,10 +135,9 @@ static void dmcu_set_backlight_level( 0, 1, 80000); } -static void dce_abm_init(struct abm *abm) +static void dce_abm_init(struct abm *abm, uint32_t backlight) { struct dce_abm *abm_dce = TO_DCE_ABM(abm); - unsigned int backlight = calculate_16_bit_backlight_from_pwm(abm_dce); REG_WRITE(DC_ABM1_HG_SAMPLE_RATE, 0x103); REG_WRITE(DC_ABM1_HG_SAMPLE_RATE, 0x101); @@ -331,86 +217,12 @@ static bool dce_abm_set_level(struct abm *abm, uint32_t level) return true; } -static bool dce_abm_immediate_disable(struct abm *abm) +static bool dce_abm_immediate_disable(struct abm *abm, uint32_t panel_inst) { - struct dce_abm *abm_dce = TO_DCE_ABM(abm); - if (abm->dmcu_is_running == false) return true; - dce_abm_set_pipe(abm, MCP_DISABLE_ABM_IMMEDIATELY); - - abm->stored_backlight_registers.BL_PWM_CNTL = - REG_READ(BL_PWM_CNTL); - abm->stored_backlight_registers.BL_PWM_CNTL2 = - REG_READ(BL_PWM_CNTL2); - abm->stored_backlight_registers.BL_PWM_PERIOD_CNTL = - REG_READ(BL_PWM_PERIOD_CNTL); - - REG_GET(LVTMA_PWRSEQ_REF_DIV, BL_PWM_REF_DIV, - &abm->stored_backlight_registers.LVTMA_PWRSEQ_REF_DIV_BL_PWM_REF_DIV); - return true; -} - -static bool dce_abm_init_backlight(struct abm *abm) -{ - struct dce_abm *abm_dce = TO_DCE_ABM(abm); - uint32_t value; - - /* It must not be 0, so we have to restore them - * Bios bug w/a - period resets to zero, - * restoring to cache values which is always correct - */ - REG_GET(BL_PWM_CNTL, BL_ACTIVE_INT_FRAC_CNT, &value); - if (value == 0 || value == 1) { - if (abm->stored_backlight_registers.BL_PWM_CNTL != 0) { - REG_WRITE(BL_PWM_CNTL, - abm->stored_backlight_registers.BL_PWM_CNTL); - REG_WRITE(BL_PWM_CNTL2, - abm->stored_backlight_registers.BL_PWM_CNTL2); - REG_WRITE(BL_PWM_PERIOD_CNTL, - abm->stored_backlight_registers.BL_PWM_PERIOD_CNTL); - REG_UPDATE(LVTMA_PWRSEQ_REF_DIV, - BL_PWM_REF_DIV, - abm->stored_backlight_registers. - LVTMA_PWRSEQ_REF_DIV_BL_PWM_REF_DIV); - } else { - /* TODO: Note: This should not really happen since VBIOS - * should have initialized PWM registers on boot. - */ - REG_WRITE(BL_PWM_CNTL, 0xC000FA00); - REG_WRITE(BL_PWM_PERIOD_CNTL, 0x000C0FA0); - } - } else { - abm->stored_backlight_registers.BL_PWM_CNTL = - REG_READ(BL_PWM_CNTL); - abm->stored_backlight_registers.BL_PWM_CNTL2 = - REG_READ(BL_PWM_CNTL2); - abm->stored_backlight_registers.BL_PWM_PERIOD_CNTL = - REG_READ(BL_PWM_PERIOD_CNTL); - - REG_GET(LVTMA_PWRSEQ_REF_DIV, BL_PWM_REF_DIV, - &abm->stored_backlight_registers. - LVTMA_PWRSEQ_REF_DIV_BL_PWM_REF_DIV); - } - - /* Have driver take backlight control - * TakeBacklightControl(true) - */ - value = REG_READ(BIOS_SCRATCH_2); - value |= ATOM_S2_VRI_BRIGHT_ENABLE; - REG_WRITE(BIOS_SCRATCH_2, value); - - /* Enable the backlight output */ - REG_UPDATE(BL_PWM_CNTL, BL_PWM_EN, 1); - - /* Disable fractional pwm if configured */ - REG_UPDATE(BL_PWM_CNTL, BL_PWM_FRACTIONAL_EN, - abm->ctx->dc->config.disable_fractional_pwm ? 0 : 1); - - /* Unlock group 2 backlight registers */ - REG_UPDATE(BL_PWM_GRP1_REG_LOCK, - BL_PWM_GRP1_REG_LOCK, 0); + dce_abm_set_pipe(abm, MCP_DISABLE_ABM_IMMEDIATELY, panel_inst); return true; } @@ -420,21 +232,18 @@ static bool dce_abm_set_backlight_level_pwm( unsigned int backlight_pwm_u16_16, unsigned int frame_ramp, unsigned int controller_id, - bool use_smooth_brightness) + unsigned int panel_inst) { struct dce_abm *abm_dce = TO_DCE_ABM(abm); DC_LOG_BACKLIGHT("New Backlight level: %d (0x%X)\n", backlight_pwm_u16_16, backlight_pwm_u16_16); - /* If DMCU is in reset state, DMCU is uninitialized */ - if (use_smooth_brightness) - dmcu_set_backlight_level(abm_dce, - backlight_pwm_u16_16, - frame_ramp, - controller_id); - else - driver_set_backlight_level(abm_dce, backlight_pwm_u16_16); + dmcu_set_backlight_level(abm_dce, + backlight_pwm_u16_16, + frame_ramp, + controller_id, + panel_inst); return true; } @@ -442,12 +251,12 @@ static bool dce_abm_set_backlight_level_pwm( static const struct abm_funcs dce_funcs = { .abm_init = dce_abm_init, .set_abm_level = dce_abm_set_level, - .init_backlight = dce_abm_init_backlight, .set_pipe = dce_abm_set_pipe, .set_backlight_level_pwm = dce_abm_set_backlight_level_pwm, .get_current_backlight = dce_abm_get_current_backlight, .get_target_backlight = dce_abm_get_target_backlight, - .set_abm_immediate_disable = dce_abm_immediate_disable + .init_abm_config = NULL, + .set_abm_immediate_disable = dce_abm_immediate_disable, }; static void dce_abm_construct( @@ -461,10 +270,6 @@ static void dce_abm_construct( base->ctx = ctx; base->funcs = &dce_funcs; - base->stored_backlight_registers.BL_PWM_CNTL = 0; - base->stored_backlight_registers.BL_PWM_CNTL2 = 0; - base->stored_backlight_registers.BL_PWM_PERIOD_CNTL = 0; - base->stored_backlight_registers.LVTMA_PWRSEQ_REF_DIV_BL_PWM_REF_DIV = 0; base->dmcu_is_running = false; abm_dce->regs = regs; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.h b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.h index ba0caaffa24b..9718a4823372 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.h @@ -30,11 +30,6 @@ #include "abm.h" #define ABM_COMMON_REG_LIST_DCE_BASE() \ - SR(BL_PWM_PERIOD_CNTL), \ - SR(BL_PWM_CNTL), \ - SR(BL_PWM_CNTL2), \ - SR(BL_PWM_GRP1_REG_LOCK), \ - SR(LVTMA_PWRSEQ_REF_DIV), \ SR(MASTER_COMM_CNTL_REG), \ SR(MASTER_COMM_CMD_REG), \ SR(MASTER_COMM_DATA_REG1) @@ -85,15 +80,6 @@ .field_name = reg_name ## __ ## field_name ## post_fix #define ABM_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(mask_sh) \ - ABM_SF(BL_PWM_PERIOD_CNTL, BL_PWM_PERIOD, mask_sh), \ - ABM_SF(BL_PWM_PERIOD_CNTL, BL_PWM_PERIOD_BITCNT, mask_sh), \ - ABM_SF(BL_PWM_CNTL, BL_ACTIVE_INT_FRAC_CNT, mask_sh), \ - ABM_SF(BL_PWM_CNTL, BL_PWM_FRACTIONAL_EN, mask_sh), \ - ABM_SF(BL_PWM_CNTL, BL_PWM_EN, mask_sh), \ - ABM_SF(BL_PWM_GRP1_REG_LOCK, BL_PWM_GRP1_IGNORE_MASTER_LOCK_EN, mask_sh), \ - ABM_SF(BL_PWM_GRP1_REG_LOCK, BL_PWM_GRP1_REG_LOCK, mask_sh), \ - ABM_SF(BL_PWM_GRP1_REG_LOCK, BL_PWM_GRP1_REG_UPDATE_PENDING, mask_sh), \ - ABM_SF(LVTMA_PWRSEQ_REF_DIV, BL_PWM_REF_DIV, mask_sh), \ ABM_SF(MASTER_COMM_CNTL_REG, MASTER_COMM_INTERRUPT, mask_sh), \ ABM_SF(MASTER_COMM_CMD_REG, MASTER_COMM_CMD_REG_BYTE0, mask_sh), \ ABM_SF(MASTER_COMM_CMD_REG, MASTER_COMM_CMD_REG_BYTE1, mask_sh), \ @@ -178,19 +164,10 @@ type ABM1_HG_REG_READ_MISSED_FRAME_CLEAR; \ type ABM1_LS_REG_READ_MISSED_FRAME_CLEAR; \ type ABM1_BL_REG_READ_MISSED_FRAME_CLEAR; \ - type BL_PWM_PERIOD; \ - type BL_PWM_PERIOD_BITCNT; \ - type BL_ACTIVE_INT_FRAC_CNT; \ - type BL_PWM_FRACTIONAL_EN; \ type MASTER_COMM_INTERRUPT; \ type MASTER_COMM_CMD_REG_BYTE0; \ type MASTER_COMM_CMD_REG_BYTE1; \ - type MASTER_COMM_CMD_REG_BYTE2; \ - type BL_PWM_REF_DIV; \ - type BL_PWM_EN; \ - type BL_PWM_GRP1_IGNORE_MASTER_LOCK_EN; \ - type BL_PWM_GRP1_REG_LOCK; \ - type BL_PWM_GRP1_REG_UPDATE_PENDING + type MASTER_COMM_CMD_REG_BYTE2 struct dce_abm_shift { ABM_REG_FIELD_LIST(uint8_t); @@ -201,10 +178,6 @@ struct dce_abm_mask { }; struct dce_abm_registers { - uint32_t BL_PWM_PERIOD_CNTL; - uint32_t BL_PWM_CNTL; - uint32_t BL_PWM_CNTL2; - uint32_t LVTMA_PWRSEQ_REF_DIV; uint32_t DC_ABM1_HG_SAMPLE_RATE; uint32_t DC_ABM1_LS_SAMPLE_RATE; uint32_t BL1_PWM_BL_UPDATE_SAMPLE_RATE; @@ -219,7 +192,6 @@ struct dce_abm_registers { uint32_t MASTER_COMM_CMD_REG; uint32_t MASTER_COMM_DATA_REG1; uint32_t BIOS_SCRATCH_2; - uint32_t BL_PWM_GRP1_REG_LOCK; }; struct dce_abm { diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c index 2e992fbc0d71..d2ad0504b0de 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c @@ -1014,39 +1014,6 @@ struct pixel_rate_range_table_entry { unsigned short div_factor; }; -static const struct pixel_rate_range_table_entry video_optimized_pixel_rates[] = { - // /1.001 rates - {25170, 25180, 25200, 1000, 1001}, //25.2MHz -> 25.17 - {59340, 59350, 59400, 1000, 1001}, //59.4Mhz -> 59.340 - {74170, 74180, 74250, 1000, 1001}, //74.25Mhz -> 74.1758 - {125870, 125880, 126000, 1000, 1001}, //126Mhz -> 125.87 - {148350, 148360, 148500, 1000, 1001}, //148.5Mhz -> 148.3516 - {167830, 167840, 168000, 1000, 1001}, //168Mhz -> 167.83 - {222520, 222530, 222750, 1000, 1001}, //222.75Mhz -> 222.527 - {257140, 257150, 257400, 1000, 1001}, //257.4Mhz -> 257.1429 - {296700, 296710, 297000, 1000, 1001}, //297Mhz -> 296.7033 - {342850, 342860, 343200, 1000, 1001}, //343.2Mhz -> 342.857 - {395600, 395610, 396000, 1000, 1001}, //396Mhz -> 395.6 - {409090, 409100, 409500, 1000, 1001}, //409.5Mhz -> 409.091 - {445050, 445060, 445500, 1000, 1001}, //445.5Mhz -> 445.055 - {467530, 467540, 468000, 1000, 1001}, //468Mhz -> 467.5325 - {519230, 519240, 519750, 1000, 1001}, //519.75Mhz -> 519.231 - {525970, 525980, 526500, 1000, 1001}, //526.5Mhz -> 525.974 - {545450, 545460, 546000, 1000, 1001}, //546Mhz -> 545.455 - {593400, 593410, 594000, 1000, 1001}, //594Mhz -> 593.4066 - {623370, 623380, 624000, 1000, 1001}, //624Mhz -> 623.377 - {692300, 692310, 693000, 1000, 1001}, //693Mhz -> 692.308 - {701290, 701300, 702000, 1000, 1001}, //702Mhz -> 701.2987 - {791200, 791210, 792000, 1000, 1001}, //792Mhz -> 791.209 - {890100, 890110, 891000, 1000, 1001}, //891Mhz -> 890.1099 - {1186810, 1186820, 1188000, 1000, 1001},//1188Mhz -> 1186.8131 - - // *1.001 rates - {27020, 27030, 27000, 1001, 1000}, //27Mhz - {54050, 54060, 54000, 1001, 1000}, //54Mhz - {108100, 108110, 108000, 1001, 1000},//108Mhz -}; - static bool dcn20_program_pix_clk( struct clock_source *clock_source, struct pixel_clk_params *pix_clk_params, diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h b/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h index c5aa1f48593a..5479d959ec62 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h @@ -27,10 +27,6 @@ #include "dc_types.h" -#define BL_REG_LIST()\ - SR(LVTMA_PWRSEQ_CNTL), \ - SR(LVTMA_PWRSEQ_STATE) - #define HWSEQ_DCEF_REG_LIST_DCE8() \ .DCFE_CLOCK_CONTROL[0] = mmCRTC0_CRTC_DCFE_CLOCK_CONTROL, \ .DCFE_CLOCK_CONTROL[1] = mmCRTC1_CRTC_DCFE_CLOCK_CONTROL, \ @@ -94,20 +90,17 @@ SRII(BLND_CONTROL, BLND, 0),\ SRII(BLND_CONTROL, BLND, 1),\ SR(BLNDV_CONTROL),\ - HWSEQ_PIXEL_RATE_REG_LIST(CRTC),\ - BL_REG_LIST() + HWSEQ_PIXEL_RATE_REG_LIST(CRTC) #define HWSEQ_DCE8_REG_LIST() \ HWSEQ_DCEF_REG_LIST_DCE8(), \ HWSEQ_BLND_REG_LIST(), \ - HWSEQ_PIXEL_RATE_REG_LIST(CRTC),\ - BL_REG_LIST() + HWSEQ_PIXEL_RATE_REG_LIST(CRTC) #define HWSEQ_DCE10_REG_LIST() \ HWSEQ_DCEF_REG_LIST(), \ HWSEQ_BLND_REG_LIST(), \ - HWSEQ_PIXEL_RATE_REG_LIST(CRTC), \ - BL_REG_LIST() + HWSEQ_PIXEL_RATE_REG_LIST(CRTC) #define HWSEQ_ST_REG_LIST() \ HWSEQ_DCE11_REG_LIST_BASE(), \ @@ -134,8 +127,7 @@ SR(DCHUB_FB_LOCATION),\ SR(DCHUB_AGP_BASE),\ SR(DCHUB_AGP_BOT),\ - SR(DCHUB_AGP_TOP), \ - BL_REG_LIST() + SR(DCHUB_AGP_TOP) #define HWSEQ_VG20_REG_LIST() \ HWSEQ_DCE120_REG_LIST(),\ @@ -144,8 +136,7 @@ #define HWSEQ_DCE112_REG_LIST() \ HWSEQ_DCE10_REG_LIST(), \ HWSEQ_PIXEL_RATE_REG_LIST(CRTC), \ - HWSEQ_PHYPLL_REG_LIST(CRTC), \ - BL_REG_LIST() + HWSEQ_PHYPLL_REG_LIST(CRTC) #define HWSEQ_DCN_REG_LIST()\ SR(REFCLK_CNTL), \ @@ -207,8 +198,7 @@ SR(D3VGA_CONTROL), \ SR(D4VGA_CONTROL), \ SR(VGA_TEST_CONTROL), \ - SR(DC_IP_REQUEST_CNTL), \ - BL_REG_LIST() + SR(DC_IP_REQUEST_CNTL) #define HWSEQ_DCN2_REG_LIST()\ HWSEQ_DCN_REG_LIST(), \ @@ -273,8 +263,7 @@ SR(D4VGA_CONTROL), \ SR(D5VGA_CONTROL), \ SR(D6VGA_CONTROL), \ - SR(DC_IP_REQUEST_CNTL), \ - BL_REG_LIST() + SR(DC_IP_REQUEST_CNTL) #define HWSEQ_DCN21_REG_LIST()\ HWSEQ_DCN_REG_LIST(), \ @@ -324,15 +313,9 @@ SR(D4VGA_CONTROL), \ SR(D5VGA_CONTROL), \ SR(D6VGA_CONTROL), \ - SR(DC_IP_REQUEST_CNTL), \ - BL_REG_LIST() + SR(DC_IP_REQUEST_CNTL) struct dce_hwseq_registers { - - /* Backlight registers */ - uint32_t LVTMA_PWRSEQ_CNTL; - uint32_t LVTMA_PWRSEQ_STATE; - uint32_t DCFE_CLOCK_CONTROL[6]; uint32_t DCFEV_CLOCK_CONTROL; uint32_t DC_MEM_GLOBAL_PWR_REQ_CNTL; @@ -465,26 +448,18 @@ struct dce_hwseq_registers { HWS_SF1(blk, PHYPLL_PIXEL_RATE_CNTL, PHYPLL_PIXEL_RATE_SOURCE, mask_sh),\ HWS_SF1(blk, PHYPLL_PIXEL_RATE_CNTL, PIXEL_RATE_PLL_SOURCE, mask_sh) -#define HWSEQ_LVTMA_MASK_SH_LIST(mask_sh)\ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_DIGON, mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_DIGON_OVRD, mask_sh),\ - HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh) - #define HWSEQ_DCE8_MASK_SH_LIST(mask_sh)\ .DCFE_CLOCK_ENABLE = CRTC_DCFE_CLOCK_CONTROL__CRTC_DCFE_CLOCK_ENABLE ## mask_sh, \ HWS_SF(BLND_, V_UPDATE_LOCK, BLND_DCP_GRPH_V_UPDATE_LOCK, mask_sh),\ HWS_SF(BLND_, V_UPDATE_LOCK, BLND_SCL_V_UPDATE_LOCK, mask_sh),\ HWS_SF(BLND_, V_UPDATE_LOCK, BLND_DCP_GRPH_SURF_V_UPDATE_LOCK, mask_sh),\ HWS_SF(BLND_, CONTROL, BLND_MODE, mask_sh),\ - HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, CRTC0_),\ - HWSEQ_LVTMA_MASK_SH_LIST(mask_sh) + HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, CRTC0_) #define HWSEQ_DCE10_MASK_SH_LIST(mask_sh)\ HWSEQ_DCEF_MASK_SH_LIST(mask_sh, DCFE_),\ HWSEQ_BLND_MASK_SH_LIST(mask_sh, BLND_),\ - HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, CRTC0_),\ - HWSEQ_LVTMA_MASK_SH_LIST(mask_sh) + HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, CRTC0_) #define HWSEQ_DCE11_MASK_SH_LIST(mask_sh)\ HWSEQ_DCE10_MASK_SH_LIST(mask_sh),\ @@ -507,8 +482,7 @@ struct dce_hwseq_registers { HWSEQ_BLND_MASK_SH_LIST(mask_sh, BLND0_BLND_),\ HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, CRTC0_),\ HWSEQ_PHYPLL_MASK_SH_LIST(mask_sh, CRTC0_),\ - HWSEQ_GFX9_DCHUB_MASK_SH_LIST(mask_sh),\ - HWSEQ_LVTMA_MASK_SH_LIST(mask_sh) + HWSEQ_GFX9_DCHUB_MASK_SH_LIST(mask_sh) #define HWSEQ_VG20_MASK_SH_LIST(mask_sh)\ HWSEQ_DCE12_MASK_SH_LIST(mask_sh),\ @@ -570,8 +544,7 @@ struct dce_hwseq_registers { HWS_SF(, D3VGA_CONTROL, D3VGA_MODE_ENABLE, mask_sh),\ HWS_SF(, D4VGA_CONTROL, D4VGA_MODE_ENABLE, mask_sh),\ HWS_SF(, VGA_TEST_CONTROL, VGA_TEST_ENABLE, mask_sh),\ - HWS_SF(, VGA_TEST_CONTROL, VGA_TEST_RENDER_START, mask_sh),\ - HWSEQ_LVTMA_MASK_SH_LIST(mask_sh) + HWS_SF(, VGA_TEST_CONTROL, VGA_TEST_RENDER_START, mask_sh) #define HWSEQ_DCN2_MASK_SH_LIST(mask_sh)\ HWSEQ_DCN_MASK_SH_LIST(mask_sh), \ @@ -630,8 +603,7 @@ struct dce_hwseq_registers { HWS_SF(, DOMAIN19_PG_STATUS, DOMAIN19_PGFSM_PWR_STATUS, mask_sh), \ HWS_SF(, DOMAIN20_PG_STATUS, DOMAIN20_PGFSM_PWR_STATUS, mask_sh), \ HWS_SF(, DOMAIN21_PG_STATUS, DOMAIN21_PGFSM_PWR_STATUS, mask_sh), \ - HWS_SF(, DC_IP_REQUEST_CNTL, IP_REQUEST_EN, mask_sh), \ - HWSEQ_LVTMA_MASK_SH_LIST(mask_sh) + HWS_SF(, DC_IP_REQUEST_CNTL, IP_REQUEST_EN, mask_sh) #define HWSEQ_DCN21_MASK_SH_LIST(mask_sh)\ HWSEQ_DCN_MASK_SH_LIST(mask_sh), \ @@ -671,10 +643,7 @@ struct dce_hwseq_registers { HWS_SF(, DOMAIN16_PG_STATUS, DOMAIN16_PGFSM_PWR_STATUS, mask_sh), \ HWS_SF(, DOMAIN17_PG_STATUS, DOMAIN17_PGFSM_PWR_STATUS, mask_sh), \ HWS_SF(, DOMAIN18_PG_STATUS, DOMAIN18_PGFSM_PWR_STATUS, mask_sh), \ - HWS_SF(, DC_IP_REQUEST_CNTL, IP_REQUEST_EN, mask_sh), \ - HWSEQ_LVTMA_MASK_SH_LIST(mask_sh), \ - HWS_SF(, LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh), \ - HWS_SF(, LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh) + HWS_SF(, DC_IP_REQUEST_CNTL, IP_REQUEST_EN, mask_sh) #define HWSEQ_REG_FIELD_LIST(type) \ type DCFE_CLOCK_ENABLE; \ @@ -706,11 +675,7 @@ struct dce_hwseq_registers { type PF_LFB_REGION;\ type PF_MAX_REGION;\ type ENABLE_L1_TLB;\ - type SYSTEM_ACCESS_MODE;\ - type LVTMA_BLON;\ - type LVTMA_DIGON;\ - type LVTMA_DIGON_OVRD;\ - type LVTMA_PWRSEQ_TARGET_STATE_R; + type SYSTEM_ACCESS_MODE; #define HWSEQ_DCN_REG_FIELD_LIST(type) \ type HUBP_VTG_SEL; \ diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c index 8527cce81c6f..8d8c84c81b34 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c @@ -118,7 +118,8 @@ static const struct link_encoder_funcs dce110_lnk_enc_funcs = { .enable_hpd = dce110_link_encoder_enable_hpd, .disable_hpd = dce110_link_encoder_disable_hpd, .is_dig_enabled = dce110_is_dig_enabled, - .destroy = dce110_link_encoder_destroy + .destroy = dce110_link_encoder_destroy, + .get_max_link_cap = dce110_link_encoder_get_max_link_cap }; static enum bp_result link_transmitter_control( @@ -1389,3 +1390,20 @@ void dce110_link_encoder_disable_hpd(struct link_encoder *enc) set_reg_field_value(value, 0, DC_HPD_CONTROL, DC_HPD_EN); } + +void dce110_link_encoder_get_max_link_cap(struct link_encoder *enc, + struct dc_link_settings *link_settings) +{ + /* Set Default link settings */ + struct dc_link_settings max_link_cap = {LANE_COUNT_FOUR, LINK_RATE_HIGH, + LINK_SPREAD_05_DOWNSPREAD_30KHZ, false, 0}; + + /* Higher link settings based on feature supported */ + if (enc->features.flags.bits.IS_HBR2_CAPABLE) + max_link_cap.link_rate = LINK_RATE_HIGH2; + + if (enc->features.flags.bits.IS_HBR3_CAPABLE) + max_link_cap.link_rate = LINK_RATE_HIGH3; + + *link_settings = max_link_cap; +} diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.h b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.h index 3c9368df4093..384389f0e2c3 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.h @@ -271,4 +271,7 @@ void dce110_psr_program_secondary_packet(struct link_encoder *enc, bool dce110_is_dig_enabled(struct link_encoder *enc); +void dce110_link_encoder_get_max_link_cap(struct link_encoder *enc, + struct dc_link_settings *link_settings); + #endif /* __DC_LINK_ENCODER__DCE110_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_panel_cntl.c b/drivers/gpu/drm/amd/display/dc/dce/dce_panel_cntl.c new file mode 100644 index 000000000000..ebff9b1e312e --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_panel_cntl.c @@ -0,0 +1,299 @@ +/* + * Copyright 2012-15 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "reg_helper.h" +#include "core_types.h" +#include "dc_dmub_srv.h" +#include "panel_cntl.h" +#include "dce_panel_cntl.h" +#include "atom.h" + +#define TO_DCE_PANEL_CNTL(panel_cntl)\ + container_of(panel_cntl, struct dce_panel_cntl, base) + +#define CTX \ + dce_panel_cntl->base.ctx + +#define DC_LOGGER \ + dce_panel_cntl->base.ctx->logger + +#define REG(reg)\ + dce_panel_cntl->regs->reg + +#undef FN +#define FN(reg_name, field_name) \ + dce_panel_cntl->shift->field_name, dce_panel_cntl->mask->field_name + +static unsigned int calculate_16_bit_backlight_from_pwm(struct dce_panel_cntl *dce_panel_cntl) +{ + uint64_t current_backlight; + uint32_t round_result; + uint32_t pwm_period_cntl, bl_period, bl_int_count; + uint32_t bl_pwm_cntl, bl_pwm, fractional_duty_cycle_en; + uint32_t bl_period_mask, bl_pwm_mask; + + pwm_period_cntl = REG_READ(BL_PWM_PERIOD_CNTL); + REG_GET(BL_PWM_PERIOD_CNTL, BL_PWM_PERIOD, &bl_period); + REG_GET(BL_PWM_PERIOD_CNTL, BL_PWM_PERIOD_BITCNT, &bl_int_count); + + bl_pwm_cntl = REG_READ(BL_PWM_CNTL); + REG_GET(BL_PWM_CNTL, BL_ACTIVE_INT_FRAC_CNT, (uint32_t *)(&bl_pwm)); + REG_GET(BL_PWM_CNTL, BL_PWM_FRACTIONAL_EN, &fractional_duty_cycle_en); + + if (bl_int_count == 0) + bl_int_count = 16; + + bl_period_mask = (1 << bl_int_count) - 1; + bl_period &= bl_period_mask; + + bl_pwm_mask = bl_period_mask << (16 - bl_int_count); + + if (fractional_duty_cycle_en == 0) + bl_pwm &= bl_pwm_mask; + else + bl_pwm &= 0xFFFF; + + current_backlight = bl_pwm << (1 + bl_int_count); + + if (bl_period == 0) + bl_period = 0xFFFF; + + current_backlight = div_u64(current_backlight, bl_period); + current_backlight = (current_backlight + 1) >> 1; + + current_backlight = (uint64_t)(current_backlight) * bl_period; + + round_result = (uint32_t)(current_backlight & 0xFFFFFFFF); + + round_result = (round_result >> (bl_int_count-1)) & 1; + + current_backlight >>= bl_int_count; + current_backlight += round_result; + + return (uint32_t)(current_backlight); +} + +uint32_t dce_panel_cntl_hw_init(struct panel_cntl *panel_cntl) +{ + struct dce_panel_cntl *dce_panel_cntl = TO_DCE_PANEL_CNTL(panel_cntl); + uint32_t value; + uint32_t current_backlight; + + /* It must not be 0, so we have to restore them + * Bios bug w/a - period resets to zero, + * restoring to cache values which is always correct + */ + REG_GET(BL_PWM_CNTL, BL_ACTIVE_INT_FRAC_CNT, &value); + + if (value == 0 || value == 1) { + if (panel_cntl->stored_backlight_registers.BL_PWM_CNTL != 0) { + REG_WRITE(BL_PWM_CNTL, + panel_cntl->stored_backlight_registers.BL_PWM_CNTL); + REG_WRITE(BL_PWM_CNTL2, + panel_cntl->stored_backlight_registers.BL_PWM_CNTL2); + REG_WRITE(BL_PWM_PERIOD_CNTL, + panel_cntl->stored_backlight_registers.BL_PWM_PERIOD_CNTL); + REG_UPDATE(PWRSEQ_REF_DIV, + BL_PWM_REF_DIV, + panel_cntl->stored_backlight_registers.LVTMA_PWRSEQ_REF_DIV_BL_PWM_REF_DIV); + } else { + /* TODO: Note: This should not really happen since VBIOS + * should have initialized PWM registers on boot. + */ + REG_WRITE(BL_PWM_CNTL, 0xC000FA00); + REG_WRITE(BL_PWM_PERIOD_CNTL, 0x000C0FA0); + } + } else { + panel_cntl->stored_backlight_registers.BL_PWM_CNTL = + REG_READ(BL_PWM_CNTL); + panel_cntl->stored_backlight_registers.BL_PWM_CNTL2 = + REG_READ(BL_PWM_CNTL2); + panel_cntl->stored_backlight_registers.BL_PWM_PERIOD_CNTL = + REG_READ(BL_PWM_PERIOD_CNTL); + + REG_GET(PWRSEQ_REF_DIV, BL_PWM_REF_DIV, + &panel_cntl->stored_backlight_registers.LVTMA_PWRSEQ_REF_DIV_BL_PWM_REF_DIV); + } + + // Have driver take backlight control + // TakeBacklightControl(true) + value = REG_READ(BIOS_SCRATCH_2); + value |= ATOM_S2_VRI_BRIGHT_ENABLE; + REG_WRITE(BIOS_SCRATCH_2, value); + + // Enable the backlight output + REG_UPDATE(BL_PWM_CNTL, BL_PWM_EN, 1); + + // Unlock group 2 backlight registers + REG_UPDATE(BL_PWM_GRP1_REG_LOCK, + BL_PWM_GRP1_REG_LOCK, 0); + + current_backlight = calculate_16_bit_backlight_from_pwm(dce_panel_cntl); + + return current_backlight; +} + +bool dce_is_panel_backlight_on(struct panel_cntl *panel_cntl) +{ + struct dce_panel_cntl *dce_panel_cntl = TO_DCE_PANEL_CNTL(panel_cntl); + uint32_t value; + + REG_GET(PWRSEQ_CNTL, LVTMA_BLON, &value); + + return value; +} + +bool dce_is_panel_powered_on(struct panel_cntl *panel_cntl) +{ + struct dce_panel_cntl *dce_panel_cntl = TO_DCE_PANEL_CNTL(panel_cntl); + uint32_t pwr_seq_state, dig_on, dig_on_ovrd; + + REG_GET(PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, &pwr_seq_state); + + REG_GET_2(PWRSEQ_CNTL, LVTMA_DIGON, &dig_on, LVTMA_DIGON_OVRD, &dig_on_ovrd); + + return (pwr_seq_state == 1) || (dig_on == 1 && dig_on_ovrd == 1); +} + +void dce_store_backlight_level(struct panel_cntl *panel_cntl) +{ + struct dce_panel_cntl *dce_panel_cntl = TO_DCE_PANEL_CNTL(panel_cntl); + + panel_cntl->stored_backlight_registers.BL_PWM_CNTL = + REG_READ(BL_PWM_CNTL); + panel_cntl->stored_backlight_registers.BL_PWM_CNTL2 = + REG_READ(BL_PWM_CNTL2); + panel_cntl->stored_backlight_registers.BL_PWM_PERIOD_CNTL = + REG_READ(BL_PWM_PERIOD_CNTL); + + REG_GET(PWRSEQ_REF_DIV, BL_PWM_REF_DIV, + &panel_cntl->stored_backlight_registers.LVTMA_PWRSEQ_REF_DIV_BL_PWM_REF_DIV); +} + +void dce_driver_set_backlight(struct panel_cntl *panel_cntl, + uint32_t backlight_pwm_u16_16) +{ + uint32_t backlight_16bit; + uint32_t masked_pwm_period; + uint8_t bit_count; + uint64_t active_duty_cycle; + uint32_t pwm_period_bitcnt; + struct dce_panel_cntl *dce_panel_cntl = TO_DCE_PANEL_CNTL(panel_cntl); + + /* + * 1. Find 16 bit backlight active duty cycle, where 0 <= backlight + * active duty cycle <= backlight period + */ + + /* 1.1 Apply bitmask for backlight period value based on value of BITCNT + */ + REG_GET_2(BL_PWM_PERIOD_CNTL, + BL_PWM_PERIOD_BITCNT, &pwm_period_bitcnt, + BL_PWM_PERIOD, &masked_pwm_period); + + if (pwm_period_bitcnt == 0) + bit_count = 16; + else + bit_count = pwm_period_bitcnt; + + /* e.g. maskedPwmPeriod = 0x24 when bitCount is 6 */ + masked_pwm_period = masked_pwm_period & ((1 << bit_count) - 1); + + /* 1.2 Calculate integer active duty cycle required upper 16 bits + * contain integer component, lower 16 bits contain fractional component + * of active duty cycle e.g. 0x21BDC0 = 0xEFF0 * 0x24 + */ + active_duty_cycle = backlight_pwm_u16_16 * masked_pwm_period; + + /* 1.3 Calculate 16 bit active duty cycle from integer and fractional + * components shift by bitCount then mask 16 bits and add rounding bit + * from MSB of fraction e.g. 0x86F7 = ((0x21BDC0 >> 6) & 0xFFF) + 0 + */ + backlight_16bit = active_duty_cycle >> bit_count; + backlight_16bit &= 0xFFFF; + backlight_16bit += (active_duty_cycle >> (bit_count - 1)) & 0x1; + + /* + * 2. Program register with updated value + */ + + /* 2.1 Lock group 2 backlight registers */ + + REG_UPDATE_2(BL_PWM_GRP1_REG_LOCK, + BL_PWM_GRP1_IGNORE_MASTER_LOCK_EN, 1, + BL_PWM_GRP1_REG_LOCK, 1); + + // 2.2 Write new active duty cycle + REG_UPDATE(BL_PWM_CNTL, BL_ACTIVE_INT_FRAC_CNT, backlight_16bit); + + /* 2.3 Unlock group 2 backlight registers */ + REG_UPDATE(BL_PWM_GRP1_REG_LOCK, + BL_PWM_GRP1_REG_LOCK, 0); + + /* 3 Wait for pending bit to be cleared */ + REG_WAIT(BL_PWM_GRP1_REG_LOCK, + BL_PWM_GRP1_REG_UPDATE_PENDING, 0, + 1, 10000); +} + +static void dce_panel_cntl_destroy(struct panel_cntl **panel_cntl) +{ + struct dce_panel_cntl *dce_panel_cntl = TO_DCE_PANEL_CNTL(*panel_cntl); + + kfree(dce_panel_cntl); + *panel_cntl = NULL; +} + +static const struct panel_cntl_funcs dce_link_panel_cntl_funcs = { + .destroy = dce_panel_cntl_destroy, + .hw_init = dce_panel_cntl_hw_init, + .is_panel_backlight_on = dce_is_panel_backlight_on, + .is_panel_powered_on = dce_is_panel_powered_on, + .store_backlight_level = dce_store_backlight_level, + .driver_set_backlight = dce_driver_set_backlight, +}; + +void dce_panel_cntl_construct( + struct dce_panel_cntl *dce_panel_cntl, + const struct panel_cntl_init_data *init_data, + const struct dce_panel_cntl_registers *regs, + const struct dce_panel_cntl_shift *shift, + const struct dce_panel_cntl_mask *mask) +{ + struct panel_cntl *base = &dce_panel_cntl->base; + + base->stored_backlight_registers.BL_PWM_CNTL = 0; + base->stored_backlight_registers.BL_PWM_CNTL2 = 0; + base->stored_backlight_registers.BL_PWM_PERIOD_CNTL = 0; + base->stored_backlight_registers.LVTMA_PWRSEQ_REF_DIV_BL_PWM_REF_DIV = 0; + + dce_panel_cntl->regs = regs; + dce_panel_cntl->shift = shift; + dce_panel_cntl->mask = mask; + + dce_panel_cntl->base.funcs = &dce_link_panel_cntl_funcs; + dce_panel_cntl->base.ctx = init_data->ctx; + dce_panel_cntl->base.inst = init_data->inst; +} diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_panel_cntl.h b/drivers/gpu/drm/amd/display/dc/dce/dce_panel_cntl.h new file mode 100644 index 000000000000..70ec691e14d2 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_panel_cntl.h @@ -0,0 +1,125 @@ +/* + * Copyright 2012-15 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DC_PANEL_CNTL__DCE_H__ +#define __DC_PANEL_CNTL__DCE_H__ + +#include "panel_cntl.h" + +/* set register offset with instance */ +#define DCE_PANEL_CNTL_SR(reg_name, block)\ + .reg_name = mm ## block ## _ ## reg_name + +#define DCE_PANEL_CNTL_REG_LIST()\ + DCE_PANEL_CNTL_SR(PWRSEQ_CNTL, LVTMA), \ + DCE_PANEL_CNTL_SR(PWRSEQ_STATE, LVTMA), \ + DCE_PANEL_CNTL_SR(PWRSEQ_REF_DIV, LVTMA), \ + SR(BL_PWM_CNTL), \ + SR(BL_PWM_CNTL2), \ + SR(BL_PWM_PERIOD_CNTL), \ + SR(BL_PWM_GRP1_REG_LOCK), \ + SR(BIOS_SCRATCH_2) + +#define DCN_PANEL_CNTL_SR(reg_name, block)\ + .reg_name = BASE(mm ## block ## _ ## reg_name ## _BASE_IDX) + \ + mm ## block ## _ ## reg_name + +#define DCN_PANEL_CNTL_REG_LIST()\ + DCN_PANEL_CNTL_SR(PWRSEQ_CNTL, LVTMA), \ + DCN_PANEL_CNTL_SR(PWRSEQ_STATE, LVTMA), \ + DCE_PANEL_CNTL_SR(PWRSEQ_REF_DIV, LVTMA), \ + SR(BL_PWM_CNTL), \ + SR(BL_PWM_CNTL2), \ + SR(BL_PWM_PERIOD_CNTL), \ + SR(BL_PWM_GRP1_REG_LOCK), \ + SR(BIOS_SCRATCH_2) + +#define DCE_PANEL_CNTL_SF(reg_name, field_name, post_fix)\ + .field_name = reg_name ## __ ## field_name ## post_fix + +#define DCE_PANEL_CNTL_MASK_SH_LIST(mask_sh) \ + DCE_PANEL_CNTL_SF(LVTMA_PWRSEQ_CNTL, LVTMA_BLON, mask_sh),\ + DCE_PANEL_CNTL_SF(LVTMA_PWRSEQ_CNTL, LVTMA_DIGON, mask_sh),\ + DCE_PANEL_CNTL_SF(LVTMA_PWRSEQ_CNTL, LVTMA_DIGON_OVRD, mask_sh),\ + DCE_PANEL_CNTL_SF(LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, mask_sh), \ + DCE_PANEL_CNTL_SF(LVTMA_PWRSEQ_REF_DIV, BL_PWM_REF_DIV, mask_sh), \ + DCE_PANEL_CNTL_SF(BL_PWM_PERIOD_CNTL, BL_PWM_PERIOD, mask_sh), \ + DCE_PANEL_CNTL_SF(BL_PWM_PERIOD_CNTL, BL_PWM_PERIOD_BITCNT, mask_sh), \ + DCE_PANEL_CNTL_SF(BL_PWM_CNTL, BL_ACTIVE_INT_FRAC_CNT, mask_sh), \ + DCE_PANEL_CNTL_SF(BL_PWM_CNTL, BL_PWM_FRACTIONAL_EN, mask_sh), \ + DCE_PANEL_CNTL_SF(BL_PWM_CNTL, BL_PWM_EN, mask_sh), \ + DCE_PANEL_CNTL_SF(BL_PWM_GRP1_REG_LOCK, BL_PWM_GRP1_IGNORE_MASTER_LOCK_EN, mask_sh), \ + DCE_PANEL_CNTL_SF(BL_PWM_GRP1_REG_LOCK, BL_PWM_GRP1_REG_LOCK, mask_sh), \ + DCE_PANEL_CNTL_SF(BL_PWM_GRP1_REG_LOCK, BL_PWM_GRP1_REG_UPDATE_PENDING, mask_sh) + +#define DCE_PANEL_CNTL_REG_FIELD_LIST(type) \ + type LVTMA_BLON;\ + type LVTMA_DIGON;\ + type LVTMA_DIGON_OVRD;\ + type LVTMA_PWRSEQ_TARGET_STATE_R; \ + type BL_PWM_REF_DIV; \ + type BL_PWM_EN; \ + type BL_ACTIVE_INT_FRAC_CNT; \ + type BL_PWM_FRACTIONAL_EN; \ + type BL_PWM_PERIOD; \ + type BL_PWM_PERIOD_BITCNT; \ + type BL_PWM_GRP1_IGNORE_MASTER_LOCK_EN; \ + type BL_PWM_GRP1_REG_LOCK; \ + type BL_PWM_GRP1_REG_UPDATE_PENDING + +struct dce_panel_cntl_shift { + DCE_PANEL_CNTL_REG_FIELD_LIST(uint8_t); +}; + +struct dce_panel_cntl_mask { + DCE_PANEL_CNTL_REG_FIELD_LIST(uint32_t); +}; + +struct dce_panel_cntl_registers { + uint32_t PWRSEQ_CNTL; + uint32_t PWRSEQ_STATE; + uint32_t BL_PWM_CNTL; + uint32_t BL_PWM_CNTL2; + uint32_t BL_PWM_PERIOD_CNTL; + uint32_t BL_PWM_GRP1_REG_LOCK; + uint32_t PWRSEQ_REF_DIV; + uint32_t BIOS_SCRATCH_2; +}; + +struct dce_panel_cntl { + struct panel_cntl base; + const struct dce_panel_cntl_registers *regs; + const struct dce_panel_cntl_shift *shift; + const struct dce_panel_cntl_mask *mask; +}; + +void dce_panel_cntl_construct( + struct dce_panel_cntl *panel_cntl, + const struct panel_cntl_init_data *init_data, + const struct dce_panel_cntl_registers *regs, + const struct dce_panel_cntl_shift *shift, + const struct dce_panel_cntl_mask *mask); + +#endif /* __DC_PANEL_CNTL__DCE_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c index 451574971b96..4cdaaf4d881c 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c @@ -1336,7 +1336,6 @@ static void dce110_se_audio_setup( { struct dce110_stream_encoder *enc110 = DCE110STRENC_FROM_STRENC(enc); - uint32_t speakers = 0; uint32_t channels = 0; ASSERT(audio_info); @@ -1344,7 +1343,6 @@ static void dce110_se_audio_setup( /* This should not happen.it does so we don't get BSOD*/ return; - speakers = audio_info->flags.info.ALLSPEAKERS; channels = speakers_to_channels(audio_info->flags.speaker_flags).all; /* setup the audio stream source select (audio -> dig mapping) */ diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c new file mode 100644 index 000000000000..da0b29abfbda --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c @@ -0,0 +1,319 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "dmub_abm.h" +#include "dce_abm.h" +#include "dc.h" +#include "dc_dmub_srv.h" +#include "dmub/dmub_srv.h" +#include "core_types.h" +#include "dm_services.h" +#include "reg_helper.h" +#include "fixed31_32.h" + +#include "atom.h" + +#define TO_DMUB_ABM(abm)\ + container_of(abm, struct dce_abm, base) + +#define REG(reg) \ + (dce_abm->regs->reg) + +#undef FN +#define FN(reg_name, field_name) \ + dce_abm->abm_shift->field_name, dce_abm->abm_mask->field_name + +#define CTX \ + dce_abm->base.ctx + +#define DISABLE_ABM_IMMEDIATELY 255 + +static bool dmub_abm_set_pipe(struct abm *abm, uint32_t otg_inst, uint32_t panel_inst) +{ + union dmub_rb_cmd cmd; + struct dc_context *dc = abm->ctx; + uint32_t ramping_boundary = 0xFFFF; + + cmd.abm_set_pipe.header.type = DMUB_CMD__ABM; + cmd.abm_set_pipe.header.sub_type = DMUB_CMD__ABM_SET_PIPE; + cmd.abm_set_pipe.abm_set_pipe_data.otg_inst = otg_inst; + cmd.abm_set_pipe.abm_set_pipe_data.panel_inst = panel_inst; + cmd.abm_set_pipe.abm_set_pipe_data.ramping_boundary = ramping_boundary; + cmd.abm_set_pipe.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_pipe_data); + + dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); + dc_dmub_srv_cmd_execute(dc->dmub_srv); + dc_dmub_srv_wait_idle(dc->dmub_srv); + + return true; +} + +static void dmcub_set_backlight_level( + struct dce_abm *dce_abm, + uint32_t backlight_pwm_u16_16, + uint32_t frame_ramp, + uint32_t otg_inst, + uint32_t panel_inst) +{ + union dmub_rb_cmd cmd; + struct dc_context *dc = dce_abm->base.ctx; + unsigned int backlight_8_bit = 0; + uint32_t s2; + + if (backlight_pwm_u16_16 & 0x10000) + // Check for max backlight condition + backlight_8_bit = 0xFF; + else + // Take MSB of fractional part since backlight is not max + backlight_8_bit = (backlight_pwm_u16_16 >> 8) & 0xFF; + + dmub_abm_set_pipe(&dce_abm->base, otg_inst, panel_inst); + + REG_UPDATE(BL1_PWM_USER_LEVEL, BL1_PWM_USER_LEVEL, backlight_pwm_u16_16); + + if (otg_inst == 0) + frame_ramp = 0; + + cmd.abm_set_backlight.header.type = DMUB_CMD__ABM; + cmd.abm_set_backlight.header.sub_type = DMUB_CMD__ABM_SET_BACKLIGHT; + cmd.abm_set_backlight.abm_set_backlight_data.frame_ramp = frame_ramp; + cmd.abm_set_backlight.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_backlight_data); + + dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); + dc_dmub_srv_cmd_execute(dc->dmub_srv); + dc_dmub_srv_wait_idle(dc->dmub_srv); + + // Update requested backlight level + s2 = REG_READ(BIOS_SCRATCH_2); + + s2 &= ~ATOM_S2_CURRENT_BL_LEVEL_MASK; + backlight_8_bit &= (ATOM_S2_CURRENT_BL_LEVEL_MASK >> + ATOM_S2_CURRENT_BL_LEVEL_SHIFT); + s2 |= (backlight_8_bit << ATOM_S2_CURRENT_BL_LEVEL_SHIFT); + + REG_WRITE(BIOS_SCRATCH_2, s2); +} + +static void dmub_abm_enable_fractional_pwm(struct dc_context *dc) +{ + union dmub_rb_cmd cmd; + uint32_t fractional_pwm = (dc->dc->config.disable_fractional_pwm == false) ? 1 : 0; + + cmd.abm_set_pwm_frac.header.type = DMUB_CMD__ABM; + cmd.abm_set_pwm_frac.header.sub_type = DMUB_CMD__ABM_SET_PWM_FRAC; + cmd.abm_set_pwm_frac.abm_set_pwm_frac_data.fractional_pwm = fractional_pwm; + cmd.abm_set_pwm_frac.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_pwm_frac_data); + + dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); + dc_dmub_srv_cmd_execute(dc->dmub_srv); + dc_dmub_srv_wait_idle(dc->dmub_srv); +} + +static void dmub_abm_init(struct abm *abm, uint32_t backlight) +{ + struct dce_abm *dce_abm = TO_DMUB_ABM(abm); + + REG_WRITE(DC_ABM1_HG_SAMPLE_RATE, 0x103); + REG_WRITE(DC_ABM1_HG_SAMPLE_RATE, 0x101); + REG_WRITE(DC_ABM1_LS_SAMPLE_RATE, 0x103); + REG_WRITE(DC_ABM1_LS_SAMPLE_RATE, 0x101); + REG_WRITE(BL1_PWM_BL_UPDATE_SAMPLE_RATE, 0x101); + + REG_SET_3(DC_ABM1_HG_MISC_CTRL, 0, + ABM1_HG_NUM_OF_BINS_SEL, 0, + ABM1_HG_VMAX_SEL, 1, + ABM1_HG_BIN_BITWIDTH_SIZE_SEL, 0); + + REG_SET_3(DC_ABM1_IPCSC_COEFF_SEL, 0, + ABM1_IPCSC_COEFF_SEL_R, 2, + ABM1_IPCSC_COEFF_SEL_G, 4, + ABM1_IPCSC_COEFF_SEL_B, 2); + + REG_UPDATE(BL1_PWM_CURRENT_ABM_LEVEL, + BL1_PWM_CURRENT_ABM_LEVEL, backlight); + + REG_UPDATE(BL1_PWM_TARGET_ABM_LEVEL, + BL1_PWM_TARGET_ABM_LEVEL, backlight); + + REG_UPDATE(BL1_PWM_USER_LEVEL, + BL1_PWM_USER_LEVEL, backlight); + + REG_UPDATE_2(DC_ABM1_LS_MIN_MAX_PIXEL_VALUE_THRES, + ABM1_LS_MIN_PIXEL_VALUE_THRES, 0, + ABM1_LS_MAX_PIXEL_VALUE_THRES, 1000); + + REG_SET_3(DC_ABM1_HGLS_REG_READ_PROGRESS, 0, + ABM1_HG_REG_READ_MISSED_FRAME_CLEAR, 1, + ABM1_LS_REG_READ_MISSED_FRAME_CLEAR, 1, + ABM1_BL_REG_READ_MISSED_FRAME_CLEAR, 1); + + dmub_abm_enable_fractional_pwm(abm->ctx); +} + +static unsigned int dmub_abm_get_current_backlight(struct abm *abm) +{ + struct dce_abm *dce_abm = TO_DMUB_ABM(abm); + unsigned int backlight = REG_READ(BL1_PWM_CURRENT_ABM_LEVEL); + + /* return backlight in hardware format which is unsigned 17 bits, with + * 1 bit integer and 16 bit fractional + */ + return backlight; +} + +static unsigned int dmub_abm_get_target_backlight(struct abm *abm) +{ + struct dce_abm *dce_abm = TO_DMUB_ABM(abm); + unsigned int backlight = REG_READ(BL1_PWM_TARGET_ABM_LEVEL); + + /* return backlight in hardware format which is unsigned 17 bits, with + * 1 bit integer and 16 bit fractional + */ + return backlight; +} + +static bool dmub_abm_set_level(struct abm *abm, uint32_t level) +{ + union dmub_rb_cmd cmd; + struct dc_context *dc = abm->ctx; + + cmd.abm_set_level.header.type = DMUB_CMD__ABM; + cmd.abm_set_level.header.sub_type = DMUB_CMD__ABM_SET_LEVEL; + cmd.abm_set_level.abm_set_level_data.level = level; + cmd.abm_set_level.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_level_data); + + dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); + dc_dmub_srv_cmd_execute(dc->dmub_srv); + dc_dmub_srv_wait_idle(dc->dmub_srv); + + return true; +} + +static bool dmub_abm_immediate_disable(struct abm *abm, uint32_t panel_inst) +{ + dmub_abm_set_pipe(abm, DISABLE_ABM_IMMEDIATELY, panel_inst); + + return true; +} + +static bool dmub_abm_set_backlight_level_pwm( + struct abm *abm, + unsigned int backlight_pwm_u16_16, + unsigned int frame_ramp, + unsigned int otg_inst, + uint32_t panel_inst) +{ + struct dce_abm *dce_abm = TO_DMUB_ABM(abm); + + dmcub_set_backlight_level(dce_abm, + backlight_pwm_u16_16, + frame_ramp, + otg_inst, + panel_inst); + + return true; +} + +static bool dmub_abm_init_config(struct abm *abm, + const char *src, + unsigned int bytes) +{ + union dmub_rb_cmd cmd; + struct dc_context *dc = abm->ctx; + + // TODO: Optimize by only reading back final 4 bytes + dmub_flush_buffer_mem(&dc->dmub_srv->dmub->scratch_mem_fb); + + // Copy iramtable into cw7 + memcpy(dc->dmub_srv->dmub->scratch_mem_fb.cpu_addr, (void *)src, bytes); + + // Fw will copy from cw7 to fw_state + cmd.abm_init_config.header.type = DMUB_CMD__ABM; + cmd.abm_init_config.header.sub_type = DMUB_CMD__ABM_INIT_CONFIG; + cmd.abm_init_config.abm_init_config_data.src.quad_part = dc->dmub_srv->dmub->scratch_mem_fb.gpu_addr; + cmd.abm_init_config.abm_init_config_data.bytes = bytes; + cmd.abm_init_config.header.payload_bytes = sizeof(struct dmub_cmd_abm_init_config_data); + + dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); + dc_dmub_srv_cmd_execute(dc->dmub_srv); + dc_dmub_srv_wait_idle(dc->dmub_srv); + + return true; +} + +static const struct abm_funcs abm_funcs = { + .abm_init = dmub_abm_init, + .set_abm_level = dmub_abm_set_level, + .set_pipe = dmub_abm_set_pipe, + .set_backlight_level_pwm = dmub_abm_set_backlight_level_pwm, + .get_current_backlight = dmub_abm_get_current_backlight, + .get_target_backlight = dmub_abm_get_target_backlight, + .set_abm_immediate_disable = dmub_abm_immediate_disable, + .init_abm_config = dmub_abm_init_config, +}; + +static void dmub_abm_construct( + struct dce_abm *abm_dce, + struct dc_context *ctx, + const struct dce_abm_registers *regs, + const struct dce_abm_shift *abm_shift, + const struct dce_abm_mask *abm_mask) +{ + struct abm *base = &abm_dce->base; + + base->ctx = ctx; + base->funcs = &abm_funcs; + base->dmcu_is_running = false; + + abm_dce->regs = regs; + abm_dce->abm_shift = abm_shift; + abm_dce->abm_mask = abm_mask; +} + +struct abm *dmub_abm_create( + struct dc_context *ctx, + const struct dce_abm_registers *regs, + const struct dce_abm_shift *abm_shift, + const struct dce_abm_mask *abm_mask) +{ + struct dce_abm *abm_dce = kzalloc(sizeof(*abm_dce), GFP_KERNEL); + + if (abm_dce == NULL) { + BREAK_TO_DEBUGGER(); + return NULL; + } + + dmub_abm_construct(abm_dce, ctx, regs, abm_shift, abm_mask); + + return &abm_dce->base; +} + +void dmub_abm_destroy(struct abm **abm) +{ + struct dce_abm *abm_dce = TO_DMUB_ABM(*abm); + + kfree(abm_dce); + *abm = NULL; +} diff --git a/drivers/gpu/drm/amd/display/dc/dml/dml_common_defs.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.h index 723af0b2dda0..3a5d5ac7a86e 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dml_common_defs.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.h @@ -1,5 +1,5 @@ /* - * Copyright 2017 Advanced Micro Devices, Inc. + * Copyright 2019 Advanced Micro Devices, Inc. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -23,21 +23,18 @@ * */ -#include "dml_common_defs.h" -#include "dcn_calc_math.h" +#ifndef __DMUB_ABM_H__ +#define __DMUB_ABM_H__ -#include "dml_inline_defs.h" +#include "abm.h" +#include "dce_abm.h" -double dml_round(double a) -{ - double round_pt = 0.5; - double ceil = dml_ceil(a, 1); - double floor = dml_floor(a, 1); - - if (a - floor >= round_pt) - return ceil; - else - return floor; -} +struct abm *dmub_abm_create( + struct dc_context *ctx, + const struct dce_abm_registers *regs, + const struct dce_abm_shift *abm_shift, + const struct dce_abm_mask *abm_mask); +void dmub_abm_destroy(struct abm **abm); +#endif diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c index bc109d4fc6e6..044a0133ebb1 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c @@ -26,13 +26,51 @@ #include "dmub_psr.h" #include "dc.h" #include "dc_dmub_srv.h" -#include "../../dmub/inc/dmub_srv.h" -#include "../../dmub/inc/dmub_gpint_cmd.h" +#include "dmub/dmub_srv.h" #include "core_types.h" #define MAX_PIPES 6 /** + * Convert dmcub psr state to dmcu psr state. + */ +static void convert_psr_state(uint32_t *psr_state) +{ + if (*psr_state == 0) + *psr_state = 0; + else if (*psr_state == 0x10) + *psr_state = 1; + else if (*psr_state == 0x11) + *psr_state = 2; + else if (*psr_state == 0x20) + *psr_state = 3; + else if (*psr_state == 0x21) + *psr_state = 4; + else if (*psr_state == 0x30) + *psr_state = 5; + else if (*psr_state == 0x31) + *psr_state = 6; + else if (*psr_state == 0x40) + *psr_state = 7; + else if (*psr_state == 0x41) + *psr_state = 8; + else if (*psr_state == 0x42) + *psr_state = 9; + else if (*psr_state == 0x43) + *psr_state = 10; + else if (*psr_state == 0x44) + *psr_state = 11; + else if (*psr_state == 0x50) + *psr_state = 12; + else if (*psr_state == 0x51) + *psr_state = 13; + else if (*psr_state == 0x52) + *psr_state = 14; + else if (*psr_state == 0x53) + *psr_state = 15; +} + +/** * Get PSR state from firmware. */ static void dmub_psr_get_state(struct dmub_psr *dmub, uint32_t *psr_state) @@ -43,6 +81,8 @@ static void dmub_psr_get_state(struct dmub_psr *dmub, uint32_t *psr_state) dmub_srv_send_gpint_command(srv, DMUB_GPINT__GET_PSR_STATE, 0, 30); dmub_srv_get_gpint_response(srv, psr_state); + + convert_psr_state(psr_state); } /** @@ -53,19 +93,23 @@ static bool dmub_psr_set_version(struct dmub_psr *dmub, struct dc_stream_state * union dmub_rb_cmd cmd; struct dc_context *dc = dmub->ctx; + if (stream->link->psr_settings.psr_version == DC_PSR_VERSION_UNSUPPORTED) + return false; + cmd.psr_set_version.header.type = DMUB_CMD__PSR; cmd.psr_set_version.header.sub_type = DMUB_CMD__PSR_SET_VERSION; - - if (stream->psr_version == 0x0) // Unsupported - return false; - else if (stream->psr_version == 0x1) + switch (stream->link->psr_settings.psr_version) { + case DC_PSR_VERSION_1: cmd.psr_set_version.psr_set_version_data.version = PSR_VERSION_1; - else if (stream->psr_version == 0x2) - cmd.psr_set_version.psr_set_version_data.version = PSR_VERSION_2; - - cmd.psr_enable.header.payload_bytes = sizeof(struct dmub_cmd_psr_set_version_data); + break; + case DC_PSR_VERSION_UNSUPPORTED: + default: + cmd.psr_set_version.psr_set_version_data.version = PSR_VERSION_UNSUPPORTED; + break; + } + cmd.psr_set_version.header.payload_bytes = sizeof(struct dmub_cmd_psr_set_version_data); - dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd.psr_enable.header); + dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); dc_dmub_srv_cmd_execute(dc->dmub_srv); dc_dmub_srv_wait_idle(dc->dmub_srv); @@ -89,7 +133,7 @@ static void dmub_psr_enable(struct dmub_psr *dmub, bool enable) cmd.psr_enable.header.payload_bytes = 0; // Send header only - dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd.psr_enable.header); + dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); dc_dmub_srv_cmd_execute(dc->dmub_srv); dc_dmub_srv_wait_idle(dc->dmub_srv); } @@ -113,7 +157,7 @@ static void dmub_psr_set_level(struct dmub_psr *dmub, uint16_t psr_level) cmd.psr_set_level.header.payload_bytes = sizeof(struct dmub_cmd_psr_set_level_data); cmd.psr_set_level.psr_set_level_data.psr_level = psr_level; - dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd.psr_set_level.header); + dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); dc_dmub_srv_cmd_execute(dc->dmub_srv); dc_dmub_srv_wait_idle(dc->dmub_srv); } @@ -162,7 +206,7 @@ static bool dmub_psr_copy_settings(struct dmub_psr *dmub, cmd.psr_copy_settings.header.payload_bytes = sizeof(struct dmub_cmd_psr_copy_settings_data); // Hw insts - copy_settings_data->dpphy_inst = psr_context->phyType; + copy_settings_data->dpphy_inst = psr_context->transmitterId; copy_settings_data->aux_inst = psr_context->channel; copy_settings_data->digfe_inst = psr_context->engineId; copy_settings_data->digbe_inst = psr_context->transmitterId; @@ -187,8 +231,10 @@ static bool dmub_psr_copy_settings(struct dmub_psr *dmub, copy_settings_data->smu_optimizations_en = psr_context->allow_smu_optimizations; copy_settings_data->frame_delay = psr_context->frame_delay; copy_settings_data->frame_cap_ind = psr_context->psrFrameCaptureIndicationReq; + copy_settings_data->debug.visual_confirm = dc->dc->debug.visual_confirm == VISUAL_CONFIRM_PSR ? + true : false; - dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd.psr_copy_settings.header); + dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); dc_dmub_srv_cmd_execute(dc->dmub_srv); dc_dmub_srv_wait_idle(dc->dmub_srv); diff --git a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c index 8f78bf9abbca..a28c4ae0f259 100644 --- a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c @@ -46,6 +46,7 @@ #include "dce/dce_audio.h" #include "dce/dce_hwseq.h" #include "dce100/dce100_hw_sequencer.h" +#include "dce/dce_panel_cntl.h" #include "reg_helper.h" @@ -249,6 +250,18 @@ static const struct dce_stream_encoder_mask se_mask = { SE_COMMON_MASK_SH_LIST_DCE80_100(_MASK) }; +static const struct dce_panel_cntl_registers panel_cntl_regs[] = { + { DCE_PANEL_CNTL_REG_LIST() } +}; + +static const struct dce_panel_cntl_shift panel_cntl_shift = { + DCE_PANEL_CNTL_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce_panel_cntl_mask panel_cntl_mask = { + DCE_PANEL_CNTL_MASK_SH_LIST(_MASK) +}; + #define opp_regs(id)\ [id] = {\ OPP_DCE_100_REG_LIST(id),\ @@ -627,6 +640,23 @@ struct link_encoder *dce100_link_encoder_create( return &enc110->base; } +static struct panel_cntl *dce100_panel_cntl_create(const struct panel_cntl_init_data *init_data) +{ + struct dce_panel_cntl *panel_cntl = + kzalloc(sizeof(struct dce_panel_cntl), GFP_KERNEL); + + if (!panel_cntl) + return NULL; + + dce_panel_cntl_construct(panel_cntl, + init_data, + &panel_cntl_regs[init_data->inst], + &panel_cntl_shift, + &panel_cntl_mask); + + return &panel_cntl->base; +} + struct output_pixel_processor *dce100_opp_create( struct dc_context *ctx, uint32_t inst) @@ -943,6 +973,7 @@ struct stream_encoder *dce100_find_first_free_match_stream_enc_for_link( static const struct resource_funcs dce100_res_pool_funcs = { .destroy = dce100_destroy_resource_pool, .link_enc_create = dce100_link_encoder_create, + .panel_cntl_create = dce100_panel_cntl_create, .validate_bandwidth = dce100_validate_bandwidth, .validate_plane = dce100_validate_plane, .add_stream_to_ctx = dce100_add_stream_to_ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c index c279982947e1..b77e9dc16086 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c @@ -53,6 +53,7 @@ #include "abm.h" #include "audio.h" #include "reg_helper.h" +#include "panel_cntl.h" /* include DCE11 register header files */ #include "dce/dce_11_0_d.h" @@ -697,31 +698,6 @@ void dce110_enable_stream(struct pipe_ctx *pipe_ctx) } -/*todo: cloned in stream enc, fix*/ -bool dce110_is_panel_backlight_on(struct dc_link *link) -{ - struct dc_context *ctx = link->ctx; - struct dce_hwseq *hws = ctx->dc->hwseq; - uint32_t value; - - REG_GET(LVTMA_PWRSEQ_CNTL, LVTMA_BLON, &value); - - return value; -} - -bool dce110_is_panel_powered_on(struct dc_link *link) -{ - struct dc_context *ctx = link->ctx; - struct dce_hwseq *hws = ctx->dc->hwseq; - uint32_t pwr_seq_state, dig_on, dig_on_ovrd; - - REG_GET(LVTMA_PWRSEQ_STATE, LVTMA_PWRSEQ_TARGET_STATE_R, &pwr_seq_state); - - REG_GET_2(LVTMA_PWRSEQ_CNTL, LVTMA_DIGON, &dig_on, LVTMA_DIGON_OVRD, &dig_on_ovrd); - - return (pwr_seq_state == 1) || (dig_on == 1 && dig_on_ovrd == 1); -} - static enum bp_result link_transmitter_control( struct dc_bios *bios, struct bp_transmitter_control *cntl) @@ -810,7 +786,6 @@ void dce110_edp_power_control( bool power_up) { struct dc_context *ctx = link->ctx; - struct dce_hwseq *hwseq = ctx->dc->hwseq; struct bp_transmitter_control cntl = { 0 }; enum bp_result bp_result; @@ -821,7 +796,11 @@ void dce110_edp_power_control( return; } - if (power_up != hwseq->funcs.is_panel_powered_on(link)) { + if (!link->panel_cntl) + return; + + if (power_up != + link->panel_cntl->funcs->is_panel_powered_on(link->panel_cntl)) { /* Send VBIOS command to prompt eDP panel power */ if (power_up) { unsigned long long current_ts = dm_get_timestamp(ctx); @@ -892,7 +871,6 @@ void dce110_edp_backlight_control( bool enable) { struct dc_context *ctx = link->ctx; - struct dce_hwseq *hws = ctx->dc->hwseq; struct bp_transmitter_control cntl = { 0 }; if (dal_graphics_object_id_get_connector_id(link->link_enc->connector) @@ -901,7 +879,8 @@ void dce110_edp_backlight_control( return; } - if (enable && hws->funcs.is_panel_backlight_on(link)) { + if (enable && link->panel_cntl && + link->panel_cntl->funcs->is_panel_backlight_on(link->panel_cntl)) { DC_LOG_HW_RESUME_S3( "%s: panel already powered up. Do nothing.\n", __func__); @@ -1087,7 +1066,7 @@ void dce110_blank_stream(struct pipe_ctx *pipe_ctx) if (link->local_sink && link->local_sink->sink_signal == SIGNAL_TYPE_EDP) { hws->funcs.edp_backlight_control(link, false); - dc_link_set_abm_disable(link); + link->dc->hwss.set_abm_immediate_disable(pipe_ctx); } if (dc_is_dp_signal(pipe_ctx->stream->signal)) @@ -1432,7 +1411,7 @@ static enum dc_status apply_single_controller_ctx_to_hw( pipe_ctx->plane_res.scl_data.lb_params.alpha_en = pipe_ctx->bottom_pipe != 0; - pipe_ctx->stream->link->psr_feature_enabled = false; + pipe_ctx->stream->link->psr_settings.psr_feature_enabled = false; return DC_OK; } @@ -1838,7 +1817,7 @@ static bool should_enable_fbc(struct dc *dc, return false; /* PSR should not be enabled */ - if (pipe_ctx->stream->link->psr_feature_enabled) + if (pipe_ctx->stream->link->psr_settings.psr_feature_enabled) return false; /* Nothing to compress */ @@ -2376,6 +2355,7 @@ static void init_hw(struct dc *dc) struct abm *abm; struct dmcu *dmcu; struct dce_hwseq *hws = dc->hwseq; + uint32_t backlight = MAX_BACKLIGHT_LEVEL; bp = dc->ctx->dc_bios; for (i = 0; i < dc->res_pool->pipe_count; i++) { @@ -2422,12 +2402,17 @@ static void init_hw(struct dc *dc) audio->funcs->hw_init(audio); } - abm = dc->res_pool->abm; - if (abm != NULL) { - abm->funcs->init_backlight(abm); - abm->funcs->abm_init(abm); + for (i = 0; i < dc->link_count; i++) { + struct dc_link *link = dc->links[i]; + + if (link->panel_cntl) + backlight = link->panel_cntl->funcs->hw_init(link->panel_cntl); } + abm = dc->res_pool->abm; + if (abm != NULL) + abm->funcs->abm_init(abm, backlight); + dmcu = dc->res_pool->dmcu; if (dmcu != NULL && abm != NULL) abm->dmcu_is_running = dmcu->funcs->is_dmcu_initialized(dmcu); @@ -2735,6 +2720,53 @@ void dce110_set_cursor_attribute(struct pipe_ctx *pipe_ctx) pipe_ctx->plane_res.xfm, attributes); } +bool dce110_set_backlight_level(struct pipe_ctx *pipe_ctx, + uint32_t backlight_pwm_u16_16, + uint32_t frame_ramp) +{ + struct dc_link *link = pipe_ctx->stream->link; + struct dc *dc = link->ctx->dc; + struct abm *abm = pipe_ctx->stream_res.abm; + struct panel_cntl *panel_cntl = link->panel_cntl; + struct dmcu *dmcu = dc->res_pool->dmcu; + bool fw_set_brightness = true; + /* DMCU -1 for all controller id values, + * therefore +1 here + */ + uint32_t controller_id = pipe_ctx->stream_res.tg->inst + 1; + + if (abm == NULL || panel_cntl == NULL || (abm->funcs->set_backlight_level_pwm == NULL)) + return false; + + if (dmcu) + fw_set_brightness = dmcu->funcs->is_dmcu_initialized(dmcu); + + if (!fw_set_brightness && panel_cntl->funcs->driver_set_backlight) + panel_cntl->funcs->driver_set_backlight(panel_cntl, backlight_pwm_u16_16); + else + abm->funcs->set_backlight_level_pwm( + abm, + backlight_pwm_u16_16, + frame_ramp, + controller_id, + link->panel_cntl->inst); + + return true; +} + +void dce110_set_abm_immediate_disable(struct pipe_ctx *pipe_ctx) +{ + struct abm *abm = pipe_ctx->stream_res.abm; + struct panel_cntl *panel_cntl = pipe_ctx->stream->link->panel_cntl; + + if (abm) + abm->funcs->set_abm_immediate_disable(abm, + pipe_ctx->stream->link->panel_cntl->inst); + + if (panel_cntl) + panel_cntl->funcs->store_backlight_level(panel_cntl); +} + static const struct hw_sequencer_funcs dce110_funcs = { .program_gamut_remap = program_gamut_remap, .program_output_csc = program_output_csc, @@ -2757,6 +2789,7 @@ static const struct hw_sequencer_funcs dce110_funcs = { .disable_plane = dce110_power_down_fe, .pipe_control_lock = dce_pipe_control_lock, .interdependent_update_lock = NULL, + .cursor_lock = dce_pipe_control_lock, .prepare_bandwidth = dce110_prepare_bandwidth, .optimize_bandwidth = dce110_optimize_bandwidth, .set_drr = set_drr, @@ -2768,7 +2801,9 @@ static const struct hw_sequencer_funcs dce110_funcs = { .edp_power_control = dce110_edp_power_control, .edp_wait_for_hpd_ready = dce110_edp_wait_for_hpd_ready, .set_cursor_position = dce110_set_cursor_position, - .set_cursor_attribute = dce110_set_cursor_attribute + .set_cursor_attribute = dce110_set_cursor_attribute, + .set_backlight_level = dce110_set_backlight_level, + .set_abm_immediate_disable = dce110_set_abm_immediate_disable, }; static const struct hwseq_private_funcs dce110_private_funcs = { @@ -2784,8 +2819,6 @@ static const struct hwseq_private_funcs dce110_private_funcs = { .disable_stream_gating = NULL, .enable_stream_gating = NULL, .edp_backlight_control = dce110_edp_backlight_control, - .is_panel_backlight_on = dce110_is_panel_backlight_on, - .is_panel_powered_on = dce110_is_panel_powered_on, }; void dce110_hw_sequencer_construct(struct dc *dc) diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h index 34be166e8ff0..fe5326df00f7 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h @@ -85,9 +85,10 @@ void dce110_edp_wait_for_hpd_ready( struct dc_link *link, bool power_up); -bool dce110_is_panel_backlight_on(struct dc_link *link); - -bool dce110_is_panel_powered_on(struct dc_link *link); +bool dce110_set_backlight_level(struct pipe_ctx *pipe_ctx, + uint32_t backlight_pwm_u16_16, + uint32_t frame_ramp); +void dce110_set_abm_immediate_disable(struct pipe_ctx *pipe_ctx); #endif /* __DC_HWSS_DCE110_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_opp_csc_v.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_opp_csc_v.c index 4245e1f818a3..e096d2b95ef9 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_opp_csc_v.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_opp_csc_v.c @@ -679,8 +679,7 @@ void dce110_opp_v_set_csc_default( if (default_adjust->force_hw_default == false) { const struct out_csc_color_matrix *elm; /* currently parameter not in use */ - enum grph_color_adjust_option option = - GRPH_COLOR_MATRIX_HW_DEFAULT; + enum grph_color_adjust_option option; uint32_t i; /* * HW default false we program locally defined matrix diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c index bf14e9ab040c..9597fc79d7fa 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c @@ -53,6 +53,7 @@ #include "dce/dce_abm.h" #include "dce/dce_dmcu.h" #include "dce/dce_i2c.h" +#include "dce/dce_panel_cntl.h" #define DC_LOGGER \ dc->ctx->logger @@ -275,6 +276,18 @@ static const struct dce_stream_encoder_mask se_mask = { SE_COMMON_MASK_SH_LIST_DCE110(_MASK) }; +static const struct dce_panel_cntl_registers panel_cntl_regs[] = { + { DCE_PANEL_CNTL_REG_LIST() } +}; + +static const struct dce_panel_cntl_shift panel_cntl_shift = { + DCE_PANEL_CNTL_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce_panel_cntl_mask panel_cntl_mask = { + DCE_PANEL_CNTL_MASK_SH_LIST(_MASK) +}; + static const struct dce110_aux_registers_shift aux_shift = { DCE_AUX_MASK_SH_LIST(__SHIFT) }; @@ -673,6 +686,23 @@ static struct link_encoder *dce110_link_encoder_create( return &enc110->base; } +static struct panel_cntl *dce110_panel_cntl_create(const struct panel_cntl_init_data *init_data) +{ + struct dce_panel_cntl *panel_cntl = + kzalloc(sizeof(struct dce_panel_cntl), GFP_KERNEL); + + if (!panel_cntl) + return NULL; + + dce_panel_cntl_construct(panel_cntl, + init_data, + &panel_cntl_regs[init_data->inst], + &panel_cntl_shift, + &panel_cntl_mask); + + return &panel_cntl->base; +} + static struct output_pixel_processor *dce110_opp_create( struct dc_context *ctx, uint32_t inst) @@ -1203,6 +1233,7 @@ struct stream_encoder *dce110_find_first_free_match_stream_enc_for_link( static const struct resource_funcs dce110_res_pool_funcs = { .destroy = dce110_destroy_resource_pool, .link_enc_create = dce110_link_encoder_create, + .panel_cntl_create = dce110_panel_cntl_create, .validate_bandwidth = dce110_validate_bandwidth, .validate_plane = dce110_validate_plane, .acquire_idle_pipe_for_layer = dce110_acquire_underlay, diff --git a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c index 700ad8b3e54b..51b3fe502670 100644 --- a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c @@ -51,6 +51,7 @@ #include "dce/dce_dmcu.h" #include "dce/dce_aux.h" #include "dce/dce_i2c.h" +#include "dce/dce_panel_cntl.h" #include "reg_helper.h" @@ -238,6 +239,18 @@ static const struct dce110_link_enc_aux_registers link_enc_aux_regs[] = { aux_regs(5) }; +static const struct dce_panel_cntl_registers panel_cntl_regs[] = { + { DCE_PANEL_CNTL_REG_LIST() } +}; + +static const struct dce_panel_cntl_shift panel_cntl_shift = { + DCE_PANEL_CNTL_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce_panel_cntl_mask panel_cntl_mask = { + DCE_PANEL_CNTL_MASK_SH_LIST(_MASK) +}; + #define hpd_regs(id)\ [id] = {\ HPD_REG_LIST(id)\ @@ -398,7 +411,7 @@ static const struct dc_plane_cap plane_cap = { .pixel_format_support = { .argb8888 = true, .nv12 = false, - .fp16 = false + .fp16 = true }, .max_upscale_factor = { @@ -631,6 +644,23 @@ struct link_encoder *dce112_link_encoder_create( return &enc110->base; } +static struct panel_cntl *dce112_panel_cntl_create(const struct panel_cntl_init_data *init_data) +{ + struct dce_panel_cntl *panel_cntl = + kzalloc(sizeof(struct dce_panel_cntl), GFP_KERNEL); + + if (!panel_cntl) + return NULL; + + dce_panel_cntl_construct(panel_cntl, + init_data, + &panel_cntl_regs[init_data->inst], + &panel_cntl_shift, + &panel_cntl_mask); + + return &panel_cntl->base; +} + static struct input_pixel_processor *dce112_ipp_create( struct dc_context *ctx, uint32_t inst) { @@ -1021,6 +1051,7 @@ static void dce112_destroy_resource_pool(struct resource_pool **pool) static const struct resource_funcs dce112_res_pool_funcs = { .destroy = dce112_destroy_resource_pool, .link_enc_create = dce112_link_encoder_create, + .panel_cntl_create = dce112_panel_cntl_create, .validate_bandwidth = dce112_validate_bandwidth, .validate_plane = dce100_validate_plane, .add_stream_to_ctx = dce112_add_stream_to_ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c index 53ab88ef71f5..8f362e8c1787 100644 --- a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c @@ -44,6 +44,7 @@ #include "dce/dce_clock_source.h" #include "dce/dce_ipp.h" #include "dce/dce_mem_input.h" +#include "dce/dce_panel_cntl.h" #include "dce110/dce110_hw_sequencer.h" #include "dce120/dce120_hw_sequencer.h" @@ -293,6 +294,18 @@ static const struct dce_stream_encoder_mask se_mask = { SE_COMMON_MASK_SH_LIST_DCE120(_MASK) }; +static const struct dce_panel_cntl_registers panel_cntl_regs[] = { + { DCE_PANEL_CNTL_REG_LIST() } +}; + +static const struct dce_panel_cntl_shift panel_cntl_shift = { + DCE_PANEL_CNTL_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce_panel_cntl_mask panel_cntl_mask = { + DCE_PANEL_CNTL_MASK_SH_LIST(_MASK) +}; + static const struct dce110_aux_registers_shift aux_shift = { DCE12_AUX_MASK_SH_LIST(__SHIFT) }; @@ -503,7 +516,7 @@ static const struct dc_plane_cap plane_cap = { .pixel_format_support = { .argb8888 = true, .nv12 = false, - .fp16 = false + .fp16 = true }, .max_upscale_factor = { @@ -715,6 +728,23 @@ static struct link_encoder *dce120_link_encoder_create( return &enc110->base; } +static struct panel_cntl *dce120_panel_cntl_create(const struct panel_cntl_init_data *init_data) +{ + struct dce_panel_cntl *panel_cntl = + kzalloc(sizeof(struct dce_panel_cntl), GFP_KERNEL); + + if (!panel_cntl) + return NULL; + + dce_panel_cntl_construct(panel_cntl, + init_data, + &panel_cntl_regs[init_data->inst], + &panel_cntl_shift, + &panel_cntl_mask); + + return &panel_cntl->base; +} + static struct input_pixel_processor *dce120_ipp_create( struct dc_context *ctx, uint32_t inst) { @@ -880,6 +910,7 @@ static void dce120_destroy_resource_pool(struct resource_pool **pool) static const struct resource_funcs dce120_res_pool_funcs = { .destroy = dce120_destroy_resource_pool, .link_enc_create = dce120_link_encoder_create, + .panel_cntl_create = dce120_panel_cntl_create, .validate_bandwidth = dce112_validate_bandwidth, .validate_plane = dce100_validate_plane, .add_stream_to_ctx = dce112_add_stream_to_ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dce80/dce80_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce80/dce80_hw_sequencer.c index 893261c81854..d2ceebdbdf51 100644 --- a/drivers/gpu/drm/amd/display/dc/dce80/dce80_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce80/dce80_hw_sequencer.c @@ -36,34 +36,6 @@ #include "dce/dce_8_0_d.h" #include "dce/dce_8_0_sh_mask.h" -struct dce80_hw_seq_reg_offsets { - uint32_t crtc; -}; - -static const struct dce80_hw_seq_reg_offsets reg_offsets[] = { -{ - .crtc = (mmCRTC0_CRTC_GSL_CONTROL - mmCRTC_GSL_CONTROL), -}, -{ - .crtc = (mmCRTC1_CRTC_GSL_CONTROL - mmCRTC_GSL_CONTROL), -}, -{ - .crtc = (mmCRTC2_CRTC_GSL_CONTROL - mmCRTC_GSL_CONTROL), -}, -{ - .crtc = (mmCRTC3_CRTC_GSL_CONTROL - mmCRTC_GSL_CONTROL), -}, -{ - .crtc = (mmCRTC4_CRTC_GSL_CONTROL - mmCRTC_GSL_CONTROL), -}, -{ - .crtc = (mmCRTC5_CRTC_GSL_CONTROL - mmCRTC_GSL_CONTROL), -} -}; - -#define HW_REG_CRTC(reg, id)\ - (reg + reg_offsets[id].crtc) - /******************************************************************************* * Private definitions ******************************************************************************/ diff --git a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c index 2ad5c28c6e66..a19be9de2df7 100644 --- a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c @@ -50,6 +50,7 @@ #include "dce/dce_hwseq.h" #include "dce80/dce80_hw_sequencer.h" #include "dce100/dce100_resource.h" +#include "dce/dce_panel_cntl.h" #include "reg_helper.h" @@ -266,6 +267,18 @@ static const struct dce_stream_encoder_mask se_mask = { SE_COMMON_MASK_SH_LIST_DCE80_100(_MASK) }; +static const struct dce_panel_cntl_registers panel_cntl_regs[] = { + { DCE_PANEL_CNTL_REG_LIST() } +}; + +static const struct dce_panel_cntl_shift panel_cntl_shift = { + DCE_PANEL_CNTL_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce_panel_cntl_mask panel_cntl_mask = { + DCE_PANEL_CNTL_MASK_SH_LIST(_MASK) +}; + #define opp_regs(id)\ [id] = {\ OPP_DCE_80_REG_LIST(id),\ @@ -728,6 +741,23 @@ struct link_encoder *dce80_link_encoder_create( return &enc110->base; } +static struct panel_cntl *dce80_panel_cntl_create(const struct panel_cntl_init_data *init_data) +{ + struct dce_panel_cntl *panel_cntl = + kzalloc(sizeof(struct dce_panel_cntl), GFP_KERNEL); + + if (!panel_cntl) + return NULL; + + dce_panel_cntl_construct(panel_cntl, + init_data, + &panel_cntl_regs[init_data->inst], + &panel_cntl_shift, + &panel_cntl_mask); + + return &panel_cntl->base; +} + struct clock_source *dce80_clock_source_create( struct dc_context *ctx, struct dc_bios *bios, @@ -909,6 +939,7 @@ static void dce80_destroy_resource_pool(struct resource_pool **pool) static const struct resource_funcs dce80_res_pool_funcs = { .destroy = dce80_destroy_resource_pool, .link_enc_create = dce80_link_encoder_create, + .panel_cntl_create = dce80_panel_cntl_create, .validate_bandwidth = dce80_validate_bandwidth, .validate_plane = dce100_validate_plane, .add_stream_to_ctx = dce100_add_stream_to_ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c index 0e682b5aa3eb..7f8456b9988b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c @@ -134,13 +134,6 @@ bool dpp1_get_optimal_number_of_taps( struct scaler_data *scl_data, const struct scaling_taps *in_taps) { - uint32_t pixel_width; - - if (scl_data->viewport.width > scl_data->recout.width) - pixel_width = scl_data->recout.width; - else - pixel_width = scl_data->viewport.width; - /* Some ASICs does not support FP16 scaling, so we reject modes require this*/ if (scl_data->format == PIXEL_FORMAT_FP16 && dpp->caps->dscl_data_proc_format == DSCL_DATA_PRCESSING_FIXED_FORMAT && diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c index deccab0228d2..75637c291e75 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c @@ -93,7 +93,6 @@ void hubbub1_wm_read_state(struct hubbub *hubbub, void hubbub1_allow_self_refresh_control(struct hubbub *hubbub, bool allow) { struct dcn10_hubbub *hubbub1 = TO_DCN10_HUBBUB(hubbub); - /* * DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE = 1 means do not allow stutter * DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE = 0 means allow stutter diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c index 31b64733d693..319366ebb44f 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c @@ -1139,6 +1139,8 @@ void hubp1_cursor_set_position( int src_y_offset = pos->y - pos->y_hotspot - param->viewport.y; int x_hotspot = pos->x_hotspot; int y_hotspot = pos->y_hotspot; + int cursor_height = (int)hubp->curs_attr.height; + int cursor_width = (int)hubp->curs_attr.width; uint32_t dst_x_offset; uint32_t cur_en = pos->enable ? 1 : 0; @@ -1152,10 +1154,16 @@ void hubp1_cursor_set_position( if (hubp->curs_attr.address.quad_part == 0) return; + // Rotated cursor width/height and hotspots tweaks for offset calculation if (param->rotation == ROTATION_ANGLE_90 || param->rotation == ROTATION_ANGLE_270) { - src_x_offset = pos->y - pos->y_hotspot - param->viewport.x; - y_hotspot = pos->x_hotspot; - x_hotspot = pos->y_hotspot; + swap(cursor_height, cursor_width); + if (param->rotation == ROTATION_ANGLE_90) { + src_x_offset = pos->x - pos->y_hotspot - param->viewport.x; + src_y_offset = pos->y - pos->x_hotspot - param->viewport.y; + } + } else if (param->rotation == ROTATION_ANGLE_180) { + src_x_offset = pos->x - param->viewport.x; + src_y_offset = pos->y - param->viewport.y; } if (param->mirror) { @@ -1177,13 +1185,13 @@ void hubp1_cursor_set_position( if (src_x_offset >= (int)param->viewport.width) cur_en = 0; /* not visible beyond right edge*/ - if (src_x_offset + (int)hubp->curs_attr.width <= 0) + if (src_x_offset + cursor_width <= 0) cur_en = 0; /* not visible beyond left edge*/ if (src_y_offset >= (int)param->viewport.height) cur_en = 0; /* not visible beyond bottom edge*/ - if (src_y_offset + (int)hubp->curs_attr.height <= 0) + if (src_y_offset + cursor_height <= 0) cur_en = 0; /* not visible beyond top edge*/ if (cur_en && REG_READ(CURSOR_SURFACE_ADDRESS) == 0) diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c index b0357546471b..77f16921e7f0 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c @@ -737,7 +737,8 @@ void dcn10_bios_golden_init(struct dc *dc) if (dc->res_pool->hubbub->funcs->allow_self_refresh_control) if (allow_self_fresh_force_enable == false && dc->res_pool->hubbub->funcs->is_allow_self_refresh_enabled(dc->res_pool->hubbub)) - dc->res_pool->hubbub->funcs->allow_self_refresh_control(dc->res_pool->hubbub, true); + dc->res_pool->hubbub->funcs->allow_self_refresh_control(dc->res_pool->hubbub, + !dc->res_pool->hubbub->ctx->dc->debug.disable_stutter); } @@ -826,6 +827,14 @@ enum dc_status dcn10_enable_stream_timing( color_space = stream->output_color_space; color_space_to_black_color(dc, color_space, &black_color); + /* + * The way 420 is packed, 2 channels carry Y component, 1 channel + * alternate between Cb and Cr, so both channels need the pixel + * value for Y + */ + if (stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR420) + black_color.color_r_cr = black_color.color_g_y; + if (pipe_ctx->stream_res.tg->funcs->set_blank_color) pipe_ctx->stream_res.tg->funcs->set_blank_color( pipe_ctx->stream_res.tg, @@ -903,7 +912,7 @@ static void dcn10_reset_back_end_for_pipe( if (pipe_ctx->top_pipe == NULL) { if (pipe_ctx->stream_res.abm) - pipe_ctx->stream_res.abm->funcs->set_abm_immediate_disable(pipe_ctx->stream_res.abm); + dc->hwss.set_abm_immediate_disable(pipe_ctx); pipe_ctx->stream_res.tg->funcs->disable_crtc(pipe_ctx->stream_res.tg); @@ -1238,12 +1247,13 @@ void dcn10_init_pipes(struct dc *dc, struct dc_state *context) void dcn10_init_hw(struct dc *dc) { - int i; + int i, j; struct abm *abm = dc->res_pool->abm; struct dmcu *dmcu = dc->res_pool->dmcu; struct dce_hwseq *hws = dc->hwseq; struct dc_bios *dcb = dc->ctx->dc_bios; struct resource_pool *res_pool = dc->res_pool; + uint32_t backlight = MAX_BACKLIGHT_LEVEL; if (dc->clk_mgr && dc->clk_mgr->funcs->init_clocks) dc->clk_mgr->funcs->init_clocks(dc->clk_mgr); @@ -1333,17 +1343,28 @@ void dcn10_init_hw(struct dc *dc) continue; /* - * core_link_read_dpcd() will invoke dm_helpers_dp_read_dpcd(), - * which needs to read dpcd info with the help of aconnector. - * If aconnector (dc->links[i]->prev) is NULL, then dpcd status - * cannot be read. + * If any of the displays are lit up turn them off. + * The reason is that some MST hubs cannot be turned off + * completely until we tell them to do so. + * If not turned off, then displays connected to MST hub + * won't light up. */ - if (dc->links[i]->priv) { - /* if any of the displays are lit up turn them off */ - status = core_link_read_dpcd(dc->links[i], DP_SET_POWER, - &dpcd_power_state, sizeof(dpcd_power_state)); - if (status == DC_OK && dpcd_power_state == DP_POWER_STATE_D0) - dp_receiver_power_ctrl(dc->links[i], false); + status = core_link_read_dpcd(dc->links[i], DP_SET_POWER, + &dpcd_power_state, sizeof(dpcd_power_state)); + if (status == DC_OK && dpcd_power_state == DP_POWER_STATE_D0) { + /* blank dp stream before power off receiver*/ + if (dc->links[i]->link_enc->funcs->get_dig_frontend) { + unsigned int fe = dc->links[i]->link_enc->funcs->get_dig_frontend(dc->links[i]->link_enc); + + for (j = 0; j < dc->res_pool->stream_enc_count; j++) { + if (fe == dc->res_pool->stream_enc[j]->id) { + dc->res_pool->stream_enc[j]->funcs->dp_blank( + dc->res_pool->stream_enc[j]); + break; + } + } + } + dp_receiver_power_ctrl(dc->links[i], false); } } } @@ -1361,17 +1382,54 @@ void dcn10_init_hw(struct dc *dc) !dc->res_pool->hubbub->ctx->dc->debug.disable_stutter); } + /* In headless boot cases, DIG may be turned + * on which causes HW/SW discrepancies. + * To avoid this, power down hardware on boot + * if DIG is turned on and seamless boot not enabled + */ + if (dc->config.power_down_display_on_boot) { + struct dc_link *edp_link = get_edp_link(dc); + + if (edp_link && + edp_link->link_enc->funcs->is_dig_enabled && + edp_link->link_enc->funcs->is_dig_enabled(edp_link->link_enc) && + dc->hwss.edp_backlight_control && + dc->hwss.power_down && + dc->hwss.edp_power_control) { + dc->hwss.edp_backlight_control(edp_link, false); + dc->hwss.power_down(dc); + dc->hwss.edp_power_control(edp_link, false); + } else { + for (i = 0; i < dc->link_count; i++) { + struct dc_link *link = dc->links[i]; + + if (link->link_enc->funcs->is_dig_enabled && + link->link_enc->funcs->is_dig_enabled(link->link_enc) && + dc->hwss.power_down) { + dc->hwss.power_down(dc); + break; + } + + } + } + } + for (i = 0; i < res_pool->audio_count; i++) { struct audio *audio = res_pool->audios[i]; audio->funcs->hw_init(audio); } - if (abm != NULL) { - abm->funcs->init_backlight(abm); - abm->funcs->abm_init(abm); + for (i = 0; i < dc->link_count; i++) { + struct dc_link *link = dc->links[i]; + + if (link->panel_cntl) + backlight = link->panel_cntl->funcs->hw_init(link->panel_cntl); } + if (abm != NULL) + abm->funcs->abm_init(abm, backlight); + if (dmcu != NULL && !dmcu->auto_load_dmcu) dmcu->funcs->dmcu_init(dmcu); @@ -1625,6 +1683,85 @@ void dcn10_pipe_control_lock( hws->funcs.verify_allow_pstate_change_high(dc); } +/** + * delay_cursor_until_vupdate() - Delay cursor update if too close to VUPDATE. + * + * Software keepout workaround to prevent cursor update locking from stalling + * out cursor updates indefinitely or from old values from being retained in + * the case where the viewport changes in the same frame as the cursor. + * + * The idea is to calculate the remaining time from VPOS to VUPDATE. If it's + * too close to VUPDATE, then stall out until VUPDATE finishes. + * + * TODO: Optimize cursor programming to be once per frame before VUPDATE + * to avoid the need for this workaround. + */ +static void delay_cursor_until_vupdate(struct dc *dc, struct pipe_ctx *pipe_ctx) +{ + struct dc_stream_state *stream = pipe_ctx->stream; + struct crtc_position position; + uint32_t vupdate_start, vupdate_end; + unsigned int lines_to_vupdate, us_to_vupdate, vpos; + unsigned int us_per_line, us_vupdate; + + if (!dc->hwss.calc_vupdate_position || !dc->hwss.get_position) + return; + + if (!pipe_ctx->stream_res.stream_enc || !pipe_ctx->stream_res.tg) + return; + + dc->hwss.calc_vupdate_position(dc, pipe_ctx, &vupdate_start, + &vupdate_end); + + dc->hwss.get_position(&pipe_ctx, 1, &position); + vpos = position.vertical_count; + + /* Avoid wraparound calculation issues */ + vupdate_start += stream->timing.v_total; + vupdate_end += stream->timing.v_total; + vpos += stream->timing.v_total; + + if (vpos <= vupdate_start) { + /* VPOS is in VACTIVE or back porch. */ + lines_to_vupdate = vupdate_start - vpos; + } else if (vpos > vupdate_end) { + /* VPOS is in the front porch. */ + return; + } else { + /* VPOS is in VUPDATE. */ + lines_to_vupdate = 0; + } + + /* Calculate time until VUPDATE in microseconds. */ + us_per_line = + stream->timing.h_total * 10000u / stream->timing.pix_clk_100hz; + us_to_vupdate = lines_to_vupdate * us_per_line; + + /* 70 us is a conservative estimate of cursor update time*/ + if (us_to_vupdate > 70) + return; + + /* Stall out until the cursor update completes. */ + if (vupdate_end < vupdate_start) + vupdate_end += stream->timing.v_total; + us_vupdate = (vupdate_end - vupdate_start + 1) * us_per_line; + udelay(us_to_vupdate + us_vupdate); +} + +void dcn10_cursor_lock(struct dc *dc, struct pipe_ctx *pipe, bool lock) +{ + /* cursor lock is per MPCC tree, so only need to lock one pipe per stream */ + if (!pipe || pipe->top_pipe) + return; + + /* Prevent cursor lock from stalling out cursor updates. */ + if (lock) + delay_cursor_until_vupdate(dc, pipe); + + dc->res_pool->mpc->funcs->cursor_lock(dc->res_pool->mpc, + pipe->stream_res.opp->inst, lock); +} + static bool wait_for_reset_trigger_to_occur( struct dc_context *dc_ctx, struct timing_generator *tg) @@ -2085,25 +2222,25 @@ void dcn10_get_surface_visual_confirm_color( switch (pipe_ctx->plane_res.scl_data.format) { case PIXEL_FORMAT_ARGB8888: - /* set boarder color to red */ + /* set border color to red */ color->color_r_cr = color_value; break; case PIXEL_FORMAT_ARGB2101010: - /* set boarder color to blue */ + /* set border color to blue */ color->color_b_cb = color_value; break; case PIXEL_FORMAT_420BPP8: - /* set boarder color to green */ + /* set border color to green */ color->color_g_y = color_value; break; case PIXEL_FORMAT_420BPP10: - /* set boarder color to yellow */ + /* set border color to yellow */ color->color_g_y = color_value; color->color_r_cr = color_value; break; case PIXEL_FORMAT_FP16: - /* set boarder color to white */ + /* set border color to white */ color->color_r_cr = color_value; color->color_b_cb = color_value; color->color_g_y = color_value; @@ -2128,25 +2265,25 @@ void dcn10_get_hdr_visual_confirm_color( switch (top_pipe_ctx->plane_res.scl_data.format) { case PIXEL_FORMAT_ARGB2101010: if (top_pipe_ctx->stream->out_transfer_func->tf == TRANSFER_FUNCTION_PQ) { - /* HDR10, ARGB2101010 - set boarder color to red */ + /* HDR10, ARGB2101010 - set border color to red */ color->color_r_cr = color_value; } else if (top_pipe_ctx->stream->out_transfer_func->tf == TRANSFER_FUNCTION_GAMMA22) { - /* FreeSync 2 ARGB2101010 - set boarder color to pink */ + /* FreeSync 2 ARGB2101010 - set border color to pink */ color->color_r_cr = color_value; color->color_b_cb = color_value; } break; case PIXEL_FORMAT_FP16: if (top_pipe_ctx->stream->out_transfer_func->tf == TRANSFER_FUNCTION_PQ) { - /* HDR10, FP16 - set boarder color to blue */ + /* HDR10, FP16 - set border color to blue */ color->color_b_cb = color_value; } else if (top_pipe_ctx->stream->out_transfer_func->tf == TRANSFER_FUNCTION_GAMMA22) { - /* FreeSync 2 HDR - set boarder color to green */ + /* FreeSync 2 HDR - set border color to green */ color->color_g_y = color_value; } break; default: - /* SDR - set boarder color to Gray */ + /* SDR - set border color to Gray */ color->color_r_cr = color_value/2; color->color_b_cb = color_value/2; color->color_g_y = color_value/2; @@ -2195,6 +2332,14 @@ void dcn10_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) &blnd_cfg.black_color); } + /* + * The way 420 is packed, 2 channels carry Y component, 1 channel + * alternate between Cb and Cr, so both channels need the pixel + * value for Y + */ + if (pipe_ctx->stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR420) + blnd_cfg.black_color.color_r_cr = blnd_cfg.black_color.color_g_y; + if (per_pixel_alpha) blnd_cfg.alpha_mode = MPCC_ALPHA_BLEND_MODE_PER_PIXEL_ALPHA; else @@ -2431,12 +2576,12 @@ void dcn10_blank_pixel_data( if (stream_res->tg->funcs->set_blank) stream_res->tg->funcs->set_blank(stream_res->tg, blank); if (stream_res->abm) { - stream_res->abm->funcs->set_pipe(stream_res->abm, stream_res->tg->inst + 1); + stream_res->abm->funcs->set_pipe(stream_res->abm, stream_res->tg->inst + 1, + stream->link->panel_cntl->inst); stream_res->abm->funcs->set_abm_level(stream_res->abm, stream->abm_level); } } else if (blank) { - if (stream_res->abm) - stream_res->abm->funcs->set_abm_immediate_disable(stream_res->abm); + dc->hwss.set_abm_immediate_disable(pipe_ctx); if (stream_res->tg->funcs->set_blank) stream_res->tg->funcs->set_blank(stream_res->tg, blank); } @@ -3226,7 +3371,7 @@ int dcn10_get_vupdate_offset_from_vsync(struct pipe_ctx *pipe_ctx) return vertical_line_start; } -static void dcn10_calc_vupdate_position( +void dcn10_calc_vupdate_position( struct dc *dc, struct pipe_ctx *pipe_ctx, uint32_t *start_line, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h index 16a50e05ffbf..42b6e016d71e 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h @@ -34,6 +34,11 @@ struct dc; void dcn10_hw_sequencer_construct(struct dc *dc); int dcn10_get_vupdate_offset_from_vsync(struct pipe_ctx *pipe_ctx); +void dcn10_calc_vupdate_position( + struct dc *dc, + struct pipe_ctx *pipe_ctx, + uint32_t *start_line, + uint32_t *end_line); void dcn10_setup_vupdate_interrupt(struct dc *dc, struct pipe_ctx *pipe_ctx); enum dc_status dcn10_enable_stream_timing( struct pipe_ctx *pipe_ctx, @@ -49,6 +54,7 @@ void dcn10_pipe_control_lock( struct dc *dc, struct pipe_ctx *pipe, bool lock); +void dcn10_cursor_lock(struct dc *dc, struct pipe_ctx *pipe, bool lock); void dcn10_blank_pixel_data( struct dc *dc, struct pipe_ctx *pipe_ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.c index dd02d3983695..7cb8c3fb2665 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_init.c @@ -50,6 +50,7 @@ static const struct hw_sequencer_funcs dcn10_funcs = { .disable_audio_stream = dce110_disable_audio_stream, .disable_plane = dcn10_disable_plane, .pipe_control_lock = dcn10_pipe_control_lock, + .cursor_lock = dcn10_cursor_lock, .interdependent_update_lock = dcn10_lock_all_pipes, .prepare_bandwidth = dcn10_prepare_bandwidth, .optimize_bandwidth = dcn10_optimize_bandwidth, @@ -71,6 +72,9 @@ static const struct hw_sequencer_funcs dcn10_funcs = { .set_clock = dcn10_set_clock, .get_clock = dcn10_get_clock, .get_vupdate_offset_from_vsync = dcn10_get_vupdate_offset_from_vsync, + .calc_vupdate_position = dcn10_calc_vupdate_position, + .set_backlight_level = dce110_set_backlight_level, + .set_abm_immediate_disable = dce110_set_abm_immediate_disable, }; static const struct hwseq_private_funcs dcn10_private_funcs = { @@ -87,8 +91,6 @@ static const struct hwseq_private_funcs dcn10_private_funcs = { .reset_hw_ctx_wrap = dcn10_reset_hw_ctx_wrap, .enable_stream_timing = dcn10_enable_stream_timing, .edp_backlight_control = dce110_edp_backlight_control, - .is_panel_backlight_on = dce110_is_panel_backlight_on, - .is_panel_powered_on = dce110_is_panel_powered_on, .disable_stream_gating = NULL, .enable_stream_gating = NULL, .setup_vupdate_interrupt = dcn10_setup_vupdate_interrupt, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c index d3617d6785a7..7fd385be3f3d 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c @@ -90,7 +90,8 @@ static const struct link_encoder_funcs dcn10_lnk_enc_funcs = { .is_dig_enabled = dcn10_is_dig_enabled, .get_dig_frontend = dcn10_get_dig_frontend, .get_dig_mode = dcn10_get_dig_mode, - .destroy = dcn10_link_encoder_destroy + .destroy = dcn10_link_encoder_destroy, + .get_max_link_cap = dcn10_link_encoder_get_max_link_cap, }; static enum bp_result link_transmitter_control( @@ -1370,7 +1371,6 @@ void dcn10_link_encoder_disable_hpd(struct link_encoder *enc) DC_HPD_EN, 0); } - #define AUX_REG(reg)\ (enc10->aux_regs->reg) @@ -1425,3 +1425,19 @@ enum signal_type dcn10_get_dig_mode( return SIGNAL_TYPE_NONE; } +void dcn10_link_encoder_get_max_link_cap(struct link_encoder *enc, + struct dc_link_settings *link_settings) +{ + /* Set Default link settings */ + struct dc_link_settings max_link_cap = {LANE_COUNT_FOUR, LINK_RATE_HIGH, + LINK_SPREAD_05_DOWNSPREAD_30KHZ, false, 0}; + + /* Higher link settings based on feature supported */ + if (enc->features.flags.bits.IS_HBR2_CAPABLE) + max_link_cap.link_rate = LINK_RATE_HIGH2; + + if (enc->features.flags.bits.IS_HBR3_CAPABLE) + max_link_cap.link_rate = LINK_RATE_HIGH3; + + *link_settings = max_link_cap; +} diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h index 762109174fb8..68395bcc24fd 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h @@ -575,4 +575,7 @@ void dcn10_aux_initialize(struct dcn10_link_encoder *enc10); enum signal_type dcn10_get_dig_mode( struct link_encoder *enc); + +void dcn10_link_encoder_get_max_link_cap(struct link_encoder *enc, + struct dc_link_settings *link_settings); #endif /* __DC_LINK_ENCODER__DCN10_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c index 04f863499cfb..3fcd408e9103 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c @@ -223,6 +223,9 @@ struct mpcc *mpc1_insert_plane( REG_SET(MPCC_TOP_SEL[mpcc_id], 0, MPCC_TOP_SEL, dpp_id); REG_SET(MPCC_OPP_ID[mpcc_id], 0, MPCC_OPP_ID, tree->opp_id); + /* Configure VUPDATE lock set for this MPCC to map to the OPP */ + REG_SET(MPCC_UPDATE_LOCK_SEL[mpcc_id], 0, MPCC_UPDATE_LOCK_SEL, tree->opp_id); + /* update mpc tree mux setting */ if (tree->opp_list == insert_above_mpcc) { /* insert the toppest mpcc */ @@ -318,6 +321,7 @@ void mpc1_remove_mpcc( REG_SET(MPCC_TOP_SEL[mpcc_id], 0, MPCC_TOP_SEL, 0xf); REG_SET(MPCC_BOT_SEL[mpcc_id], 0, MPCC_BOT_SEL, 0xf); REG_SET(MPCC_OPP_ID[mpcc_id], 0, MPCC_OPP_ID, 0xf); + REG_SET(MPCC_UPDATE_LOCK_SEL[mpcc_id], 0, MPCC_UPDATE_LOCK_SEL, 0xf); /* mark this mpcc as not in use */ mpc10->mpcc_in_use_mask &= ~(1 << mpcc_id); @@ -328,6 +332,7 @@ void mpc1_remove_mpcc( REG_SET(MPCC_TOP_SEL[mpcc_id], 0, MPCC_TOP_SEL, 0xf); REG_SET(MPCC_BOT_SEL[mpcc_id], 0, MPCC_BOT_SEL, 0xf); REG_SET(MPCC_OPP_ID[mpcc_id], 0, MPCC_OPP_ID, 0xf); + REG_SET(MPCC_UPDATE_LOCK_SEL[mpcc_id], 0, MPCC_UPDATE_LOCK_SEL, 0xf); } } @@ -361,6 +366,7 @@ void mpc1_mpc_init(struct mpc *mpc) REG_SET(MPCC_TOP_SEL[mpcc_id], 0, MPCC_TOP_SEL, 0xf); REG_SET(MPCC_BOT_SEL[mpcc_id], 0, MPCC_BOT_SEL, 0xf); REG_SET(MPCC_OPP_ID[mpcc_id], 0, MPCC_OPP_ID, 0xf); + REG_SET(MPCC_UPDATE_LOCK_SEL[mpcc_id], 0, MPCC_UPDATE_LOCK_SEL, 0xf); mpc1_init_mpcc(&(mpc->mpcc_array[mpcc_id]), mpcc_id); } @@ -381,6 +387,7 @@ void mpc1_mpc_init_single_inst(struct mpc *mpc, unsigned int mpcc_id) REG_SET(MPCC_TOP_SEL[mpcc_id], 0, MPCC_TOP_SEL, 0xf); REG_SET(MPCC_BOT_SEL[mpcc_id], 0, MPCC_BOT_SEL, 0xf); REG_SET(MPCC_OPP_ID[mpcc_id], 0, MPCC_OPP_ID, 0xf); + REG_SET(MPCC_UPDATE_LOCK_SEL[mpcc_id], 0, MPCC_UPDATE_LOCK_SEL, 0xf); mpc1_init_mpcc(&(mpc->mpcc_array[mpcc_id]), mpcc_id); @@ -453,6 +460,13 @@ void mpc1_read_mpcc_state( MPCC_BUSY, &s->busy); } +void mpc1_cursor_lock(struct mpc *mpc, int opp_id, bool lock) +{ + struct dcn10_mpc *mpc10 = TO_DCN10_MPC(mpc); + + REG_SET(CUR[opp_id], 0, CUR_VUPDATE_LOCK_SET, lock ? 1 : 0); +} + static const struct mpc_funcs dcn10_mpc_funcs = { .read_mpcc_state = mpc1_read_mpcc_state, .insert_plane = mpc1_insert_plane, @@ -464,6 +478,7 @@ static const struct mpc_funcs dcn10_mpc_funcs = { .assert_mpcc_idle_before_connect = mpc1_assert_mpcc_idle_before_connect, .init_mpcc_list_from_hw = mpc1_init_mpcc_list_from_hw, .update_blending = mpc1_update_blending, + .cursor_lock = mpc1_cursor_lock, .set_denorm = NULL, .set_denorm_clamp = NULL, .set_output_csc = NULL, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.h index 962a68e322ee..66a4719c22a0 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.h @@ -39,11 +39,12 @@ SRII(MPCC_BG_G_Y, MPCC, inst),\ SRII(MPCC_BG_R_CR, MPCC, inst),\ SRII(MPCC_BG_B_CB, MPCC, inst),\ - SRII(MPCC_BG_B_CB, MPCC, inst),\ - SRII(MPCC_SM_CONTROL, MPCC, inst) + SRII(MPCC_SM_CONTROL, MPCC, inst),\ + SRII(MPCC_UPDATE_LOCK_SEL, MPCC, inst) #define MPC_OUT_MUX_COMMON_REG_LIST_DCN1_0(inst) \ - SRII(MUX, MPC_OUT, inst) + SRII(MUX, MPC_OUT, inst),\ + VUPDATE_SRII(CUR, VUPDATE_LOCK_SET, inst) #define MPC_COMMON_REG_VARIABLE_LIST \ uint32_t MPCC_TOP_SEL[MAX_MPCC]; \ @@ -55,7 +56,9 @@ uint32_t MPCC_BG_R_CR[MAX_MPCC]; \ uint32_t MPCC_BG_B_CB[MAX_MPCC]; \ uint32_t MPCC_SM_CONTROL[MAX_MPCC]; \ - uint32_t MUX[MAX_OPP]; + uint32_t MUX[MAX_OPP]; \ + uint32_t MPCC_UPDATE_LOCK_SEL[MAX_MPCC]; \ + uint32_t CUR[MAX_OPP]; #define MPC_COMMON_MASK_SH_LIST_DCN1_0(mask_sh)\ SF(MPCC0_MPCC_TOP_SEL, MPCC_TOP_SEL, mask_sh),\ @@ -78,7 +81,8 @@ SF(MPCC0_MPCC_SM_CONTROL, MPCC_SM_FIELD_ALT, mask_sh),\ SF(MPCC0_MPCC_SM_CONTROL, MPCC_SM_FORCE_NEXT_FRAME_POL, mask_sh),\ SF(MPCC0_MPCC_SM_CONTROL, MPCC_SM_FORCE_NEXT_TOP_POL, mask_sh),\ - SF(MPC_OUT0_MUX, MPC_OUT_MUX, mask_sh) + SF(MPC_OUT0_MUX, MPC_OUT_MUX, mask_sh),\ + SF(MPCC0_MPCC_UPDATE_LOCK_SEL, MPCC_UPDATE_LOCK_SEL, mask_sh) #define MPC_REG_FIELD_LIST(type) \ type MPCC_TOP_SEL;\ @@ -101,7 +105,9 @@ type MPCC_SM_FIELD_ALT;\ type MPCC_SM_FORCE_NEXT_FRAME_POL;\ type MPCC_SM_FORCE_NEXT_TOP_POL;\ - type MPC_OUT_MUX; + type MPC_OUT_MUX;\ + type MPCC_UPDATE_LOCK_SEL;\ + type CUR_VUPDATE_LOCK_SET; struct dcn_mpc_registers { MPC_COMMON_REG_VARIABLE_LIST @@ -192,4 +198,6 @@ void mpc1_read_mpcc_state( int mpcc_inst, struct mpcc_state *s); +void mpc1_cursor_lock(struct mpc *mpc, int opp_id, bool lock); + #endif diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c index 17d96ec6acd8..ec0ab42becba 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c @@ -299,6 +299,7 @@ void optc1_set_vtg_params(struct timing_generator *optc, uint32_t asic_blank_end; uint32_t v_init; uint32_t v_fp2 = 0; + int32_t vertical_line_start; struct optc *optc1 = DCN10TG_FROM_TG(optc); @@ -315,8 +316,9 @@ void optc1_set_vtg_params(struct timing_generator *optc, patched_crtc_timing.v_border_top; /* if VSTARTUP is before VSYNC, FP2 is the offset, otherwise 0 */ - if (optc1->vstartup_start > asic_blank_end) - v_fp2 = optc1->vstartup_start - asic_blank_end; + vertical_line_start = asic_blank_end - optc1->vstartup_start + 1; + if (vertical_line_start < 0) + v_fp2 = -vertical_line_start; /* Interlace */ if (REG(OTG_INTERLACE_CONTROL)) { diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h index 9a459a8fe8a0..8d1e52fb0393 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h @@ -158,6 +158,7 @@ struct dcn_optc_registers { uint32_t OTG_GSL_WINDOW_Y; uint32_t OTG_VUPDATE_KEEPOUT; uint32_t OTG_CRC_CNTL; + uint32_t OTG_CRC_CNTL2; uint32_t OTG_CRC0_DATA_RG; uint32_t OTG_CRC0_DATA_B; uint32_t OTG_CRC0_WINDOWA_X_CONTROL; @@ -475,7 +476,11 @@ struct dcn_optc_registers { type OPTC_DSC_SLICE_WIDTH;\ type OPTC_SEGMENT_WIDTH;\ type OPTC_DWB0_SOURCE_SELECT;\ - type OPTC_DWB1_SOURCE_SELECT; + type OPTC_DWB1_SOURCE_SELECT;\ + type OTG_CRC_DSC_MODE;\ + type OTG_CRC_DATA_STREAM_COMBINE_MODE;\ + type OTG_CRC_DATA_STREAM_SPLIT_MODE;\ + type OTG_CRC_DATA_FORMAT; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c index 07265ca7d28c..17d5cb422025 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c @@ -51,6 +51,7 @@ #include "dce112/dce112_resource.h" #include "dcn10_hubp.h" #include "dcn10_hubbub.h" +#include "dce/dce_panel_cntl.h" #include "soc15_hw_ip.h" #include "vega10_ip_offset.h" @@ -181,6 +182,14 @@ enum dcn10_clk_src_array_id { .reg_name[id] = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \ mm ## block ## id ## _ ## reg_name +#define VUPDATE_SRII(reg_name, block, id)\ + .reg_name[id] = BASE(mm ## reg_name ## 0 ## _ ## block ## id ## _BASE_IDX) + \ + mm ## reg_name ## 0 ## _ ## block ## id + +/* set field/register/bitfield name */ +#define SFRB(field_name, reg_name, bitfield, post_fix)\ + .field_name = reg_name ## __ ## bitfield ## post_fix + /* NBIO */ #define NBIO_BASE_INNER(seg) \ NBIF_BASE__INST0_SEG ## seg @@ -321,6 +330,18 @@ static const struct dcn10_link_enc_mask le_mask = { LINK_ENCODER_MASK_SH_LIST_DCN10(_MASK) }; +static const struct dce_panel_cntl_registers panel_cntl_regs[] = { + { DCN_PANEL_CNTL_REG_LIST() } +}; + +static const struct dce_panel_cntl_shift panel_cntl_shift = { + DCE_PANEL_CNTL_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce_panel_cntl_mask panel_cntl_mask = { + DCE_PANEL_CNTL_MASK_SH_LIST(_MASK) +}; + static const struct dce110_aux_registers_shift aux_shift = { DCN10_AUX_MASK_SH_LIST(__SHIFT) }; @@ -419,11 +440,13 @@ static const struct dcn_mpc_registers mpc_regs = { }; static const struct dcn_mpc_shift mpc_shift = { - MPC_COMMON_MASK_SH_LIST_DCN1_0(__SHIFT) + MPC_COMMON_MASK_SH_LIST_DCN1_0(__SHIFT),\ + SFRB(CUR_VUPDATE_LOCK_SET, CUR0_VUPDATE_LOCK_SET0, CUR0_VUPDATE_LOCK_SET, __SHIFT) }; static const struct dcn_mpc_mask mpc_mask = { - MPC_COMMON_MASK_SH_LIST_DCN1_0(_MASK), + MPC_COMMON_MASK_SH_LIST_DCN1_0(_MASK),\ + SFRB(CUR_VUPDATE_LOCK_SET, CUR0_VUPDATE_LOCK_SET0, CUR0_VUPDATE_LOCK_SET, _MASK) }; #define tg_regs(id)\ @@ -807,6 +830,23 @@ struct link_encoder *dcn10_link_encoder_create( return &enc10->base; } +static struct panel_cntl *dcn10_panel_cntl_create(const struct panel_cntl_init_data *init_data) +{ + struct dce_panel_cntl *panel_cntl = + kzalloc(sizeof(struct dce_panel_cntl), GFP_KERNEL); + + if (!panel_cntl) + return NULL; + + dce_panel_cntl_construct(panel_cntl, + init_data, + &panel_cntl_regs[init_data->inst], + &panel_cntl_shift, + &panel_cntl_mask); + + return &panel_cntl->base; +} + struct clock_source *dcn10_clock_source_create( struct dc_context *ctx, struct dc_bios *bios, @@ -1081,24 +1121,6 @@ static enum dc_status build_mapped_resource( { struct pipe_ctx *pipe_ctx = resource_get_head_pipe_for_stream(&context->res_ctx, stream); - /*TODO Seems unneeded anymore */ - /* if (old_context && resource_is_stream_unchanged(old_context, stream)) { - if (stream != NULL && old_context->streams[i] != NULL) { - todo: shouldn't have to copy missing parameter here - resource_build_bit_depth_reduction_params(stream, - &stream->bit_depth_params); - stream->clamping.pixel_encoding = - stream->timing.pixel_encoding; - - resource_build_bit_depth_reduction_params(stream, - &stream->bit_depth_params); - build_clamping_params(stream); - - continue; - } - } - */ - if (!pipe_ctx) return DC_ERROR_UNEXPECTED; @@ -1291,6 +1313,7 @@ static const struct dc_cap_funcs cap_funcs = { static const struct resource_funcs dcn10_res_pool_funcs = { .destroy = dcn10_destroy_resource_pool, .link_enc_create = dcn10_link_encoder_create, + .panel_cntl_create = dcn10_panel_cntl_create, .validate_bandwidth = dcn_validate_bandwidth, .acquire_idle_pipe_for_layer = dcn10_acquire_idle_pipe_for_layer, .validate_plane = dcn10_validate_plane, @@ -1353,6 +1376,40 @@ static bool dcn10_resource_construct( /* Raven DP PHY HBR2 eye diagram pattern is not stable. Use TP4 */ dc->caps.force_dp_tps4_for_cp2520 = true; + /* Color pipeline capabilities */ + dc->caps.color.dpp.dcn_arch = 1; + dc->caps.color.dpp.input_lut_shared = 1; + dc->caps.color.dpp.icsc = 1; + dc->caps.color.dpp.dgam_ram = 1; + dc->caps.color.dpp.dgam_rom_caps.srgb = 1; + dc->caps.color.dpp.dgam_rom_caps.bt2020 = 1; + dc->caps.color.dpp.dgam_rom_caps.gamma2_2 = 0; + dc->caps.color.dpp.dgam_rom_caps.pq = 0; + dc->caps.color.dpp.dgam_rom_caps.hlg = 0; + dc->caps.color.dpp.post_csc = 0; + dc->caps.color.dpp.gamma_corr = 0; + + dc->caps.color.dpp.hw_3d_lut = 0; + dc->caps.color.dpp.ogam_ram = 1; // RGAM on DCN1 + dc->caps.color.dpp.ogam_rom_caps.srgb = 1; + dc->caps.color.dpp.ogam_rom_caps.bt2020 = 1; + dc->caps.color.dpp.ogam_rom_caps.gamma2_2 = 0; + dc->caps.color.dpp.ogam_rom_caps.pq = 0; + dc->caps.color.dpp.ogam_rom_caps.hlg = 0; + dc->caps.color.dpp.ocsc = 1; + + /* no post-blend color operations */ + dc->caps.color.mpc.gamut_remap = 0; + dc->caps.color.mpc.num_3dluts = 0; + dc->caps.color.mpc.shared_3d_lut = 0; + dc->caps.color.mpc.ogam_ram = 0; + dc->caps.color.mpc.ogam_rom_caps.srgb = 0; + dc->caps.color.mpc.ogam_rom_caps.bt2020 = 0; + dc->caps.color.mpc.ogam_rom_caps.gamma2_2 = 0; + dc->caps.color.mpc.ogam_rom_caps.pq = 0; + dc->caps.color.mpc.ogam_rom_caps.hlg = 0; + dc->caps.color.mpc.ocsc = 0; + if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) dc->debug = debug_defaults_drv; else diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c index 7eba9333c328..07b2f9399671 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c @@ -1274,7 +1274,6 @@ static void enc1_se_audio_setup( { struct dcn10_stream_encoder *enc1 = DCN10STRENC_FROM_STRENC(enc); - uint32_t speakers = 0; uint32_t channels = 0; ASSERT(audio_info); @@ -1282,7 +1281,6 @@ static void enc1_se_audio_setup( /* This should not happen.it does so we don't get BSOD*/ return; - speakers = audio_info->flags.info.ALLSPEAKERS; channels = speakers_to_channels(audio_info->flags.speaker_flags).all; /* setup the audio stream source select (audio -> dig mapping) */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.h index 501532dd523a..c478213ba7ad 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.h @@ -80,6 +80,7 @@ struct dcn20_hubbub { const struct dcn_hubbub_mask *masks; unsigned int debug_test_index_pstate; struct dcn_watermark_set watermarks; + int num_vmid; struct dcn20_vmid vmid[16]; unsigned int detile_buf_size; }; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c index 22f421e82733..da5333d165ac 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c @@ -961,8 +961,7 @@ void dcn20_blank_pixel_data( width = width / odm_cnt; if (blank) { - if (stream_res->abm) - stream_res->abm->funcs->set_abm_immediate_disable(stream_res->abm); + dc->hwss.set_abm_immediate_disable(pipe_ctx); if (dc->debug.visual_confirm != VISUAL_CONFIRM_DISABLE) { test_pattern = CONTROLLER_DP_TEST_PATTERN_COLORSQUARES; @@ -997,7 +996,8 @@ void dcn20_blank_pixel_data( if (!blank) if (stream_res->abm) { - stream_res->abm->funcs->set_pipe(stream_res->abm, stream_res->tg->inst + 1); + stream_res->abm->funcs->set_pipe(stream_res->abm, stream_res->tg->inst + 1, + stream->link->panel_cntl->inst); stream_res->abm->funcs->set_abm_level(stream_res->abm, stream->abm_level); } } @@ -1478,8 +1478,11 @@ static void dcn20_program_pipe( if (pipe_ctx->update_flags.bits.odm) hws->funcs.update_odm(dc, context, pipe_ctx); - if (pipe_ctx->update_flags.bits.enable) + if (pipe_ctx->update_flags.bits.enable) { dcn20_enable_plane(dc, pipe_ctx, context); + if (dc->res_pool->hubbub->funcs->force_wm_propagate_to_pipes) + dc->res_pool->hubbub->funcs->force_wm_propagate_to_pipes(dc->res_pool->hubbub); + } if (pipe_ctx->update_flags.raw || pipe_ctx->plane_state->update_flags.raw || pipe_ctx->stream->update_flags.raw) dcn20_update_dchubp_dpp(dc, pipe_ctx, context); @@ -2037,8 +2040,7 @@ static void dcn20_reset_back_end_for_pipe( */ if (pipe_ctx->top_pipe == NULL) { - if (pipe_ctx->stream_res.abm) - pipe_ctx->stream_res.abm->funcs->set_abm_immediate_disable(pipe_ctx->stream_res.abm); + dc->hwss.set_abm_immediate_disable(pipe_ctx); pipe_ctx->stream_res.tg->funcs->disable_crtc(pipe_ctx->stream_res.tg); @@ -2171,6 +2173,13 @@ void dcn20_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) */ mpcc_id = hubp->inst; + /* If there is no full update, don't need to touch MPC tree*/ + if (!pipe_ctx->plane_state->update_flags.bits.full_update && + !pipe_ctx->update_flags.bits.mpcc) { + mpc->funcs->update_blending(mpc, &blnd_cfg, mpcc_id); + return; + } + /* check if this MPCC is already being used */ new_mpcc = mpc->funcs->get_mpcc_for_dpp(mpc_tree_params, mpcc_id); /* remove MPCC if being used */ @@ -2294,7 +2303,8 @@ void dcn20_fpga_init_hw(struct dc *dc) REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_REFDIV, 2); REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1); - REG_WRITE(REFCLK_CNTL, 0); + if (REG(REFCLK_CNTL)) + REG_WRITE(REFCLK_CNTL, 0); // diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c index 1e73357eda34..2fbde4241559 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c @@ -52,6 +52,7 @@ static const struct hw_sequencer_funcs dcn20_funcs = { .disable_plane = dcn20_disable_plane, .pipe_control_lock = dcn20_pipe_control_lock, .interdependent_update_lock = dcn10_lock_all_pipes, + .cursor_lock = dcn10_cursor_lock, .prepare_bandwidth = dcn20_prepare_bandwidth, .optimize_bandwidth = dcn20_optimize_bandwidth, .update_bandwidth = dcn20_update_bandwidth, @@ -82,6 +83,9 @@ static const struct hw_sequencer_funcs dcn20_funcs = { .init_vm_ctx = dcn20_init_vm_ctx, .set_flip_control_gsl = dcn20_set_flip_control_gsl, .get_vupdate_offset_from_vsync = dcn10_get_vupdate_offset_from_vsync, + .calc_vupdate_position = dcn10_calc_vupdate_position, + .set_backlight_level = dce110_set_backlight_level, + .set_abm_immediate_disable = dce110_set_abm_immediate_disable, }; static const struct hwseq_private_funcs dcn20_private_funcs = { @@ -97,8 +101,6 @@ static const struct hwseq_private_funcs dcn20_private_funcs = { .reset_hw_ctx_wrap = dcn20_reset_hw_ctx_wrap, .enable_stream_timing = dcn20_enable_stream_timing, .edp_backlight_control = dce110_edp_backlight_control, - .is_panel_backlight_on = dce110_is_panel_backlight_on, - .is_panel_powered_on = dce110_is_panel_powered_on, .disable_stream_gating = dcn20_disable_stream_gating, .enable_stream_gating = dcn20_enable_stream_gating, .setup_vupdate_interrupt = dcn20_setup_vupdate_interrupt, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.c index e4ac73035c84..8d209dae66e6 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.c @@ -49,6 +49,12 @@ #define IND_REG(index) \ (enc10->link_regs->index) +#ifndef MAX +#define MAX(X, Y) ((X) > (Y) ? (X) : (Y)) +#endif +#ifndef MIN +#define MIN(X, Y) ((X) < (Y) ? (X) : (Y)) +#endif static struct mpll_cfg dcn2_mpll_cfg[] = { // RBR @@ -260,6 +266,38 @@ void dcn20_link_encoder_enable_dp_output( } +void dcn20_link_encoder_get_max_link_cap(struct link_encoder *enc, + struct dc_link_settings *link_settings) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + uint32_t is_in_usb_c_dp4_mode = 0; + + dcn10_link_encoder_get_max_link_cap(enc, link_settings); + + /* in usb c dp2 mode, max lane count is 2 */ + if (enc->funcs->is_in_alt_mode && enc->funcs->is_in_alt_mode(enc)) { + REG_GET(RDPCSTX_PHY_CNTL6, RDPCS_PHY_DPALT_DP4, &is_in_usb_c_dp4_mode); + if (!is_in_usb_c_dp4_mode) + link_settings->lane_count = MIN(LANE_COUNT_TWO, link_settings->lane_count); + } + +} + +bool dcn20_link_encoder_is_in_alt_mode(struct link_encoder *enc) +{ + struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); + uint32_t dp_alt_mode_disable = 0; + bool is_usb_c_alt_mode = false; + + if (enc->features.flags.bits.DP_IS_USB_C) { + /* if value == 1 alt mode is disabled, otherwise it is enabled */ + REG_GET(RDPCSTX_PHY_CNTL6, RDPCS_PHY_DPALT_DISABLE, &dp_alt_mode_disable); + is_usb_c_alt_mode = (dp_alt_mode_disable == 0); + } + + return is_usb_c_alt_mode; +} + #define AUX_REG(reg)\ (enc10->aux_regs->reg) @@ -338,6 +376,8 @@ static const struct link_encoder_funcs dcn20_link_enc_funcs = { .fec_is_active = enc2_fec_is_active, .get_dig_mode = dcn10_get_dig_mode, .get_dig_frontend = dcn10_get_dig_frontend, + .is_in_alt_mode = dcn20_link_encoder_is_in_alt_mode, + .get_max_link_cap = dcn20_link_encoder_get_max_link_cap, }; void dcn20_link_encoder_construct( diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.h index 8cab8107fd94..284a1ee4d249 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_link_encoder.h @@ -343,6 +343,10 @@ void dcn20_link_encoder_enable_dp_output( const struct dc_link_settings *link_settings, enum clock_source_id clock_source); +bool dcn20_link_encoder_is_in_alt_mode(struct link_encoder *enc); +void dcn20_link_encoder_get_max_link_cap(struct link_encoder *enc, + struct dc_link_settings *link_settings); + void dcn20_link_encoder_construct( struct dcn20_link_encoder *enc20, const struct encoder_init_data *init_data, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c index de9c857ab3e9..99cc095dc33c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c @@ -452,7 +452,7 @@ void mpc2_set_output_gamma( next_mode = LUT_RAM_A; mpc20_power_on_ogam_lut(mpc, mpcc_id, true); - mpc20_configure_ogam_lut(mpc, mpcc_id, next_mode == LUT_RAM_A ? true:false); + mpc20_configure_ogam_lut(mpc, mpcc_id, next_mode == LUT_RAM_A); if (next_mode == LUT_RAM_A) mpc2_program_luta(mpc, mpcc_id, params); @@ -545,6 +545,7 @@ const struct mpc_funcs dcn20_mpc_funcs = { .mpc_init = mpc1_mpc_init, .mpc_init_single_inst = mpc1_mpc_init_single_inst, .update_blending = mpc2_update_blending, + .cursor_lock = mpc1_cursor_lock, .get_mpcc_for_dpp = mpc2_get_mpcc_for_dpp, .wait_for_idle = mpc2_assert_idle_mpcc, .assert_mpcc_idle_before_connect = mpc2_assert_mpcc_idle_before_connect, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.h index c78fd5123497..496658f420db 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.h @@ -179,7 +179,8 @@ SF(MPC_OUT0_DENORM_CLAMP_G_Y, MPC_OUT_DENORM_CLAMP_MAX_G_Y, mask_sh),\ SF(MPC_OUT0_DENORM_CLAMP_G_Y, MPC_OUT_DENORM_CLAMP_MIN_G_Y, mask_sh),\ SF(MPC_OUT0_DENORM_CLAMP_B_CB, MPC_OUT_DENORM_CLAMP_MAX_B_CB, mask_sh),\ - SF(MPC_OUT0_DENORM_CLAMP_B_CB, MPC_OUT_DENORM_CLAMP_MIN_B_CB, mask_sh) + SF(MPC_OUT0_DENORM_CLAMP_B_CB, MPC_OUT_DENORM_CLAMP_MIN_B_CB, mask_sh),\ + SF(CUR_VUPDATE_LOCK_SET0, CUR_VUPDATE_LOCK_SET, mask_sh) /* * DCN2 MPC_OCSC debug status register: diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c index d875b0c38fde..8c16967fe018 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c @@ -409,6 +409,18 @@ void optc2_program_manual_trigger(struct timing_generator *optc) OTG_TRIGA_MANUAL_TRIG, 1); } +bool optc2_configure_crc(struct timing_generator *optc, + const struct crc_params *params) +{ + struct optc *optc1 = DCN10TG_FROM_TG(optc); + + REG_SET_2(OTG_CRC_CNTL2, 0, + OTG_CRC_DSC_MODE, params->dsc_mode, + OTG_CRC_DATA_STREAM_COMBINE_MODE, params->odm_mode); + + return optc1_configure_crc(optc, params); +} + static struct timing_generator_funcs dcn20_tg_funcs = { .validate_timing = optc1_validate_timing, .program_timing = optc1_program_timing, @@ -452,7 +464,7 @@ static struct timing_generator_funcs dcn20_tg_funcs = { .clear_optc_underflow = optc1_clear_optc_underflow, .setup_global_swap_lock = NULL, .get_crc = optc1_get_crc, - .configure_crc = optc1_configure_crc, + .configure_crc = optc2_configure_crc, .set_dsc_config = optc2_set_dsc_config, .set_dwb_source = optc2_set_dwb_source, .set_odm_bypass = optc2_set_odm_bypass, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.h index 239cc40ae474..e0a0a8a8e2c6 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.h @@ -36,6 +36,7 @@ SRI(OTG_GSL_WINDOW_Y, OTG, inst),\ SRI(OTG_VUPDATE_KEEPOUT, OTG, inst),\ SRI(OTG_DSC_START_POSITION, OTG, inst),\ + SRI(OTG_CRC_CNTL2, OTG, inst),\ SRI(OPTC_DATA_FORMAT_CONTROL, ODM, inst),\ SRI(OPTC_BYTES_PER_PIXEL, ODM, inst),\ SRI(OPTC_WIDTH_CONTROL, ODM, inst),\ @@ -62,6 +63,10 @@ SF(OTG0_OTG_GSL_CONTROL, OTG_MASTER_UPDATE_LOCK_GSL_EN, mask_sh), \ SF(OTG0_OTG_DSC_START_POSITION, OTG_DSC_START_POSITION_X, mask_sh), \ SF(OTG0_OTG_DSC_START_POSITION, OTG_DSC_START_POSITION_LINE_NUM, mask_sh),\ + SF(OTG0_OTG_CRC_CNTL2, OTG_CRC_DSC_MODE, mask_sh),\ + SF(OTG0_OTG_CRC_CNTL2, OTG_CRC_DATA_STREAM_COMBINE_MODE, mask_sh),\ + SF(OTG0_OTG_CRC_CNTL2, OTG_CRC_DATA_STREAM_SPLIT_MODE, mask_sh),\ + SF(OTG0_OTG_CRC_CNTL2, OTG_CRC_DATA_FORMAT, mask_sh),\ SF(ODM0_OPTC_DATA_SOURCE_SELECT, OPTC_SEG0_SRC_SEL, mask_sh),\ SF(ODM0_OPTC_DATA_SOURCE_SELECT, OPTC_SEG1_SRC_SEL, mask_sh),\ SF(ODM0_OPTC_DATA_SOURCE_SELECT, OPTC_NUM_OF_INPUT_SEGMENT, mask_sh),\ @@ -109,4 +114,6 @@ void optc2_lock_doublebuffer_enable(struct timing_generator *optc); void optc2_setup_manual_trigger(struct timing_generator *optc); void optc2_program_manual_trigger(struct timing_generator *optc); bool optc2_is_two_pixels_per_containter(const struct dc_crtc_timing *timing); +bool optc2_configure_crc(struct timing_generator *optc, + const struct crc_params *params); #endif /* __DC_OPTC_DCN20_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c index 5cdbba0cd873..cef1aa938ab5 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c @@ -61,6 +61,7 @@ #include "dcn20_dccg.h" #include "dcn20_vmid.h" #include "dc_link_ddc.h" +#include "dce/dce_panel_cntl.h" #include "navi10_ip_offset.h" @@ -508,6 +509,10 @@ enum dcn20_clk_src_array_id { .block ## _ ## reg_name[id] = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \ mm ## block ## id ## _ ## reg_name +#define VUPDATE_SRII(reg_name, block, id)\ + .reg_name[id] = BASE(mm ## reg_name ## _ ## block ## id ## _BASE_IDX) + \ + mm ## reg_name ## _ ## block ## id + /* NBIO */ #define NBIO_BASE_INNER(seg) \ NBIO_BASE__INST0_SEG ## seg @@ -687,6 +692,18 @@ static const struct dcn10_link_enc_mask le_mask = { DPCS_DCN2_MASK_SH_LIST(_MASK) }; +static const struct dce_panel_cntl_registers panel_cntl_regs[] = { + { DCN_PANEL_CNTL_REG_LIST() } +}; + +static const struct dce_panel_cntl_shift panel_cntl_shift = { + DCE_PANEL_CNTL_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce_panel_cntl_mask panel_cntl_mask = { + DCE_PANEL_CNTL_MASK_SH_LIST(_MASK) +}; + #define ipp_regs(id)\ [id] = {\ IPP_REG_LIST_DCN20(id),\ @@ -1289,6 +1306,23 @@ struct link_encoder *dcn20_link_encoder_create( return &enc20->enc10.base; } +static struct panel_cntl *dcn20_panel_cntl_create(const struct panel_cntl_init_data *init_data) +{ + struct dce_panel_cntl *panel_cntl = + kzalloc(sizeof(struct dce_panel_cntl), GFP_KERNEL); + + if (!panel_cntl) + return NULL; + + dce_panel_cntl_construct(panel_cntl, + init_data, + &panel_cntl_regs[init_data->inst], + &panel_cntl_shift, + &panel_cntl_mask); + + return &panel_cntl->base; +} + struct clock_source *dcn20_clock_source_create( struct dc_context *ctx, struct dc_bios *bios, @@ -1619,24 +1653,6 @@ enum dc_status dcn20_build_mapped_resource(const struct dc *dc, struct dc_state enum dc_status status = DC_OK; struct pipe_ctx *pipe_ctx = resource_get_head_pipe_for_stream(&context->res_ctx, stream); - /*TODO Seems unneeded anymore */ - /* if (old_context && resource_is_stream_unchanged(old_context, stream)) { - if (stream != NULL && old_context->streams[i] != NULL) { - todo: shouldn't have to copy missing parameter here - resource_build_bit_depth_reduction_params(stream, - &stream->bit_depth_params); - stream->clamping.pixel_encoding = - stream->timing.pixel_encoding; - - resource_build_bit_depth_reduction_params(stream, - &stream->bit_depth_params); - build_clamping_params(stream); - - continue; - } - } - */ - if (!pipe_ctx) return DC_ERROR_UNEXPECTED; @@ -1647,22 +1663,32 @@ enum dc_status dcn20_build_mapped_resource(const struct dc *dc, struct dc_state } -static void acquire_dsc(struct resource_context *res_ctx, - const struct resource_pool *pool, +void dcn20_acquire_dsc(const struct dc *dc, + struct resource_context *res_ctx, struct display_stream_compressor **dsc, int pipe_idx) { int i; + const struct resource_pool *pool = dc->res_pool; + struct display_stream_compressor *dsc_old = dc->current_state->res_ctx.pipe_ctx[pipe_idx].stream_res.dsc; - ASSERT(*dsc == NULL); + ASSERT(*dsc == NULL); /* If this ASSERT fails, dsc was not released properly */ *dsc = NULL; + /* Always do 1-to-1 mapping when number of DSCs is same as number of pipes */ if (pool->res_cap->num_dsc == pool->res_cap->num_opp) { *dsc = pool->dscs[pipe_idx]; res_ctx->is_dsc_acquired[pipe_idx] = true; return; } + /* Return old DSC to avoid the need for re-programming */ + if (dsc_old && !res_ctx->is_dsc_acquired[dsc_old->inst]) { + *dsc = dsc_old; + res_ctx->is_dsc_acquired[dsc_old->inst] = true; + return ; + } + /* Find first free DSC */ for (i = 0; i < pool->res_cap->num_dsc; i++) if (!res_ctx->is_dsc_acquired[i]) { @@ -1694,7 +1720,6 @@ enum dc_status dcn20_add_dsc_to_stream_resource(struct dc *dc, { enum dc_status result = DC_OK; int i; - const struct resource_pool *pool = dc->res_pool; /* Get a DSC if required and available */ for (i = 0; i < dc->res_pool->pipe_count; i++) { @@ -1706,7 +1731,7 @@ enum dc_status dcn20_add_dsc_to_stream_resource(struct dc *dc, if (pipe_ctx->stream_res.dsc) continue; - acquire_dsc(&dc_ctx->res_ctx, pool, &pipe_ctx->stream_res.dsc, i); + dcn20_acquire_dsc(dc, &dc_ctx->res_ctx, &pipe_ctx->stream_res.dsc, i); /* The number of DSCs can be less than the number of pipes */ if (!pipe_ctx->stream_res.dsc) { @@ -1834,12 +1859,13 @@ static void swizzle_to_dml_params( } bool dcn20_split_stream_for_odm( + const struct dc *dc, struct resource_context *res_ctx, - const struct resource_pool *pool, struct pipe_ctx *prev_odm_pipe, struct pipe_ctx *next_odm_pipe) { int pipe_idx = next_odm_pipe->pipe_idx; + const struct resource_pool *pool = dc->res_pool; *next_odm_pipe = *prev_odm_pipe; @@ -1897,7 +1923,7 @@ bool dcn20_split_stream_for_odm( } next_odm_pipe->stream_res.opp = pool->opps[next_odm_pipe->pipe_idx]; if (next_odm_pipe->stream->timing.flags.DSC == 1) { - acquire_dsc(res_ctx, pool, &next_odm_pipe->stream_res.dsc, next_odm_pipe->pipe_idx); + dcn20_acquire_dsc(dc, res_ctx, &next_odm_pipe->stream_res.dsc, next_odm_pipe->pipe_idx); ASSERT(next_odm_pipe->stream_res.dsc); if (next_odm_pipe->stream_res.dsc == NULL) return false; @@ -1935,8 +1961,6 @@ void dcn20_split_stream_for_mpc( secondary_pipe->top_pipe = primary_pipe; ASSERT(primary_pipe->plane_state); - resource_build_scaling_params(primary_pipe); - resource_build_scaling_params(secondary_pipe); } void dcn20_populate_dml_writeback_from_context( @@ -2212,12 +2236,12 @@ int dcn20_populate_dml_pipes_from_context( || pipes[pipe_cnt].pipe.dest.odm_combine != dm_odm_combine_mode_disabled; pipes[pipe_cnt].pipe.src.source_scan = pln->rotation == ROTATION_ANGLE_90 || pln->rotation == ROTATION_ANGLE_270 ? dm_vert : dm_horz; - pipes[pipe_cnt].pipe.src.viewport_y_y = scl->viewport.y; - pipes[pipe_cnt].pipe.src.viewport_y_c = scl->viewport_c.y; - pipes[pipe_cnt].pipe.src.viewport_width = scl->viewport.width; - pipes[pipe_cnt].pipe.src.viewport_width_c = scl->viewport_c.width; - pipes[pipe_cnt].pipe.src.viewport_height = scl->viewport.height; - pipes[pipe_cnt].pipe.src.viewport_height_c = scl->viewport_c.height; + pipes[pipe_cnt].pipe.src.viewport_y_y = scl->viewport_unadjusted.y; + pipes[pipe_cnt].pipe.src.viewport_y_c = scl->viewport_c_unadjusted.y; + pipes[pipe_cnt].pipe.src.viewport_width = scl->viewport_unadjusted.width; + pipes[pipe_cnt].pipe.src.viewport_width_c = scl->viewport_c_unadjusted.width; + pipes[pipe_cnt].pipe.src.viewport_height = scl->viewport_unadjusted.height; + pipes[pipe_cnt].pipe.src.viewport_height_c = scl->viewport_c_unadjusted.height; pipes[pipe_cnt].pipe.src.surface_width_y = pln->plane_size.surface_size.width; pipes[pipe_cnt].pipe.src.surface_height_y = pln->plane_size.surface_size.height; pipes[pipe_cnt].pipe.src.surface_width_c = pln->plane_size.chroma_size.width; @@ -2566,13 +2590,15 @@ int dcn20_validate_apply_pipe_split_flags( struct dc *dc, struct dc_state *context, int vlevel, - bool *split, + int *split, bool *merge) { int i, pipe_idx, vlevel_split; int plane_count = 0; bool force_split = false; bool avoid_split = dc->debug.pipe_split_policy == MPC_SPLIT_AVOID; + struct vba_vars_st *v = &context->bw_ctx.dml.vba; + int max_mpc_comb = v->maxMpcComb; if (context->stream_count > 1) { if (dc->debug.pipe_split_policy == MPC_SPLIT_AVOID_MULT_DISP) @@ -2580,10 +2606,22 @@ int dcn20_validate_apply_pipe_split_flags( } else if (dc->debug.force_single_disp_pipe_split) force_split = true; - /* TODO: fix dc bugs and remove this split threshold thing */ for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + /** + * Workaround for avoiding pipe-split in cases where we'd split + * planes that are too small, resulting in splits that aren't + * valid for the scaler. + */ + if (pipe->plane_state && + (pipe->plane_state->dst_rect.width <= 16 || + pipe->plane_state->dst_rect.height <= 16 || + pipe->plane_state->src_rect.width <= 16 || + pipe->plane_state->src_rect.height <= 16)) + avoid_split = true; + + /* TODO: fix dc bugs and remove this split threshold thing */ if (pipe->stream && !pipe->prev_odm_pipe && (!pipe->top_pipe || pipe->top_pipe->plane_state != pipe->plane_state)) ++plane_count; @@ -2598,26 +2636,35 @@ int dcn20_validate_apply_pipe_split_flags( continue; for (vlevel_split = vlevel; vlevel <= context->bw_ctx.dml.soc.num_states; vlevel++) - if (context->bw_ctx.dml.vba.NoOfDPP[vlevel][0][pipe_idx] == 1) + if (v->NoOfDPP[vlevel][0][pipe_idx] == 1 && + v->ModeSupport[vlevel][0]) break; /* Impossible to not split this pipe */ if (vlevel > context->bw_ctx.dml.soc.num_states) vlevel = vlevel_split; + else + max_mpc_comb = 0; pipe_idx++; } - context->bw_ctx.dml.vba.maxMpcComb = 0; + v->maxMpcComb = max_mpc_comb; } /* Split loop sets which pipe should be split based on dml outputs and dc flags */ for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; - int pipe_plane = context->bw_ctx.dml.vba.pipe_plane[pipe_idx]; + int pipe_plane = v->pipe_plane[pipe_idx]; + bool split4mpc = context->stream_count == 1 && plane_count == 1 + && dc->config.enable_4to1MPC && dc->res_pool->pipe_count >= 4; if (!context->res_ctx.pipe_ctx[i].stream) continue; - if (force_split || context->bw_ctx.dml.vba.NoOfDPP[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_plane] > 1) - split[i] = true; + if (force_split || v->NoOfDPP[vlevel][max_mpc_comb][pipe_plane] > 1) { + if (split4mpc) + split[i] = 4; + else + split[i] = 2; + } if ((pipe->stream->view_format == VIEW_3D_FORMAT_SIDE_BY_SIDE || pipe->stream->view_format == @@ -2626,50 +2673,75 @@ int dcn20_validate_apply_pipe_split_flags( TIMING_3D_FORMAT_TOP_AND_BOTTOM || pipe->stream->timing.timing_3d_format == TIMING_3D_FORMAT_SIDE_BY_SIDE)) - split[i] = true; + split[i] = 2; if (dc->debug.force_odm_combine & (1 << pipe->stream_res.tg->inst)) { - split[i] = true; - context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel][pipe_plane] = dm_odm_combine_mode_2to1; + split[i] = 2; + v->ODMCombineEnablePerState[vlevel][pipe_plane] = dm_odm_combine_mode_2to1; } - context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_plane] = - context->bw_ctx.dml.vba.ODMCombineEnablePerState[vlevel][pipe_plane]; - - if (pipe->prev_odm_pipe && context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_plane] != dm_odm_combine_mode_disabled) { - /*Already split odm pipe tree, don't try to split again*/ - split[i] = false; - split[pipe->prev_odm_pipe->pipe_idx] = false; - } else if (pipe->top_pipe && pipe->plane_state == pipe->top_pipe->plane_state - && context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_plane] == dm_odm_combine_mode_disabled) { - /*Already split mpc tree, don't try to split again, assumes only 2x mpc combine*/ - split[i] = false; - split[pipe->top_pipe->pipe_idx] = false; - } else if (pipe->prev_odm_pipe || (pipe->top_pipe && pipe->plane_state == pipe->top_pipe->plane_state)) { - if (split[i] == false) { - /*Exiting mpc/odm combine*/ - merge[i] = true; - if (pipe->prev_odm_pipe) { - ASSERT(0); /*should not actually happen yet*/ - merge[pipe->prev_odm_pipe->pipe_idx] = true; - } else - merge[pipe->top_pipe->pipe_idx] = true; - } else { - /*Transition from mpc combine to odm combine or vice versa*/ - ASSERT(0); /*should not actually happen yet*/ - split[i] = true; - merge[i] = true; + v->ODMCombineEnabled[pipe_plane] = + v->ODMCombineEnablePerState[vlevel][pipe_plane]; + + if (v->ODMCombineEnabled[pipe_plane] == dm_odm_combine_mode_disabled) { + if (get_num_mpc_splits(pipe) == 1) { + /*If need split for mpc but 2 way split already*/ + if (split[i] == 4) + split[i] = 2; /* 2 -> 4 MPC */ + else if (split[i] == 2) + split[i] = 0; /* 2 -> 2 MPC */ + else if (pipe->top_pipe && pipe->top_pipe->plane_state == pipe->plane_state) + merge[i] = true; /* 2 -> 1 MPC */ + } else if (get_num_mpc_splits(pipe) == 3) { + /*If need split for mpc but 4 way split already*/ + if (split[i] == 2 && ((pipe->top_pipe && !pipe->top_pipe->top_pipe) + || !pipe->bottom_pipe)) { + merge[i] = true; /* 4 -> 2 MPC */ + } else if (split[i] == 0 && pipe->top_pipe && + pipe->top_pipe->plane_state == pipe->plane_state) + merge[i] = true; /* 4 -> 1 MPC */ + split[i] = 0; + } else if (get_num_odm_splits(pipe)) { + /* ODM -> MPC transition */ + ASSERT(0); /* NOT expected yet */ if (pipe->prev_odm_pipe) { - split[pipe->prev_odm_pipe->pipe_idx] = true; - merge[pipe->prev_odm_pipe->pipe_idx] = true; - } else { - split[pipe->top_pipe->pipe_idx] = true; - merge[pipe->top_pipe->pipe_idx] = true; + split[i] = 0; + merge[i] = true; + } + } + } else { + if (get_num_odm_splits(pipe) == 1) { + /*If need split for odm but 2 way split already*/ + if (split[i] == 4) + split[i] = 2; /* 2 -> 4 ODM */ + else if (split[i] == 2) + split[i] = 0; /* 2 -> 2 ODM */ + else if (pipe->prev_odm_pipe) { + ASSERT(0); /* NOT expected yet */ + merge[i] = true; /* exit ODM */ + } + } else if (get_num_odm_splits(pipe) == 3) { + /*If need split for odm but 4 way split already*/ + if (split[i] == 2 && ((pipe->prev_odm_pipe && !pipe->prev_odm_pipe->prev_odm_pipe) + || !pipe->next_odm_pipe)) { + ASSERT(0); /* NOT expected yet */ + merge[i] = true; /* 4 -> 2 ODM */ + } else if (split[i] == 0 && pipe->prev_odm_pipe) { + ASSERT(0); /* NOT expected yet */ + merge[i] = true; /* exit ODM */ + } + split[i] = 0; + } else if (get_num_mpc_splits(pipe)) { + /* MPC -> ODM transition */ + ASSERT(0); /* NOT expected yet */ + if (pipe->top_pipe && pipe->top_pipe->plane_state == pipe->plane_state) { + split[i] = 0; + merge[i] = true; } } } /* Adjust dppclk when split is forced, do not bother with dispclk */ - if (split[i] && context->bw_ctx.dml.vba.NoOfDPP[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] == 1) - context->bw_ctx.dml.vba.RequiredDPPCLK[vlevel][context->bw_ctx.dml.vba.maxMpcComb][pipe_idx] /= 2; + if (split[i] != 0 && v->NoOfDPP[vlevel][max_mpc_comb][pipe_idx] == 1) + v->RequiredDPPCLK[vlevel][max_mpc_comb][pipe_idx] /= 2; pipe_idx++; } @@ -2685,7 +2757,7 @@ bool dcn20_fast_validate_bw( int *vlevel_out) { bool out = false; - bool split[MAX_PIPES] = { false }; + int split[MAX_PIPES] = { 0 }; int pipe_cnt, i, pipe_idx, vlevel; ASSERT(pipes); @@ -2727,7 +2799,7 @@ bool dcn20_fast_validate_bw( hsplit_pipe = dcn20_find_secondary_pipe(dc, &context->res_ctx, dc->res_pool, pipe); ASSERT(hsplit_pipe); if (!dcn20_split_stream_for_odm( - &context->res_ctx, dc->res_pool, + dc, &context->res_ctx, pipe, hsplit_pipe)) goto validate_fail; pipe_split_from[hsplit_pipe->pipe_idx] = pipe_idx; @@ -2745,7 +2817,7 @@ bool dcn20_fast_validate_bw( && context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_idx]) goto validate_fail; - if (split[i]) { + if (split[i] == 2) { if (!hsplit_pipe || hsplit_pipe->plane_state != pipe->plane_state) { /* pipe not split previously needs split */ hsplit_pipe = dcn20_find_secondary_pipe(dc, &context->res_ctx, dc->res_pool, pipe); @@ -2756,14 +2828,17 @@ bool dcn20_fast_validate_bw( } if (context->bw_ctx.dml.vba.ODMCombineEnabled[pipe_idx]) { if (!dcn20_split_stream_for_odm( - &context->res_ctx, dc->res_pool, + dc, &context->res_ctx, pipe, hsplit_pipe)) goto validate_fail; dcn20_build_mapped_resource(dc, context, pipe->stream); - } else + } else { dcn20_split_stream_for_mpc( - &context->res_ctx, dc->res_pool, - pipe, hsplit_pipe); + &context->res_ctx, dc->res_pool, + pipe, hsplit_pipe); + if (!resource_build_scaling_params(pipe) || !resource_build_scaling_params(hsplit_pipe)) + goto validate_fail; + } pipe_split_from[hsplit_pipe->pipe_idx] = pipe_idx; } } else if (hsplit_pipe && hsplit_pipe->plane_state == pipe->plane_state) { @@ -3003,7 +3078,7 @@ void dcn20_calculate_dlg_params( pipe_idx, cstate_en, context->bw_ctx.bw.dcn.clk.p_state_change_support, - false, false, false); + false, false, true); context->bw_ctx.dml.funcs.rq_dlg_get_rq_reg(&context->bw_ctx.dml, &context->res_ctx.pipe_ctx[i].rq_regs, @@ -3064,25 +3139,34 @@ validate_out: return out; } - -bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) +/* + * This must be noinline to ensure anything that deals with FP registers + * is contained within this call; previously our compiling with hard-float + * would result in fp instructions being emitted outside of the boundaries + * of the DC_FP_START/END macros, which makes sense as the compiler has no + * idea about what is wrapped and what is not + * + * This is largely just a workaround to avoid breakage introduced with 5.6, + * ideally all fp-using code should be moved into its own file, only that + * should be compiled with hard-float, and all code exported from there + * should be strictly wrapped with DC_FP_START/END + */ +static noinline bool dcn20_validate_bandwidth_fp(struct dc *dc, + struct dc_state *context, bool fast_validate) { bool voltage_supported = false; bool full_pstate_supported = false; bool dummy_pstate_supported = false; double p_state_latency_us; - DC_FP_START(); p_state_latency_us = context->bw_ctx.dml.soc.dram_clock_change_latency_us; context->bw_ctx.dml.soc.disable_dram_clock_change_vactive_support = dc->debug.disable_dram_clock_change_vactive_support; + context->bw_ctx.dml.soc.allow_dram_clock_one_display_vactive = + dc->debug.enable_dram_clock_change_one_display_vactive; if (fast_validate) { - voltage_supported = dcn20_validate_bandwidth_internal(dc, context, true); - - DC_FP_END(); - return voltage_supported; + return dcn20_validate_bandwidth_internal(dc, context, true); } // Best case, we support full UCLK switch latency @@ -3111,7 +3195,15 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, restore_dml_state: context->bw_ctx.dml.soc.dram_clock_change_latency_us = p_state_latency_us; + return voltage_supported; +} +bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, + bool fast_validate) +{ + bool voltage_supported = false; + DC_FP_START(); + voltage_supported = dcn20_validate_bandwidth_fp(dc, context, fast_validate); DC_FP_END(); return voltage_supported; } @@ -3170,8 +3262,6 @@ static struct dc_cap_funcs cap_funcs = { enum dc_status dcn20_patch_unknown_plane_state(struct dc_plane_state *plane_state) { - enum dc_status result = DC_OK; - enum surface_pixel_format surf_pix_format = plane_state->format; unsigned int bpp = resource_pixel_format_to_bpp(surf_pix_format); @@ -3183,12 +3273,13 @@ enum dc_status dcn20_patch_unknown_plane_state(struct dc_plane_state *plane_stat swizzle = DC_SW_64KB_S; plane_state->tiling_info.gfx9.swizzle = swizzle; - return result; + return DC_OK; } static struct resource_funcs dcn20_res_pool_funcs = { .destroy = dcn20_destroy_resource_pool, .link_enc_create = dcn20_link_encoder_create, + .panel_cntl_create = dcn20_panel_cntl_create, .validate_bandwidth = dcn20_validate_bandwidth, .acquire_idle_pipe_for_layer = dcn20_acquire_idle_pipe_for_layer, .add_stream_to_ctx = dcn20_add_stream_to_ctx, @@ -3427,6 +3518,13 @@ void dcn20_patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st bb->dram_clock_change_latency_us = dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; } + + if ((int)(bb->dummy_pstate_latency_us * 1000) + != dc->bb_overrides.dummy_clock_change_latency_ns + && dc->bb_overrides.dummy_clock_change_latency_ns) { + bb->dummy_pstate_latency_us = + dc->bb_overrides.dummy_clock_change_latency_ns / 1000.0; + } } static struct _vcs_dpi_soc_bounding_box_st *get_asic_rev_soc_bb( @@ -3662,9 +3760,42 @@ static bool dcn20_resource_construct( dc->caps.max_slave_planes = 1; dc->caps.post_blend_color_processing = true; dc->caps.force_dp_tps4_for_cp2520 = true; - dc->caps.hw_3d_lut = true; dc->caps.extended_aux_timeout_support = true; + /* Color pipeline capabilities */ + dc->caps.color.dpp.dcn_arch = 1; + dc->caps.color.dpp.input_lut_shared = 0; + dc->caps.color.dpp.icsc = 1; + dc->caps.color.dpp.dgam_ram = 1; + dc->caps.color.dpp.dgam_rom_caps.srgb = 1; + dc->caps.color.dpp.dgam_rom_caps.bt2020 = 1; + dc->caps.color.dpp.dgam_rom_caps.gamma2_2 = 0; + dc->caps.color.dpp.dgam_rom_caps.pq = 0; + dc->caps.color.dpp.dgam_rom_caps.hlg = 0; + dc->caps.color.dpp.post_csc = 0; + dc->caps.color.dpp.gamma_corr = 0; + + dc->caps.color.dpp.hw_3d_lut = 1; + dc->caps.color.dpp.ogam_ram = 1; + // no OGAM ROM on DCN2, only MPC ROM + dc->caps.color.dpp.ogam_rom_caps.srgb = 0; + dc->caps.color.dpp.ogam_rom_caps.bt2020 = 0; + dc->caps.color.dpp.ogam_rom_caps.gamma2_2 = 0; + dc->caps.color.dpp.ogam_rom_caps.pq = 0; + dc->caps.color.dpp.ogam_rom_caps.hlg = 0; + dc->caps.color.dpp.ocsc = 0; + + dc->caps.color.mpc.gamut_remap = 0; + dc->caps.color.mpc.num_3dluts = 0; + dc->caps.color.mpc.shared_3d_lut = 0; + dc->caps.color.mpc.ogam_ram = 1; + dc->caps.color.mpc.ogam_rom_caps.srgb = 0; + dc->caps.color.mpc.ogam_rom_caps.bt2020 = 0; + dc->caps.color.mpc.ogam_rom_caps.gamma2_2 = 0; + dc->caps.color.mpc.ogam_rom_caps.pq = 0; + dc->caps.color.mpc.ogam_rom_caps.hlg = 0; + dc->caps.color.mpc.ocsc = 1; + if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) { dc->debug = debug_defaults_drv; } else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) { diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h index 9d5bff9455fd..2c1959845c29 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h @@ -123,7 +123,7 @@ int dcn20_validate_apply_pipe_split_flags( struct dc *dc, struct dc_state *context, int vlevel, - bool *split, + int *split, bool *merge); void dcn20_release_dsc(struct resource_context *res_ctx, const struct resource_pool *pool, @@ -135,10 +135,14 @@ void dcn20_split_stream_for_mpc( struct pipe_ctx *primary_pipe, struct pipe_ctx *secondary_pipe); bool dcn20_split_stream_for_odm( + const struct dc *dc, struct resource_context *res_ctx, - const struct resource_pool *pool, struct pipe_ctx *prev_odm_pipe, struct pipe_ctx *next_odm_pipe); +void dcn20_acquire_dsc(const struct dc *dc, + struct resource_context *res_ctx, + struct display_stream_compressor **dsc, + int pipe_idx); struct pipe_ctx *dcn20_find_secondary_pipe(struct dc *dc, struct resource_context *res_ctx, const struct resource_pool *pool, diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c index 5e2d14b897af..129f0b62f751 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c @@ -49,11 +49,6 @@ #define FN(reg_name, field_name) \ hubbub1->shifts->field_name, hubbub1->masks->field_name -#ifdef NUM_VMID -#undef NUM_VMID -#endif -#define NUM_VMID 16 - static uint32_t convert_and_clamp( uint32_t wm_ns, uint32_t refclk_mhz, @@ -138,7 +133,7 @@ int hubbub21_init_dchub(struct hubbub *hubbub, dcn21_dchvm_init(hubbub); - return NUM_VMID; + return hubbub1->num_vmid; } bool hubbub21_program_urgent_watermarks( diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c index d285ba622d61..960a0716dde5 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c @@ -778,21 +778,28 @@ void dmcub_PLAT_54186_wa(struct hubp *hubp, struct surface_flip_registers *flip_ { struct dc_dmub_srv *dmcub = hubp->ctx->dmub_srv; struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); - struct dmub_rb_cmd_PLAT_54186_wa PLAT_54186_wa = { 0 }; - - PLAT_54186_wa.header.type = DMUB_CMD__PLAT_54186_WA; - PLAT_54186_wa.flip.DCSURF_PRIMARY_SURFACE_ADDRESS = flip_regs->DCSURF_PRIMARY_SURFACE_ADDRESS; - PLAT_54186_wa.flip.DCSURF_PRIMARY_SURFACE_ADDRESS_C = flip_regs->DCSURF_PRIMARY_SURFACE_ADDRESS_C; - PLAT_54186_wa.flip.DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH = flip_regs->DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH; - PLAT_54186_wa.flip.DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH_C = flip_regs->DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH_C; - PLAT_54186_wa.flip.flip_params.grph_stereo = flip_regs->grph_stereo; - PLAT_54186_wa.flip.flip_params.hubp_inst = hubp->inst; - PLAT_54186_wa.flip.flip_params.immediate = flip_regs->immediate; - PLAT_54186_wa.flip.flip_params.tmz_surface = flip_regs->tmz_surface; - PLAT_54186_wa.flip.flip_params.vmid = flip_regs->vmid; + union dmub_rb_cmd cmd; + + memset(&cmd, 0, sizeof(cmd)); + + cmd.PLAT_54186_wa.header.type = DMUB_CMD__PLAT_54186_WA; + cmd.PLAT_54186_wa.header.payload_bytes = sizeof(cmd.PLAT_54186_wa.flip); + cmd.PLAT_54186_wa.flip.DCSURF_PRIMARY_SURFACE_ADDRESS = + flip_regs->DCSURF_PRIMARY_SURFACE_ADDRESS; + cmd.PLAT_54186_wa.flip.DCSURF_PRIMARY_SURFACE_ADDRESS_C = + flip_regs->DCSURF_PRIMARY_SURFACE_ADDRESS_C; + cmd.PLAT_54186_wa.flip.DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH = + flip_regs->DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH; + cmd.PLAT_54186_wa.flip.DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH_C = + flip_regs->DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH_C; + cmd.PLAT_54186_wa.flip.flip_params.grph_stereo = flip_regs->grph_stereo; + cmd.PLAT_54186_wa.flip.flip_params.hubp_inst = hubp->inst; + cmd.PLAT_54186_wa.flip.flip_params.immediate = flip_regs->immediate; + cmd.PLAT_54186_wa.flip.flip_params.tmz_surface = flip_regs->tmz_surface; + cmd.PLAT_54186_wa.flip.flip_params.vmid = flip_regs->vmid; PERF_TRACE(); // TODO: remove after performance is stable. - dc_dmub_srv_cmd_queue(dmcub, &PLAT_54186_wa.header); + dc_dmub_srv_cmd_queue(dmcub, &cmd); PERF_TRACE(); // TODO: remove after performance is stable. dc_dmub_srv_cmd_execute(dmcub); PERF_TRACE(); // TODO: remove after performance is stable. diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c index b9ff9767e08f..a5baef7e7a7d 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c @@ -53,6 +53,7 @@ static const struct hw_sequencer_funcs dcn21_funcs = { .disable_plane = dcn20_disable_plane, .pipe_control_lock = dcn20_pipe_control_lock, .interdependent_update_lock = dcn10_lock_all_pipes, + .cursor_lock = dcn10_cursor_lock, .prepare_bandwidth = dcn20_prepare_bandwidth, .optimize_bandwidth = dcn20_optimize_bandwidth, .update_bandwidth = dcn20_update_bandwidth, @@ -85,11 +86,10 @@ static const struct hw_sequencer_funcs dcn21_funcs = { .optimize_pwr_state = dcn21_optimize_pwr_state, .exit_optimized_pwr_state = dcn21_exit_optimized_pwr_state, .get_vupdate_offset_from_vsync = dcn10_get_vupdate_offset_from_vsync, - .set_cursor_position = dcn10_set_cursor_position, - .set_cursor_attribute = dcn10_set_cursor_attribute, - .set_cursor_sdr_white_level = dcn10_set_cursor_sdr_white_level, - .optimize_pwr_state = dcn21_optimize_pwr_state, - .exit_optimized_pwr_state = dcn21_exit_optimized_pwr_state, + .calc_vupdate_position = dcn10_calc_vupdate_position, + .power_down = dce110_power_down, + .set_backlight_level = dce110_set_backlight_level, + .set_abm_immediate_disable = dce110_set_abm_immediate_disable, }; static const struct hwseq_private_funcs dcn21_private_funcs = { @@ -105,8 +105,6 @@ static const struct hwseq_private_funcs dcn21_private_funcs = { .reset_hw_ctx_wrap = dcn20_reset_hw_ctx_wrap, .enable_stream_timing = dcn20_enable_stream_timing, .edp_backlight_control = dce110_edp_backlight_control, - .is_panel_backlight_on = dce110_is_panel_backlight_on, - .is_panel_powered_on = dce110_is_panel_powered_on, .disable_stream_gating = dcn20_disable_stream_gating, .enable_stream_gating = dcn20_enable_stream_gating, .setup_vupdate_interrupt = dcn20_setup_vupdate_interrupt, diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_link_encoder.c index e45683ac871a..aa46c35b05a2 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_link_encoder.c @@ -203,29 +203,6 @@ static bool update_cfg_data( return true; } -void dcn21_link_encoder_get_max_link_cap(struct link_encoder *enc, - struct dc_link_settings *link_settings) -{ - struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); - uint32_t value; - - REG_GET(RDPCSTX_PHY_CNTL6, RDPCS_PHY_DPALT_DP4, &value); - - if (!value && link_settings->lane_count > LANE_COUNT_TWO) - link_settings->lane_count = LANE_COUNT_TWO; -} - -bool dcn21_link_encoder_is_in_alt_mode(struct link_encoder *enc) -{ - struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); - uint32_t value; - - REG_GET(RDPCSTX_PHY_CNTL6, RDPCS_PHY_DPALT_DISABLE, &value); - - // if value == 1 alt mode is disabled, otherwise it is enabled - return !value; -} - bool dcn21_link_encoder_acquire_phy(struct link_encoder *enc) { struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); @@ -348,8 +325,8 @@ static const struct link_encoder_funcs dcn21_link_enc_funcs = { .fec_set_ready = enc2_fec_set_ready, .fec_is_active = enc2_fec_is_active, .get_dig_frontend = dcn10_get_dig_frontend, - .is_in_alt_mode = dcn21_link_encoder_is_in_alt_mode, - .get_max_link_cap = dcn21_link_encoder_get_max_link_cap, + .is_in_alt_mode = dcn20_link_encoder_is_in_alt_mode, + .get_max_link_cap = dcn20_link_encoder_get_max_link_cap, }; void dcn21_link_encoder_construct( diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c index b25484aa8222..f00a56835084 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c @@ -61,6 +61,7 @@ #include "dcn21_hubbub.h" #include "dcn10/dcn10_resource.h" #include "dce110/dce110_resource.h" +#include "dce/dce_panel_cntl.h" #include "dcn20/dcn20_dwb.h" #include "dcn20/dcn20_mmhubbub.h" @@ -85,6 +86,7 @@ #include "vm_helper.h" #include "dcn20/dcn20_vmid.h" #include "dce/dmub_psr.h" +#include "dce/dmub_abm.h" #define SOC_BOUNDING_BOX_VALID false #define DC_LOGGER_INIT(logger) @@ -284,7 +286,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc = { .dram_channel_width_bytes = 4, .fabric_datapath_to_dcn_data_return_bytes = 32, .dcn_downspread_percent = 0.5, - .downspread_percent = 0.5, + .downspread_percent = 0.38, .dram_page_open_time_ns = 50.0, .dram_rw_turnaround_time_ns = 17.5, .dram_return_buffer_per_channel_bytes = 8192, @@ -340,6 +342,10 @@ struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc = { .block ## _ ## reg_name[id] = BASE(mm ## block ## id ## _ ## reg_name ## _BASE_IDX) + \ mm ## block ## id ## _ ## reg_name +#define VUPDATE_SRII(reg_name, block, id)\ + .reg_name[id] = BASE(mm ## reg_name ## _ ## block ## id ## _BASE_IDX) + \ + mm ## reg_name ## _ ## block ## id + /* NBIO */ #define NBIO_BASE_INNER(seg) \ NBIF0_BASE__INST0_SEG ## seg @@ -799,7 +805,7 @@ static const struct resource_caps res_cap_rn = { .num_pll = 5, // maybe 3 because the last two used for USB-c .num_dwb = 1, .num_ddc = 5, - .num_vmid = 1, + .num_vmid = 16, .num_dsc = 3, }; @@ -991,9 +997,12 @@ static void dcn21_resource_destruct(struct dcn21_resource_pool *pool) pool->base.dp_clock_source = NULL; } - - if (pool->base.abm != NULL) - dce_abm_destroy(&pool->base.abm); + if (pool->base.abm != NULL) { + if (pool->base.abm->ctx->dc->config.disable_dmcu) + dmub_abm_destroy(&pool->base.abm); + else + dce_abm_destroy(&pool->base.abm); + } if (pool->base.dmcu != NULL) dce_dmcu_destroy(&pool->base.dmcu); @@ -1286,6 +1295,7 @@ static struct hubbub *dcn21_hubbub_create(struct dc_context *ctx) vmid->shifts = &vmid_shifts; vmid->masks = &vmid_masks; } + hubbub->num_vmid = res_cap_rn.num_vmid; return &hubbub->base; } @@ -1374,64 +1384,50 @@ static void update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_param { struct dcn21_resource_pool *pool = TO_DCN21_RES_POOL(dc->res_pool); struct clk_limit_table *clk_table = &bw_params->clk_table; - unsigned int i, j, k; - int closest_clk_lvl; + struct _vcs_dpi_voltage_scaling_st clock_limits[DC__VOLTAGE_STATES]; + unsigned int i, closest_clk_lvl; + int j; // Default clock levels are used for diags, which may lead to overclocking. - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) && !IS_DIAG_DC(dc->ctx->dce_environment)) { + if (!IS_DIAG_DC(dc->ctx->dce_environment)) { dcn2_1_ip.max_num_otg = pool->base.res_cap->num_timing_generator; dcn2_1_ip.max_num_dpp = pool->base.pipe_count; dcn2_1_soc.num_chans = bw_params->num_channels; - /* Vmin: leave lowest DCN clocks, override with dcfclk, fclk, memclk from fuse */ - dcn2_1_soc.clock_limits[0].state = 0; - dcn2_1_soc.clock_limits[0].dcfclk_mhz = clk_table->entries[0].dcfclk_mhz; - dcn2_1_soc.clock_limits[0].fabricclk_mhz = clk_table->entries[0].fclk_mhz; - dcn2_1_soc.clock_limits[0].socclk_mhz = clk_table->entries[0].socclk_mhz; - dcn2_1_soc.clock_limits[0].dram_speed_mts = clk_table->entries[0].memclk_mhz * 2; - - /* - * Other levels: find closest DCN clocks that fit the given clock limit using dcfclk - * as indicator - */ - - closest_clk_lvl = -1; - /* index currently being filled */ - k = 1; - for (i = 1; i < clk_table->num_entries; i++) { - /* loop backwards, skip duplicate state*/ - for (j = dcn2_1_soc.num_states - 1; j >= k; j--) { + ASSERT(clk_table->num_entries); + for (i = 0; i < clk_table->num_entries; i++) { + /* loop backwards*/ + for (closest_clk_lvl = 0, j = dcn2_1_soc.num_states - 1; j >= 0; j--) { if ((unsigned int) dcn2_1_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) { closest_clk_lvl = j; break; } } - /* if found a lvl that fits, use the DCN clks from it, if not, go to next clk limit*/ - if (closest_clk_lvl != -1) { - dcn2_1_soc.clock_limits[k].state = i; - dcn2_1_soc.clock_limits[k].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz; - dcn2_1_soc.clock_limits[k].fabricclk_mhz = clk_table->entries[i].fclk_mhz; - dcn2_1_soc.clock_limits[k].socclk_mhz = clk_table->entries[i].socclk_mhz; - dcn2_1_soc.clock_limits[k].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2; - - dcn2_1_soc.clock_limits[k].dispclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dispclk_mhz; - dcn2_1_soc.clock_limits[k].dppclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dppclk_mhz; - dcn2_1_soc.clock_limits[k].dram_bw_per_chan_gbps = dcn2_1_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps; - dcn2_1_soc.clock_limits[k].dscclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dscclk_mhz; - dcn2_1_soc.clock_limits[k].dtbclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dtbclk_mhz; - dcn2_1_soc.clock_limits[k].phyclk_d18_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz; - dcn2_1_soc.clock_limits[k].phyclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].phyclk_mhz; - k++; - } + clock_limits[i].state = i; + clock_limits[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz; + clock_limits[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz; + clock_limits[i].socclk_mhz = clk_table->entries[i].socclk_mhz; + clock_limits[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2; + + clock_limits[i].dispclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dispclk_mhz; + clock_limits[i].dppclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dppclk_mhz; + clock_limits[i].dram_bw_per_chan_gbps = dcn2_1_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps; + clock_limits[i].dscclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dscclk_mhz; + clock_limits[i].dtbclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].dtbclk_mhz; + clock_limits[i].phyclk_d18_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz; + clock_limits[i].phyclk_mhz = dcn2_1_soc.clock_limits[closest_clk_lvl].phyclk_mhz; + } + for (i = 0; i < clk_table->num_entries; i++) + dcn2_1_soc.clock_limits[i] = clock_limits[i]; + if (clk_table->num_entries) { + dcn2_1_soc.num_states = clk_table->num_entries; + /* duplicate last level */ + dcn2_1_soc.clock_limits[dcn2_1_soc.num_states] = dcn2_1_soc.clock_limits[dcn2_1_soc.num_states - 1]; + dcn2_1_soc.clock_limits[dcn2_1_soc.num_states].state = dcn2_1_soc.num_states; } - dcn2_1_soc.num_states = k; } - /* duplicate last level */ - dcn2_1_soc.clock_limits[dcn2_1_soc.num_states] = dcn2_1_soc.clock_limits[dcn2_1_soc.num_states - 1]; - dcn2_1_soc.clock_limits[dcn2_1_soc.num_states].state = dcn2_1_soc.num_states; - dml_init_instance(&dc->dml, &dcn2_1_soc, &dcn2_1_ip, DML_PROJECT_DCN21); } @@ -1602,6 +1598,18 @@ static const struct dcn10_link_enc_registers link_enc_regs[] = { link_regs(4, E), }; +static const struct dce_panel_cntl_registers panel_cntl_regs[] = { + { DCN_PANEL_CNTL_REG_LIST() } +}; + +static const struct dce_panel_cntl_shift panel_cntl_shift = { + DCE_PANEL_CNTL_MASK_SH_LIST(__SHIFT) +}; + +static const struct dce_panel_cntl_mask panel_cntl_mask = { + DCE_PANEL_CNTL_MASK_SH_LIST(_MASK) +}; + #define aux_regs(id)\ [id] = {\ DCN2_AUX_REG_LIST(id)\ @@ -1687,6 +1695,24 @@ static struct link_encoder *dcn21_link_encoder_create( return &enc21->enc10.base; } + +static struct panel_cntl *dcn21_panel_cntl_create(const struct panel_cntl_init_data *init_data) +{ + struct dce_panel_cntl *panel_cntl = + kzalloc(sizeof(struct dce_panel_cntl), GFP_KERNEL); + + if (!panel_cntl) + return NULL; + + dce_panel_cntl_construct(panel_cntl, + init_data, + &panel_cntl_regs[init_data->inst], + &panel_cntl_shift, + &panel_cntl_mask); + + return &panel_cntl->base; +} + #define CTX ctx #define REG(reg_name) \ @@ -1705,12 +1731,8 @@ static int dcn21_populate_dml_pipes_from_context( { uint32_t pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, context, pipes); int i; - struct resource_context *res_ctx = &context->res_ctx; - - for (i = 0; i < dc->res_pool->pipe_count; i++) { - if (!res_ctx->pipe_ctx[i].stream) - continue; + for (i = 0; i < pipe_cnt; i++) { pipes[i].pipe.src.hostvm = 1; pipes[i].pipe.src.gpuvm = 1; @@ -1735,6 +1757,7 @@ enum dc_status dcn21_patch_unknown_plane_state(struct dc_plane_state *plane_stat static struct resource_funcs dcn21_res_pool_funcs = { .destroy = dcn21_destroy_resource_pool, .link_enc_create = dcn21_link_encoder_create, + .panel_cntl_create = dcn21_panel_cntl_create, .validate_bandwidth = dcn21_validate_bandwidth, .populate_dml_pipes = dcn21_populate_dml_pipes_from_context, .add_stream_to_ctx = dcn20_add_stream_to_ctx, @@ -1781,7 +1804,6 @@ static bool dcn21_resource_construct( dc->caps.i2c_speed_in_khz = 100; dc->caps.max_cursor_size = 256; dc->caps.dmdata_alloc_size = 2048; - dc->caps.hw_3d_lut = true; dc->caps.max_slave_planes = 1; dc->caps.post_blend_color_processing = true; @@ -1790,6 +1812,40 @@ static bool dcn21_resource_construct( dc->caps.dmcub_support = true; dc->caps.is_apu = true; + /* Color pipeline capabilities */ + dc->caps.color.dpp.dcn_arch = 1; + dc->caps.color.dpp.input_lut_shared = 0; + dc->caps.color.dpp.icsc = 1; + dc->caps.color.dpp.dgam_ram = 1; + dc->caps.color.dpp.dgam_rom_caps.srgb = 1; + dc->caps.color.dpp.dgam_rom_caps.bt2020 = 1; + dc->caps.color.dpp.dgam_rom_caps.gamma2_2 = 0; + dc->caps.color.dpp.dgam_rom_caps.pq = 0; + dc->caps.color.dpp.dgam_rom_caps.hlg = 0; + dc->caps.color.dpp.post_csc = 0; + dc->caps.color.dpp.gamma_corr = 0; + + dc->caps.color.dpp.hw_3d_lut = 1; + dc->caps.color.dpp.ogam_ram = 1; + // no OGAM ROM on DCN2 + dc->caps.color.dpp.ogam_rom_caps.srgb = 0; + dc->caps.color.dpp.ogam_rom_caps.bt2020 = 0; + dc->caps.color.dpp.ogam_rom_caps.gamma2_2 = 0; + dc->caps.color.dpp.ogam_rom_caps.pq = 0; + dc->caps.color.dpp.ogam_rom_caps.hlg = 0; + dc->caps.color.dpp.ocsc = 0; + + dc->caps.color.mpc.gamut_remap = 0; + dc->caps.color.mpc.num_3dluts = 0; + dc->caps.color.mpc.shared_3d_lut = 0; + dc->caps.color.mpc.ogam_ram = 1; + dc->caps.color.mpc.ogam_rom_caps.srgb = 0; + dc->caps.color.mpc.ogam_rom_caps.bt2020 = 0; + dc->caps.color.mpc.ogam_rom_caps.gamma2_2 = 0; + dc->caps.color.mpc.ogam_rom_caps.pq = 0; + dc->caps.color.mpc.ogam_rom_caps.hlg = 0; + dc->caps.color.mpc.ocsc = 1; + if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) dc->debug = debug_defaults_drv; else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) { @@ -1842,17 +1898,19 @@ static bool dcn21_resource_construct( goto create_fail; } - pool->base.dmcu = dcn21_dmcu_create(ctx, - &dmcu_regs, - &dmcu_shift, - &dmcu_mask); - if (pool->base.dmcu == NULL) { - dm_error("DC: failed to create dmcu!\n"); - BREAK_TO_DEBUGGER(); - goto create_fail; + if (!dc->config.disable_dmcu) { + pool->base.dmcu = dcn21_dmcu_create(ctx, + &dmcu_regs, + &dmcu_shift, + &dmcu_mask); + if (pool->base.dmcu == NULL) { + dm_error("DC: failed to create dmcu!\n"); + BREAK_TO_DEBUGGER(); + goto create_fail; + } } - if (dc->debug.disable_dmcu) { + if (dc->config.disable_dmcu) { pool->base.psr = dmub_psr_create(ctx); if (pool->base.psr == NULL) { @@ -1862,15 +1920,16 @@ static bool dcn21_resource_construct( } } - pool->base.abm = dce_abm_create(ctx, + if (dc->config.disable_dmcu) + pool->base.abm = dmub_abm_create(ctx, + &abm_regs, + &abm_shift, + &abm_mask); + else + pool->base.abm = dce_abm_create(ctx, &abm_regs, &abm_shift, &abm_mask); - if (pool->base.abm == NULL) { - dm_error("DC: failed to create abm!\n"); - BREAK_TO_DEBUGGER(); - goto create_fail; - } pool->base.pp_smu = dcn21_pp_smu_create(ctx); diff --git a/drivers/gpu/drm/amd/display/dc/dml/Makefile b/drivers/gpu/drm/amd/display/dc/dml/Makefile index 7ee8b8460a9b..e34c3376efc1 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile @@ -63,10 +63,8 @@ CFLAGS_$(AMDDALPATH)/dc/dml/dcn21/display_rq_dlg_calc_21.o := $(dml_ccflags) endif CFLAGS_$(AMDDALPATH)/dc/dml/dml1_display_rq_dlg_calc.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/display_rq_dlg_helpers.o := $(dml_ccflags) -CFLAGS_$(AMDDALPATH)/dc/dml/dml_common_defs.o := $(dml_ccflags) DML = display_mode_lib.o display_rq_dlg_helpers.o dml1_display_rq_dlg_calc.o \ - dml_common_defs.o ifdef CONFIG_DRM_AMD_DC_DCN DML += display_mode_vba.o dcn20/display_rq_dlg_calc_20.o dcn20/display_mode_vba_20.o diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c index 5bbbafacc720..80170f9721ce 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c @@ -2599,21 +2599,44 @@ static void dml20v2_DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndP } } + { + float SecondMinActiveDRAMClockChangeMarginOneDisplayInVBLank = 999999; + int PlaneWithMinActiveDRAMClockChangeMargin = -1; + mode_lib->vba.MinActiveDRAMClockChangeMargin = 999999; for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { if (mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] < mode_lib->vba.MinActiveDRAMClockChangeMargin) { mode_lib->vba.MinActiveDRAMClockChangeMargin = mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k]; + if (mode_lib->vba.BlendingAndTiming[k] == k) { + PlaneWithMinActiveDRAMClockChangeMargin = k; + } else { + for (j = 0; j < mode_lib->vba.NumberOfActivePlanes; ++j) { + if (mode_lib->vba.BlendingAndTiming[k] == j) { + PlaneWithMinActiveDRAMClockChangeMargin = j; + } + } + } } } mode_lib->vba.MinActiveDRAMClockChangeLatencySupported = mode_lib->vba.MinActiveDRAMClockChangeMargin + mode_lib->vba.DRAMClockChangeLatency; + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { + if (!((k == PlaneWithMinActiveDRAMClockChangeMargin) && (mode_lib->vba.BlendingAndTiming[k] == k)) + && !(mode_lib->vba.BlendingAndTiming[k] == PlaneWithMinActiveDRAMClockChangeMargin) + && mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k] + < SecondMinActiveDRAMClockChangeMarginOneDisplayInVBLank) { + SecondMinActiveDRAMClockChangeMarginOneDisplayInVBLank = + mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k]; + } + } if (mode_lib->vba.DRAMClockChangeSupportsVActive && mode_lib->vba.MinActiveDRAMClockChangeMargin > 60) { + mode_lib->vba.DRAMClockChangeWatermark += 25; for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { if (mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb] == 0) { @@ -2622,13 +2645,17 @@ static void dml20v2_DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndP mode_lib->vba.MinTTUVBlank[k] += 25; } } - mode_lib->vba.DRAMClockChangeWatermark += 25; + mode_lib->vba.DRAMClockChangeSupport[0][0] = dm_dram_clock_change_vactive; } else if (mode_lib->vba.DummyPStateCheck && mode_lib->vba.MinActiveDRAMClockChangeMargin > 0) { mode_lib->vba.DRAMClockChangeSupport[0][0] = dm_dram_clock_change_vactive; } else { - if (mode_lib->vba.SynchronizedVBlank || mode_lib->vba.NumberOfActivePlanes == 1) { + if ((mode_lib->vba.SynchronizedVBlank + || mode_lib->vba.NumberOfActivePlanes == 1 + || (SecondMinActiveDRAMClockChangeMarginOneDisplayInVBLank > 0 && + mode_lib->vba.AllowDramClockChangeOneDisplayVactive)) + && mode_lib->vba.PrefetchMode[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb] == 0) { mode_lib->vba.DRAMClockChangeSupport[0][0] = dm_dram_clock_change_vblank; for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; ++k) { if (!mode_lib->vba.AllowDRAMClockChangeDuringVBlank[k]) { @@ -2640,6 +2667,7 @@ static void dml20v2_DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndP mode_lib->vba.DRAMClockChangeSupport[0][0] = dm_dram_clock_change_unsupported; } } + } for (k = 0; k <= mode_lib->vba.soc.num_states; k++) for (j = 0; j < 2; j++) mode_lib->vba.DRAMClockChangeSupport[k][j] = mode_lib->vba.DRAMClockChangeSupport[0][0]; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.h b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.h index 8c86b63ddf07..1e557ddcb638 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.h @@ -26,7 +26,6 @@ #ifndef __DML20_DISPLAY_RQ_DLG_CALC_H__ #define __DML20_DISPLAY_RQ_DLG_CALC_H__ -#include "../dml_common_defs.h" #include "../display_rq_dlg_helpers.h" struct display_mode_lib; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.h b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.h index 0378406bf7e7..0d53e871a9d1 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.h @@ -26,7 +26,6 @@ #ifndef __DML20V2_DISPLAY_RQ_DLG_CALC_H__ #define __DML20V2_DISPLAY_RQ_DLG_CALC_H__ -#include "../dml_common_defs.h" #include "../display_rq_dlg_helpers.h" struct display_mode_lib; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c index e6617c958bb8..a576eed94d9b 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c @@ -3190,6 +3190,7 @@ static void CalculateFlipSchedule( double TimeForFetchingRowInVBlankImmediateFlip; double ImmediateFlipBW; double HostVMInefficiencyFactor; + double VRatioClamped; if (GPUVMEnable == true && HostVMEnable == true) { HostVMInefficiencyFactor = @@ -3222,31 +3223,32 @@ static void CalculateFlipSchedule( *DestinationLinesToRequestRowInImmediateFlip = dml_ceil(4.0 * (TimeForFetchingRowInVBlankImmediateFlip / LineTime), 1) / 4.0; *final_flip_bw = dml_max(PDEAndMetaPTEBytesPerFrame * HostVMInefficiencyFactor / (*DestinationLinesToRequestVMInImmediateFlip * LineTime), (MetaRowBytes + DPTEBytesPerRow) * HostVMInefficiencyFactor / (*DestinationLinesToRequestRowInImmediateFlip * LineTime)); + VRatioClamped = (VRatio < 1.0) ? 1.0 : VRatio; if (SourcePixelFormat == dm_420_8 || SourcePixelFormat == dm_420_10) { if (GPUVMEnable == true && DCCEnable != true) { min_row_time = dml_min( - dpte_row_height * LineTime / VRatio, - dpte_row_height_chroma * LineTime / (VRatio / 2)); + dpte_row_height * LineTime / VRatioClamped, + dpte_row_height_chroma * LineTime / (VRatioClamped / 2)); } else if (GPUVMEnable != true && DCCEnable == true) { min_row_time = dml_min( - meta_row_height * LineTime / VRatio, - meta_row_height_chroma * LineTime / (VRatio / 2)); + meta_row_height * LineTime / VRatioClamped, + meta_row_height_chroma * LineTime / (VRatioClamped / 2)); } else { min_row_time = dml_min4( - dpte_row_height * LineTime / VRatio, - meta_row_height * LineTime / VRatio, - dpte_row_height_chroma * LineTime / (VRatio / 2), - meta_row_height_chroma * LineTime / (VRatio / 2)); + dpte_row_height * LineTime / VRatioClamped, + meta_row_height * LineTime / VRatioClamped, + dpte_row_height_chroma * LineTime / (VRatioClamped / 2), + meta_row_height_chroma * LineTime / (VRatioClamped / 2)); } } else { if (GPUVMEnable == true && DCCEnable != true) { - min_row_time = dpte_row_height * LineTime / VRatio; + min_row_time = dpte_row_height * LineTime / VRatioClamped; } else if (GPUVMEnable != true && DCCEnable == true) { - min_row_time = meta_row_height * LineTime / VRatio; + min_row_time = meta_row_height * LineTime / VRatioClamped; } else { min_row_time = dml_min( - dpte_row_height * LineTime / VRatio, - meta_row_height * LineTime / VRatio); + dpte_row_height * LineTime / VRatioClamped, + meta_row_height * LineTime / VRatioClamped); } } @@ -5944,7 +5946,7 @@ static void CalculateMetaAndPTETimes( * PixelPTEReqHeightY[k]; } dpte_groups_per_row_luma_ub = dml_ceil( - dpte_row_width_luma_ub[k] / dpte_group_width_luma, + (float) dpte_row_width_luma_ub[k] / dpte_group_width_luma, 1); time_per_pte_group_nom_luma[k] = DST_Y_PER_PTE_ROW_NOM_L[k] * HTotal[k] / PixelClock[k] / dpte_groups_per_row_luma_ub; @@ -5968,7 +5970,7 @@ static void CalculateMetaAndPTETimes( * PixelPTEReqHeightC[k]; } dpte_groups_per_row_chroma_ub = dml_ceil( - dpte_row_width_chroma_ub[k] + (float) dpte_row_width_chroma_ub[k] / dpte_group_width_chroma, 1); time_per_pte_group_nom_chroma[k] = DST_Y_PER_PTE_ROW_NOM_C[k] diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c index a38baa73d484..90a5fefef05b 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c @@ -1200,7 +1200,7 @@ static void dml_rq_dlg_get_dlg_params( min_hratio_fact_l = 1.0; min_hratio_fact_c = 1.0; - if (htaps_l <= 1) + if (hratio_l <= 1) min_hratio_fact_l = 2.0; else if (htaps_l <= 6) { if ((hratio_l * 2.0) > 4.0) @@ -1216,7 +1216,7 @@ static void dml_rq_dlg_get_dlg_params( hscale_pixel_rate_l = min_hratio_fact_l * dppclk_freq_in_mhz; - if (htaps_c <= 1) + if (hratio_c <= 1) min_hratio_fact_c = 2.0; else if (htaps_c <= 6) { if ((hratio_c * 2.0) > 4.0) @@ -1490,19 +1490,30 @@ static void dml_rq_dlg_get_dlg_params( disp_dlg_regs->refcyc_per_pte_group_vblank_l = (unsigned int) (dst_y_per_row_vblank * (double) htotal * ref_freq_to_pix_freq / (double) dpte_groups_per_row_ub_l); - ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_l < (unsigned int)dml_pow(2, 13)); + if ((refclk_freq_in_mhz / ref_freq_to_pix_freq < 28) && + disp_dlg_regs->refcyc_per_pte_group_vblank_l >= (unsigned int)dml_pow(2, 13)) + disp_dlg_regs->refcyc_per_pte_group_vblank_l = (1 << 13) - 1; + else + ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_l < (unsigned int)dml_pow(2, 13)); if (dual_plane) { disp_dlg_regs->refcyc_per_pte_group_vblank_c = (unsigned int) (dst_y_per_row_vblank * (double) htotal * ref_freq_to_pix_freq / (double) dpte_groups_per_row_ub_c); - ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_c + if ((refclk_freq_in_mhz / ref_freq_to_pix_freq < 28) && + disp_dlg_regs->refcyc_per_pte_group_vblank_c >= (unsigned int)dml_pow(2, 13)) + disp_dlg_regs->refcyc_per_pte_group_vblank_c = (1 << 13) - 1; + else + ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_c < (unsigned int)dml_pow(2, 13)); } - disp_dlg_regs->refcyc_per_meta_chunk_vblank_l = + if (src->dcc) + disp_dlg_regs->refcyc_per_meta_chunk_vblank_l = (unsigned int) (dst_y_per_row_vblank * (double) htotal * ref_freq_to_pix_freq / (double) meta_chunks_per_row_ub_l); + else + disp_dlg_regs->refcyc_per_meta_chunk_vblank_l = 0; ASSERT(disp_dlg_regs->refcyc_per_meta_chunk_vblank_l < (unsigned int)dml_pow(2, 13)); disp_dlg_regs->refcyc_per_meta_chunk_vblank_c = @@ -1522,8 +1533,8 @@ static void dml_rq_dlg_get_dlg_params( disp_dlg_regs->refcyc_per_vm_group_vblank = get_refcyc_per_vm_group_vblank(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz; disp_dlg_regs->refcyc_per_vm_group_flip = get_refcyc_per_vm_group_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz; - disp_dlg_regs->refcyc_per_vm_req_vblank = get_refcyc_per_vm_req_vblank(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz; - disp_dlg_regs->refcyc_per_vm_req_flip = get_refcyc_per_vm_req_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz; + disp_dlg_regs->refcyc_per_vm_req_vblank = get_refcyc_per_vm_req_vblank(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz * dml_pow(2, 10); + disp_dlg_regs->refcyc_per_vm_req_flip = get_refcyc_per_vm_req_flip(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz * dml_pow(2, 10); // Clamp to max for now if (disp_dlg_regs->refcyc_per_vm_group_vblank >= (unsigned int)dml_pow(2, 23)) diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.h b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.h index 83e95f8cbff2..e8f7785e3fc6 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.h @@ -26,7 +26,7 @@ #ifndef __DML21_DISPLAY_RQ_DLG_CALC_H__ #define __DML21_DISPLAY_RQ_DLG_CALC_H__ -#include "../dml_common_defs.h" +#include "dm_services.h" #include "../display_rq_dlg_helpers.h" struct display_mode_lib; diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h index cf2758ca5b02..c77c3d827e4a 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h @@ -25,8 +25,10 @@ #ifndef __DISPLAY_MODE_LIB_H__ #define __DISPLAY_MODE_LIB_H__ - -#include "dml_common_defs.h" +#include "dm_services.h" +#include "dc_features.h" +#include "display_mode_structs.h" +#include "display_mode_enums.h" #include "display_mode_vba.h" enum dml_project { diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h index 687010c17324..439ffd04be34 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h @@ -118,9 +118,11 @@ struct _vcs_dpi_soc_bounding_box_st { double urgent_latency_adjustment_fabric_clock_component_us; double urgent_latency_adjustment_fabric_clock_reference_mhz; bool disable_dram_clock_change_vactive_support; + bool allow_dram_clock_one_display_vactive; }; struct _vcs_dpi_ip_params_st { + bool use_min_dcfclk; bool gpuvm_enable; bool hostvm_enable; unsigned int gpuvm_max_page_table_levels; diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c index 6b525c52124c..b19988f54721 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c @@ -224,6 +224,7 @@ static void fetch_socbb_params(struct display_mode_lib *mode_lib) mode_lib->vba.DummyPStateCheck = soc->dram_clock_change_latency_us == soc->dummy_pstate_latency_us; mode_lib->vba.DRAMClockChangeSupportsVActive = !soc->disable_dram_clock_change_vactive_support || mode_lib->vba.DummyPStateCheck; + mode_lib->vba.AllowDramClockChangeOneDisplayVactive = soc->allow_dram_clock_one_display_vactive; mode_lib->vba.Downspreading = soc->downspread_percent; mode_lib->vba.DRAMChannelWidth = soc->dram_channel_width_bytes; // new! @@ -280,6 +281,7 @@ static void fetch_ip_params(struct display_mode_lib *mode_lib) ip_params_st *ip = &mode_lib->vba.ip; // IP Parameters + mode_lib->vba.UseMinimumRequiredDCFCLK = ip->use_min_dcfclk; mode_lib->vba.MaxNumDPP = ip->max_num_dpp; mode_lib->vba.MaxNumOTG = ip->max_num_otg; mode_lib->vba.MaxNumHDMIFRLOutputs = ip->max_num_hdmi_frl_outputs; diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h index 5d82fc5a7ed7..3f559e725ab1 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h @@ -27,8 +27,6 @@ #ifndef __DML2_DISPLAY_MODE_VBA_H__ #define __DML2_DISPLAY_MODE_VBA_H__ -#include "dml_common_defs.h" - struct display_mode_lib; void ModeSupportAndSystemConfiguration(struct display_mode_lib *mode_lib); @@ -898,6 +896,8 @@ struct vba_vars_st { bool dummystring[DC__NUM_DPP__MAX]; double BPP; enum odm_combine_policy ODMCombinePolicy; + bool UseMinimumRequiredDCFCLK; + bool AllowDramClockChangeOneDisplayVactive; }; bool CalculateMinAndMaxPrefetchMode( diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_rq_dlg_helpers.h b/drivers/gpu/drm/amd/display/dc/dml/display_rq_dlg_helpers.h index 1f24db830737..2555ef0358c2 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_rq_dlg_helpers.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_rq_dlg_helpers.h @@ -26,7 +26,6 @@ #ifndef __DISPLAY_RQ_DLG_HELPERS_H__ #define __DISPLAY_RQ_DLG_HELPERS_H__ -#include "dml_common_defs.h" #include "display_mode_lib.h" /* Function: Printer functions diff --git a/drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.h b/drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.h index 304164986bd8..9c06913ad767 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.h @@ -26,8 +26,6 @@ #ifndef __DISPLAY_RQ_DLG_CALC_H__ #define __DISPLAY_RQ_DLG_CALC_H__ -#include "dml_common_defs.h" - struct display_mode_lib; #include "display_rq_dlg_helpers.h" diff --git a/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h b/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h index ded71ea82413..02e06c9b3230 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dml_inline_defs.h @@ -26,7 +26,6 @@ #ifndef __DML_INLINE_DEFS_H__ #define __DML_INLINE_DEFS_H__ -#include "dml_common_defs.h" #include "dcn_calc_math.h" #include "dml_logger.h" @@ -75,6 +74,18 @@ static inline double dml_floor(double a, double granularity) return (double) dcn_bw_floor2(a, granularity); } +static inline double dml_round(double a) +{ + double round_pt = 0.5; + double ceil = dml_ceil(a, 1); + double floor = dml_floor(a, 1); + + if (a - floor >= round_pt) + return ceil; + else + return floor; +} + static inline int dml_log2(double x) { return dml_round((double)dcn_bw_log(x, 2)); @@ -112,7 +123,7 @@ static inline double dml_log(double x, double base) static inline unsigned int dml_round_to_multiple(unsigned int num, unsigned int multiple, - bool up) + unsigned char up) { unsigned int remainder; diff --git a/drivers/gpu/drm/amd/display/dc/dsc/Makefile b/drivers/gpu/drm/amd/display/dc/dsc/Makefile index 3f66868df171..ea29cf95d470 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dsc/Makefile @@ -28,8 +28,6 @@ endif endif CFLAGS_$(AMDDALPATH)/dc/dsc/rc_calc.o := $(dsc_ccflags) -CFLAGS_$(AMDDALPATH)/dc/dsc/rc_calc_dpi.o := $(dsc_ccflags) -CFLAGS_$(AMDDALPATH)/dc/dsc/dc_dsc.o := $(dsc_ccflags) DSC = dc_dsc.o rc_calc.o rc_calc_dpi.o diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c index 0ea6662a1563..0c7f247bb7de 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c +++ b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c @@ -22,10 +22,12 @@ * Author: AMD */ +#include <drm/drm_dsc.h> #include "dc_hw_types.h" #include "dsc.h" #include <drm/drm_dp_helper.h> #include "dc.h" +#include "rc_calc.h" /* This module's internal functions */ @@ -304,22 +306,6 @@ static inline uint32_t dsc_div_by_10_round_up(uint32_t value) return (value + 9) / 10; } -static inline uint32_t calc_dsc_bpp_x16(uint32_t stream_bandwidth_kbps, uint32_t pix_clk_100hz, uint32_t bpp_increment_div) -{ - uint32_t dsc_target_bpp_x16; - float f_dsc_target_bpp; - float f_stream_bandwidth_100bps = stream_bandwidth_kbps * 10.0f; - uint32_t precision = bpp_increment_div; // bpp_increment_div is actually precision - - f_dsc_target_bpp = f_stream_bandwidth_100bps / pix_clk_100hz; - - // Round down to the nearest precision stop to bring it into DSC spec range - dsc_target_bpp_x16 = (uint32_t)(f_dsc_target_bpp * precision); - dsc_target_bpp_x16 = (dsc_target_bpp_x16 * 16) / precision; - - return dsc_target_bpp_x16; -} - /* Get DSC bandwidth range based on [min_bpp, max_bpp] target bitrate range, and timing's pixel clock * and uncompressed bandwidth. */ diff --git a/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.c b/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.c index 03ae15946c6d..667afbc260f9 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.c +++ b/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.c @@ -23,6 +23,7 @@ * Authors: AMD * */ +#include <drm/drm_dsc.h> #include "os_types.h" #include "rc_calc.h" @@ -40,7 +41,8 @@ break -void get_qp_set(qp_set qps, enum colour_mode cm, enum bits_per_comp bpc, enum max_min max_min, float bpp) +static void get_qp_set(qp_set qps, enum colour_mode cm, enum bits_per_comp bpc, + enum max_min max_min, float bpp) { int mode = MODE_SELECT(444, 422, 420); int sel = table_hash(mode, bpc, max_min); @@ -85,7 +87,7 @@ void get_qp_set(qp_set qps, enum colour_mode cm, enum bits_per_comp bpc, enum ma memcpy(qps, table[index].qps, sizeof(qp_set)); } -double dsc_roundf(double num) +static double dsc_roundf(double num) { if (num < 0.0) num = num - 0.5; @@ -95,7 +97,7 @@ double dsc_roundf(double num) return (int)(num); } -double dsc_ceil(double num) +static double dsc_ceil(double num) { double retval = (int)num; @@ -105,7 +107,7 @@ double dsc_ceil(double num) return (int)retval; } -void get_ofs_set(qp_set ofs, enum colour_mode mode, float bpp) +static void get_ofs_set(qp_set ofs, enum colour_mode mode, float bpp) { int *p = ofs; @@ -160,7 +162,7 @@ void get_ofs_set(qp_set ofs, enum colour_mode mode, float bpp) } } -int median3(int a, int b, int c) +static int median3(int a, int b, int c) { if (a > b) swap(a, b); @@ -172,13 +174,25 @@ int median3(int a, int b, int c) return b; } -void calc_rc_params(struct rc_params *rc, enum colour_mode cm, enum bits_per_comp bpc, float bpp, int slice_width, int slice_height, int minor_version) +static void _do_calc_rc_params(struct rc_params *rc, enum colour_mode cm, + enum bits_per_comp bpc, u8 drm_bpp, + bool is_navite_422_or_420, + int slice_width, int slice_height, + int minor_version) { + float bpp; float bpp_group; float initial_xmit_delay_factor; int padding_pixels; int i; + bpp = ((float)drm_bpp / 16.0); + /* in native_422 or native_420 modes, the bits_per_pixel is double the + * target bpp (the latter is what calc_rc_params expects) + */ + if (is_navite_422_or_420) + bpp /= 2.0; + rc->rc_quant_incr_limit0 = ((bpc == BPC_8) ? 11 : (bpc == BPC_10 ? 15 : 19)) - ((minor_version == 1 && cm == CM_444) ? 1 : 0); rc->rc_quant_incr_limit1 = ((bpc == BPC_8) ? 11 : (bpc == BPC_10 ? 15 : 19)) - ((minor_version == 1 && cm == CM_444) ? 1 : 0); @@ -251,3 +265,128 @@ void calc_rc_params(struct rc_params *rc, enum colour_mode cm, enum bits_per_com rc->rc_buf_thresh[13] = 8064; } +static u32 _do_bytes_per_pixel_calc(int slice_width, u8 drm_bpp, + bool is_navite_422_or_420) +{ + float bpp; + u32 bytes_per_pixel; + double d_bytes_per_pixel; + + bpp = ((float)drm_bpp / 16.0); + d_bytes_per_pixel = dsc_ceil(bpp * slice_width / 8.0) / slice_width; + // TODO: Make sure the formula for calculating this is precise (ceiling + // vs. floor, and at what point they should be applied) + if (is_navite_422_or_420) + d_bytes_per_pixel /= 2; + + bytes_per_pixel = (u32)dsc_ceil(d_bytes_per_pixel * 0x10000000); + + return bytes_per_pixel; +} + +static u32 _do_calc_dsc_bpp_x16(u32 stream_bandwidth_kbps, u32 pix_clk_100hz, + u32 bpp_increment_div) +{ + u32 dsc_target_bpp_x16; + float f_dsc_target_bpp; + float f_stream_bandwidth_100bps; + // bpp_increment_div is actually precision + u32 precision = bpp_increment_div; + + f_stream_bandwidth_100bps = stream_bandwidth_kbps * 10.0f; + f_dsc_target_bpp = f_stream_bandwidth_100bps / pix_clk_100hz; + + // Round down to the nearest precision stop to bring it into DSC spec + // range + dsc_target_bpp_x16 = (u32)(f_dsc_target_bpp * precision); + dsc_target_bpp_x16 = (dsc_target_bpp_x16 * 16) / precision; + + return dsc_target_bpp_x16; +} + +/** + * calc_rc_params - reads the user's cmdline mode + * @rc: DC internal DSC parameters + * @pps: DRM struct with all required DSC values + * + * This function expects a drm_dsc_config data struct with all the required DSC + * values previously filled out by our driver and based on this information it + * computes some of the DSC values. + * + * @note This calculation requires float point operation, most of it executes + * under kernel_fpu_{begin,end}. + */ +void calc_rc_params(struct rc_params *rc, const struct drm_dsc_config *pps) +{ + enum colour_mode mode; + enum bits_per_comp bpc; + bool is_navite_422_or_420; + u8 drm_bpp = pps->bits_per_pixel; + int slice_width = pps->slice_width; + int slice_height = pps->slice_height; + + mode = pps->convert_rgb ? CM_RGB : (pps->simple_422 ? CM_444 : + (pps->native_422 ? CM_422 : + pps->native_420 ? CM_420 : CM_444)); + bpc = (pps->bits_per_component == 8) ? BPC_8 : (pps->bits_per_component == 10) + ? BPC_10 : BPC_12; + + is_navite_422_or_420 = pps->native_422 || pps->native_420; + + DC_FP_START(); + _do_calc_rc_params(rc, mode, bpc, drm_bpp, is_navite_422_or_420, + slice_width, slice_height, + pps->dsc_version_minor); + DC_FP_END(); +} + +/** + * calc_dsc_bytes_per_pixel - calculate bytes per pixel + * @pps: DRM struct with all required DSC values + * + * Based on the information inside drm_dsc_config, this function calculates the + * total of bytes per pixel. + * + * @note This calculation requires float point operation, most of it executes + * under kernel_fpu_{begin,end}. + * + * Return: + * Return the number of bytes per pixel + */ +u32 calc_dsc_bytes_per_pixel(const struct drm_dsc_config *pps) + +{ + u32 ret; + u8 drm_bpp = pps->bits_per_pixel; + int slice_width = pps->slice_width; + bool is_navite_422_or_420 = pps->native_422 || pps->native_420; + + DC_FP_START(); + ret = _do_bytes_per_pixel_calc(slice_width, drm_bpp, + is_navite_422_or_420); + DC_FP_END(); + return ret; +} + +/** + * calc_dsc_bpp_x16 - retrieve the dsc bits per pixel + * @stream_bandwidth_kbps: + * @pix_clk_100hz: + * @bpp_increment_div: + * + * Calculate the total of bits per pixel for DSC configuration. + * + * @note This calculation requires float point operation, most of it executes + * under kernel_fpu_{begin,end}. + */ +u32 calc_dsc_bpp_x16(u32 stream_bandwidth_kbps, u32 pix_clk_100hz, + u32 bpp_increment_div) +{ + u32 dsc_bpp; + + DC_FP_START(); + dsc_bpp = _do_calc_dsc_bpp_x16(stream_bandwidth_kbps, pix_clk_100hz, + bpp_increment_div); + DC_FP_END(); + return dsc_bpp; +} diff --git a/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.h b/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.h index b6b1f09c2009..21723fa6561e 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.h +++ b/drivers/gpu/drm/amd/display/dc/dsc/rc_calc.h @@ -77,7 +77,10 @@ struct qp_entry { typedef struct qp_entry qp_table[]; -void calc_rc_params(struct rc_params *rc, enum colour_mode cm, enum bits_per_comp bpc, float bpp, int slice_width, int slice_height, int minor_version); +void calc_rc_params(struct rc_params *rc, const struct drm_dsc_config *pps); +u32 calc_dsc_bytes_per_pixel(const struct drm_dsc_config *pps); +u32 calc_dsc_bpp_x16(u32 stream_bandwidth_kbps, u32 pix_clk_100hz, + u32 bpp_increment_div); #endif diff --git a/drivers/gpu/drm/amd/display/dc/dsc/rc_calc_dpi.c b/drivers/gpu/drm/amd/display/dc/dsc/rc_calc_dpi.c index 1f6e63b71456..ef830aded5b1 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/rc_calc_dpi.c +++ b/drivers/gpu/drm/amd/display/dc/dsc/rc_calc_dpi.c @@ -27,8 +27,6 @@ #include "dscc_types.h" #include "rc_calc.h" -double dsc_ceil(double num); - static void copy_pps_fields(struct drm_dsc_config *to, const struct drm_dsc_config *from) { to->line_buf_depth = from->line_buf_depth; @@ -100,34 +98,13 @@ static void copy_rc_to_cfg(struct drm_dsc_config *dsc_cfg, const struct rc_param int dscc_compute_dsc_parameters(const struct drm_dsc_config *pps, struct dsc_parameters *dsc_params) { - enum colour_mode mode = pps->convert_rgb ? CM_RGB : - (pps->simple_422 ? CM_444 : - (pps->native_422 ? CM_422 : - pps->native_420 ? CM_420 : CM_444)); - enum bits_per_comp bpc = (pps->bits_per_component == 8) ? BPC_8 : - (pps->bits_per_component == 10) ? BPC_10 : BPC_12; - float bpp = ((float) pps->bits_per_pixel / 16.0); - int slice_width = pps->slice_width; - int slice_height = pps->slice_height; int ret; struct rc_params rc; struct drm_dsc_config dsc_cfg; - double d_bytes_per_pixel = dsc_ceil(bpp * slice_width / 8.0) / slice_width; - - // TODO: Make sure the formula for calculating this is precise (ceiling vs. floor, and at what point they should be applied) - if (pps->native_422 || pps->native_420) - d_bytes_per_pixel /= 2; - - dsc_params->bytes_per_pixel = (uint32_t)dsc_ceil(d_bytes_per_pixel * 0x10000000); - - /* in native_422 or native_420 modes, the bits_per_pixel is double the target bpp - * (the latter is what calc_rc_params expects) - */ - if (pps->native_422 || pps->native_420) - bpp /= 2.0; + dsc_params->bytes_per_pixel = calc_dsc_bytes_per_pixel(pps); - calc_rc_params(&rc, mode, bpc, bpp, slice_width, slice_height, pps->dsc_version_minor); + calc_rc_params(&rc, pps); dsc_params->pps = *pps; dsc_params->pps.initial_scale_value = 8 * rc.rc_model_size / (rc.rc_model_size - rc.initial_fullness_offset); diff --git a/drivers/gpu/drm/amd/display/dc/hdcp/hdcp_msg.c b/drivers/gpu/drm/amd/display/dc/hdcp/hdcp_msg.c index 6f730b5bfe42..5e384a8a83dc 100644 --- a/drivers/gpu/drm/amd/display/dc/hdcp/hdcp_msg.c +++ b/drivers/gpu/drm/amd/display/dc/hdcp/hdcp_msg.c @@ -322,3 +322,92 @@ static const struct protection_properties dp_11_protection = { .process_transaction = dp_11_process_transaction }; +static const struct protection_properties *get_protection_properties_by_signal( + struct dc_link *link, + enum signal_type st, + enum hdcp_version version) +{ + switch (version) { + case HDCP_VERSION_14: + switch (st) { + case SIGNAL_TYPE_DVI_SINGLE_LINK: + case SIGNAL_TYPE_DVI_DUAL_LINK: + case SIGNAL_TYPE_HDMI_TYPE_A: + return &hdmi_14_protection; + case SIGNAL_TYPE_DISPLAY_PORT: + if (link && + (link->dpcd_caps.dongle_type == DISPLAY_DONGLE_DP_VGA_CONVERTER || + link->dpcd_caps.dongle_caps.dongle_type == DISPLAY_DONGLE_DP_VGA_CONVERTER)) { + return &non_supported_protection; + } + return &dp_11_protection; + case SIGNAL_TYPE_DISPLAY_PORT_MST: + case SIGNAL_TYPE_EDP: + return &dp_11_protection; + default: + return &non_supported_protection; + } + break; + case HDCP_VERSION_22: + switch (st) { + case SIGNAL_TYPE_DVI_SINGLE_LINK: + case SIGNAL_TYPE_DVI_DUAL_LINK: + case SIGNAL_TYPE_HDMI_TYPE_A: + return &hdmi_14_protection; //todo version2.2 + case SIGNAL_TYPE_DISPLAY_PORT: + case SIGNAL_TYPE_DISPLAY_PORT_MST: + case SIGNAL_TYPE_EDP: + return &dp_11_protection; //todo version2.2 + default: + return &non_supported_protection; + } + break; + default: + return &non_supported_protection; + } +} + +enum hdcp_message_status dc_process_hdcp_msg( + enum signal_type signal, + struct dc_link *link, + struct hdcp_protection_message *message_info) +{ + enum hdcp_message_status status = HDCP_MESSAGE_FAILURE; + uint32_t i = 0; + + const struct protection_properties *protection_props; + + if (!message_info) + return HDCP_MESSAGE_UNSUPPORTED; + + if (message_info->msg_id < HDCP_MESSAGE_ID_READ_BKSV || + message_info->msg_id >= HDCP_MESSAGE_ID_MAX) + return HDCP_MESSAGE_UNSUPPORTED; + + protection_props = + get_protection_properties_by_signal( + link, + signal, + message_info->version); + + if (!protection_props->supported) + return HDCP_MESSAGE_UNSUPPORTED; + + if (protection_props->process_transaction( + link, + message_info)) { + status = HDCP_MESSAGE_SUCCESS; + } else { + for (i = 0; i < message_info->max_retries; i++) { + if (protection_props->process_transaction( + link, + message_info)) { + status = HDCP_MESSAGE_SUCCESS; + break; + } + } + } + + return status; +} + diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h index d523fc9547e7..c7fd702a4a87 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h @@ -38,6 +38,7 @@ #endif #include "dwb.h" #include "mcif_wb.h" +#include "panel_cntl.h" #define MAX_CLOCK_SOURCES 7 @@ -92,6 +93,8 @@ struct clk_bw_params; struct resource_funcs { void (*destroy)(struct resource_pool **pool); void (*link_init)(struct dc_link *link); + struct panel_cntl*(*panel_cntl_create)( + const struct panel_cntl_init_data *panel_cntl_init_data); struct link_encoder *(*link_enc_create)( const struct encoder_init_data *init); bool (*validate_bandwidth)( diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h b/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h index d607b3191954..e8ce8c85adf1 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h @@ -27,27 +27,17 @@ #include "dm_services_types.h" -struct abm_backlight_registers { - unsigned int BL_PWM_CNTL; - unsigned int BL_PWM_CNTL2; - unsigned int BL_PWM_PERIOD_CNTL; - unsigned int LVTMA_PWRSEQ_REF_DIV_BL_PWM_REF_DIV; -}; - struct abm { struct dc_context *ctx; const struct abm_funcs *funcs; bool dmcu_is_running; - /* registers setting needs to be saved and restored at InitBacklight */ - struct abm_backlight_registers stored_backlight_registers; }; struct abm_funcs { - void (*abm_init)(struct abm *abm); + void (*abm_init)(struct abm *abm, uint32_t back_light); bool (*set_abm_level)(struct abm *abm, unsigned int abm_level); - bool (*set_abm_immediate_disable)(struct abm *abm); - bool (*set_pipe)(struct abm *abm, unsigned int controller_id); - bool (*init_backlight)(struct abm *abm); + bool (*set_abm_immediate_disable)(struct abm *abm, unsigned int panel_inst); + bool (*set_pipe)(struct abm *abm, unsigned int controller_id, unsigned int panel_inst); /* backlight_pwm_u16_16 is unsigned 32 bit, * 16 bit integer + 16 fractional, where 1.0 is max backlight value. @@ -56,10 +46,13 @@ struct abm_funcs { unsigned int backlight_pwm_u16_16, unsigned int frame_ramp, unsigned int controller_id, - bool use_smooth_brightness); + unsigned int panel_inst); unsigned int (*get_current_backlight)(struct abm *abm); unsigned int (*get_target_backlight)(struct abm *abm); + bool (*init_abm_config)(struct abm *abm, + const char *src, + unsigned int bytes); }; #endif diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h index f5dd0cc73c63..47a566d82d6e 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h @@ -144,6 +144,8 @@ struct hubbub_funcs { void (*allow_self_refresh_control)(struct hubbub *hubbub, bool allow); void (*apply_DEDCN21_147_wa)(struct hubbub *hubbub); + + void (*force_wm_propagate_to_pipes)(struct hubbub *hubbub); }; struct hubbub { diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h index 094afc4c8173..50ee8aa7ec3b 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h @@ -210,6 +210,22 @@ struct mpc_funcs { struct mpcc_blnd_cfg *blnd_cfg, int mpcc_id); + /* + * Lock cursor updates for the specified OPP. + * OPP defines the set of MPCC that are locked together for cursor. + * + * Parameters: + * [in] mpc - MPC context. + * [in] opp_id - The OPP to lock cursor updates on + * [in] lock - lock/unlock the OPP + * + * Return: void + */ + void (*cursor_lock)( + struct mpc *mpc, + int opp_id, + bool lock); + struct mpcc* (*get_mpcc_for_dpp)( struct mpc_tree *tree, int dpp_id); diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/panel_cntl.h b/drivers/gpu/drm/amd/display/dc/inc/hw/panel_cntl.h new file mode 100644 index 000000000000..f9ab5abb6462 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/panel_cntl.h @@ -0,0 +1,67 @@ +/* + * Copyright 2017 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ +/* + * panel_cntl.h + * + * Created on: Oct 6, 2015 + * Author: yonsun + */ + +#ifndef DC_PANEL_CNTL_H_ +#define DC_PANEL_CNTL_H_ + +#include "dc_types.h" + +#define MAX_BACKLIGHT_LEVEL 0xFFFF + +struct panel_cntl_backlight_registers { + unsigned int BL_PWM_CNTL; + unsigned int BL_PWM_CNTL2; + unsigned int BL_PWM_PERIOD_CNTL; + unsigned int LVTMA_PWRSEQ_REF_DIV_BL_PWM_REF_DIV; +}; + +struct panel_cntl_funcs { + void (*destroy)(struct panel_cntl **panel_cntl); + uint32_t (*hw_init)(struct panel_cntl *panel_cntl); + bool (*is_panel_backlight_on)(struct panel_cntl *panel_cntl); + bool (*is_panel_powered_on)(struct panel_cntl *panel_cntl); + void (*store_backlight_level)(struct panel_cntl *panel_cntl); + void (*driver_set_backlight)(struct panel_cntl *panel_cntl, + uint32_t backlight_pwm_u16_16); +}; + +struct panel_cntl_init_data { + struct dc_context *ctx; + uint32_t inst; +}; + +struct panel_cntl { + const struct panel_cntl_funcs *funcs; + struct dc_context *ctx; + uint32_t inst; + /* registers setting needs to be saved and restored at InitBacklight */ + struct panel_cntl_backlight_registers stored_backlight_registers; +}; + +#endif /* DC_PANEL_CNTL_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h index e5e7d94026fc..f803191e3134 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h @@ -117,6 +117,9 @@ struct crc_params { enum crc_selection selection; + uint8_t dsc_mode; + uint8_t odm_mode; + bool continuous_mode; bool enable; }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/transform.h b/drivers/gpu/drm/amd/display/dc/inc/hw/transform.h index fecc80c47c26..2947d1b15512 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/transform.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/transform.h @@ -173,6 +173,8 @@ struct scaler_data { struct scaling_taps taps; struct rect viewport; struct rect viewport_c; + struct rect viewport_unadjusted; + struct rect viewport_c_unadjusted; struct rect recout; struct scaling_ratios ratios; struct scl_inits inits; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h index d4c1fb242c63..8e72f077e552 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h @@ -75,9 +75,13 @@ struct hw_sequencer_funcs { void (*wait_for_mpcc_disconnect)(struct dc *dc, struct resource_pool *res_pool, struct pipe_ctx *pipe_ctx); + void (*edp_backlight_control)( + struct dc_link *link, + bool enable); void (*program_triplebuffer)(const struct dc *dc, struct pipe_ctx *pipe_ctx, bool enableTripleBuffer); void (*update_pending_status)(struct pipe_ctx *pipe_ctx); + void (*power_down)(struct dc *dc); /* Pipe Lock Related */ void (*pipe_control_lock)(struct dc *dc, @@ -86,11 +90,17 @@ struct hw_sequencer_funcs { struct dc_state *context, bool lock); void (*set_flip_control_gsl)(struct pipe_ctx *pipe_ctx, bool flip_immediate); + void (*cursor_lock)(struct dc *dc, struct pipe_ctx *pipe, bool lock); /* Timing Related */ void (*get_position)(struct pipe_ctx **pipe_ctx, int num_pipes, struct crtc_position *position); int (*get_vupdate_offset_from_vsync)(struct pipe_ctx *pipe_ctx); + void (*calc_vupdate_position)( + struct dc *dc, + struct pipe_ctx *pipe_ctx, + uint32_t *start_line, + uint32_t *end_line); void (*enable_per_frame_crtc_position_reset)(struct dc *dc, int group_size, struct pipe_ctx *grouped_pipes[]); void (*enable_timing_synchronization)(struct dc *dc, @@ -187,6 +197,12 @@ struct hw_sequencer_funcs { unsigned int bufSize, unsigned int mask); void (*clear_status_bits)(struct dc *dc, unsigned int mask); + bool (*set_backlight_level)(struct pipe_ctx *pipe_ctx, + uint32_t backlight_pwm_u16_16, + uint32_t frame_ramp); + + void (*set_abm_immediate_disable)(struct pipe_ctx *pipe_ctx); + }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h index 52a26e6be066..36e906bb6bfc 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h @@ -100,8 +100,6 @@ struct hwseq_private_funcs { struct dc *dc); void (*edp_backlight_control)(struct dc_link *link, bool enable); - bool (*is_panel_backlight_on)(struct dc_link *link); - bool (*is_panel_powered_on)(struct dc_link *link); void (*setup_vupdate_interrupt)(struct dc *dc, struct pipe_ctx *pipe_ctx); bool (*did_underflow_occur)(struct dc *dc, struct pipe_ctx *pipe_ctx); diff --git a/drivers/gpu/drm/amd/display/dc/inc/resource.h b/drivers/gpu/drm/amd/display/dc/inc/resource.h index ca4c36c0c9bc..a9be495af922 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/resource.h +++ b/drivers/gpu/drm/amd/display/dc/inc/resource.h @@ -138,9 +138,6 @@ struct pipe_ctx *find_idle_secondary_pipe( const struct resource_pool *pool, const struct pipe_ctx *primary_pipe); -bool resource_is_stream_unchanged( - struct dc_state *old_context, struct dc_stream_state *stream); - bool resource_validate_attach_surfaces( const struct dc_validation_set set[], int set_count, @@ -180,6 +177,8 @@ unsigned int resource_pixel_format_to_bpp(enum surface_pixel_format format); void get_audio_check(struct audio_info *aud_modes, struct audio_check *aud_chk); +int get_num_mpc_splits(struct pipe_ctx *pipe); + int get_num_odm_splits(struct pipe_ctx *pipe); #endif /* DRIVERS_GPU_DRM_AMD_DC_DEV_DC_INC_RESOURCE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/os_types.h b/drivers/gpu/drm/amd/display/dc/os_types.h index c34eba19860a..6d7bca562eec 100644 --- a/drivers/gpu/drm/amd/display/dc/os_types.h +++ b/drivers/gpu/drm/amd/display/dc/os_types.h @@ -108,7 +108,7 @@ #define ASSERT(expr) ASSERT_CRITICAL(expr) #else -#define ASSERT(expr) WARN_ON(!(expr)) +#define ASSERT(expr) WARN_ON_ONCE(!(expr)) #endif #define BREAK_TO_DEBUGGER() ASSERT(0) diff --git a/drivers/gpu/drm/amd/display/dc/virtual/virtual_link_encoder.c b/drivers/gpu/drm/amd/display/dc/virtual/virtual_link_encoder.c index 3464b2d5b89a..348e9a600a72 100644 --- a/drivers/gpu/drm/amd/display/dc/virtual/virtual_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/virtual/virtual_link_encoder.c @@ -84,6 +84,14 @@ static void virtual_link_encoder_destroy(struct link_encoder **enc) *enc = NULL; } +static void virtual_link_encoder_get_max_link_cap(struct link_encoder *enc, + struct dc_link_settings *link_settings) +{ + /* Set Default link settings */ + struct dc_link_settings max_link_cap = {LANE_COUNT_FOUR, LINK_RATE_HIGH, + LINK_SPREAD_05_DOWNSPREAD_30KHZ, false, 0}; + *link_settings = max_link_cap; +} static const struct link_encoder_funcs virtual_lnk_enc_funcs = { .validate_output_with_stream = @@ -94,6 +102,7 @@ static const struct link_encoder_funcs virtual_lnk_enc_funcs = { .enable_dp_output = virtual_link_encoder_enable_dp_output, .enable_dp_mst_output = virtual_link_encoder_enable_dp_mst_output, .disable_output = virtual_link_encoder_disable_output, + .get_max_link_cap = virtual_link_encoder_get_max_link_cap, .dp_set_lane_settings = virtual_link_encoder_dp_set_lane_settings, .dp_set_phy_pattern = virtual_link_encoder_dp_set_phy_pattern, .update_mst_stream_allocation_table = diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h index c2671f2616c8..26d94eb5ab58 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_srv.h +++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h @@ -64,10 +64,11 @@ * other component within DAL. */ -#include "dmub_types.h" -#include "dmub_cmd.h" -#include "dmub_gpint_cmd.h" -#include "dmub_rb.h" +#include "inc/dmub_types.h" +#include "inc/dmub_cmd.h" +#include "inc/dmub_gpint_cmd.h" +#include "inc/dmub_cmd_dal.h" +#include "inc/dmub_rb.h" #if defined(__cplusplus) extern "C" { @@ -75,7 +76,6 @@ extern "C" { /* Forward declarations */ struct dmub_srv; -struct dmub_cmd_header; struct dmub_srv_common_regs; /* enum dmub_status - return code for dmcub functions */ @@ -151,6 +151,7 @@ struct dmub_srv_region_params { uint32_t inst_const_size; uint32_t bss_data_size; uint32_t vbios_size; + const uint8_t *fw_inst_const; const uint8_t *fw_bss_data; }; @@ -457,7 +458,7 @@ enum dmub_status dmub_srv_hw_reset(struct dmub_srv *dmub); * DMUB_STATUS_INVALID - unspecified error */ enum dmub_status dmub_srv_cmd_queue(struct dmub_srv *dmub, - const struct dmub_cmd_header *cmd); + const union dmub_rb_cmd *cmd); /** * dmub_srv_cmd_execute() - Executes a queued sequence to the dmub @@ -565,6 +566,16 @@ dmub_srv_send_gpint_command(struct dmub_srv *dmub, enum dmub_status dmub_srv_get_gpint_response(struct dmub_srv *dmub, uint32_t *response); +/** + * dmub_flush_buffer_mem() - Read back entire frame buffer region. + * This ensures that the write from x86 has been flushed and will not + * hang the DMCUB. + * @fb: frame buffer to flush + * + * Can be called after software initialization. + */ +void dmub_flush_buffer_mem(const struct dmub_fb *fb); + #if defined(__cplusplus) } #endif diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h index 10b5fa9d2588..599bf2055bcb 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h @@ -228,6 +228,7 @@ struct dmub_cmd_psr_copy_settings_data { uint8_t smu_optimizations_en; uint8_t frame_delay; uint8_t frame_cap_ind; + struct dmub_psr_debug_flags debug; }; struct dmub_rb_cmd_psr_copy_settings { @@ -260,6 +261,8 @@ struct dmub_rb_cmd_psr_set_version { struct dmub_cmd_abm_set_pipe_data { uint32_t ramping_boundary; uint32_t otg_inst; + uint32_t panel_inst; + uint32_t set_pipe_option; }; struct dmub_rb_cmd_abm_set_pipe { @@ -303,6 +306,16 @@ struct dmub_rb_cmd_abm_set_pwm_frac { struct dmub_cmd_abm_set_pwm_frac_data abm_set_pwm_frac_data; }; +struct dmub_cmd_abm_init_config_data { + union dmub_addr src; + uint16_t bytes; +}; + +struct dmub_rb_cmd_abm_init_config { + struct dmub_cmd_header header; + struct dmub_cmd_abm_init_config_data abm_init_config_data; +}; + union dmub_rb_cmd { struct dmub_rb_cmd_read_modify_write read_modify_write; struct dmub_rb_cmd_reg_field_update_sequence reg_field_update_seq; @@ -324,6 +337,7 @@ union dmub_rb_cmd { struct dmub_rb_cmd_abm_set_level abm_set_level; struct dmub_rb_cmd_abm_set_ambient_level abm_set_ambient_level; struct dmub_rb_cmd_abm_set_pwm_frac abm_set_pwm_frac; + struct dmub_rb_cmd_abm_init_config abm_init_config; }; #pragma pack(pop) diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd_dal.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd_dal.h index d37535d21928..e42de9ded275 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd_dal.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd_dal.h @@ -32,17 +32,16 @@ */ enum dmub_cmd_psr_type { - DMUB_CMD__PSR_SET_VERSION = 0, - DMUB_CMD__PSR_COPY_SETTINGS = 1, - DMUB_CMD__PSR_ENABLE = 2, - DMUB_CMD__PSR_DISABLE = 3, - DMUB_CMD__PSR_SET_LEVEL = 4, + DMUB_CMD__PSR_SET_VERSION = 0, + DMUB_CMD__PSR_COPY_SETTINGS = 1, + DMUB_CMD__PSR_ENABLE = 2, + DMUB_CMD__PSR_DISABLE = 3, + DMUB_CMD__PSR_SET_LEVEL = 4, }; enum psr_version { - PSR_VERSION_1 = 0x10, // PSR Version 1 - PSR_VERSION_2 = 0x20, // PSR Version 2, includes selective update - PSR_VERSION_2_1 = 0x21, // PSR Version 2, includes Y-coordinate support for SU + PSR_VERSION_1 = 0, + PSR_VERSION_UNSUPPORTED = 0xFFFFFFFF, }; enum dmub_cmd_abm_type { diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_rb.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_rb.h index df875fdd2ab0..2ae48c18bb5b 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_rb.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_rb.h @@ -33,8 +33,6 @@ extern "C" { #endif -struct dmub_cmd_header; - struct dmub_rb_init_params { void *ctx; void *base_address; @@ -71,7 +69,7 @@ static inline bool dmub_rb_full(struct dmub_rb *rb) } static inline bool dmub_rb_push_front(struct dmub_rb *rb, - const struct dmub_cmd_header *cmd) + const union dmub_rb_cmd *cmd) { uint64_t volatile *dst = (uint64_t volatile *)(rb->base_address) + rb->wrpt / sizeof(uint64_t); const uint64_t *src = (const uint64_t *)cmd; @@ -93,7 +91,7 @@ static inline bool dmub_rb_push_front(struct dmub_rb *rb, } static inline bool dmub_rb_front(struct dmub_rb *rb, - struct dmub_cmd_header *cmd) + union dmub_rb_cmd *cmd) { uint8_t *rd_ptr = (uint8_t *)rb->base_address + rb->rptr; diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_types.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_types.h index 41d524b0db2f..bed5b023a396 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_types.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_types.h @@ -49,6 +49,12 @@ extern "C" { #define dmub_udelay(microseconds) udelay(microseconds) #endif +/* Maximum number of streams on any ASIC. */ +#define DMUB_MAX_STREAMS 6 + +/* Maximum number of planes on any ASIC. */ +#define DMUB_MAX_PLANES 6 + union dmub_addr { struct { uint32_t low_part; @@ -57,6 +63,11 @@ union dmub_addr { uint64_t quad_part; }; +struct dmub_psr_debug_flags { + uint8_t visual_confirm : 1; + uint8_t reserved : 7; +}; + #if defined(__cplusplus) } #endif diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.c index 63bb9e2c81de..edc73d6d7ba2 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.c @@ -23,7 +23,7 @@ * */ -#include "../inc/dmub_srv.h" +#include "../dmub_srv.h" #include "dmub_reg.h" #include "dmub_dcn20.h" @@ -186,14 +186,22 @@ void dmub_dcn20_setup_windows(struct dmub_srv *dmub, dmub_dcn20_get_fb_base_offset(dmub, &fb_base, &fb_offset); - dmub_dcn20_translate_addr(&cw2->offset, fb_base, fb_offset, &offset); - - REG_WRITE(DMCUB_REGION3_CW2_OFFSET, offset.u.low_part); - REG_WRITE(DMCUB_REGION3_CW2_OFFSET_HIGH, offset.u.high_part); - REG_WRITE(DMCUB_REGION3_CW2_BASE_ADDRESS, cw2->region.base); - REG_SET_2(DMCUB_REGION3_CW2_TOP_ADDRESS, 0, - DMCUB_REGION3_CW2_TOP_ADDRESS, cw2->region.top, - DMCUB_REGION3_CW2_ENABLE, 1); + if (cw2->region.base != cw2->region.top) { + dmub_dcn20_translate_addr(&cw2->offset, fb_base, fb_offset, + &offset); + + REG_WRITE(DMCUB_REGION3_CW2_OFFSET, offset.u.low_part); + REG_WRITE(DMCUB_REGION3_CW2_OFFSET_HIGH, offset.u.high_part); + REG_WRITE(DMCUB_REGION3_CW2_BASE_ADDRESS, cw2->region.base); + REG_SET_2(DMCUB_REGION3_CW2_TOP_ADDRESS, 0, + DMCUB_REGION3_CW2_TOP_ADDRESS, cw2->region.top, + DMCUB_REGION3_CW2_ENABLE, 1); + } else { + REG_WRITE(DMCUB_REGION3_CW2_OFFSET, 0); + REG_WRITE(DMCUB_REGION3_CW2_OFFSET_HIGH, 0); + REG_WRITE(DMCUB_REGION3_CW2_BASE_ADDRESS, 0); + REG_WRITE(DMCUB_REGION3_CW2_TOP_ADDRESS, 0); + } dmub_dcn20_translate_addr(&cw3->offset, fb_base, fb_offset, &offset); diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.c index 5bed9fcd6b5c..e8f488232e34 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.c @@ -23,7 +23,7 @@ * */ -#include "../inc/dmub_srv.h" +#include "../dmub_srv.h" #include "dmub_reg.h" #include "dmub_dcn21.h" diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_reg.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_reg.c index 4094eca212f0..ca0c8a54b635 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_reg.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_reg.c @@ -24,7 +24,7 @@ */ #include "dmub_reg.h" -#include "../inc/dmub_srv.h" +#include "../dmub_srv.h" struct dmub_reg_value_masks { uint32_t value; diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c index ce32cc7933c4..0e3751d94cb0 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c @@ -23,7 +23,7 @@ * */ -#include "../inc/dmub_srv.h" +#include "../dmub_srv.h" #include "dmub_dcn20.h" #include "dmub_dcn21.h" #include "dmub_fw_meta.h" @@ -70,7 +70,7 @@ static inline uint32_t dmub_align(uint32_t val, uint32_t factor) return (val + factor - 1) / factor * factor; } -static void dmub_flush_buffer_mem(const struct dmub_fb *fb) +void dmub_flush_buffer_mem(const struct dmub_fb *fb) { const uint8_t *base = (const uint8_t *)fb->cpu_addr; uint8_t buf[64]; @@ -91,18 +91,32 @@ static void dmub_flush_buffer_mem(const struct dmub_fb *fb) } static const struct dmub_fw_meta_info * -dmub_get_fw_meta_info(const uint8_t *fw_bss_data, uint32_t fw_bss_data_size) +dmub_get_fw_meta_info(const struct dmub_srv_region_params *params) { const union dmub_fw_meta *meta; + const uint8_t *blob = NULL; + uint32_t blob_size = 0; + uint32_t meta_offset = 0; + + if (params->fw_bss_data) { + /* Legacy metadata region. */ + blob = params->fw_bss_data; + blob_size = params->bss_data_size; + meta_offset = DMUB_FW_META_OFFSET; + } else if (params->fw_inst_const) { + /* Combined metadata region. */ + blob = params->fw_inst_const; + blob_size = params->inst_const_size; + meta_offset = 0; + } - if (fw_bss_data == NULL) + if (!blob || !blob_size) return NULL; - if (fw_bss_data_size < sizeof(union dmub_fw_meta) + DMUB_FW_META_OFFSET) + if (blob_size < sizeof(union dmub_fw_meta) + meta_offset) return NULL; - meta = (const union dmub_fw_meta *)(fw_bss_data + fw_bss_data_size - - DMUB_FW_META_OFFSET - + meta = (const union dmub_fw_meta *)(blob + blob_size - meta_offset - sizeof(union dmub_fw_meta)); if (meta->info.magic_value != DMUB_FW_META_MAGIC) @@ -247,8 +261,7 @@ dmub_srv_calc_region_info(struct dmub_srv *dmub, mail->base = dmub_align(bios->top, 256); mail->top = mail->base + DMUB_MAILBOX_SIZE; - fw_info = dmub_get_fw_meta_info(params->fw_bss_data, - params->bss_data_size); + fw_info = dmub_get_fw_meta_info(params); if (fw_info) { fw_state_size = fw_info->fw_region_size; @@ -449,7 +462,7 @@ enum dmub_status dmub_srv_hw_reset(struct dmub_srv *dmub) } enum dmub_status dmub_srv_cmd_queue(struct dmub_srv *dmub, - const struct dmub_cmd_header *cmd) + const union dmub_rb_cmd *cmd) { if (!dmub->hw_init) return DMUB_STATUS_INVALID; diff --git a/drivers/gpu/drm/amd/display/include/hdcp_types.h b/drivers/gpu/drm/amd/display/include/hdcp_types.h index f31e6befc8d6..42229b4effdc 100644 --- a/drivers/gpu/drm/amd/display/include/hdcp_types.h +++ b/drivers/gpu/drm/amd/display/include/hdcp_types.h @@ -83,6 +83,12 @@ enum hdcp_link { HDCP_LINK_SECONDARY }; +enum hdcp_message_status { + HDCP_MESSAGE_SUCCESS, + HDCP_MESSAGE_FAILURE, + HDCP_MESSAGE_UNSUPPORTED +}; + struct hdcp_protection_message { enum hdcp_version version; /* relevant only for DVI */ @@ -91,6 +97,7 @@ struct hdcp_protection_message { uint32_t length; uint8_t max_retries; uint8_t *data; + enum hdcp_message_status status; }; #endif diff --git a/drivers/gpu/drm/amd/display/include/logger_interface.h b/drivers/gpu/drm/amd/display/include/logger_interface.h index 6e008de25629..02c23b04d34b 100644 --- a/drivers/gpu/drm/amd/display/include/logger_interface.h +++ b/drivers/gpu/drm/amd/display/include/logger_interface.h @@ -40,8 +40,6 @@ struct dc_state; * */ -void dc_conn_log_hex_linux(const uint8_t *hex_data, int hex_data_count); - void pre_surface_trace( struct dc *dc, const struct dc_plane_state *const *plane_states, @@ -102,14 +100,12 @@ void context_clock_trace( #define CONN_DATA_DETECT(link, hex_data, hex_len, ...) \ do { \ (void)(link); \ - dc_conn_log_hex_linux(hex_data, hex_len); \ DC_LOG_EVENT_DETECTION(__VA_ARGS__); \ } while (0) #define CONN_DATA_LINK_LOSS(link, hex_data, hex_len, ...) \ do { \ (void)(link); \ - dc_conn_log_hex_linux(hex_data, hex_len); \ DC_LOG_EVENT_LINK_LOSS(__VA_ARGS__); \ } while (0) diff --git a/drivers/gpu/drm/amd/display/modules/color/color_gamma.c b/drivers/gpu/drm/amd/display/modules/color/color_gamma.c index cac09d500fda..bcfe34ef8c28 100644 --- a/drivers/gpu/drm/amd/display/modules/color/color_gamma.c +++ b/drivers/gpu/drm/amd/display/modules/color/color_gamma.c @@ -843,7 +843,7 @@ static bool build_regamma(struct pwl_float_data_ex *rgb_regamma, pow_buffer_ptr = -1; // reset back to no optimize ret = true; release: - kfree(coeff); + kvfree(coeff); return ret; } @@ -1777,12 +1777,13 @@ bool calculate_user_regamma_ramp(struct dc_transfer_func *output_tf, kfree(rgb_regamma); rgb_regamma_alloc_fail: - kvfree(rgb_user); + kfree(rgb_user); rgb_user_alloc_fail: return ret; } -bool mod_color_calculate_degamma_params(struct dc_transfer_func *input_tf, +bool mod_color_calculate_degamma_params(struct dc_color_caps *dc_caps, + struct dc_transfer_func *input_tf, const struct dc_gamma *ramp, bool mapUserRamp) { struct dc_transfer_func_distributed_points *tf_pts = &input_tf->tf_pts; @@ -1801,11 +1802,29 @@ bool mod_color_calculate_degamma_params(struct dc_transfer_func *input_tf, /* we can use hardcoded curve for plain SRGB TF * If linear, it's bypass if on user ramp */ - if (input_tf->type == TF_TYPE_PREDEFINED && - (input_tf->tf == TRANSFER_FUNCTION_SRGB || - input_tf->tf == TRANSFER_FUNCTION_LINEAR) && - !mapUserRamp) - return true; + if (input_tf->type == TF_TYPE_PREDEFINED) { + if ((input_tf->tf == TRANSFER_FUNCTION_SRGB || + input_tf->tf == TRANSFER_FUNCTION_LINEAR) && + !mapUserRamp) + return true; + + if (dc_caps != NULL && + dc_caps->dpp.dcn_arch == 1) { + + if (input_tf->tf == TRANSFER_FUNCTION_PQ && + dc_caps->dpp.dgam_rom_caps.pq == 1) + return true; + + if (input_tf->tf == TRANSFER_FUNCTION_GAMMA22 && + dc_caps->dpp.dgam_rom_caps.gamma2_2 == 1) + return true; + + // HLG OOTF not accounted for + if (input_tf->tf == TRANSFER_FUNCTION_HLG && + dc_caps->dpp.dgam_rom_caps.hlg == 1) + return true; + } + } input_tf->type = TF_TYPE_DISTRIBUTED_POINTS; @@ -1902,7 +1921,7 @@ bool mod_color_calculate_degamma_params(struct dc_transfer_func *input_tf, - if (ramp->type == GAMMA_CUSTOM) + if (ramp && ramp->type == GAMMA_CUSTOM) apply_lut_1d(ramp, MAX_HW_POINTS, tf_pts); ret = true; diff --git a/drivers/gpu/drm/amd/display/modules/color/color_gamma.h b/drivers/gpu/drm/amd/display/modules/color/color_gamma.h index 9994817a9a03..7f56226ba77a 100644 --- a/drivers/gpu/drm/amd/display/modules/color/color_gamma.h +++ b/drivers/gpu/drm/amd/display/modules/color/color_gamma.h @@ -30,6 +30,7 @@ struct dc_transfer_func; struct dc_gamma; struct dc_transfer_func_distributed_points; struct dc_rgb_fixed; +struct dc_color_caps; enum dc_transfer_func_predefined; /* For SetRegamma ADL interface support @@ -100,7 +101,8 @@ bool mod_color_calculate_regamma_params(struct dc_transfer_func *output_tf, const struct dc_gamma *ramp, bool mapUserRamp, bool canRomBeUsed, const struct freesync_hdr_tf_params *fs_params); -bool mod_color_calculate_degamma_params(struct dc_transfer_func *output_tf, +bool mod_color_calculate_degamma_params(struct dc_color_caps *dc_caps, + struct dc_transfer_func *output_tf, const struct dc_gamma *ramp, bool mapUserRamp); bool mod_color_calculate_degamma_curve(enum dc_transfer_func_predefined trans, diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c index c33454a9e0b4..eb7421e83b86 100644 --- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c +++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c @@ -443,7 +443,7 @@ static bool vrr_settings_require_update(struct core_freesync *core_freesync, return true; } else if (in_vrr->state == VRR_STATE_ACTIVE_FIXED && in_vrr->fixed.target_refresh_in_uhz != - in_config->min_refresh_in_uhz) { + in_config->fixed_refresh_in_uhz) { return true; } else if (in_vrr->min_refresh_in_uhz != min_refresh_in_uhz) { return true; @@ -491,7 +491,7 @@ bool mod_freesync_get_v_position(struct mod_freesync *mod_freesync, return false; } -static void build_vrr_infopacket_data(const struct mod_vrr_params *vrr, +static void build_vrr_infopacket_data_v1(const struct mod_vrr_params *vrr, struct dc_info_packet *infopacket) { /* PB1 = 0x1A (24bit AMD IEEE OUI (0x00001A) - Byte 0) */ @@ -523,14 +523,74 @@ static void build_vrr_infopacket_data(const struct mod_vrr_params *vrr, vrr->state == VRR_STATE_ACTIVE_FIXED) infopacket->sb[6] |= 0x04; + // For v1 & 2 infoframes program nominal if non-fs mode, otherwise full range /* PB7 = FreeSync Minimum refresh rate (Hz) */ - infopacket->sb[7] = (unsigned char)((vrr->min_refresh_in_uhz + 500000) / 1000000); + if (vrr->state == VRR_STATE_ACTIVE_VARIABLE || + vrr->state == VRR_STATE_ACTIVE_FIXED) { + infopacket->sb[7] = (unsigned char)((vrr->min_refresh_in_uhz + 500000) / 1000000); + } else { + infopacket->sb[7] = (unsigned char)((vrr->max_refresh_in_uhz + 500000) / 1000000); + } /* PB8 = FreeSync Maximum refresh rate (Hz) * Note: We should never go above the field rate of the mode timing set. */ infopacket->sb[8] = (unsigned char)((vrr->max_refresh_in_uhz + 500000) / 1000000); + //FreeSync HDR + infopacket->sb[9] = 0; + infopacket->sb[10] = 0; +} + +static void build_vrr_infopacket_data_v3(const struct mod_vrr_params *vrr, + struct dc_info_packet *infopacket) +{ + /* PB1 = 0x1A (24bit AMD IEEE OUI (0x00001A) - Byte 0) */ + infopacket->sb[1] = 0x1A; + + /* PB2 = 0x00 (24bit AMD IEEE OUI (0x00001A) - Byte 1) */ + infopacket->sb[2] = 0x00; + + /* PB3 = 0x00 (24bit AMD IEEE OUI (0x00001A) - Byte 2) */ + infopacket->sb[3] = 0x00; + + /* PB4 = Reserved */ + + /* PB5 = Reserved */ + + /* PB6 = [Bits 7:3 = Reserved] */ + + /* PB6 = [Bit 0 = FreeSync Supported] */ + if (vrr->state != VRR_STATE_UNSUPPORTED) + infopacket->sb[6] |= 0x01; + + /* PB6 = [Bit 1 = FreeSync Enabled] */ + if (vrr->state != VRR_STATE_DISABLED && + vrr->state != VRR_STATE_UNSUPPORTED) + infopacket->sb[6] |= 0x02; + + /* PB6 = [Bit 2 = FreeSync Active] */ + if (vrr->state == VRR_STATE_ACTIVE_VARIABLE || + vrr->state == VRR_STATE_ACTIVE_FIXED) + infopacket->sb[6] |= 0x04; + + if (vrr->state == VRR_STATE_ACTIVE_FIXED) { + /* PB7 = FreeSync Minimum refresh rate (Hz) */ + infopacket->sb[7] = (unsigned char)((vrr->fixed_refresh_in_uhz + 500000) / 1000000); + /* PB8 = FreeSync Maximum refresh rate (Hz) */ + infopacket->sb[8] = (unsigned char)((vrr->fixed_refresh_in_uhz + 500000) / 1000000); + } else if (vrr->state == VRR_STATE_ACTIVE_VARIABLE) { + /* PB7 = FreeSync Minimum refresh rate (Hz) */ + infopacket->sb[7] = (unsigned char)((vrr->min_refresh_in_uhz + 500000) / 1000000); + /* PB8 = FreeSync Maximum refresh rate (Hz) */ + infopacket->sb[8] = (unsigned char)((vrr->max_refresh_in_uhz + 500000) / 1000000); + } else { + // Non-fs case, program nominal range + /* PB7 = FreeSync Minimum refresh rate (Hz) */ + infopacket->sb[7] = (unsigned char)((vrr->max_refresh_in_uhz + 500000) / 1000000); + /* PB8 = FreeSync Maximum refresh rate (Hz) */ + infopacket->sb[8] = (unsigned char)((vrr->max_refresh_in_uhz + 500000) / 1000000); + } //FreeSync HDR infopacket->sb[9] = 0; @@ -678,7 +738,7 @@ static void build_vrr_infopacket_v1(enum signal_type signal, unsigned int payload_size = 0; build_vrr_infopacket_header_v1(signal, infopacket, &payload_size); - build_vrr_infopacket_data(vrr, infopacket); + build_vrr_infopacket_data_v1(vrr, infopacket); build_vrr_infopacket_checksum(&payload_size, infopacket); infopacket->valid = true; @@ -692,7 +752,24 @@ static void build_vrr_infopacket_v2(enum signal_type signal, unsigned int payload_size = 0; build_vrr_infopacket_header_v2(signal, infopacket, &payload_size); - build_vrr_infopacket_data(vrr, infopacket); + build_vrr_infopacket_data_v1(vrr, infopacket); + + build_vrr_infopacket_fs2_data(app_tf, infopacket); + + build_vrr_infopacket_checksum(&payload_size, infopacket); + + infopacket->valid = true; +} + +static void build_vrr_infopacket_v3(enum signal_type signal, + const struct mod_vrr_params *vrr, + enum color_transfer_func app_tf, + struct dc_info_packet *infopacket) +{ + unsigned int payload_size = 0; + + build_vrr_infopacket_header_v2(signal, infopacket, &payload_size); + build_vrr_infopacket_data_v3(vrr, infopacket); build_vrr_infopacket_fs2_data(app_tf, infopacket); @@ -717,11 +794,14 @@ void mod_freesync_build_vrr_infopacket(struct mod_freesync *mod_freesync, return; switch (packet_type) { - case PACKET_TYPE_FS2: + case PACKET_TYPE_FS_V3: + build_vrr_infopacket_v3(stream->signal, vrr, app_tf, infopacket); + break; + case PACKET_TYPE_FS_V2: build_vrr_infopacket_v2(stream->signal, vrr, app_tf, infopacket); break; case PACKET_TYPE_VRR: - case PACKET_TYPE_FS1: + case PACKET_TYPE_FS_V1: default: build_vrr_infopacket_v1(stream->signal, vrr, infopacket); } @@ -793,6 +873,11 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync, calc_duration_in_us_from_refresh_in_uhz( (unsigned int)max_refresh_in_uhz); + if (in_config->state == VRR_STATE_ACTIVE_FIXED) + in_out_vrr->fixed_refresh_in_uhz = in_config->fixed_refresh_in_uhz; + else + in_out_vrr->fixed_refresh_in_uhz = 0; + refresh_range = in_out_vrr->max_refresh_in_uhz - in_out_vrr->min_refresh_in_uhz; @@ -843,7 +928,7 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync, in_out_vrr->min_refresh_in_uhz); } else if (in_out_vrr->state == VRR_STATE_ACTIVE_FIXED) { in_out_vrr->fixed.target_refresh_in_uhz = - in_out_vrr->min_refresh_in_uhz; + in_out_vrr->fixed_refresh_in_uhz; if (in_out_vrr->fixed.ramping_active && in_out_vrr->fixed.fixed_active) { /* Do not update vtotals if ramping is already active diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c index cc1d3f470b99..e9fbd94f8635 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c @@ -328,7 +328,8 @@ enum mod_hdcp_status mod_hdcp_add_display(struct mod_hdcp *hdcp, /* add display to connection */ hdcp->connection.link = *link; *display_container = *display; - status = mod_hdcp_add_display_to_topology(hdcp, display->index); + status = mod_hdcp_add_display_to_topology(hdcp, display_container); + if (status != MOD_HDCP_STATUS_SUCCESS) goto out; @@ -374,7 +375,7 @@ enum mod_hdcp_status mod_hdcp_remove_display(struct mod_hdcp *hdcp, status = mod_hdcp_remove_display_from_topology(hdcp, index); if (status != MOD_HDCP_STATUS_SUCCESS) goto out; - display->state = MOD_HDCP_DISPLAY_INACTIVE; + memset(display, 0, sizeof(struct mod_hdcp_display)); /* request authentication when connection is not reset */ if (current_state(hdcp) != HDCP_UNINITIALIZED) diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h index 5cb4546be0ef..b0cefed2eb02 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h @@ -328,7 +328,7 @@ void mod_hdcp_dump_binary_message(uint8_t *msg, uint32_t msg_size, /* psp functions */ enum mod_hdcp_status mod_hdcp_add_display_to_topology( - struct mod_hdcp *hdcp, uint8_t index); + struct mod_hdcp *hdcp, struct mod_hdcp_display *display); enum mod_hdcp_status mod_hdcp_remove_display_from_topology( struct mod_hdcp *hdcp, uint8_t index); enum mod_hdcp_status mod_hdcp_hdcp1_create_session(struct mod_hdcp *hdcp); @@ -357,8 +357,6 @@ enum mod_hdcp_status mod_hdcp_hdcp2_prepare_stream_management( struct mod_hdcp *hdcp); enum mod_hdcp_status mod_hdcp_hdcp2_validate_stream_ready( struct mod_hdcp *hdcp); -enum mod_hdcp_status mod_hdcp_hdcp2_get_link_encryption_status(struct mod_hdcp *hdcp, - enum mod_hdcp_encryption_status *encryption_status); /* ddc functions */ enum mod_hdcp_status mod_hdcp_read_bksv(struct mod_hdcp *hdcp); @@ -503,11 +501,6 @@ static inline uint8_t is_display_active(struct mod_hdcp_display *display) return display->state >= MOD_HDCP_DISPLAY_ACTIVE; } -static inline uint8_t is_display_added(struct mod_hdcp_display *display) -{ - return display->state >= MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED; -} - static inline uint8_t is_display_encryption_enabled(struct mod_hdcp_display *display) { return display->state >= MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED; @@ -515,34 +508,23 @@ static inline uint8_t is_display_encryption_enabled(struct mod_hdcp_display *dis static inline uint8_t get_active_display_count(struct mod_hdcp *hdcp) { - uint8_t added_count = 0; + uint8_t active_count = 0; uint8_t i; for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) if (is_display_active(&hdcp->displays[i])) - added_count++; - return added_count; -} - -static inline uint8_t get_added_display_count(struct mod_hdcp *hdcp) -{ - uint8_t added_count = 0; - uint8_t i; - - for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) - if (is_display_added(&hdcp->displays[i])) - added_count++; - return added_count; + active_count++; + return active_count; } -static inline struct mod_hdcp_display *get_first_added_display( +static inline struct mod_hdcp_display *get_first_active_display( struct mod_hdcp *hdcp) { uint8_t i; struct mod_hdcp_display *display = NULL; for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) - if (is_display_added(&hdcp->displays[i])) { + if (is_display_active(&hdcp->displays[i])) { display = &hdcp->displays[i]; break; } diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c index 37c8c05497d6..f244b72e74e0 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp1_execution.c @@ -129,7 +129,7 @@ static inline uint8_t get_device_count(struct mod_hdcp *hdcp) static inline enum mod_hdcp_status check_device_count(struct mod_hdcp *hdcp) { /* device count must be greater than or equal to tracked hdcp displays */ - return (get_device_count(hdcp) < get_added_display_count(hdcp)) ? + return (get_device_count(hdcp) < get_active_display_count(hdcp)) ? MOD_HDCP_STATUS_HDCP1_DEVICE_COUNT_MISMATCH_FAILURE : MOD_HDCP_STATUS_SUCCESS; } diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c index 491c00f48026..549c113abcf7 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp2_execution.c @@ -208,7 +208,7 @@ static inline uint8_t get_device_count(struct mod_hdcp *hdcp) static enum mod_hdcp_status check_device_count(struct mod_hdcp *hdcp) { /* device count must be greater than or equal to tracked hdcp displays */ - return (get_device_count(hdcp) < get_added_display_count(hdcp)) ? + return (get_device_count(hdcp) < get_active_display_count(hdcp)) ? MOD_HDCP_STATUS_HDCP2_DEVICE_COUNT_MISMATCH_FAILURE : MOD_HDCP_STATUS_SUCCESS; } diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.c index 44956f9ba178..fb6a19d020f9 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_log.c @@ -98,8 +98,8 @@ char *mod_hdcp_status_to_str(int32_t status) return "MOD_HDCP_STATUS_HDCP1_VALIDATE_KSV_LIST_FAILURE"; case MOD_HDCP_STATUS_HDCP1_KSV_LIST_REVOKED: return "MOD_HDCP_STATUS_HDCP1_KSV_LIST_REVOKED"; - case MOD_HDCP_STATUS_HDCP1_ENABLE_ENCRYPTION: - return "MOD_HDCP_STATUS_HDCP1_ENABLE_ENCRYPTION"; + case MOD_HDCP_STATUS_HDCP1_ENABLE_ENCRYPTION_FAILURE: + return "MOD_HDCP_STATUS_HDCP1_ENABLE_ENCRYPTION_FAILURE"; case MOD_HDCP_STATUS_HDCP1_ENABLE_STREAM_ENCRYPTION_FAILURE: return "MOD_HDCP_STATUS_HDCP1_ENABLE_STREAM_ENCRYPTION_FAILURE"; case MOD_HDCP_STATUS_HDCP1_MAX_CASCADE_EXCEEDED_FAILURE: @@ -158,8 +158,8 @@ char *mod_hdcp_status_to_str(int32_t status) return "MOD_HDCP_STATUS_HDCP2_RX_ID_LIST_REVOKED"; case MOD_HDCP_STATUS_HDCP2_RX_ID_LIST_NOT_READY: return "MOD_HDCP_STATUS_HDCP2_RX_ID_LIST_NOT_READY"; - case MOD_HDCP_STATUS_HDCP2_ENABLE_STREAM_ENCRYPTION: - return "MOD_HDCP_STATUS_HDCP2_ENABLE_STREAM_ENCRYPTION"; + case MOD_HDCP_STATUS_HDCP2_ENABLE_STREAM_ENCRYPTION_FAILURE: + return "MOD_HDCP_STATUS_HDCP2_ENABLE_STREAM_ENCRYPTION_FAILURE"; case MOD_HDCP_STATUS_HDCP2_STREAM_READY_PENDING: return "MOD_HDCP_STATUS_HDCP2_STREAM_READY_PENDING"; case MOD_HDCP_STATUS_HDCP2_VALIDATE_STREAM_READY_FAILURE: diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c index c2929815c3ee..fb1161dd7ea8 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c @@ -51,12 +51,15 @@ enum mod_hdcp_status mod_hdcp_remove_display_from_topology( struct ta_dtm_shared_memory *dtm_cmd; struct mod_hdcp_display *display = get_active_display_at_index(hdcp, index); + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; dtm_cmd = (struct ta_dtm_shared_memory *)psp->dtm_context.dtm_shared_buf; - if (!display || !is_display_added(display)) + if (!display || !is_display_active(display)) return MOD_HDCP_STATUS_DISPLAY_NOT_FOUND; + mutex_lock(&psp->dtm_context.mutex); + memset(dtm_cmd, 0, sizeof(struct ta_dtm_shared_memory)); dtm_cmd->cmd_id = TA_DTM_COMMAND__TOPOLOGY_UPDATE_V2; @@ -66,34 +69,33 @@ enum mod_hdcp_status mod_hdcp_remove_display_from_topology( psp_dtm_invoke(psp, dtm_cmd->cmd_id); - if (dtm_cmd->dtm_status != TA_DTM_STATUS__SUCCESS) - return MOD_HDCP_STATUS_UPDATE_TOPOLOGY_FAILURE; + if (dtm_cmd->dtm_status != TA_DTM_STATUS__SUCCESS) { + status = MOD_HDCP_STATUS_UPDATE_TOPOLOGY_FAILURE; + } else { + display->state = MOD_HDCP_DISPLAY_ACTIVE; + HDCP_TOP_REMOVE_DISPLAY_TRACE(hdcp, display->index); + } - display->state = MOD_HDCP_DISPLAY_ACTIVE; - HDCP_TOP_REMOVE_DISPLAY_TRACE(hdcp, display->index); - - return MOD_HDCP_STATUS_SUCCESS; - + mutex_unlock(&psp->dtm_context.mutex); + return status; } enum mod_hdcp_status mod_hdcp_add_display_to_topology(struct mod_hdcp *hdcp, - uint8_t index) + struct mod_hdcp_display *display) { struct psp_context *psp = hdcp->config.psp.handle; struct ta_dtm_shared_memory *dtm_cmd; - struct mod_hdcp_display *display = - get_active_display_at_index(hdcp, index); struct mod_hdcp_link *link = &hdcp->connection.link; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; if (!psp->dtm_context.dtm_initialized) { DRM_ERROR("Failed to add display topology, DTM TA is not initialized."); + display->state = MOD_HDCP_DISPLAY_INACTIVE; return MOD_HDCP_STATUS_FAILURE; } - if (!display || is_display_added(display)) - return MOD_HDCP_STATUS_UPDATE_TOPOLOGY_FAILURE; - dtm_cmd = (struct ta_dtm_shared_memory *)psp->dtm_context.dtm_shared_buf; + mutex_lock(&psp->dtm_context.mutex); memset(dtm_cmd, 0, sizeof(struct ta_dtm_shared_memory)); dtm_cmd->cmd_id = TA_DTM_COMMAND__TOPOLOGY_UPDATE_V2; @@ -113,21 +115,24 @@ enum mod_hdcp_status mod_hdcp_add_display_to_topology(struct mod_hdcp *hdcp, psp_dtm_invoke(psp, dtm_cmd->cmd_id); - if (dtm_cmd->dtm_status != TA_DTM_STATUS__SUCCESS) - return MOD_HDCP_STATUS_UPDATE_TOPOLOGY_FAILURE; - - display->state = MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED; - HDCP_TOP_ADD_DISPLAY_TRACE(hdcp, display->index); + if (dtm_cmd->dtm_status != TA_DTM_STATUS__SUCCESS) { + display->state = MOD_HDCP_DISPLAY_INACTIVE; + status = MOD_HDCP_STATUS_UPDATE_TOPOLOGY_FAILURE; + } else { + HDCP_TOP_ADD_DISPLAY_TRACE(hdcp, display->index); + } - return MOD_HDCP_STATUS_SUCCESS; + mutex_unlock(&psp->dtm_context.mutex); + return status; } enum mod_hdcp_status mod_hdcp_hdcp1_create_session(struct mod_hdcp *hdcp) { struct psp_context *psp = hdcp->config.psp.handle; - struct mod_hdcp_display *display = get_first_added_display(hdcp); + struct mod_hdcp_display *display = get_first_active_display(hdcp); struct ta_hdcp_shared_memory *hdcp_cmd; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; if (!psp->hdcp_context.hdcp_initialized) { DRM_ERROR("Failed to create hdcp session. HDCP TA is not initialized."); @@ -135,6 +140,8 @@ enum mod_hdcp_status mod_hdcp_hdcp1_create_session(struct mod_hdcp *hdcp) } hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + + mutex_lock(&psp->hdcp_context.mutex); memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); hdcp_cmd->in_msg.hdcp1_create_session.display_handle = display->index; @@ -144,16 +151,18 @@ enum mod_hdcp_status mod_hdcp_hdcp1_create_session(struct mod_hdcp *hdcp) hdcp->auth.id = hdcp_cmd->out_msg.hdcp1_create_session.session_handle; - if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP1_CREATE_SESSION_FAILURE; - - hdcp->auth.msg.hdcp1.ainfo = hdcp_cmd->out_msg.hdcp1_create_session.ainfo_primary; - memcpy(hdcp->auth.msg.hdcp1.aksv, hdcp_cmd->out_msg.hdcp1_create_session.aksv_primary, - sizeof(hdcp->auth.msg.hdcp1.aksv)); - memcpy(hdcp->auth.msg.hdcp1.an, hdcp_cmd->out_msg.hdcp1_create_session.an_primary, - sizeof(hdcp->auth.msg.hdcp1.an)); + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) { + status = MOD_HDCP_STATUS_HDCP1_CREATE_SESSION_FAILURE; + } else { + hdcp->auth.msg.hdcp1.ainfo = hdcp_cmd->out_msg.hdcp1_create_session.ainfo_primary; + memcpy(hdcp->auth.msg.hdcp1.aksv, hdcp_cmd->out_msg.hdcp1_create_session.aksv_primary, + sizeof(hdcp->auth.msg.hdcp1.aksv)); + memcpy(hdcp->auth.msg.hdcp1.an, hdcp_cmd->out_msg.hdcp1_create_session.an_primary, + sizeof(hdcp->auth.msg.hdcp1.an)); + } - return MOD_HDCP_STATUS_SUCCESS; + mutex_unlock(&psp->hdcp_context.mutex); + return status; } enum mod_hdcp_status mod_hdcp_hdcp1_destroy_session(struct mod_hdcp *hdcp) @@ -162,7 +171,9 @@ enum mod_hdcp_status mod_hdcp_hdcp1_destroy_session(struct mod_hdcp *hdcp) struct psp_context *psp = hdcp->config.psp.handle; struct ta_hdcp_shared_memory *hdcp_cmd; uint8_t i = 0; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + mutex_lock(&psp->hdcp_context.mutex); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -171,27 +182,30 @@ enum mod_hdcp_status mod_hdcp_hdcp1_destroy_session(struct mod_hdcp *hdcp) psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); - if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP1_DESTROY_SESSION_FAILURE; - - HDCP_TOP_HDCP1_DESTROY_SESSION_TRACE(hdcp); - for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) - if (is_display_encryption_enabled( - &hdcp->displays[i])) { - hdcp->displays[i].state = - MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED; - HDCP_HDCP1_DISABLED_TRACE(hdcp, - hdcp->displays[i].index); - } + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) { + status = MOD_HDCP_STATUS_HDCP1_DESTROY_SESSION_FAILURE; + } else { + HDCP_TOP_HDCP1_DESTROY_SESSION_TRACE(hdcp); + for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) + if (is_display_encryption_enabled(&hdcp->displays[i])) { + hdcp->displays[i].state = + MOD_HDCP_DISPLAY_ACTIVE; + HDCP_HDCP1_DISABLED_TRACE( + hdcp, hdcp->displays[i].index); + } + } - return MOD_HDCP_STATUS_SUCCESS; + mutex_unlock(&psp->hdcp_context.mutex); + return status; } enum mod_hdcp_status mod_hdcp_hdcp1_validate_rx(struct mod_hdcp *hdcp) { struct psp_context *psp = hdcp->config.psp.handle; struct ta_hdcp_shared_memory *hdcp_cmd; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + mutex_lock(&psp->hdcp_context.mutex); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -206,10 +220,9 @@ enum mod_hdcp_status mod_hdcp_hdcp1_validate_rx(struct mod_hdcp *hdcp) psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); - if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP1_VALIDATE_RX_FAILURE; - - if (hdcp_cmd->out_msg.hdcp1_first_part_authentication.authentication_status == + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) { + status = MOD_HDCP_STATUS_HDCP1_VALIDATE_RX_FAILURE; + } else if (hdcp_cmd->out_msg.hdcp1_first_part_authentication.authentication_status == TA_HDCP_AUTHENTICATION_STATUS__HDCP1_FIRST_PART_COMPLETE) { /* needs second part of authentication */ hdcp->connection.is_repeater = 1; @@ -219,20 +232,22 @@ enum mod_hdcp_status mod_hdcp_hdcp1_validate_rx(struct mod_hdcp *hdcp) } else if (hdcp_cmd->out_msg.hdcp1_first_part_authentication.authentication_status == TA_HDCP_AUTHENTICATION_STATUS__HDCP1_KSV_REVOKED) { hdcp->connection.is_hdcp1_revoked = 1; - return MOD_HDCP_STATUS_HDCP1_BKSV_REVOKED; + status = MOD_HDCP_STATUS_HDCP1_BKSV_REVOKED; } else - return MOD_HDCP_STATUS_HDCP1_VALIDATE_RX_FAILURE; - + status = MOD_HDCP_STATUS_HDCP1_VALIDATE_RX_FAILURE; - return MOD_HDCP_STATUS_SUCCESS; + mutex_unlock(&psp->hdcp_context.mutex); + return status; } enum mod_hdcp_status mod_hdcp_hdcp1_enable_encryption(struct mod_hdcp *hdcp) { struct psp_context *psp = hdcp->config.psp.handle; struct ta_hdcp_shared_memory *hdcp_cmd; - struct mod_hdcp_display *display = get_first_added_display(hdcp); + struct mod_hdcp_display *display = get_first_active_display(hdcp); + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + mutex_lock(&psp->hdcp_context.mutex); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -241,14 +256,15 @@ enum mod_hdcp_status mod_hdcp_hdcp1_enable_encryption(struct mod_hdcp *hdcp) psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); - if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP1_ENABLE_ENCRYPTION; - - if (!is_dp_mst_hdcp(hdcp)) { + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) { + status = MOD_HDCP_STATUS_HDCP1_ENABLE_ENCRYPTION_FAILURE; + } else if (!is_dp_mst_hdcp(hdcp)) { display->state = MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED; HDCP_HDCP1_ENABLED_TRACE(hdcp, display->index); } - return MOD_HDCP_STATUS_SUCCESS; + + mutex_unlock(&psp->hdcp_context.mutex); + return status; } enum mod_hdcp_status mod_hdcp_hdcp1_validate_ksvlist_vp(struct mod_hdcp *hdcp) @@ -257,6 +273,7 @@ enum mod_hdcp_status mod_hdcp_hdcp1_validate_ksvlist_vp(struct mod_hdcp *hdcp) struct ta_hdcp_shared_memory *hdcp_cmd; enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + mutex_lock(&psp->hdcp_context.mutex); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -287,6 +304,7 @@ enum mod_hdcp_status mod_hdcp_hdcp1_validate_ksvlist_vp(struct mod_hdcp *hdcp) status = MOD_HDCP_STATUS_HDCP1_VALIDATE_KSV_LIST_FAILURE; } + mutex_unlock(&psp->hdcp_context.mutex); return status; } @@ -296,14 +314,15 @@ enum mod_hdcp_status mod_hdcp_hdcp1_enable_dp_stream_encryption(struct mod_hdcp struct psp_context *psp = hdcp->config.psp.handle; struct ta_hdcp_shared_memory *hdcp_cmd; int i = 0; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + mutex_lock(&psp->hdcp_context.mutex); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { - if (hdcp->displays[i].state != MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED || - hdcp->displays[i].adjust.disable) - continue; + if (hdcp->displays[i].adjust.disable || hdcp->displays[i].state != MOD_HDCP_DISPLAY_ACTIVE) + continue; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -313,21 +332,26 @@ enum mod_hdcp_status mod_hdcp_hdcp1_enable_dp_stream_encryption(struct mod_hdcp psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); - if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP1_ENABLE_STREAM_ENCRYPTION_FAILURE; + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) { + status = MOD_HDCP_STATUS_HDCP1_ENABLE_STREAM_ENCRYPTION_FAILURE; + break; + } hdcp->displays[i].state = MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED; HDCP_HDCP1_ENABLED_TRACE(hdcp, hdcp->displays[i].index); } - return MOD_HDCP_STATUS_SUCCESS; + mutex_unlock(&psp->hdcp_context.mutex); + return status; } enum mod_hdcp_status mod_hdcp_hdcp1_link_maintenance(struct mod_hdcp *hdcp) { struct psp_context *psp = hdcp->config.psp.handle; struct ta_hdcp_shared_memory *hdcp_cmd; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + mutex_lock(&psp->hdcp_context.mutex); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -339,12 +363,12 @@ enum mod_hdcp_status mod_hdcp_hdcp1_link_maintenance(struct mod_hdcp *hdcp) psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); - if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP1_LINK_MAINTENANCE_FAILURE; + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS || + hdcp_cmd->out_msg.hdcp1_get_encryption_status.protection_level != 1) + status = MOD_HDCP_STATUS_HDCP1_LINK_MAINTENANCE_FAILURE; - return (hdcp_cmd->out_msg.hdcp1_get_encryption_status.protection_level == 1) - ? MOD_HDCP_STATUS_SUCCESS - : MOD_HDCP_STATUS_HDCP1_LINK_MAINTENANCE_FAILURE; + mutex_unlock(&psp->hdcp_context.mutex); + return status; } enum mod_hdcp_status mod_hdcp_hdcp1_get_link_encryption_status(struct mod_hdcp *hdcp, @@ -364,19 +388,23 @@ enum mod_hdcp_status mod_hdcp_hdcp2_create_session(struct mod_hdcp *hdcp) { struct psp_context *psp = hdcp->config.psp.handle; struct ta_hdcp_shared_memory *hdcp_cmd; - struct mod_hdcp_display *display = get_first_added_display(hdcp); + struct mod_hdcp_display *display = get_first_active_display(hdcp); + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + if (!psp->hdcp_context.hdcp_initialized) { DRM_ERROR("Failed to create hdcp session, HDCP TA is not initialized"); return MOD_HDCP_STATUS_FAILURE; } - hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; - memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); - if (!display) return MOD_HDCP_STATUS_DISPLAY_NOT_FOUND; + mutex_lock(&psp->hdcp_context.mutex); + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + hdcp_cmd->in_msg.hdcp2_create_session_v2.display_handle = display->index; if (hdcp->connection.link.adjust.hdcp2.force_type == MOD_HDCP_FORCE_TYPE_0) @@ -393,12 +421,14 @@ enum mod_hdcp_status mod_hdcp_hdcp2_create_session(struct mod_hdcp *hdcp) psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); - if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP2_CREATE_SESSION_FAILURE; - hdcp->auth.id = hdcp_cmd->out_msg.hdcp2_create_session_v2.session_handle; + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) + status = MOD_HDCP_STATUS_HDCP2_CREATE_SESSION_FAILURE; + else + hdcp->auth.id = hdcp_cmd->out_msg.hdcp2_create_session_v2.session_handle; - return MOD_HDCP_STATUS_SUCCESS; + mutex_unlock(&psp->hdcp_context.mutex); + return status; } enum mod_hdcp_status mod_hdcp_hdcp2_destroy_session(struct mod_hdcp *hdcp) @@ -406,7 +436,9 @@ enum mod_hdcp_status mod_hdcp_hdcp2_destroy_session(struct mod_hdcp *hdcp) struct psp_context *psp = hdcp->config.psp.handle; struct ta_hdcp_shared_memory *hdcp_cmd; uint8_t i = 0; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + mutex_lock(&psp->hdcp_context.mutex); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -415,20 +447,21 @@ enum mod_hdcp_status mod_hdcp_hdcp2_destroy_session(struct mod_hdcp *hdcp) psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); - if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP2_DESTROY_SESSION_FAILURE; - - HDCP_TOP_HDCP2_DESTROY_SESSION_TRACE(hdcp); - for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) - if (is_display_encryption_enabled( - &hdcp->displays[i])) { - hdcp->displays[i].state = - MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED; - HDCP_HDCP2_DISABLED_TRACE(hdcp, - hdcp->displays[i].index); - } + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) { + status = MOD_HDCP_STATUS_HDCP2_DESTROY_SESSION_FAILURE; + } else { + HDCP_TOP_HDCP2_DESTROY_SESSION_TRACE(hdcp); + for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) + if (is_display_encryption_enabled(&hdcp->displays[i])) { + hdcp->displays[i].state = + MOD_HDCP_DISPLAY_ACTIVE; + HDCP_HDCP2_DISABLED_TRACE( + hdcp, hdcp->displays[i].index); + } + } - return MOD_HDCP_STATUS_SUCCESS; + mutex_unlock(&psp->hdcp_context.mutex); + return status; } enum mod_hdcp_status mod_hdcp_hdcp2_prepare_ake_init(struct mod_hdcp *hdcp) @@ -437,7 +470,9 @@ enum mod_hdcp_status mod_hdcp_hdcp2_prepare_ake_init(struct mod_hdcp *hdcp) struct ta_hdcp_shared_memory *hdcp_cmd; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 *msg_out; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + mutex_lock(&psp->hdcp_context.mutex); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -452,12 +487,13 @@ enum mod_hdcp_status mod_hdcp_hdcp2_prepare_ake_init(struct mod_hdcp *hdcp) psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP2_PREP_AKE_INIT_FAILURE; - - memcpy(&hdcp->auth.msg.hdcp2.ake_init[0], &msg_out->prepare.transmitter_message[0], - sizeof(hdcp->auth.msg.hdcp2.ake_init)); + status = MOD_HDCP_STATUS_HDCP2_PREP_AKE_INIT_FAILURE; + else + memcpy(&hdcp->auth.msg.hdcp2.ake_init[0], &msg_out->prepare.transmitter_message[0], + sizeof(hdcp->auth.msg.hdcp2.ake_init)); - return MOD_HDCP_STATUS_SUCCESS; + mutex_unlock(&psp->hdcp_context.mutex); + return status; } enum mod_hdcp_status mod_hdcp_hdcp2_validate_ake_cert(struct mod_hdcp *hdcp) @@ -466,7 +502,9 @@ enum mod_hdcp_status mod_hdcp_hdcp2_validate_ake_cert(struct mod_hdcp *hdcp) struct ta_hdcp_shared_memory *hdcp_cmd; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 *msg_out; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + mutex_lock(&psp->hdcp_context.mutex); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -488,26 +526,32 @@ enum mod_hdcp_status mod_hdcp_hdcp2_validate_ake_cert(struct mod_hdcp *hdcp) psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); - if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP2_VALIDATE_AKE_CERT_FAILURE; - - memcpy(hdcp->auth.msg.hdcp2.ake_no_stored_km, &msg_out->prepare.transmitter_message[0], - sizeof(hdcp->auth.msg.hdcp2.ake_no_stored_km)); - - memcpy(hdcp->auth.msg.hdcp2.ake_stored_km, - &msg_out->prepare.transmitter_message[sizeof(hdcp->auth.msg.hdcp2.ake_no_stored_km)], - sizeof(hdcp->auth.msg.hdcp2.ake_stored_km)); - - if (msg_out->process.msg1_status == TA_HDCP2_MSG_AUTHENTICATION_STATUS__SUCCESS) { - hdcp->connection.is_km_stored = msg_out->process.is_km_stored ? 1 : 0; - hdcp->connection.is_repeater = msg_out->process.is_repeater ? 1 : 0; - return MOD_HDCP_STATUS_SUCCESS; - } else if (msg_out->process.msg1_status == TA_HDCP2_MSG_AUTHENTICATION_STATUS__RECEIVERID_REVOKED) { - hdcp->connection.is_hdcp2_revoked = 1; - return MOD_HDCP_STATUS_HDCP2_AKE_CERT_REVOKED; + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) { + status = MOD_HDCP_STATUS_HDCP2_VALIDATE_AKE_CERT_FAILURE; + } else { + memcpy(hdcp->auth.msg.hdcp2.ake_no_stored_km, + &msg_out->prepare.transmitter_message[0], + sizeof(hdcp->auth.msg.hdcp2.ake_no_stored_km)); + + memcpy(hdcp->auth.msg.hdcp2.ake_stored_km, + &msg_out->prepare.transmitter_message[sizeof(hdcp->auth.msg.hdcp2.ake_no_stored_km)], + sizeof(hdcp->auth.msg.hdcp2.ake_stored_km)); + + if (msg_out->process.msg1_status == + TA_HDCP2_MSG_AUTHENTICATION_STATUS__SUCCESS) { + hdcp->connection.is_km_stored = + msg_out->process.is_km_stored ? 1 : 0; + hdcp->connection.is_repeater = + msg_out->process.is_repeater ? 1 : 0; + status = MOD_HDCP_STATUS_SUCCESS; + } else if (msg_out->process.msg1_status == + TA_HDCP2_MSG_AUTHENTICATION_STATUS__RECEIVERID_REVOKED) { + hdcp->connection.is_hdcp2_revoked = 1; + status = MOD_HDCP_STATUS_HDCP2_AKE_CERT_REVOKED; + } } - - return MOD_HDCP_STATUS_HDCP2_VALIDATE_AKE_CERT_FAILURE; + mutex_unlock(&psp->hdcp_context.mutex); + return status; } enum mod_hdcp_status mod_hdcp_hdcp2_validate_h_prime(struct mod_hdcp *hdcp) @@ -516,7 +560,9 @@ enum mod_hdcp_status mod_hdcp_hdcp2_validate_h_prime(struct mod_hdcp *hdcp) struct ta_hdcp_shared_memory *hdcp_cmd; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 *msg_out; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + mutex_lock(&psp->hdcp_context.mutex); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -543,16 +589,15 @@ enum mod_hdcp_status mod_hdcp_hdcp2_validate_h_prime(struct mod_hdcp *hdcp) psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP2_VALIDATE_H_PRIME_FAILURE; - - if (msg_out->process.msg1_status != TA_HDCP2_MSG_AUTHENTICATION_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP2_VALIDATE_H_PRIME_FAILURE; + status = MOD_HDCP_STATUS_HDCP2_VALIDATE_H_PRIME_FAILURE; + else if (msg_out->process.msg1_status != TA_HDCP2_MSG_AUTHENTICATION_STATUS__SUCCESS) + status = MOD_HDCP_STATUS_HDCP2_VALIDATE_H_PRIME_FAILURE; else if (!hdcp->connection.is_km_stored && - msg_out->process.msg2_status != TA_HDCP2_MSG_AUTHENTICATION_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP2_VALIDATE_PAIRING_INFO_FAILURE; - + msg_out->process.msg2_status != TA_HDCP2_MSG_AUTHENTICATION_STATUS__SUCCESS) + status = MOD_HDCP_STATUS_HDCP2_VALIDATE_PAIRING_INFO_FAILURE; - return MOD_HDCP_STATUS_SUCCESS; + mutex_unlock(&psp->hdcp_context.mutex); + return status; } enum mod_hdcp_status mod_hdcp_hdcp2_prepare_lc_init(struct mod_hdcp *hdcp) @@ -561,7 +606,9 @@ enum mod_hdcp_status mod_hdcp_hdcp2_prepare_lc_init(struct mod_hdcp *hdcp) struct ta_hdcp_shared_memory *hdcp_cmd; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 *msg_out; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + mutex_lock(&psp->hdcp_context.mutex); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -577,12 +624,13 @@ enum mod_hdcp_status mod_hdcp_hdcp2_prepare_lc_init(struct mod_hdcp *hdcp) psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP2_PREP_LC_INIT_FAILURE; - - memcpy(hdcp->auth.msg.hdcp2.lc_init, &msg_out->prepare.transmitter_message[0], - sizeof(hdcp->auth.msg.hdcp2.lc_init)); + status = MOD_HDCP_STATUS_HDCP2_PREP_LC_INIT_FAILURE; + else + memcpy(hdcp->auth.msg.hdcp2.lc_init, &msg_out->prepare.transmitter_message[0], + sizeof(hdcp->auth.msg.hdcp2.lc_init)); - return MOD_HDCP_STATUS_SUCCESS; + mutex_unlock(&psp->hdcp_context.mutex); + return status; } enum mod_hdcp_status mod_hdcp_hdcp2_validate_l_prime(struct mod_hdcp *hdcp) @@ -591,7 +639,9 @@ enum mod_hdcp_status mod_hdcp_hdcp2_validate_l_prime(struct mod_hdcp *hdcp) struct ta_hdcp_shared_memory *hdcp_cmd; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 *msg_out; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + mutex_lock(&psp->hdcp_context.mutex); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -610,13 +660,12 @@ enum mod_hdcp_status mod_hdcp_hdcp2_validate_l_prime(struct mod_hdcp *hdcp) psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); - if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP2_VALIDATE_L_PRIME_FAILURE; - - if (msg_out->process.msg1_status != TA_HDCP2_MSG_AUTHENTICATION_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP2_VALIDATE_L_PRIME_FAILURE; + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS || + msg_out->process.msg1_status != TA_HDCP2_MSG_AUTHENTICATION_STATUS__SUCCESS) + status = MOD_HDCP_STATUS_HDCP2_VALIDATE_L_PRIME_FAILURE; - return MOD_HDCP_STATUS_SUCCESS; + mutex_unlock(&psp->hdcp_context.mutex); + return status; } enum mod_hdcp_status mod_hdcp_hdcp2_prepare_eks(struct mod_hdcp *hdcp) @@ -625,7 +674,9 @@ enum mod_hdcp_status mod_hdcp_hdcp2_prepare_eks(struct mod_hdcp *hdcp) struct ta_hdcp_shared_memory *hdcp_cmd; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 *msg_out; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + mutex_lock(&psp->hdcp_context.mutex); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -642,48 +693,55 @@ enum mod_hdcp_status mod_hdcp_hdcp2_prepare_eks(struct mod_hdcp *hdcp) hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_PREPARE_PROCESS_AUTHENTICATION_MSG_V2; psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); - if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP2_PREP_EKS_FAILURE; - - memcpy(hdcp->auth.msg.hdcp2.ske_eks, &msg_out->prepare.transmitter_message[0], - sizeof(hdcp->auth.msg.hdcp2.ske_eks)); - msg_out->prepare.msg1_desc.msg_size = sizeof(hdcp->auth.msg.hdcp2.ske_eks); - - if (is_dp_hdcp(hdcp)) { - memcpy(hdcp->auth.msg.hdcp2.content_stream_type_dp, - &msg_out->prepare.transmitter_message[sizeof(hdcp->auth.msg.hdcp2.ske_eks)], - sizeof(hdcp->auth.msg.hdcp2.content_stream_type_dp)); + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) { + status = MOD_HDCP_STATUS_HDCP2_PREP_EKS_FAILURE; + } else { + memcpy(hdcp->auth.msg.hdcp2.ske_eks, + &msg_out->prepare.transmitter_message[0], + sizeof(hdcp->auth.msg.hdcp2.ske_eks)); + msg_out->prepare.msg1_desc.msg_size = + sizeof(hdcp->auth.msg.hdcp2.ske_eks); + + if (is_dp_hdcp(hdcp)) { + memcpy(hdcp->auth.msg.hdcp2.content_stream_type_dp, + &msg_out->prepare.transmitter_message[sizeof(hdcp->auth.msg.hdcp2.ske_eks)], + sizeof(hdcp->auth.msg.hdcp2.content_stream_type_dp)); + } } + mutex_unlock(&psp->hdcp_context.mutex); - return MOD_HDCP_STATUS_SUCCESS; + return status; } enum mod_hdcp_status mod_hdcp_hdcp2_enable_encryption(struct mod_hdcp *hdcp) { struct psp_context *psp = hdcp->config.psp.handle; struct ta_hdcp_shared_memory *hdcp_cmd; - struct mod_hdcp_display *display = get_first_added_display(hdcp); - - hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; - memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + struct mod_hdcp_display *display = get_first_active_display(hdcp); + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; if (!display) return MOD_HDCP_STATUS_DISPLAY_NOT_FOUND; + mutex_lock(&psp->hdcp_context.mutex); + + hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; + memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); + hdcp_cmd->in_msg.hdcp2_set_encryption.session_handle = hdcp->auth.id; hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_SET_ENCRYPTION; psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); - if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP2_ENABLE_ENCRYPTION_FAILURE; - - if (!is_dp_mst_hdcp(hdcp)) { + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) { + status = MOD_HDCP_STATUS_HDCP2_ENABLE_ENCRYPTION_FAILURE; + } else if (!is_dp_mst_hdcp(hdcp)) { display->state = MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED; HDCP_HDCP2_ENABLED_TRACE(hdcp, display->index); } - return MOD_HDCP_STATUS_SUCCESS; + mutex_unlock(&psp->hdcp_context.mutex); + return status; } enum mod_hdcp_status mod_hdcp_hdcp2_validate_rx_id_list(struct mod_hdcp *hdcp) @@ -692,6 +750,9 @@ enum mod_hdcp_status mod_hdcp_hdcp2_validate_rx_id_list(struct mod_hdcp *hdcp) struct ta_hdcp_shared_memory *hdcp_cmd; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 *msg_out; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + + mutex_lock(&psp->hdcp_context.mutex); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -712,23 +773,26 @@ enum mod_hdcp_status mod_hdcp_hdcp2_validate_rx_id_list(struct mod_hdcp *hdcp) psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); - if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP2_VALIDATE_RX_ID_LIST_FAILURE; - - memcpy(hdcp->auth.msg.hdcp2.repeater_auth_ack, &msg_out->prepare.transmitter_message[0], - sizeof(hdcp->auth.msg.hdcp2.repeater_auth_ack)); - - if (msg_out->process.msg1_status == TA_HDCP2_MSG_AUTHENTICATION_STATUS__SUCCESS) { - hdcp->connection.is_km_stored = msg_out->process.is_km_stored ? 1 : 0; - hdcp->connection.is_repeater = msg_out->process.is_repeater ? 1 : 0; - return MOD_HDCP_STATUS_SUCCESS; - } else if (msg_out->process.msg1_status == TA_HDCP2_MSG_AUTHENTICATION_STATUS__RECEIVERID_REVOKED) { - hdcp->connection.is_hdcp2_revoked = 1; - return MOD_HDCP_STATUS_HDCP2_RX_ID_LIST_REVOKED; + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) { + status = MOD_HDCP_STATUS_HDCP2_VALIDATE_RX_ID_LIST_FAILURE; + } else { + memcpy(hdcp->auth.msg.hdcp2.repeater_auth_ack, + &msg_out->prepare.transmitter_message[0], + sizeof(hdcp->auth.msg.hdcp2.repeater_auth_ack)); + + if (msg_out->process.msg1_status == + TA_HDCP2_MSG_AUTHENTICATION_STATUS__SUCCESS) { + hdcp->connection.is_km_stored = msg_out->process.is_km_stored ? 1 : 0; + hdcp->connection.is_repeater = msg_out->process.is_repeater ? 1 : 0; + status = MOD_HDCP_STATUS_SUCCESS; + } else if (msg_out->process.msg1_status == + TA_HDCP2_MSG_AUTHENTICATION_STATUS__RECEIVERID_REVOKED) { + hdcp->connection.is_hdcp2_revoked = 1; + status = MOD_HDCP_STATUS_HDCP2_RX_ID_LIST_REVOKED; + } } - - - return MOD_HDCP_STATUS_HDCP2_VALIDATE_RX_ID_LIST_FAILURE; + mutex_unlock(&psp->hdcp_context.mutex); + return status; } enum mod_hdcp_status mod_hdcp_hdcp2_enable_dp_stream_encryption(struct mod_hdcp *hdcp) @@ -737,7 +801,9 @@ enum mod_hdcp_status mod_hdcp_hdcp2_enable_dp_stream_encryption(struct mod_hdcp struct ta_hdcp_shared_memory *hdcp_cmd; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; uint8_t i; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + mutex_lock(&psp->hdcp_context.mutex); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -747,9 +813,9 @@ enum mod_hdcp_status mod_hdcp_hdcp2_enable_dp_stream_encryption(struct mod_hdcp for (i = 0; i < MAX_NUM_OF_DISPLAYS; i++) { - if (hdcp->displays[i].state != MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED || - hdcp->displays[i].adjust.disable) - continue; + if (hdcp->displays[i].adjust.disable || hdcp->displays[i].state != MOD_HDCP_DISPLAY_ACTIVE) + continue; + hdcp_cmd->in_msg.hdcp2_enable_dp_stream_encryption.display_handle = hdcp->displays[i].index; hdcp_cmd->in_msg.hdcp2_enable_dp_stream_encryption.session_handle = hdcp->auth.id; @@ -763,8 +829,13 @@ enum mod_hdcp_status mod_hdcp_hdcp2_enable_dp_stream_encryption(struct mod_hdcp HDCP_HDCP2_ENABLED_TRACE(hdcp, hdcp->displays[i].index); } - return (hdcp_cmd->hdcp_status == TA_HDCP_STATUS__SUCCESS) ? MOD_HDCP_STATUS_SUCCESS - : MOD_HDCP_STATUS_HDCP2_ENABLE_STREAM_ENCRYPTION; + if (hdcp_cmd->hdcp_status == TA_HDCP_STATUS__SUCCESS) + status = MOD_HDCP_STATUS_SUCCESS; + else + status = MOD_HDCP_STATUS_HDCP2_ENABLE_STREAM_ENCRYPTION_FAILURE; + + mutex_unlock(&psp->hdcp_context.mutex); + return status; } enum mod_hdcp_status mod_hdcp_hdcp2_prepare_stream_management(struct mod_hdcp *hdcp) @@ -774,7 +845,9 @@ enum mod_hdcp_status mod_hdcp_hdcp2_prepare_stream_management(struct mod_hdcp *h struct ta_hdcp_shared_memory *hdcp_cmd; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 *msg_out; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + mutex_lock(&psp->hdcp_context.mutex); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -789,15 +862,17 @@ enum mod_hdcp_status mod_hdcp_hdcp2_prepare_stream_management(struct mod_hdcp *h hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_PREPARE_PROCESS_AUTHENTICATION_MSG_V2; psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); - if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) - return MOD_HDCP_STATUS_HDCP2_PREPARE_STREAM_MANAGEMENT_FAILURE; - - hdcp->auth.msg.hdcp2.stream_manage_size = msg_out->prepare.msg1_desc.msg_size; - - memcpy(hdcp->auth.msg.hdcp2.repeater_auth_stream_manage, &msg_out->prepare.transmitter_message[0], - sizeof(hdcp->auth.msg.hdcp2.repeater_auth_stream_manage)); + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) { + status = MOD_HDCP_STATUS_HDCP2_PREPARE_STREAM_MANAGEMENT_FAILURE; + } else { + hdcp->auth.msg.hdcp2.stream_manage_size = msg_out->prepare.msg1_desc.msg_size; - return MOD_HDCP_STATUS_SUCCESS; + memcpy(hdcp->auth.msg.hdcp2.repeater_auth_stream_manage, + &msg_out->prepare.transmitter_message[0], + sizeof(hdcp->auth.msg.hdcp2.repeater_auth_stream_manage)); + } + mutex_unlock(&psp->hdcp_context.mutex); + return status; } enum mod_hdcp_status mod_hdcp_hdcp2_validate_stream_ready(struct mod_hdcp *hdcp) @@ -806,7 +881,9 @@ enum mod_hdcp_status mod_hdcp_hdcp2_validate_stream_ready(struct mod_hdcp *hdcp) struct ta_hdcp_shared_memory *hdcp_cmd; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_input_v2 *msg_in; struct ta_hdcp_cmd_hdcp2_process_prepare_authentication_message_output_v2 *msg_out; + enum mod_hdcp_status status = MOD_HDCP_STATUS_SUCCESS; + mutex_lock(&psp->hdcp_context.mutex); hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); @@ -825,38 +902,13 @@ enum mod_hdcp_status mod_hdcp_hdcp2_validate_stream_ready(struct mod_hdcp *hdcp) hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_PREPARE_PROCESS_AUTHENTICATION_MSG_V2; psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); - return (hdcp_cmd->hdcp_status == TA_HDCP_STATUS__SUCCESS) && - (msg_out->process.msg1_status == TA_HDCP2_MSG_AUTHENTICATION_STATUS__SUCCESS) - ? MOD_HDCP_STATUS_SUCCESS - : MOD_HDCP_STATUS_HDCP2_VALIDATE_STREAM_READY_FAILURE; -} - -enum mod_hdcp_status mod_hdcp_hdcp2_get_link_encryption_status(struct mod_hdcp *hdcp, - enum mod_hdcp_encryption_status *encryption_status) -{ - struct psp_context *psp = hdcp->config.psp.handle; - struct ta_hdcp_shared_memory *hdcp_cmd; - - hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.hdcp_shared_buf; - - memset(hdcp_cmd, 0, sizeof(struct ta_hdcp_shared_memory)); - - hdcp_cmd->in_msg.hdcp2_get_encryption_status.session_handle = hdcp->auth.id; - hdcp_cmd->out_msg.hdcp2_get_encryption_status.protection_level = 0; - hdcp_cmd->cmd_id = TA_HDCP_COMMAND__HDCP2_GET_ENCRYPTION_STATUS; - *encryption_status = MOD_HDCP_ENCRYPTION_STATUS_HDCP_OFF; - - psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); - - if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS) - return MOD_HDCP_STATUS_FAILURE; - - if (hdcp_cmd->out_msg.hdcp2_get_encryption_status.protection_level == 1) { - if (hdcp_cmd->out_msg.hdcp2_get_encryption_status.hdcp2_type == TA_HDCP2_CONTENT_TYPE__TYPE1) - *encryption_status = MOD_HDCP_ENCRYPTION_STATUS_HDCP2_TYPE1_ON; - else - *encryption_status = MOD_HDCP_ENCRYPTION_STATUS_HDCP2_TYPE0_ON; - } + if (hdcp_cmd->hdcp_status == TA_HDCP_STATUS__SUCCESS && + msg_out->process.msg1_status == TA_HDCP2_MSG_AUTHENTICATION_STATUS__SUCCESS) + status = MOD_HDCP_STATUS_SUCCESS; + else + status = MOD_HDCP_STATUS_HDCP2_VALIDATE_STREAM_READY_FAILURE; - return MOD_HDCP_STATUS_SUCCESS; + mutex_unlock(&psp->hdcp_context.mutex); + return status; } + diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h b/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h index dbe7835aabcf..0ba3cf7f336a 100644 --- a/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h +++ b/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h @@ -83,6 +83,8 @@ struct mod_freesync_config { bool btr; unsigned int min_refresh_in_uhz; unsigned int max_refresh_in_uhz; + unsigned int fixed_refresh_in_uhz; + }; struct mod_vrr_params_btr { @@ -112,6 +114,7 @@ struct mod_vrr_params { uint32_t max_duration_in_us; uint32_t max_refresh_in_uhz; uint32_t min_duration_in_us; + uint32_t fixed_refresh_in_uhz; struct dc_crtc_timing_adjust adjust; diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h b/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h index c088602bc1a0..eed560eecbab 100644 --- a/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h +++ b/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h @@ -60,7 +60,7 @@ enum mod_hdcp_status { MOD_HDCP_STATUS_HDCP1_KSV_LIST_NOT_READY, MOD_HDCP_STATUS_HDCP1_VALIDATE_KSV_LIST_FAILURE, MOD_HDCP_STATUS_HDCP1_KSV_LIST_REVOKED, - MOD_HDCP_STATUS_HDCP1_ENABLE_ENCRYPTION, + MOD_HDCP_STATUS_HDCP1_ENABLE_ENCRYPTION_FAILURE, MOD_HDCP_STATUS_HDCP1_ENABLE_STREAM_ENCRYPTION_FAILURE, MOD_HDCP_STATUS_HDCP1_MAX_CASCADE_EXCEEDED_FAILURE, MOD_HDCP_STATUS_HDCP1_MAX_DEVS_EXCEEDED_FAILURE, @@ -90,7 +90,7 @@ enum mod_hdcp_status { MOD_HDCP_STATUS_HDCP2_RX_ID_LIST_NOT_READY, MOD_HDCP_STATUS_HDCP2_VALIDATE_RX_ID_LIST_FAILURE, MOD_HDCP_STATUS_HDCP2_RX_ID_LIST_REVOKED, - MOD_HDCP_STATUS_HDCP2_ENABLE_STREAM_ENCRYPTION, + MOD_HDCP_STATUS_HDCP2_ENABLE_STREAM_ENCRYPTION_FAILURE, MOD_HDCP_STATUS_HDCP2_STREAM_READY_PENDING, MOD_HDCP_STATUS_HDCP2_VALIDATE_STREAM_READY_FAILURE, MOD_HDCP_STATUS_HDCP2_PREPARE_STREAM_MANAGEMENT_FAILURE, @@ -117,7 +117,6 @@ enum mod_hdcp_operation_mode { enum mod_hdcp_display_state { MOD_HDCP_DISPLAY_INACTIVE = 0, MOD_HDCP_DISPLAY_ACTIVE, - MOD_HDCP_DISPLAY_ACTIVE_AND_ADDED, MOD_HDCP_DISPLAY_ENCRYPTION_ENABLED }; diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_shared.h b/drivers/gpu/drm/amd/display/modules/inc/mod_shared.h index fe2117904329..198c0e64d13a 100644 --- a/drivers/gpu/drm/amd/display/modules/inc/mod_shared.h +++ b/drivers/gpu/drm/amd/display/modules/inc/mod_shared.h @@ -40,8 +40,9 @@ enum color_transfer_func { enum vrr_packet_type { PACKET_TYPE_VRR, - PACKET_TYPE_FS1, - PACKET_TYPE_FS2, + PACKET_TYPE_FS_V1, + PACKET_TYPE_FS_V2, + PACKET_TYPE_FS_V3, PACKET_TYPE_VTEM }; diff --git a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c index cff3ab15fc0c..7cd8a43d1889 100644 --- a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c +++ b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c @@ -144,7 +144,7 @@ void mod_build_vsc_infopacket(const struct dc_stream_state *stream, } /*VSC packet set to 2 when DP revision >= 1.2*/ - if (stream->psr_version != 0) + if (stream->link->psr_settings.psr_version != DC_PSR_VERSION_UNSUPPORTED) vsc_packet_revision = vsc_packet_rev2; /* Update to revision 5 for extended colorimetry support */ diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c index e75a4bb94488..8c37bcc27132 100644 --- a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c +++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c @@ -24,6 +24,9 @@ #include "power_helpers.h" #include "dc/inc/hw/dmcu.h" +#include "dc/inc/hw/abm.h" +#include "dc.h" +#include "core_types.h" #define DIV_ROUNDUP(a, b) (((a)+((b)/2))/(b)) @@ -237,7 +240,7 @@ static void fill_backlight_transform_table(struct dmcu_iram_parameters params, } static void fill_backlight_transform_table_v_2_2(struct dmcu_iram_parameters params, - struct iram_table_v_2_2 *table) + struct iram_table_v_2_2 *table, bool big_endian) { unsigned int i; unsigned int num_entries = NUM_BL_CURVE_SEGS; @@ -261,10 +264,12 @@ static void fill_backlight_transform_table_v_2_2(struct dmcu_iram_parameters par lut_index = (params.backlight_lut_array_size - 1) * i / (num_entries - 1); ASSERT(lut_index < params.backlight_lut_array_size); - table->backlight_thresholds[i] = - cpu_to_be16(DIV_ROUNDUP((i * 65536), num_entries)); - table->backlight_offsets[i] = - cpu_to_be16(params.backlight_lut_array[lut_index]); + table->backlight_thresholds[i] = (big_endian) ? + cpu_to_be16(DIV_ROUNDUP((i * 65536), num_entries)) : + cpu_to_le16(DIV_ROUNDUP((i * 65536), num_entries)); + table->backlight_offsets[i] = (big_endian) ? + cpu_to_be16(params.backlight_lut_array[lut_index]) : + cpu_to_le16(params.backlight_lut_array[lut_index]); } } @@ -584,18 +589,18 @@ void fill_iram_v_2_2(struct iram_table_v_2_2 *ram_table, struct dmcu_iram_parame ram_table->crgb_slope[7] = cpu_to_be16(0x1910); fill_backlight_transform_table_v_2_2( - params, ram_table); + params, ram_table, true); } -void fill_iram_v_2_3(struct iram_table_v_2_2 *ram_table, struct dmcu_iram_parameters params) +void fill_iram_v_2_3(struct iram_table_v_2_2 *ram_table, struct dmcu_iram_parameters params, bool big_endian) { unsigned int i, j; unsigned int set = params.set; ram_table->flags = 0x0; - - ram_table->min_abm_backlight = - cpu_to_be16(params.min_abm_backlight); + ram_table->min_abm_backlight = (big_endian) ? + cpu_to_be16(params.min_abm_backlight) : + cpu_to_le16(params.min_abm_backlight); for (i = 0; i < NUM_AGGR_LEVEL; i++) { ram_table->hybrid_factor[i] = abm_settings[set][i].brightness_gain; @@ -619,33 +624,51 @@ void fill_iram_v_2_3(struct iram_table_v_2_2 *ram_table, struct dmcu_iram_parame ram_table->iir_curve[4] = 0x65; //Gamma 2.2 - ram_table->crgb_thresh[0] = cpu_to_be16(0x127c); - ram_table->crgb_thresh[1] = cpu_to_be16(0x151b); - ram_table->crgb_thresh[2] = cpu_to_be16(0x17d5); - ram_table->crgb_thresh[3] = cpu_to_be16(0x1a56); - ram_table->crgb_thresh[4] = cpu_to_be16(0x1c83); - ram_table->crgb_thresh[5] = cpu_to_be16(0x1e72); - ram_table->crgb_thresh[6] = cpu_to_be16(0x20f0); - ram_table->crgb_thresh[7] = cpu_to_be16(0x232b); - ram_table->crgb_offset[0] = cpu_to_be16(0x2999); - ram_table->crgb_offset[1] = cpu_to_be16(0x3999); - ram_table->crgb_offset[2] = cpu_to_be16(0x4666); - ram_table->crgb_offset[3] = cpu_to_be16(0x5999); - ram_table->crgb_offset[4] = cpu_to_be16(0x6333); - ram_table->crgb_offset[5] = cpu_to_be16(0x7800); - ram_table->crgb_offset[6] = cpu_to_be16(0x8c00); - ram_table->crgb_offset[7] = cpu_to_be16(0xa000); - ram_table->crgb_slope[0] = cpu_to_be16(0x3609); - ram_table->crgb_slope[1] = cpu_to_be16(0x2dfa); - ram_table->crgb_slope[2] = cpu_to_be16(0x27ea); - ram_table->crgb_slope[3] = cpu_to_be16(0x235d); - ram_table->crgb_slope[4] = cpu_to_be16(0x2042); - ram_table->crgb_slope[5] = cpu_to_be16(0x1dc3); - ram_table->crgb_slope[6] = cpu_to_be16(0x1b1a); - ram_table->crgb_slope[7] = cpu_to_be16(0x1910); + ram_table->crgb_thresh[0] = (big_endian) ? cpu_to_be16(0x127c) : cpu_to_le16(0x127c); + ram_table->crgb_thresh[1] = (big_endian) ? cpu_to_be16(0x151b) : cpu_to_le16(0x151b); + ram_table->crgb_thresh[2] = (big_endian) ? cpu_to_be16(0x17d5) : cpu_to_le16(0x17d5); + ram_table->crgb_thresh[3] = (big_endian) ? cpu_to_be16(0x1a56) : cpu_to_le16(0x1a56); + ram_table->crgb_thresh[4] = (big_endian) ? cpu_to_be16(0x1c83) : cpu_to_le16(0x1c83); + ram_table->crgb_thresh[5] = (big_endian) ? cpu_to_be16(0x1e72) : cpu_to_le16(0x1e72); + ram_table->crgb_thresh[6] = (big_endian) ? cpu_to_be16(0x20f0) : cpu_to_le16(0x20f0); + ram_table->crgb_thresh[7] = (big_endian) ? cpu_to_be16(0x232b) : cpu_to_le16(0x232b); + ram_table->crgb_offset[0] = (big_endian) ? cpu_to_be16(0x2999) : cpu_to_le16(0x2999); + ram_table->crgb_offset[1] = (big_endian) ? cpu_to_be16(0x3999) : cpu_to_le16(0x3999); + ram_table->crgb_offset[2] = (big_endian) ? cpu_to_be16(0x4666) : cpu_to_le16(0x4666); + ram_table->crgb_offset[3] = (big_endian) ? cpu_to_be16(0x5999) : cpu_to_le16(0x5999); + ram_table->crgb_offset[4] = (big_endian) ? cpu_to_be16(0x6333) : cpu_to_le16(0x6333); + ram_table->crgb_offset[5] = (big_endian) ? cpu_to_be16(0x7800) : cpu_to_le16(0x7800); + ram_table->crgb_offset[6] = (big_endian) ? cpu_to_be16(0x8c00) : cpu_to_le16(0x8c00); + ram_table->crgb_offset[7] = (big_endian) ? cpu_to_be16(0xa000) : cpu_to_le16(0xa000); + ram_table->crgb_slope[0] = (big_endian) ? cpu_to_be16(0x3609) : cpu_to_le16(0x3609); + ram_table->crgb_slope[1] = (big_endian) ? cpu_to_be16(0x2dfa) : cpu_to_le16(0x2dfa); + ram_table->crgb_slope[2] = (big_endian) ? cpu_to_be16(0x27ea) : cpu_to_le16(0x27ea); + ram_table->crgb_slope[3] = (big_endian) ? cpu_to_be16(0x235d) : cpu_to_le16(0x235d); + ram_table->crgb_slope[4] = (big_endian) ? cpu_to_be16(0x2042) : cpu_to_le16(0x2042); + ram_table->crgb_slope[5] = (big_endian) ? cpu_to_be16(0x1dc3) : cpu_to_le16(0x1dc3); + ram_table->crgb_slope[6] = (big_endian) ? cpu_to_be16(0x1b1a) : cpu_to_le16(0x1b1a); + ram_table->crgb_slope[7] = (big_endian) ? cpu_to_be16(0x1910) : cpu_to_le16(0x1910); fill_backlight_transform_table_v_2_2( - params, ram_table); + params, ram_table, big_endian); +} + +bool dmub_init_abm_config(struct abm *abm, + struct dmcu_iram_parameters params) +{ + unsigned char ram_table[IRAM_SIZE]; + bool result = false; + + if (abm == NULL) + return false; + + memset(&ram_table, 0, sizeof(ram_table)); + + fill_iram_v_2_3((struct iram_table_v_2_2 *)ram_table, params, false); + result = abm->funcs->init_abm_config( + abm, (char *)(&ram_table), IRAM_RESERVE_AREA_START_V2_2); + + return result; } bool dmcu_load_iram(struct dmcu *dmcu, @@ -657,17 +680,17 @@ bool dmcu_load_iram(struct dmcu *dmcu, if (dmcu == NULL) return false; - if (!dmcu->funcs->is_dmcu_initialized(dmcu)) + if (dmcu && !dmcu->funcs->is_dmcu_initialized(dmcu)) return true; memset(&ram_table, 0, sizeof(ram_table)); if (dmcu->dmcu_version.abm_version == 0x24) { - fill_iram_v_2_3((struct iram_table_v_2_2 *)ram_table, params); - result = dmcu->funcs->load_iram( - dmcu, 0, (char *)(&ram_table), IRAM_RESERVE_AREA_START_V2_2); + fill_iram_v_2_3((struct iram_table_v_2_2 *)ram_table, params, true); + result = dmcu->funcs->load_iram( + dmcu, 0, (char *)(&ram_table), IRAM_RESERVE_AREA_START_V2_2); } else if (dmcu->dmcu_version.abm_version == 0x23) { - fill_iram_v_2_3((struct iram_table_v_2_2 *)ram_table, params); + fill_iram_v_2_3((struct iram_table_v_2_2 *)ram_table, params, true); result = dmcu->funcs->load_iram( dmcu, 0, (char *)(&ram_table), IRAM_RESERVE_AREA_START_V2_2); diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.h b/drivers/gpu/drm/amd/display/modules/power/power_helpers.h index e54157026330..46fbca2e2cd1 100644 --- a/drivers/gpu/drm/amd/display/modules/power/power_helpers.h +++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.h @@ -26,6 +26,7 @@ #define MODULES_POWER_POWER_HELPERS_H_ #include "dc/inc/hw/dmcu.h" +#include "dc/inc/hw/abm.h" enum abm_defines { @@ -44,5 +45,7 @@ struct dmcu_iram_parameters { bool dmcu_load_iram(struct dmcu *dmcu, struct dmcu_iram_parameters params); +bool dmub_init_abm_config(struct abm *abm, + struct dmcu_iram_parameters params); #endif /* MODULES_POWER_POWER_HELPERS_H_ */ diff --git a/drivers/gpu/drm/amd/display/modules/stats/stats.c b/drivers/gpu/drm/amd/display/modules/stats/stats.c deleted file mode 100644 index 03121ca64fe4..000000000000 --- a/drivers/gpu/drm/amd/display/modules/stats/stats.c +++ /dev/null @@ -1,448 +0,0 @@ -/* - * Copyright 2016 Advanced Micro Devices, Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: AMD - * - */ - -#include "mod_stats.h" -#include "dm_services.h" -#include "dc.h" -#include "core_types.h" - -#define DAL_STATS_ENABLE_REGKEY "DalStatsEnable" -#define DAL_STATS_ENABLE_REGKEY_DEFAULT 0x00000000 -#define DAL_STATS_ENABLE_REGKEY_ENABLED 0x00000001 - -#define DAL_STATS_ENTRIES_REGKEY "DalStatsEntries" -#define DAL_STATS_ENTRIES_REGKEY_DEFAULT 0x00350000 -#define DAL_STATS_ENTRIES_REGKEY_MAX 0x01000000 - -#define DAL_STATS_EVENT_ENTRIES_DEFAULT 0x00000100 - -#define MOD_STATS_NUM_VSYNCS 5 -#define MOD_STATS_EVENT_STRING_MAX 512 - -struct stats_time_cache { - unsigned int entry_id; - - unsigned long flip_timestamp_in_ns; - unsigned long vupdate_timestamp_in_ns; - - unsigned int render_time_in_us; - unsigned int avg_render_time_in_us_last_ten; - unsigned int v_sync_time_in_us[MOD_STATS_NUM_VSYNCS]; - unsigned int num_vsync_between_flips; - - unsigned int flip_to_vsync_time_in_us; - unsigned int vsync_to_flip_time_in_us; - - unsigned int min_window; - unsigned int max_window; - unsigned int v_total_min; - unsigned int v_total_max; - unsigned int event_triggers; - - unsigned int lfc_mid_point_in_us; - unsigned int num_frames_inserted; - unsigned int inserted_duration_in_us; - - unsigned int flags; -}; - -struct stats_event_cache { - unsigned int entry_id; - char event_string[MOD_STATS_EVENT_STRING_MAX]; -}; - -struct core_stats { - struct mod_stats public; - struct dc *dc; - - bool enabled; - unsigned int entries; - unsigned int event_entries; - unsigned int entry_id; - - struct stats_time_cache *time; - unsigned int index; - - struct stats_event_cache *events; - unsigned int event_index; - -}; - -#define MOD_STATS_TO_CORE(mod_stats)\ - container_of(mod_stats, struct core_stats, public) - -bool mod_stats_init(struct mod_stats *mod_stats) -{ - bool result = false; - struct core_stats *core_stats = NULL; - struct dc *dc = NULL; - - if (mod_stats == NULL) - return false; - - core_stats = MOD_STATS_TO_CORE(mod_stats); - dc = core_stats->dc; - - return result; -} - -struct mod_stats *mod_stats_create(struct dc *dc) -{ - struct core_stats *core_stats = NULL; - struct persistent_data_flag flag; - unsigned int reg_data; - int i = 0; - - if (dc == NULL) - goto fail_construct; - - core_stats = kzalloc(sizeof(struct core_stats), GFP_KERNEL); - - if (core_stats == NULL) - goto fail_construct; - - core_stats->dc = dc; - - core_stats->enabled = DAL_STATS_ENABLE_REGKEY_DEFAULT; - if (dm_read_persistent_data(dc->ctx, NULL, NULL, - DAL_STATS_ENABLE_REGKEY, - ®_data, sizeof(unsigned int), &flag)) - core_stats->enabled = reg_data; - - if (core_stats->enabled) { - core_stats->entries = DAL_STATS_ENTRIES_REGKEY_DEFAULT; - if (dm_read_persistent_data(dc->ctx, NULL, NULL, - DAL_STATS_ENTRIES_REGKEY, - ®_data, sizeof(unsigned int), &flag)) { - if (reg_data > DAL_STATS_ENTRIES_REGKEY_MAX) - core_stats->entries = DAL_STATS_ENTRIES_REGKEY_MAX; - else - core_stats->entries = reg_data; - } - core_stats->time = kcalloc(core_stats->entries, - sizeof(struct stats_time_cache), - GFP_KERNEL); - - if (core_stats->time == NULL) - goto fail_construct_time; - - core_stats->event_entries = DAL_STATS_EVENT_ENTRIES_DEFAULT; - core_stats->events = kcalloc(core_stats->event_entries, - sizeof(struct stats_event_cache), - GFP_KERNEL); - - if (core_stats->events == NULL) - goto fail_construct_events; - - } else { - core_stats->entries = 0; - } - - /* Purposely leave index 0 unused so we don't need special logic to - * handle calculation cases that depend on previous flip data. - */ - core_stats->index = 1; - core_stats->event_index = 0; - - // Keeps track of ordering within the different stats structures - core_stats->entry_id = 0; - - return &core_stats->public; - -fail_construct_events: - kfree(core_stats->time); - -fail_construct_time: - kfree(core_stats); - -fail_construct: - return NULL; -} - -void mod_stats_destroy(struct mod_stats *mod_stats) -{ - if (mod_stats != NULL) { - struct core_stats *core_stats = MOD_STATS_TO_CORE(mod_stats); - - kfree(core_stats->time); - kfree(core_stats->events); - kfree(core_stats); - } -} - -void mod_stats_dump(struct mod_stats *mod_stats) -{ - struct dc *dc = NULL; - struct dal_logger *logger = NULL; - struct core_stats *core_stats = NULL; - struct stats_time_cache *time = NULL; - struct stats_event_cache *events = NULL; - unsigned int time_index = 1; - unsigned int event_index = 0; - unsigned int index = 0; - struct log_entry log_entry; - - if (mod_stats == NULL) - return; - - core_stats = MOD_STATS_TO_CORE(mod_stats); - dc = core_stats->dc; - logger = dc->ctx->logger; - time = core_stats->time; - events = core_stats->events; - - DISPLAY_STATS_BEGIN(log_entry); - - DISPLAY_STATS("==Display Caps==\n"); - - DISPLAY_STATS("==Display Stats==\n"); - - DISPLAY_STATS("%10s %10s %10s %10s %10s" - " %11s %11s %17s %10s %14s" - " %10s %10s %10s %10s %10s" - " %10s %10s %10s %10s\n", - "render", "avgRender", - "minWindow", "midPoint", "maxWindow", - "vsyncToFlip", "flipToVsync", "vsyncsBetweenFlip", - "numFrame", "insertDuration", - "vTotalMin", "vTotalMax", "eventTrigs", - "vSyncTime1", "vSyncTime2", "vSyncTime3", - "vSyncTime4", "vSyncTime5", "flags"); - - for (int i = 0; i < core_stats->entry_id; i++) { - if (event_index < core_stats->event_index && - i == events[event_index].entry_id) { - DISPLAY_STATS("==Event==%s\n", events[event_index].event_string); - event_index++; - } else if (time_index < core_stats->index && - i == time[time_index].entry_id) { - DISPLAY_STATS("%10u %10u %10u %10u %10u" - " %11u %11u %17u %10u %14u" - " %10u %10u %10u %10u %10u" - " %10u %10u %10u %10u\n", - time[time_index].render_time_in_us, - time[time_index].avg_render_time_in_us_last_ten, - time[time_index].min_window, - time[time_index].lfc_mid_point_in_us, - time[time_index].max_window, - time[time_index].vsync_to_flip_time_in_us, - time[time_index].flip_to_vsync_time_in_us, - time[time_index].num_vsync_between_flips, - time[time_index].num_frames_inserted, - time[time_index].inserted_duration_in_us, - time[time_index].v_total_min, - time[time_index].v_total_max, - time[time_index].event_triggers, - time[time_index].v_sync_time_in_us[0], - time[time_index].v_sync_time_in_us[1], - time[time_index].v_sync_time_in_us[2], - time[time_index].v_sync_time_in_us[3], - time[time_index].v_sync_time_in_us[4], - time[time_index].flags); - - time_index++; - } - } - - DISPLAY_STATS_END(log_entry); -} - -void mod_stats_reset_data(struct mod_stats *mod_stats) -{ - struct core_stats *core_stats = NULL; - struct stats_time_cache *time = NULL; - unsigned int index = 0; - - if (mod_stats == NULL) - return; - - core_stats = MOD_STATS_TO_CORE(mod_stats); - - memset(core_stats->time, 0, - sizeof(struct stats_time_cache) * core_stats->entries); - - memset(core_stats->events, 0, - sizeof(struct stats_event_cache) * core_stats->event_entries); - - core_stats->index = 1; - core_stats->event_index = 0; - - // Keeps track of ordering within the different stats structures - core_stats->entry_id = 0; -} - -void mod_stats_update_event(struct mod_stats *mod_stats, - char *event_string, - unsigned int length) -{ - struct core_stats *core_stats = NULL; - struct stats_event_cache *events = NULL; - unsigned int index = 0; - unsigned int copy_length = 0; - - if (mod_stats == NULL) - return; - - core_stats = MOD_STATS_TO_CORE(mod_stats); - - if (core_stats->event_index >= core_stats->event_entries) - return; - - events = core_stats->events; - index = core_stats->event_index; - - copy_length = length; - if (length > MOD_STATS_EVENT_STRING_MAX) - copy_length = MOD_STATS_EVENT_STRING_MAX; - - memcpy(&events[index].event_string, event_string, copy_length); - events[index].event_string[copy_length - 1] = '\0'; - - events[index].entry_id = core_stats->entry_id; - core_stats->event_index++; - core_stats->entry_id++; -} - -void mod_stats_update_flip(struct mod_stats *mod_stats, - unsigned long timestamp_in_ns) -{ - struct core_stats *core_stats = NULL; - struct stats_time_cache *time = NULL; - unsigned int index = 0; - - if (mod_stats == NULL) - return; - - core_stats = MOD_STATS_TO_CORE(mod_stats); - - if (core_stats->index >= core_stats->entries) - return; - - time = core_stats->time; - index = core_stats->index; - - time[index].flip_timestamp_in_ns = timestamp_in_ns; - time[index].render_time_in_us = - (timestamp_in_ns - time[index - 1].flip_timestamp_in_ns) / 1000; - - if (index >= 10) { - for (unsigned int i = 0; i < 10; i++) - time[index].avg_render_time_in_us_last_ten += - time[index - i].render_time_in_us; - time[index].avg_render_time_in_us_last_ten /= 10; - } - - if (time[index].num_vsync_between_flips > 0) - time[index].vsync_to_flip_time_in_us = - (timestamp_in_ns - - time[index].vupdate_timestamp_in_ns) / 1000; - else - time[index].vsync_to_flip_time_in_us = - (timestamp_in_ns - - time[index - 1].vupdate_timestamp_in_ns) / 1000; - - time[index].entry_id = core_stats->entry_id; - core_stats->index++; - core_stats->entry_id++; -} - -void mod_stats_update_vupdate(struct mod_stats *mod_stats, - unsigned long timestamp_in_ns) -{ - struct core_stats *core_stats = NULL; - struct stats_time_cache *time = NULL; - unsigned int index = 0; - unsigned int num_vsyncs = 0; - unsigned int prev_vsync_in_ns = 0; - - if (mod_stats == NULL) - return; - - core_stats = MOD_STATS_TO_CORE(mod_stats); - - if (core_stats->index >= core_stats->entries) - return; - - time = core_stats->time; - index = core_stats->index; - num_vsyncs = time[index].num_vsync_between_flips; - - if (num_vsyncs < MOD_STATS_NUM_VSYNCS) { - if (num_vsyncs == 0) { - prev_vsync_in_ns = - time[index - 1].vupdate_timestamp_in_ns; - - time[index].flip_to_vsync_time_in_us = - (timestamp_in_ns - - time[index - 1].flip_timestamp_in_ns) / - 1000; - } else { - prev_vsync_in_ns = - time[index].vupdate_timestamp_in_ns; - } - - time[index].v_sync_time_in_us[num_vsyncs] = - (timestamp_in_ns - prev_vsync_in_ns) / 1000; - } - - time[index].vupdate_timestamp_in_ns = timestamp_in_ns; - time[index].num_vsync_between_flips++; -} - -void mod_stats_update_freesync(struct mod_stats *mod_stats, - unsigned int v_total_min, - unsigned int v_total_max, - unsigned int event_triggers, - unsigned int window_min, - unsigned int window_max, - unsigned int lfc_mid_point_in_us, - unsigned int inserted_frames, - unsigned int inserted_duration_in_us) -{ - struct core_stats *core_stats = NULL; - struct stats_time_cache *time = NULL; - unsigned int index = 0; - - if (mod_stats == NULL) - return; - - core_stats = MOD_STATS_TO_CORE(mod_stats); - - if (core_stats->index >= core_stats->entries) - return; - - time = core_stats->time; - index = core_stats->index; - - time[index].v_total_min = v_total_min; - time[index].v_total_max = v_total_max; - time[index].event_triggers = event_triggers; - time[index].min_window = window_min; - time[index].max_window = window_max; - time[index].lfc_mid_point_in_us = lfc_mid_point_in_us; - time[index].num_frames_inserted = inserted_frames; - time[index].inserted_duration_in_us = inserted_duration_in_us; -} - diff --git a/drivers/gpu/drm/amd/display/modules/vmid/vmid.c b/drivers/gpu/drm/amd/display/modules/vmid/vmid.c index 00f132f8ad55..61ee4be35d27 100644 --- a/drivers/gpu/drm/amd/display/modules/vmid/vmid.c +++ b/drivers/gpu/drm/amd/display/modules/vmid/vmid.c @@ -112,9 +112,12 @@ uint8_t mod_vmid_get_for_ptb(struct mod_vmid *mod_vmid, uint64_t ptb) evict_vmids(core_vmid); vmid = get_next_available_vmid(core_vmid); - add_ptb_to_table(core_vmid, vmid, ptb); + if (vmid != -1) { + add_ptb_to_table(core_vmid, vmid, ptb); - dc_setup_vm_context(core_vmid->dc, &va_config, vmid); + dc_setup_vm_context(core_vmid->dc, &va_config, vmid); + } else + ASSERT(0); } return vmid; diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h index d655a76bedc6..e98c84ef206f 100644 --- a/drivers/gpu/drm/amd/include/amd_shared.h +++ b/drivers/gpu/drm/amd/include/amd_shared.h @@ -40,6 +40,13 @@ enum amd_chip_flags { AMD_EXP_HW_SUPPORT = 0x00080000UL, }; +enum amd_apu_flags { + AMD_APU_IS_RAVEN = 0x00000001UL, + AMD_APU_IS_RAVEN2 = 0x00000002UL, + AMD_APU_IS_PICASSO = 0x00000004UL, + AMD_APU_IS_RENOIR = 0x00000008UL, +}; + enum amd_ip_block_type { AMD_IP_BLOCK_TYPE_COMMON, AMD_IP_BLOCK_TYPE_GMC, @@ -150,6 +157,13 @@ enum DC_FEATURE_MASK { DC_PSR_MASK = 0x8, }; +enum DC_DEBUG_MASK { + DC_DISABLE_PIPE_SPLIT = 0x1, + DC_DISABLE_STUTTER = 0x2, + DC_DISABLE_DSC = 0x4, + DC_DISABLE_CLOCK_GATING = 0x8 +}; + enum amd_dpm_forced_level; /** * struct amd_ip_funcs - general hooks for managing amdgpu IP Blocks diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_sh_mask.h index e7db6f9f9c86..8b0b9a2a8fed 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_sh_mask.h +++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_sh_mask.h @@ -5599,6 +5599,7 @@ #define GRBM_PWR_CNTL__ALL_REQ_EN_MASK 0x00008000L //GRBM_STATUS #define GRBM_STATUS__ME0PIPE0_CMDFIFO_AVAIL__SHIFT 0x0 +#define GRBM_STATUS__RSMU_RQ_PENDING__SHIFT 0x5 #define GRBM_STATUS__ME0PIPE0_CF_RQ_PENDING__SHIFT 0x7 #define GRBM_STATUS__ME0PIPE0_PF_RQ_PENDING__SHIFT 0x8 #define GRBM_STATUS__GDS_DMA_RQ_PENDING__SHIFT 0x9 @@ -5619,6 +5620,7 @@ #define GRBM_STATUS__CB_BUSY__SHIFT 0x1e #define GRBM_STATUS__GUI_ACTIVE__SHIFT 0x1f #define GRBM_STATUS__ME0PIPE0_CMDFIFO_AVAIL_MASK 0x0000000FL +#define GRBM_STATUS__RSMU_RQ_PENDING_MASK 0x00000020L #define GRBM_STATUS__ME0PIPE0_CF_RQ_PENDING_MASK 0x00000080L #define GRBM_STATUS__ME0PIPE0_PF_RQ_PENDING_MASK 0x00000100L #define GRBM_STATUS__GDS_DMA_RQ_PENDING_MASK 0x00000200L @@ -5832,6 +5834,7 @@ #define GRBM_READ_ERROR__READ_ERROR_MASK 0x80000000L //GRBM_READ_ERROR2 #define GRBM_READ_ERROR2__READ_REQUESTER_CPF__SHIFT 0x10 +#define GRBM_READ_ERROR2__READ_REQUESTER_RSMU__SHIFT 0x11 #define GRBM_READ_ERROR2__READ_REQUESTER_RLC__SHIFT 0x12 #define GRBM_READ_ERROR2__READ_REQUESTER_GDS_DMA__SHIFT 0x13 #define GRBM_READ_ERROR2__READ_REQUESTER_ME0PIPE0_CF__SHIFT 0x14 @@ -5847,6 +5850,7 @@ #define GRBM_READ_ERROR2__READ_REQUESTER_ME2PIPE2__SHIFT 0x1e #define GRBM_READ_ERROR2__READ_REQUESTER_ME2PIPE3__SHIFT 0x1f #define GRBM_READ_ERROR2__READ_REQUESTER_CPF_MASK 0x00010000L +#define GRBM_READ_ERROR2__READ_REQUESTER_RSMU_MASK 0x00020000L #define GRBM_READ_ERROR2__READ_REQUESTER_RLC_MASK 0x00040000L #define GRBM_READ_ERROR2__READ_REQUESTER_GDS_DMA_MASK 0x00080000L #define GRBM_READ_ERROR2__READ_REQUESTER_ME0PIPE0_CF_MASK 0x00100000L diff --git a/drivers/gpu/drm/amd/include/asic_reg/nbif/nbif_6_1_offset.h b/drivers/gpu/drm/amd/include/asic_reg/nbif/nbif_6_1_offset.h index 68d0ffad28c7..92fd27c26a77 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/nbif/nbif_6_1_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/nbif/nbif_6_1_offset.h @@ -1162,8 +1162,10 @@ #define mmRCC_CONFIG_MEMSIZE_BASE_IDX 0 #define mmRCC_CONFIG_RESERVED 0x0de4 // duplicate #define mmRCC_CONFIG_RESERVED_BASE_IDX 0 +#ifndef mmRCC_IOV_FUNC_IDENTIFIER #define mmRCC_IOV_FUNC_IDENTIFIER 0x0de5 // duplicate #define mmRCC_IOV_FUNC_IDENTIFIER_BASE_IDX 0 +#endif // addressBlock: syshub_mmreg_ind_syshubdec diff --git a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_0_offset.h index 435462294fbc..a7cd760ebf8f 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_0_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_0_offset.h @@ -4251,8 +4251,10 @@ #define mmRCC_CONFIG_MEMSIZE_BASE_IDX 2 #define mmRCC_CONFIG_RESERVED 0x00c4 #define mmRCC_CONFIG_RESERVED_BASE_IDX 2 +#ifndef mmRCC_IOV_FUNC_IDENTIFIER #define mmRCC_IOV_FUNC_IDENTIFIER 0x00c5 #define mmRCC_IOV_FUNC_IDENTIFIER_BASE_IDX 2 +#endif // addressBlock: nbio_nbif0_rcc_dev0_BIFDEC1 diff --git a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_4_offset.h b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_4_offset.h index ce5830ebe095..0c5a08bc034a 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_4_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_4_offset.h @@ -2687,8 +2687,10 @@ #define mmRCC_CONFIG_MEMSIZE_BASE_IDX 2 #define mmRCC_CONFIG_RESERVED 0x00c4 #define mmRCC_CONFIG_RESERVED_BASE_IDX 2 +#ifndef mmRCC_IOV_FUNC_IDENTIFIER #define mmRCC_IOV_FUNC_IDENTIFIER 0x00c5 #define mmRCC_IOV_FUNC_IDENTIFIER_BASE_IDX 2 +#endif // addressBlock: nbio_nbif0_rcc_dev0_BIFDEC1 diff --git a/drivers/gpu/drm/amd/include/asic_reg/pwr/pwr_10_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/pwr/pwr_10_0_offset.h new file mode 100644 index 000000000000..e87c359ea1fe --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/pwr/pwr_10_0_offset.h @@ -0,0 +1,27 @@ +/* + * Copyright (C) 2020 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included + * in all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS + * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN + * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + */ +#ifndef _pwr_10_0_OFFSET_HEADER +#define _pwr_10_0_OFFSET_HEADER + +#define mmPWR_MISC_CNTL_STATUS 0x0183 +#define mmPWR_MISC_CNTL_STATUS_BASE_IDX 0 + +#endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/pwr/pwr_10_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/pwr/pwr_10_0_sh_mask.h new file mode 100644 index 000000000000..8a000c21651c --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/pwr/pwr_10_0_sh_mask.h @@ -0,0 +1,30 @@ +/* + * Copyright (C) 2020 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included + * in all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS + * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN + * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + */ +#ifndef _pwr_10_0_SH_MASK_HEADER +#define _pwr_10_0_SH_MASK_HEADER + +//PWR_MISC_CNTL_STATUS +#define PWR_MISC_CNTL_STATUS__PWR_GFX_RLC_CGPG_EN__SHIFT 0x0 +#define PWR_MISC_CNTL_STATUS__PWR_GFXOFF_STATUS__SHIFT 0x1 +#define PWR_MISC_CNTL_STATUS__PWR_GFX_RLC_CGPG_EN_MASK 0x00000001L +#define PWR_MISC_CNTL_STATUS__PWR_GFXOFF_STATUS_MASK 0x00000006L + +#endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_12_0_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_12_0_0_offset.h new file mode 100644 index 000000000000..9bf73284ad73 --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_12_0_0_offset.h @@ -0,0 +1,30 @@ +/* + * Copyright (C) 2020 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included + * in all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS + * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN + * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + */ +#ifndef _smuio_12_0_0_OFFSET_HEADER +#define _smuio_12_0_0_OFFSET_HEADER + +#define mmSMUIO_GFX_MISC_CNTL 0x00c8 +#define mmSMUIO_GFX_MISC_CNTL_BASE_IDX 0 + +#define mmPWR_MISC_CNTL_STATUS 0x0183 +#define mmPWR_MISC_CNTL_STATUS_BASE_IDX 1 + +#endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_12_0_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_12_0_0_sh_mask.h new file mode 100644 index 000000000000..26556fa3d054 --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_12_0_0_sh_mask.h @@ -0,0 +1,33 @@ +/* + * Copyright (C) 2020 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included + * in all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS + * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN + * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + */ +#ifndef _smuio_12_0_0_SH_MASK_HEADER +#define _smuio_12_0_0_SH_MASK_HEADER + +//SMUIO_GFX_MISC_CNTL +#define SMUIO_GFX_MISC_CNTL__PWR_GFXOFF_STATUS_MASK 0x00000006L +#define SMUIO_GFX_MISC_CNTL__PWR_GFXOFF_STATUS__SHIFT 0x1 +//PWR_MISC_CNTL_STATUS +#define PWR_MISC_CNTL_STATUS__PWR_GFX_RLC_CGPG_EN__SHIFT 0x0 +#define PWR_MISC_CNTL_STATUS__PWR_GFXOFF_STATUS__SHIFT 0x1 +#define PWR_MISC_CNTL_STATUS__PWR_GFX_RLC_CGPG_EN_MASK 0x00000001L +#define PWR_MISC_CNTL_STATUS__PWR_GFXOFF_STATUS_MASK 0x00000006L + +#endif diff --git a/drivers/gpu/drm/amd/include/atomfirmware.h b/drivers/gpu/drm/amd/include/atomfirmware.h index 70146518174c..b36ea8340afa 100644 --- a/drivers/gpu/drm/amd/include/atomfirmware.h +++ b/drivers/gpu/drm/amd/include/atomfirmware.h @@ -972,11 +972,13 @@ struct atom_ext_display_path }; //usCaps -enum ext_display_path_cap_def -{ - EXT_DISPLAY_PATH_CAPS__HBR2_DISABLE =0x0001, - EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN =0x0002, - EXT_DISPLAY_PATH_CAPS__EXT_CHIP_MASK =0x007C, +enum ext_display_path_cap_def { + EXT_DISPLAY_PATH_CAPS__HBR2_DISABLE = 0x0001, + EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN = 0x0002, + EXT_DISPLAY_PATH_CAPS__EXT_CHIP_MASK = 0x007C, + EXT_DISPLAY_PATH_CAPS__HDMI20_PI3EQX1204 = (0x01 << 2), //PI redriver chip + EXT_DISPLAY_PATH_CAPS__HDMI20_TISN65DP159RSBT = (0x02 << 2), //TI retimer chip + EXT_DISPLAY_PATH_CAPS__HDMI20_PARADE_PS175 = (0x03 << 2) //Parade DP->HDMI recoverter chip }; struct atom_external_display_connection_info @@ -1876,6 +1878,108 @@ struct atom_smc_dpm_info_v4_6 uint32_t boardreserved[10]; }; +struct atom_smc_dpm_info_v4_7 +{ + struct atom_common_table_header table_header; + // SECTION: BOARD PARAMETERS + // I2C Control + struct smudpm_i2c_controller_config_v2 I2cControllers[8]; + + // SVI2 Board Parameters + uint16_t MaxVoltageStepGfx; // In mV(Q2) Max voltage step that SMU will request. Multiple steps are taken if voltage change exceeds this value. + uint16_t MaxVoltageStepSoc; // In mV(Q2) Max voltage step that SMU will request. Multiple steps are taken if voltage change exceeds this value. + + uint8_t VddGfxVrMapping; // Use VR_MAPPING* bitfields + uint8_t VddSocVrMapping; // Use VR_MAPPING* bitfields + uint8_t VddMem0VrMapping; // Use VR_MAPPING* bitfields + uint8_t VddMem1VrMapping; // Use VR_MAPPING* bitfields + + uint8_t GfxUlvPhaseSheddingMask; // set this to 1 to set PSI0/1 to 1 in ULV mode + uint8_t SocUlvPhaseSheddingMask; // set this to 1 to set PSI0/1 to 1 in ULV mode + uint8_t ExternalSensorPresent; // External RDI connected to TMON (aka TEMP IN) + uint8_t Padding8_V; + + // Telemetry Settings + uint16_t GfxMaxCurrent; // in Amps + uint8_t GfxOffset; // in Amps + uint8_t Padding_TelemetryGfx; + uint16_t SocMaxCurrent; // in Amps + uint8_t SocOffset; // in Amps + uint8_t Padding_TelemetrySoc; + + uint16_t Mem0MaxCurrent; // in Amps + uint8_t Mem0Offset; // in Amps + uint8_t Padding_TelemetryMem0; + + uint16_t Mem1MaxCurrent; // in Amps + uint8_t Mem1Offset; // in Amps + uint8_t Padding_TelemetryMem1; + + // GPIO Settings + uint8_t AcDcGpio; // GPIO pin configured for AC/DC switching + uint8_t AcDcPolarity; // GPIO polarity for AC/DC switching + uint8_t VR0HotGpio; // GPIO pin configured for VR0 HOT event + uint8_t VR0HotPolarity; // GPIO polarity for VR0 HOT event + + uint8_t VR1HotGpio; // GPIO pin configured for VR1 HOT event + uint8_t VR1HotPolarity; // GPIO polarity for VR1 HOT event + uint8_t GthrGpio; // GPIO pin configured for GTHR Event + uint8_t GthrPolarity; // replace GPIO polarity for GTHR + + // LED Display Settings + uint8_t LedPin0; // GPIO number for LedPin[0] + uint8_t LedPin1; // GPIO number for LedPin[1] + uint8_t LedPin2; // GPIO number for LedPin[2] + uint8_t padding8_4; + + // GFXCLK PLL Spread Spectrum + uint8_t PllGfxclkSpreadEnabled; // on or off + uint8_t PllGfxclkSpreadPercent; // Q4.4 + uint16_t PllGfxclkSpreadFreq; // kHz + + // GFXCLK DFLL Spread Spectrum + uint8_t DfllGfxclkSpreadEnabled; // on or off + uint8_t DfllGfxclkSpreadPercent; // Q4.4 + uint16_t DfllGfxclkSpreadFreq; // kHz + + // UCLK Spread Spectrum + uint8_t UclkSpreadEnabled; // on or off + uint8_t UclkSpreadPercent; // Q4.4 + uint16_t UclkSpreadFreq; // kHz + + // SOCCLK Spread Spectrum + uint8_t SoclkSpreadEnabled; // on or off + uint8_t SocclkSpreadPercent; // Q4.4 + uint16_t SocclkSpreadFreq; // kHz + + // Total board power + uint16_t TotalBoardPower; //Only needed for TCP Estimated case, where TCP = TGP+Total Board Power + uint16_t BoardPadding; + + // Mvdd Svi2 Div Ratio Setting + uint32_t MvddRatio; // This is used for MVDD Vid workaround. It has 16 fractional bits (Q16.16) + + // GPIO pins for I2C communications with 2nd controller for Input Telemetry Sequence + uint8_t GpioI2cScl; // Serial Clock + uint8_t GpioI2cSda; // Serial Data + uint16_t GpioPadding; + + // Additional LED Display Settings + uint8_t LedPin3; // GPIO number for LedPin[3] - PCIE GEN Speed + uint8_t LedPin4; // GPIO number for LedPin[4] - PMFW Error Status + uint16_t LedEnableMask; + + // Power Limit Scalars + uint8_t PowerLimitScalar[4]; //[PPT_THROTTLER_COUNT] + + uint8_t MvddUlvPhaseSheddingMask; + uint8_t VddciUlvPhaseSheddingMask; + uint8_t Padding8_Psi1; + uint8_t Padding8_Psi2; + + uint32_t BoardReserved[5]; +}; + /* *************************************************************************** Data Table asic_profiling_info structure diff --git a/drivers/gpu/drm/amd/include/cgs_common.h b/drivers/gpu/drm/amd/include/cgs_common.h index a69deb3a2ac0..60a6536ff656 100644 --- a/drivers/gpu/drm/amd/include/cgs_common.h +++ b/drivers/gpu/drm/amd/include/cgs_common.h @@ -32,7 +32,6 @@ struct cgs_device; * enum cgs_ind_reg - Indirect register spaces */ enum cgs_ind_reg { - CGS_IND_REG__MMIO, CGS_IND_REG__PCIE, CGS_IND_REG__SMC, CGS_IND_REG__UVD_CTX, diff --git a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c index 2a12614a12c2..7e6dcdf7df73 100644 --- a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c +++ b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c @@ -50,6 +50,7 @@ static int amd_powerplay_create(struct amdgpu_device *adev) hwmgr->not_vf = !amdgpu_sriov_vf(adev); hwmgr->device = amdgpu_cgs_create_device(adev); mutex_init(&hwmgr->smu_lock); + mutex_init(&hwmgr->msg_lock); hwmgr->chip_family = adev->family; hwmgr->chip_id = adev->asic_type; hwmgr->feature_mask = adev->pm.pp_feature; @@ -64,6 +65,8 @@ static void amd_powerplay_destroy(struct amdgpu_device *adev) { struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle; + mutex_destroy(&hwmgr->msg_lock); + kfree(hwmgr->hardcode_pp_table); hwmgr->hardcode_pp_table = NULL; @@ -319,12 +322,12 @@ static void pp_dpm_en_umd_pstate(struct pp_hwmgr *hwmgr, if (*level & profile_mode_mask) { hwmgr->saved_dpm_level = hwmgr->dpm_level; hwmgr->en_umd_pstate = true; - amdgpu_device_ip_set_clockgating_state(hwmgr->adev, - AMD_IP_BLOCK_TYPE_GFX, - AMD_CG_STATE_UNGATE); amdgpu_device_ip_set_powergating_state(hwmgr->adev, AMD_IP_BLOCK_TYPE_GFX, AMD_PG_STATE_UNGATE); + amdgpu_device_ip_set_clockgating_state(hwmgr->adev, + AMD_IP_BLOCK_TYPE_GFX, + AMD_CG_STATE_UNGATE); } } else { /* exit umd pstate, restore level, enable gfx cg*/ @@ -1435,7 +1438,8 @@ static int pp_get_asic_baco_capability(void *handle, bool *cap) if (!hwmgr) return -EINVAL; - if (!hwmgr->pm_en || !hwmgr->hwmgr_func->get_asic_baco_capability) + if (!(hwmgr->not_vf && amdgpu_dpm) || + !hwmgr->hwmgr_func->get_asic_baco_capability) return 0; mutex_lock(&hwmgr->smu_lock); @@ -1452,8 +1456,7 @@ static int pp_get_asic_baco_state(void *handle, int *state) if (!hwmgr) return -EINVAL; - if (!(hwmgr->not_vf && amdgpu_dpm) || - !hwmgr->hwmgr_func->get_asic_baco_state) + if (!hwmgr->pm_en || !hwmgr->hwmgr_func->get_asic_baco_state) return 0; mutex_lock(&hwmgr->smu_lock); @@ -1470,7 +1473,8 @@ static int pp_set_asic_baco_state(void *handle, int state) if (!hwmgr) return -EINVAL; - if (!hwmgr->pm_en || !hwmgr->hwmgr_func->set_asic_baco_state) + if (!(hwmgr->not_vf && amdgpu_dpm) || + !hwmgr->hwmgr_func->set_asic_baco_state) return 0; mutex_lock(&hwmgr->smu_lock); diff --git a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c index e8b27fab6aa1..8c684a6e0156 100644 --- a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c +++ b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c @@ -62,6 +62,7 @@ const char *smu_get_feature_name(struct smu_context *smu, enum smu_feature_mask size_t smu_sys_get_pp_feature_mask(struct smu_context *smu, char *buf) { + struct amdgpu_device *adev = smu->adev; size_t size = 0; int ret = 0, i = 0; uint32_t feature_mask[2] = { 0 }; @@ -70,6 +71,9 @@ size_t smu_sys_get_pp_feature_mask(struct smu_context *smu, char *buf) uint32_t sort_feature[SMU_FEATURE_COUNT]; uint64_t hw_feature_count = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); ret = smu_feature_get_enabled_mask(smu, feature_mask, 2); @@ -110,9 +114,6 @@ static int smu_feature_update_enable_state(struct smu_context *smu, uint32_t feature_low = 0, feature_high = 0; int ret = 0; - if (!smu->pm_enabled) - return ret; - feature_low = (feature_mask >> 0 ) & 0xffffffff; feature_high = (feature_mask >> 32) & 0xffffffff; @@ -155,6 +156,10 @@ int smu_sys_set_pp_feature_mask(struct smu_context *smu, uint64_t new_mask) uint64_t feature_2_enabled = 0; uint64_t feature_2_disabled = 0; uint64_t feature_enables = 0; + struct amdgpu_device *adev = smu->adev; + + if (!adev->pm.dpm_enabled) + return -EINVAL; mutex_lock(&smu->mutex); @@ -191,16 +196,31 @@ int smu_get_smc_version(struct smu_context *smu, uint32_t *if_version, uint32_t if (!if_version && !smu_version) return -EINVAL; + if (smu->smc_fw_if_version && smu->smc_fw_version) + { + if (if_version) + *if_version = smu->smc_fw_if_version; + + if (smu_version) + *smu_version = smu->smc_fw_version; + + return 0; + } + if (if_version) { ret = smu_send_smc_msg(smu, SMU_MSG_GetDriverIfVersion, if_version); if (ret) return ret; + + smu->smc_fw_if_version = *if_version; } if (smu_version) { ret = smu_send_smc_msg(smu, SMU_MSG_GetSmuVersion, smu_version); if (ret) return ret; + + smu->smc_fw_version = *smu_version; } return ret; @@ -327,13 +347,13 @@ int smu_get_dpm_freq_by_index(struct smu_context *smu, enum smu_clk_type clk_typ param = (uint32_t)(((clk_id & 0xffff) << 16) | (level & 0xffff)); ret = smu_send_smc_msg_with_param(smu, SMU_MSG_GetDpmFreqByIndex, - param, ¶m); + param, value); if (ret) return ret; /* BIT31: 0 - Fine grained DPM, 1 - Dicrete DPM * now, we un-support it */ - *value = param & 0x7fffffff; + *value = *value & 0x7fffffff; return ret; } @@ -417,8 +437,12 @@ bool smu_clk_dpm_is_enabled(struct smu_context *smu, enum smu_clk_type clk_type) int smu_dpm_set_power_gate(struct smu_context *smu, uint32_t block_type, bool gate) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + switch (block_type) { case AMD_IP_BLOCK_TYPE_UVD: ret = smu_dpm_set_uvd_enable(smu, !gate); @@ -511,7 +535,6 @@ int smu_update_table(struct smu_context *smu, enum smu_table_id table_index, int int table_id = smu_table_get_index(smu, table_index); uint32_t table_size; int ret = 0; - if (!table_data || table_id >= SMU_TABLE_COUNT || table_id < 0) return -EINVAL; @@ -547,12 +570,10 @@ bool is_support_sw_smu(struct amdgpu_device *adev) if (adev->asic_type == CHIP_VEGA20) return (amdgpu_dpm == 2) ? true : false; else if (adev->asic_type >= CHIP_ARCTURUS) { - if (amdgpu_sriov_vf(adev)&& !amdgpu_sriov_is_pp_one_vf(adev)) - return false; - else + if (amdgpu_sriov_is_pp_one_vf(adev) || !amdgpu_sriov_vf(adev)) return true; - } else - return false; + } + return false; } bool is_support_sw_smu_xgmi(struct amdgpu_device *adev) @@ -569,8 +590,12 @@ bool is_support_sw_smu_xgmi(struct amdgpu_device *adev) int smu_sys_get_pp_table(struct smu_context *smu, void **table) { struct smu_table_context *smu_table = &smu->smu_table; + struct amdgpu_device *adev = smu->adev; uint32_t powerplay_table_size; + if (!adev->pm.dpm_enabled) + return -EINVAL; + if (!smu_table->power_play_table && !smu_table->hardcode_pptable) return -EINVAL; @@ -591,11 +616,13 @@ int smu_sys_get_pp_table(struct smu_context *smu, void **table) int smu_sys_set_pp_table(struct smu_context *smu, void *buf, size_t size) { struct smu_table_context *smu_table = &smu->smu_table; + struct amdgpu_device *adev = smu->adev; ATOM_COMMON_TABLE_HEADER *header = (ATOM_COMMON_TABLE_HEADER *)buf; int ret = 0; - if (!smu->pm_enabled) + if (!adev->pm.dpm_enabled) return -EINVAL; + if (header->usStructureSize != size) { pr_err("pp table size not matched !\n"); return -EIO; @@ -636,8 +663,6 @@ int smu_feature_init_dpm(struct smu_context *smu) int ret = 0; uint32_t allowed_feature_mask[SMU_FEATURE_MAX/32]; - if (!smu->pm_enabled) - return ret; mutex_lock(&feature->mutex); bitmap_zero(feature->allowed, SMU_FEATURE_MAX); mutex_unlock(&feature->mutex); @@ -665,7 +690,6 @@ int smu_feature_is_enabled(struct smu_context *smu, enum smu_feature_mask mask) if (smu->is_apu) return 1; - feature_id = smu_feature_get_index(smu, mask); if (feature_id < 0) return 0; @@ -932,13 +956,6 @@ static int smu_sw_init(void *handle) return ret; } - if (adev->smu.ppt_funcs->i2c_eeprom_init) { - ret = smu_i2c_eeprom_init(smu, &adev->pm.smu_i2c); - - if (ret) - return ret; - } - return 0; } @@ -948,9 +965,6 @@ static int smu_sw_fini(void *handle) struct smu_context *smu = &adev->smu; int ret; - if (adev->smu.ppt_funcs->i2c_eeprom_fini) - smu_i2c_eeprom_fini(smu, &adev->pm.smu_i2c); - kfree(smu->irq_source); smu->irq_source = NULL; @@ -1323,6 +1337,9 @@ static int smu_hw_init(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; struct smu_context *smu = &adev->smu; + if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) + return 0; + ret = smu_start_smc_engine(smu); if (ret) { pr_err("SMU is not ready yet!\n"); @@ -1336,9 +1353,6 @@ static int smu_hw_init(void *handle) smu_set_gfx_cgpg(&adev->smu, true); } - if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) - return 0; - if (!smu->pm_enabled) return 0; @@ -1366,10 +1380,11 @@ static int smu_hw_init(void *handle) if (ret) goto failed; - if (!smu->pm_enabled) - adev->pm.dpm_enabled = false; - else - adev->pm.dpm_enabled = true; /* TODO: will set dpm_enabled flag while VCN and DAL DPM is workable */ + ret = smu_i2c_eeprom_init(smu, &adev->pm.smu_i2c); + if (ret) + goto failed; + + adev->pm.dpm_enabled = true; pr_info("SMU is initialized successfully!\n"); @@ -1381,6 +1396,9 @@ failed: static int smu_stop_dpms(struct smu_context *smu) { + if (amdgpu_sriov_vf(smu->adev)) + return 0; + return smu_system_features_control(smu, false); } @@ -1403,6 +1421,10 @@ static int smu_hw_fini(void *handle) if (!smu->pm_enabled) return 0; + adev->pm.dpm_enabled = false; + + smu_i2c_eeprom_fini(smu, &adev->pm.smu_i2c); + if (!amdgpu_sriov_vf(adev)){ ret = smu_stop_thermal_control(smu); if (ret) { @@ -1476,7 +1498,7 @@ static int smu_disable_dpm(struct smu_context *smu) bool use_baco = !smu->is_apu && ((adev->in_gpu_reset && (amdgpu_asic_reset_method(adev) == AMD_RESET_METHOD_BACO)) || - (adev->in_runpm && amdgpu_asic_supports_baco(adev))); + ((adev->in_runpm || adev->in_hibernate) && amdgpu_asic_supports_baco(adev))); ret = smu_get_smc_version(smu, NULL, &smu_version); if (ret) { @@ -1542,6 +1564,10 @@ static int smu_suspend(void *handle) if (!smu->pm_enabled) return 0; + adev->pm.dpm_enabled = false; + + smu_i2c_eeprom_fini(smu, &adev->pm.smu_i2c); + if(!amdgpu_sriov_vf(adev)) { ret = smu_disable_dpm(smu); if (ret) @@ -1587,11 +1613,17 @@ static int smu_resume(void *handle) if (ret) goto failed; + ret = smu_i2c_eeprom_init(smu, &adev->pm.smu_i2c); + if (ret) + goto failed; + if (smu->is_apu) smu_set_gfx_cgpg(&adev->smu, true); smu->disable_uclk_switch = 0; + adev->pm.dpm_enabled = true; + pr_info("SMU is resumed successfully!\n"); return 0; @@ -1603,10 +1635,14 @@ failed: int smu_display_configuration_change(struct smu_context *smu, const struct amd_pp_display_configuration *display_config) { + struct amdgpu_device *adev = smu->adev; int index = 0; int num_of_active_display = 0; - if (!smu->pm_enabled || !is_support_sw_smu(smu->adev)) + if (!adev->pm.dpm_enabled) + return -EINVAL; + + if (!is_support_sw_smu(smu->adev)) return -EINVAL; if (!display_config) @@ -1668,12 +1704,16 @@ int smu_get_current_clocks(struct smu_context *smu, struct amd_pp_clock_info *clocks) { struct amd_pp_simple_clock_info simple_clocks = {0}; + struct amdgpu_device *adev = smu->adev; struct smu_clock_info hw_clocks; int ret = 0; if (!is_support_sw_smu(smu->adev)) return -EINVAL; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); smu_get_dal_power_level(smu, &simple_clocks); @@ -1736,7 +1776,7 @@ static int smu_enable_umd_pstate(void *handle, struct smu_context *smu = (struct smu_context*)(handle); struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm); - if (!smu->is_apu && (!smu->pm_enabled || !smu_dpm_ctx->dpm_context)) + if (!smu->is_apu && !smu_dpm_ctx->dpm_context) return -EINVAL; if (!(smu_dpm_ctx->dpm_level & profile_mode_mask)) { @@ -1744,12 +1784,12 @@ static int smu_enable_umd_pstate(void *handle, if (*level & profile_mode_mask) { smu_dpm_ctx->saved_dpm_level = smu_dpm_ctx->dpm_level; smu_dpm_ctx->enable_umd_pstate = true; - amdgpu_device_ip_set_clockgating_state(smu->adev, - AMD_IP_BLOCK_TYPE_GFX, - AMD_CG_STATE_UNGATE); amdgpu_device_ip_set_powergating_state(smu->adev, AMD_IP_BLOCK_TYPE_GFX, AMD_PG_STATE_UNGATE); + amdgpu_device_ip_set_clockgating_state(smu->adev, + AMD_IP_BLOCK_TYPE_GFX, + AMD_CG_STATE_UNGATE); } } else { /* exit umd pstate, restore level, enable gfx cg*/ @@ -1778,9 +1818,6 @@ int smu_adjust_power_state_dynamic(struct smu_context *smu, long workload; struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm); - if (!smu->pm_enabled) - return -EINVAL; - if (!skip_display_settings) { ret = smu_display_config_changed(smu); if (ret) { @@ -1831,8 +1868,12 @@ int smu_handle_task(struct smu_context *smu, enum amd_pp_task task_id, bool lock_needed) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + if (lock_needed) mutex_lock(&smu->mutex); @@ -1866,10 +1907,11 @@ int smu_switch_power_profile(struct smu_context *smu, bool en) { struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm); + struct amdgpu_device *adev = smu->adev; long workload; uint32_t index; - if (!smu->pm_enabled) + if (!adev->pm.dpm_enabled) return -EINVAL; if (!(type < PP_SMC_POWER_PROFILE_CUSTOM)) @@ -1900,8 +1942,12 @@ int smu_switch_power_profile(struct smu_context *smu, enum amd_dpm_forced_level smu_get_performance_level(struct smu_context *smu) { struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm); + struct amdgpu_device *adev = smu->adev; enum amd_dpm_forced_level level; + if (!adev->pm.dpm_enabled) + return -EINVAL; + if (!smu->is_apu && !smu_dpm_ctx->dpm_context) return -EINVAL; @@ -1915,8 +1961,12 @@ enum amd_dpm_forced_level smu_get_performance_level(struct smu_context *smu) int smu_force_performance_level(struct smu_context *smu, enum amd_dpm_forced_level level) { struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm); + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + if (!smu->is_apu && !smu_dpm_ctx->dpm_context) return -EINVAL; @@ -1939,8 +1989,12 @@ int smu_force_performance_level(struct smu_context *smu, enum amd_dpm_forced_lev int smu_set_display_count(struct smu_context *smu, uint32_t count) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); ret = smu_init_display_count(smu, count); mutex_unlock(&smu->mutex); @@ -1954,8 +2008,12 @@ int smu_force_clk_levels(struct smu_context *smu, bool lock_needed) { struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm); + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + if (smu_dpm_ctx->dpm_level != AMD_DPM_FORCED_LEVEL_MANUAL) { pr_debug("force clock level is for dpm manual mode only.\n"); return -EINVAL; @@ -1973,20 +2031,19 @@ int smu_force_clk_levels(struct smu_context *smu, return ret; } +/* + * On system suspending or resetting, the dpm_enabled + * flag will be cleared. So that those SMU services which + * are not supported will be gated. + * However, the mp1 state setting should still be granted + * even if the dpm_enabled cleared. + */ int smu_set_mp1_state(struct smu_context *smu, enum pp_mp1_state mp1_state) { uint16_t msg; int ret; - /* - * The SMC is not fully ready. That may be - * expected as the IP may be masked. - * So, just return without error. - */ - if (!smu->pm_enabled) - return 0; - mutex_lock(&smu->mutex); switch (mp1_state) { @@ -2023,15 +2080,11 @@ int smu_set_mp1_state(struct smu_context *smu, int smu_set_df_cstate(struct smu_context *smu, enum pp_df_cstate state) { + struct amdgpu_device *adev = smu->adev; int ret = 0; - /* - * The SMC is not fully ready. That may be - * expected as the IP may be masked. - * So, just return without error. - */ - if (!smu->pm_enabled) - return 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; if (!smu->ppt_funcs || !smu->ppt_funcs->set_df_cstate) return 0; @@ -2047,6 +2100,28 @@ int smu_set_df_cstate(struct smu_context *smu, return ret; } +int smu_allow_xgmi_power_down(struct smu_context *smu, bool en) +{ + struct amdgpu_device *adev = smu->adev; + int ret = 0; + + if (!adev->pm.dpm_enabled) + return -EINVAL; + + if (!smu->ppt_funcs || !smu->ppt_funcs->allow_xgmi_power_down) + return 0; + + mutex_lock(&smu->mutex); + + ret = smu->ppt_funcs->allow_xgmi_power_down(smu, en); + if (ret) + pr_err("[AllowXgmiPowerDown] failed!\n"); + + mutex_unlock(&smu->mutex); + + return ret; +} + int smu_write_watermarks_table(struct smu_context *smu) { void *watermarks_table = smu->smu_table.watermarks_table; @@ -2065,6 +2140,10 @@ int smu_set_watermarks_for_clock_ranges(struct smu_context *smu, struct dm_pp_wm_sets_with_clock_ranges_soc15 *clock_ranges) { void *table = smu->smu_table.watermarks_table; + struct amdgpu_device *adev = smu->adev; + + if (!adev->pm.dpm_enabled) + return -EINVAL; if (!table) return -EINVAL; @@ -2089,8 +2168,12 @@ int smu_set_watermarks_for_clock_ranges(struct smu_context *smu, int smu_set_ac_dc(struct smu_context *smu) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + /* controlled by firmware */ if (smu->dc_controlled_by_gpio) return 0; @@ -2149,8 +2232,12 @@ const struct amdgpu_ip_block_version smu_v12_0_ip_block = int smu_load_microcode(struct smu_context *smu) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->load_microcode) @@ -2163,8 +2250,12 @@ int smu_load_microcode(struct smu_context *smu) int smu_check_fw_status(struct smu_context *smu) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->check_fw_status) @@ -2191,8 +2282,12 @@ int smu_set_gfx_cgpg(struct smu_context *smu, bool enabled) int smu_set_fan_speed_rpm(struct smu_context *smu, uint32_t speed) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->set_fan_speed_rpm) @@ -2208,10 +2303,15 @@ int smu_get_power_limit(struct smu_context *smu, bool def, bool lock_needed) { + struct amdgpu_device *adev = smu->adev; int ret = 0; - if (lock_needed) + if (lock_needed) { + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); + } if (smu->ppt_funcs->get_power_limit) ret = smu->ppt_funcs->get_power_limit(smu, limit, def); @@ -2224,8 +2324,12 @@ int smu_get_power_limit(struct smu_context *smu, int smu_set_power_limit(struct smu_context *smu, uint32_t limit) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->set_power_limit) @@ -2238,8 +2342,12 @@ int smu_set_power_limit(struct smu_context *smu, uint32_t limit) int smu_print_clk_levels(struct smu_context *smu, enum smu_clk_type clk_type, char *buf) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->print_clk_levels) @@ -2252,8 +2360,12 @@ int smu_print_clk_levels(struct smu_context *smu, enum smu_clk_type clk_type, ch int smu_get_od_percentage(struct smu_context *smu, enum smu_clk_type type) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->get_od_percentage) @@ -2266,8 +2378,12 @@ int smu_get_od_percentage(struct smu_context *smu, enum smu_clk_type type) int smu_set_od_percentage(struct smu_context *smu, enum smu_clk_type type, uint32_t value) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->set_od_percentage) @@ -2282,8 +2398,12 @@ int smu_od_edit_dpm_table(struct smu_context *smu, enum PP_OD_DPM_TABLE_COMMAND type, long *input, uint32_t size) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->od_edit_dpm_table) @@ -2298,8 +2418,12 @@ int smu_read_sensor(struct smu_context *smu, enum amd_pp_sensors sensor, void *data, uint32_t *size) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->read_sensor) @@ -2312,8 +2436,12 @@ int smu_read_sensor(struct smu_context *smu, int smu_get_power_profile_mode(struct smu_context *smu, char *buf) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->get_power_profile_mode) @@ -2329,8 +2457,12 @@ int smu_set_power_profile_mode(struct smu_context *smu, uint32_t param_size, bool lock_needed) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + if (lock_needed) mutex_lock(&smu->mutex); @@ -2346,8 +2478,12 @@ int smu_set_power_profile_mode(struct smu_context *smu, int smu_get_fan_control_mode(struct smu_context *smu) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->get_fan_control_mode) @@ -2360,8 +2496,12 @@ int smu_get_fan_control_mode(struct smu_context *smu) int smu_set_fan_control_mode(struct smu_context *smu, int value) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->set_fan_control_mode) @@ -2374,8 +2514,12 @@ int smu_set_fan_control_mode(struct smu_context *smu, int value) int smu_get_fan_speed_percent(struct smu_context *smu, uint32_t *speed) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->get_fan_speed_percent) @@ -2388,8 +2532,12 @@ int smu_get_fan_speed_percent(struct smu_context *smu, uint32_t *speed) int smu_set_fan_speed_percent(struct smu_context *smu, uint32_t speed) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->set_fan_speed_percent) @@ -2402,8 +2550,12 @@ int smu_set_fan_speed_percent(struct smu_context *smu, uint32_t speed) int smu_get_fan_speed_rpm(struct smu_context *smu, uint32_t *speed) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->get_fan_speed_rpm) @@ -2416,8 +2568,12 @@ int smu_get_fan_speed_rpm(struct smu_context *smu, uint32_t *speed) int smu_set_deep_sleep_dcefclk(struct smu_context *smu, int clk) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->set_deep_sleep_dcefclk) @@ -2430,8 +2586,12 @@ int smu_set_deep_sleep_dcefclk(struct smu_context *smu, int clk) int smu_set_active_display_count(struct smu_context *smu, uint32_t count) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + if (smu->ppt_funcs->set_active_display_count) ret = smu->ppt_funcs->set_active_display_count(smu, count); @@ -2442,8 +2602,12 @@ int smu_get_clock_by_type(struct smu_context *smu, enum amd_pp_clock_type type, struct amd_pp_clocks *clocks) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->get_clock_by_type) @@ -2457,8 +2621,12 @@ int smu_get_clock_by_type(struct smu_context *smu, int smu_get_max_high_clocks(struct smu_context *smu, struct amd_pp_simple_clock_info *clocks) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->get_max_high_clocks) @@ -2473,8 +2641,12 @@ int smu_get_clock_by_type_with_latency(struct smu_context *smu, enum smu_clk_type clk_type, struct pp_clock_levels_with_latency *clocks) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->get_clock_by_type_with_latency) @@ -2489,8 +2661,12 @@ int smu_get_clock_by_type_with_voltage(struct smu_context *smu, enum amd_pp_clock_type type, struct pp_clock_levels_with_voltage *clocks) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->get_clock_by_type_with_voltage) @@ -2505,8 +2681,12 @@ int smu_get_clock_by_type_with_voltage(struct smu_context *smu, int smu_display_clock_voltage_request(struct smu_context *smu, struct pp_display_clock_request *clock_req) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->display_clock_voltage_request) @@ -2520,8 +2700,12 @@ int smu_display_clock_voltage_request(struct smu_context *smu, int smu_display_disable_memory_clock_switch(struct smu_context *smu, bool disable_memory_clock_switch) { + struct amdgpu_device *adev = smu->adev; int ret = -EINVAL; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->display_disable_memory_clock_switch) @@ -2534,8 +2718,12 @@ int smu_display_disable_memory_clock_switch(struct smu_context *smu, bool disabl int smu_notify_smu_enable_pwe(struct smu_context *smu) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->notify_smu_enable_pwe) @@ -2549,8 +2737,12 @@ int smu_notify_smu_enable_pwe(struct smu_context *smu) int smu_set_xgmi_pstate(struct smu_context *smu, uint32_t pstate) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->set_xgmi_pstate) @@ -2563,8 +2755,12 @@ int smu_set_xgmi_pstate(struct smu_context *smu, int smu_set_azalia_d3_pme(struct smu_context *smu) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->set_azalia_d3_pme) @@ -2575,6 +2771,14 @@ int smu_set_azalia_d3_pme(struct smu_context *smu) return ret; } +/* + * On system suspending or resetting, the dpm_enabled + * flag will be cleared. So that those SMU services which + * are not supported will be gated. + * + * However, the baco/mode1 reset should still be granted + * as they are still supported and necessary. + */ bool smu_baco_is_support(struct smu_context *smu) { bool ret = false; @@ -2646,8 +2850,12 @@ int smu_mode2_reset(struct smu_context *smu) int smu_get_max_sustainable_clocks_by_dc(struct smu_context *smu, struct pp_smu_nv_clock_table *max_clocks) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->get_max_sustainable_clocks_by_dc) @@ -2662,8 +2870,12 @@ int smu_get_uclk_dpm_states(struct smu_context *smu, unsigned int *clock_values_in_khz, unsigned int *num_states) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->get_uclk_dpm_states) @@ -2677,6 +2889,10 @@ int smu_get_uclk_dpm_states(struct smu_context *smu, enum amd_pm_state_type smu_get_current_power_state(struct smu_context *smu) { enum amd_pm_state_type pm_state = POWER_STATE_TYPE_DEFAULT; + struct amdgpu_device *adev = smu->adev; + + if (!adev->pm.dpm_enabled) + return -EINVAL; mutex_lock(&smu->mutex); @@ -2691,8 +2907,12 @@ enum amd_pm_state_type smu_get_current_power_state(struct smu_context *smu) int smu_get_dpm_clock_table(struct smu_context *smu, struct dpm_clocks *clock_table) { + struct amdgpu_device *adev = smu->adev; int ret = 0; + if (!adev->pm.dpm_enabled) + return -EINVAL; + mutex_lock(&smu->mutex); if (smu->ppt_funcs->get_dpm_clock_table) diff --git a/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c b/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c index 1ef0923f7190..27c5fc9572b2 100644 --- a/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c +++ b/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c @@ -128,6 +128,7 @@ static struct smu_11_0_cmn2aisc_mapping arcturus_message_map[SMU_MSG_MAX_COUNT] MSG_MAP(SetXgmiMode, PPSMC_MSG_SetXgmiMode), MSG_MAP(SetMemoryChannelEnable, PPSMC_MSG_SetMemoryChannelEnable), MSG_MAP(DFCstateControl, PPSMC_MSG_DFCstateControl), + MSG_MAP(GmiPwrDnControl, PPSMC_MSG_GmiPwrDnControl), }; static struct smu_11_0_cmn2aisc_mapping arcturus_clk_map[SMU_CLK_COUNT] = { @@ -622,6 +623,9 @@ static int arcturus_print_clk_levels(struct smu_context *smu, struct smu_dpm_context *smu_dpm = &smu->smu_dpm; struct arcturus_dpm_table *dpm_table = NULL; + if (amdgpu_ras_intr_triggered()) + return snprintf(buf, PAGE_SIZE, "unavailable\n"); + dpm_table = smu_dpm->dpm_context; switch (type) { @@ -997,6 +1001,9 @@ static int arcturus_read_sensor(struct smu_context *smu, PPTable_t *pptable = table_context->driver_pptable; int ret = 0; + if (amdgpu_ras_intr_triggered()) + return 0; + if (!data || !size) return -EINVAL; @@ -2226,12 +2233,8 @@ static const struct i2c_algorithm arcturus_i2c_eeprom_i2c_algo = { static int arcturus_i2c_eeprom_control_init(struct i2c_adapter *control) { struct amdgpu_device *adev = to_amdgpu_device(control); - struct smu_context *smu = &adev->smu; int res; - if (!smu->pm_enabled) - return -EOPNOTSUPP; - control->owner = THIS_MODULE; control->class = I2C_CLASS_SPD; control->dev.parent = &adev->pdev->dev; @@ -2247,12 +2250,6 @@ static int arcturus_i2c_eeprom_control_init(struct i2c_adapter *control) static void arcturus_i2c_eeprom_control_fini(struct i2c_adapter *control) { - struct amdgpu_device *adev = to_amdgpu_device(control); - struct smu_context *smu = &adev->smu; - - if (!smu->pm_enabled) - return; - i2c_del_adapter(control); } @@ -2261,7 +2258,7 @@ static bool arcturus_is_baco_supported(struct smu_context *smu) struct amdgpu_device *adev = smu->adev; uint32_t val; - if (!smu_v11_0_baco_is_support(smu)) + if (!smu_v11_0_baco_is_support(smu) || amdgpu_sriov_vf(adev)) return false; val = RREG32_SOC15(NBIO, 0, mmRCC_BIF_STRAP0); @@ -2296,6 +2293,35 @@ static int arcturus_set_df_cstate(struct smu_context *smu, return smu_send_smc_msg_with_param(smu, SMU_MSG_DFCstateControl, state, NULL); } +static int arcturus_allow_xgmi_power_down(struct smu_context *smu, bool en) +{ + uint32_t smu_version; + int ret; + + ret = smu_get_smc_version(smu, NULL, &smu_version); + if (ret) { + pr_err("Failed to get smu version!\n"); + return ret; + } + + /* PPSMC_MSG_GmiPwrDnControl is supported by 54.23.0 and onwards */ + if (smu_version < 0x00361700) { + pr_err("XGMI power down control is only supported by PMFW 54.23.0 and onwards\n"); + return -EINVAL; + } + + if (en) + return smu_send_smc_msg_with_param(smu, + SMU_MSG_GmiPwrDnControl, + 1, + NULL); + + return smu_send_smc_msg_with_param(smu, + SMU_MSG_GmiPwrDnControl, + 0, + NULL); +} + static const struct pptable_funcs arcturus_ppt_funcs = { /* translate smu index into arcturus specific index */ .get_smu_msg_index = arcturus_get_smu_msg_index, @@ -2389,6 +2415,7 @@ static const struct pptable_funcs arcturus_ppt_funcs = { .override_pcie_parameters = smu_v11_0_override_pcie_parameters, .get_pptable_power_limit = arcturus_get_pptable_power_limit, .set_df_cstate = arcturus_set_df_cstate, + .allow_xgmi_power_down = arcturus_allow_xgmi_power_down, }; void arcturus_set_ppt_funcs(struct smu_context *smu) diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/processpptables.c b/drivers/gpu/drm/amd/powerplay/hwmgr/processpptables.c index 77c14671866c..719597c5d27d 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/processpptables.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/processpptables.c @@ -984,6 +984,32 @@ static int init_thermal_controller( struct pp_hwmgr *hwmgr, const ATOM_PPLIB_POWERPLAYTABLE *powerplay_table) { + hwmgr->thermal_controller.ucType = + powerplay_table->sThermalController.ucType; + hwmgr->thermal_controller.ucI2cLine = + powerplay_table->sThermalController.ucI2cLine; + hwmgr->thermal_controller.ucI2cAddress = + powerplay_table->sThermalController.ucI2cAddress; + + hwmgr->thermal_controller.fanInfo.bNoFan = + (0 != (powerplay_table->sThermalController.ucFanParameters & + ATOM_PP_FANPARAMETERS_NOFAN)); + + hwmgr->thermal_controller.fanInfo.ucTachometerPulsesPerRevolution = + powerplay_table->sThermalController.ucFanParameters & + ATOM_PP_FANPARAMETERS_TACHOMETER_PULSES_PER_REVOLUTION_MASK; + + hwmgr->thermal_controller.fanInfo.ulMinRPM + = powerplay_table->sThermalController.ucFanMinRPM * 100UL; + hwmgr->thermal_controller.fanInfo.ulMaxRPM + = powerplay_table->sThermalController.ucFanMaxRPM * 100UL; + + set_hw_cap(hwmgr, + ATOM_PP_THERMALCONTROLLER_NONE != hwmgr->thermal_controller.ucType, + PHM_PlatformCaps_ThermalController); + + hwmgr->thermal_controller.use_hw_fan_control = 1; + return 0; } diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c index 689072a312a7..c9cfe90a2947 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c @@ -36,6 +36,8 @@ #include "power_state.h" #include "soc15_common.h" #include "smu10.h" +#include "asic_reg/pwr/pwr_10_0_offset.h" +#include "asic_reg/pwr/pwr_10_0_sh_mask.h" #define SMU10_MAX_DEEPSLEEP_DIVIDER_ID 5 #define SMU10_MINIMUM_ENGINE_CLOCK 800 /* 8Mhz, the low boundary of engine clock allowed on this chip */ @@ -43,13 +45,6 @@ #define SMU10_DISPCLK_BYPASS_THRESHOLD 10000 /* 100Mhz */ #define SMC_RAM_END 0x40000 -#define mmPWR_MISC_CNTL_STATUS 0x0183 -#define mmPWR_MISC_CNTL_STATUS_BASE_IDX 0 -#define PWR_MISC_CNTL_STATUS__PWR_GFX_RLC_CGPG_EN__SHIFT 0x0 -#define PWR_MISC_CNTL_STATUS__PWR_GFXOFF_STATUS__SHIFT 0x1 -#define PWR_MISC_CNTL_STATUS__PWR_GFX_RLC_CGPG_EN_MASK 0x00000001L -#define PWR_MISC_CNTL_STATUS__PWR_GFXOFF_STATUS_MASK 0x00000006L - static const unsigned long SMU10_Magic = (unsigned long) PHM_Rv_Magic; @@ -81,7 +76,7 @@ static int smu10_display_clock_voltage_request(struct pp_hwmgr *hwmgr, pr_info("[DisplayClockVoltageRequest]Invalid Clock Type!"); return -EINVAL; } - smum_send_msg_to_smc_with_parameter(hwmgr, msg, clk_freq); + smum_send_msg_to_smc_with_parameter(hwmgr, msg, clk_freq, NULL); return 0; } @@ -214,7 +209,8 @@ static int smu10_set_min_deep_sleep_dcefclk(struct pp_hwmgr *hwmgr, uint32_t clo smu10_data->deep_sleep_dcefclk = clock; smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetMinDeepSleepDcefclk, - smu10_data->deep_sleep_dcefclk); + smu10_data->deep_sleep_dcefclk, + NULL); } return 0; } @@ -228,7 +224,8 @@ static int smu10_set_hard_min_dcefclk_by_freq(struct pp_hwmgr *hwmgr, uint32_t c smu10_data->dcf_actual_hard_min_freq = clock; smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinDcefclkByFreq, - smu10_data->dcf_actual_hard_min_freq); + smu10_data->dcf_actual_hard_min_freq, + NULL); } return 0; } @@ -242,7 +239,8 @@ static int smu10_set_hard_min_fclk_by_freq(struct pp_hwmgr *hwmgr, uint32_t cloc smu10_data->f_actual_hard_min_freq = clock; smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinFclkByFreq, - smu10_data->f_actual_hard_min_freq); + smu10_data->f_actual_hard_min_freq, + NULL); } return 0; } @@ -255,7 +253,8 @@ static int smu10_set_active_display_count(struct pp_hwmgr *hwmgr, uint32_t count smu10_data->num_active_display = count; smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDisplayCount, - smu10_data->num_active_display); + smu10_data->num_active_display, + NULL); } return 0; @@ -278,7 +277,8 @@ static int smu10_init_power_gate_state(struct pp_hwmgr *hwmgr) if (adev->pg_flags & AMD_PG_SUPPORT_GFX_PG) return smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetGfxCGPG, - true); + true, + NULL); else return 0; } @@ -324,7 +324,7 @@ static int smu10_disable_gfx_off(struct pp_hwmgr *hwmgr) struct amdgpu_device *adev = hwmgr->adev; if (adev->pm.pp_feature & PP_GFXOFF_MASK) { - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_DisableGfxOff); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_DisableGfxOff, NULL); /* confirm gfx is back to "on" state */ while (!smu10_is_gfx_on(hwmgr)) @@ -344,7 +344,7 @@ static int smu10_enable_gfx_off(struct pp_hwmgr *hwmgr) struct amdgpu_device *adev = hwmgr->adev; if (adev->pm.pp_feature & PP_GFXOFF_MASK) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnableGfxOff); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnableGfxOff, NULL); return 0; } @@ -410,12 +410,10 @@ static int smu10_get_clock_voltage_dependency_table(struct pp_hwmgr *hwmgr, struct smu10_voltage_dependency_table **pptable, uint32_t num_entry, const DpmClock_t *pclk_dependency_table) { - uint32_t table_size, i; + uint32_t i; struct smu10_voltage_dependency_table *ptable; - table_size = sizeof(uint32_t) + sizeof(struct smu10_voltage_dependency_table) * num_entry; - ptable = kzalloc(table_size, GFP_KERNEL); - + ptable = kzalloc(struct_size(ptable, entries, num_entry), GFP_KERNEL); if (NULL == ptable) return -ENOMEM; @@ -479,12 +477,10 @@ static int smu10_populate_clock_table(struct pp_hwmgr *hwmgr) smu10_get_clock_voltage_dependency_table(hwmgr, &pinfo->vdd_dep_on_phyclk, ARRAY_SIZE(VddPhyClk), &VddPhyClk[0]); - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetMinGfxclkFrequency); - result = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetMinGfxclkFrequency, &result); smu10_data->gfx_min_freq_limit = result / 10 * 1000; - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetMaxGfxclkFrequency); - result = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetMaxGfxclkFrequency, &result); smu10_data->gfx_max_freq_limit = result / 10 * 1000; return 0; @@ -588,116 +584,148 @@ static int smu10_dpm_force_dpm_level(struct pp_hwmgr *hwmgr, case AMD_DPM_FORCED_LEVEL_PROFILE_PEAK: smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinGfxClk, - data->gfx_max_freq_limit/100); + data->gfx_max_freq_limit/100, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinFclkByFreq, - SMU10_UMD_PSTATE_PEAK_FCLK); + SMU10_UMD_PSTATE_PEAK_FCLK, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinSocclkByFreq, - SMU10_UMD_PSTATE_PEAK_SOCCLK); + SMU10_UMD_PSTATE_PEAK_SOCCLK, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinVcn, - SMU10_UMD_PSTATE_VCE); + SMU10_UMD_PSTATE_VCE, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxGfxClk, - data->gfx_max_freq_limit/100); + data->gfx_max_freq_limit/100, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxFclkByFreq, - SMU10_UMD_PSTATE_PEAK_FCLK); + SMU10_UMD_PSTATE_PEAK_FCLK, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxSocclkByFreq, - SMU10_UMD_PSTATE_PEAK_SOCCLK); + SMU10_UMD_PSTATE_PEAK_SOCCLK, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxVcn, - SMU10_UMD_PSTATE_VCE); + SMU10_UMD_PSTATE_VCE, + NULL); break; case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK: smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinGfxClk, - min_sclk); + min_sclk, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxGfxClk, - min_sclk); + min_sclk, + NULL); break; case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK: smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinFclkByFreq, - min_mclk); + min_mclk, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxFclkByFreq, - min_mclk); + min_mclk, + NULL); break; case AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD: smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinGfxClk, - SMU10_UMD_PSTATE_GFXCLK); + SMU10_UMD_PSTATE_GFXCLK, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinFclkByFreq, - SMU10_UMD_PSTATE_FCLK); + SMU10_UMD_PSTATE_FCLK, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinSocclkByFreq, - SMU10_UMD_PSTATE_SOCCLK); + SMU10_UMD_PSTATE_SOCCLK, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinVcn, - SMU10_UMD_PSTATE_VCE); + SMU10_UMD_PSTATE_VCE, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxGfxClk, - SMU10_UMD_PSTATE_GFXCLK); + SMU10_UMD_PSTATE_GFXCLK, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxFclkByFreq, - SMU10_UMD_PSTATE_FCLK); + SMU10_UMD_PSTATE_FCLK, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxSocclkByFreq, - SMU10_UMD_PSTATE_SOCCLK); + SMU10_UMD_PSTATE_SOCCLK, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxVcn, - SMU10_UMD_PSTATE_VCE); + SMU10_UMD_PSTATE_VCE, + NULL); break; case AMD_DPM_FORCED_LEVEL_AUTO: smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinGfxClk, - min_sclk); + min_sclk, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinFclkByFreq, hwmgr->display_config->num_display > 3 ? SMU10_UMD_PSTATE_PEAK_FCLK : - min_mclk); + min_mclk, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinSocclkByFreq, - SMU10_UMD_PSTATE_MIN_SOCCLK); + SMU10_UMD_PSTATE_MIN_SOCCLK, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinVcn, - SMU10_UMD_PSTATE_MIN_VCE); + SMU10_UMD_PSTATE_MIN_VCE, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxGfxClk, - data->gfx_max_freq_limit/100); + data->gfx_max_freq_limit/100, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxFclkByFreq, - SMU10_UMD_PSTATE_PEAK_FCLK); + SMU10_UMD_PSTATE_PEAK_FCLK, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxSocclkByFreq, - SMU10_UMD_PSTATE_PEAK_SOCCLK); + SMU10_UMD_PSTATE_PEAK_SOCCLK, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxVcn, - SMU10_UMD_PSTATE_VCE); + SMU10_UMD_PSTATE_VCE, + NULL); break; case AMD_DPM_FORCED_LEVEL_LOW: smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinGfxClk, - data->gfx_min_freq_limit/100); + data->gfx_min_freq_limit/100, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxGfxClk, - data->gfx_min_freq_limit/100); + data->gfx_min_freq_limit/100, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinFclkByFreq, - min_mclk); + min_mclk, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxFclkByFreq, - min_mclk); + min_mclk, + NULL); break; case AMD_DPM_FORCED_LEVEL_MANUAL: case AMD_DPM_FORCED_LEVEL_PROFILE_EXIT: @@ -849,13 +877,15 @@ static int smu10_force_clock_level(struct pp_hwmgr *hwmgr, PPSMC_MSG_SetHardMinGfxClk, low == 2 ? data->gfx_max_freq_limit/100 : low == 1 ? SMU10_UMD_PSTATE_GFXCLK : - data->gfx_min_freq_limit/100); + data->gfx_min_freq_limit/100, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxGfxClk, high == 0 ? data->gfx_min_freq_limit/100 : high == 1 ? SMU10_UMD_PSTATE_GFXCLK : - data->gfx_max_freq_limit/100); + data->gfx_max_freq_limit/100, + NULL); break; case PP_MCLK: @@ -864,11 +894,13 @@ static int smu10_force_clock_level(struct pp_hwmgr *hwmgr, smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinFclkByFreq, - mclk_table->entries[low].clk/100); + mclk_table->entries[low].clk/100, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxFclkByFreq, - mclk_table->entries[high].clk/100); + mclk_table->entries[high].clk/100, + NULL); break; case PP_PCIE: @@ -888,8 +920,7 @@ static int smu10_print_clock_levels(struct pp_hwmgr *hwmgr, switch (type) { case PP_SCLK: - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetGfxclkFrequency); - now = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetGfxclkFrequency, &now); /* driver only know min/max gfx_clk, Add level 1 for all other gfx clks */ if (now == data->gfx_max_freq_limit/100) @@ -910,8 +941,7 @@ static int smu10_print_clock_levels(struct pp_hwmgr *hwmgr, i == 2 ? "*" : ""); break; case PP_MCLK: - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetFclkFrequency); - now = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetFclkFrequency, &now); for (i = 0; i < mclk_table->count; i++) size += sprintf(buf + size, "%d: %uMhz %s\n", @@ -1122,15 +1152,13 @@ static int smu10_read_sensor(struct pp_hwmgr *hwmgr, int idx, switch (idx) { case AMDGPU_PP_SENSOR_GFX_SCLK: - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetGfxclkFrequency); - sclk = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetGfxclkFrequency, &sclk); /* in units of 10KHZ */ *((uint32_t *)value) = sclk * 100; *size = 4; break; case AMDGPU_PP_SENSOR_GFX_MCLK: - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetFclkFrequency); - mclk = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetFclkFrequency, &mclk); /* in units of 10KHZ */ *((uint32_t *)value) = mclk * 100; *size = 4; @@ -1166,20 +1194,20 @@ static int smu10_set_watermarks_for_clocks_ranges(struct pp_hwmgr *hwmgr, static int smu10_smus_notify_pwe(struct pp_hwmgr *hwmgr) { - return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_SetRccPfcPmeRestoreRegister); + return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_SetRccPfcPmeRestoreRegister, NULL); } static int smu10_powergate_mmhub(struct pp_hwmgr *hwmgr) { - return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_PowerGateMmHub); + return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_PowerGateMmHub, NULL); } static int smu10_powergate_sdma(struct pp_hwmgr *hwmgr, bool gate) { if (gate) - return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_PowerDownSdma); + return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_PowerDownSdma, NULL); else - return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_PowerUpSdma); + return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_PowerUpSdma, NULL); } static void smu10_powergate_vcn(struct pp_hwmgr *hwmgr, bool bgate) @@ -1191,11 +1219,11 @@ static void smu10_powergate_vcn(struct pp_hwmgr *hwmgr, bool bgate) AMD_IP_BLOCK_TYPE_VCN, AMD_PG_STATE_GATE); smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_PowerDownVcn, 0); + PPSMC_MSG_PowerDownVcn, 0, NULL); smu10_data->vcn_power_gated = true; } else { smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_PowerUpVcn, 0); + PPSMC_MSG_PowerUpVcn, 0, NULL); amdgpu_device_ip_set_powergating_state(hwmgr->adev, AMD_IP_BLOCK_TYPE_VCN, AMD_PG_STATE_UNGATE); @@ -1274,8 +1302,7 @@ static int smu10_get_power_profile_mode(struct pp_hwmgr *hwmgr, char *buf) static bool smu10_is_raven1_refresh(struct pp_hwmgr *hwmgr) { struct amdgpu_device *adev = hwmgr->adev; - if ((adev->asic_type == CHIP_RAVEN) && - (adev->rev_id != 0x15d8) && + if ((adev->apu_flags & AMD_APU_IS_RAVEN) && (hwmgr->smu_version >= 0x41e2b)) return true; else @@ -1304,7 +1331,8 @@ static int smu10_set_power_profile_mode(struct pp_hwmgr *hwmgr, long *input, uin hwmgr->gfxoff_state_changed_by_workload = true; } result = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_ActiveProcessNotify, - 1 << workload_type); + 1 << workload_type, + NULL); if (!result) hwmgr->power_profile_mode = input[size]; if (workload_type && hwmgr->gfxoff_state_changed_by_workload) { @@ -1319,13 +1347,13 @@ static int smu10_asic_reset(struct pp_hwmgr *hwmgr, enum SMU_ASIC_RESET_MODE mod { return smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DeviceDriverReset, - mode); + mode, + NULL); } static const struct pp_hwmgr_func smu10_hwmgr_funcs = { .backend_init = smu10_hwmgr_backend_init, .backend_fini = smu10_hwmgr_backend_fini, - .asic_setup = NULL, .apply_state_adjust_rules = smu10_apply_state_adjust_rules, .force_dpm_level = smu10_dpm_force_dpm_level, .get_power_state_size = smu10_get_power_state_size, diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.h b/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.h index 1fb296a996f3..0f969de10fab 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.h +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.h @@ -192,7 +192,7 @@ struct smu10_clock_voltage_dependency_record { struct smu10_voltage_dependency_table { uint32_t count; - struct smu10_clock_voltage_dependency_record entries[1]; + struct smu10_clock_voltage_dependency_record entries[]; }; struct smu10_clock_voltage_information { diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_clockpowergating.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_clockpowergating.c index 683b29a99366..f2bda3bcbbde 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_clockpowergating.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_clockpowergating.c @@ -29,14 +29,16 @@ static int smu7_enable_disable_uvd_dpm(struct pp_hwmgr *hwmgr, bool enable) { return smum_send_msg_to_smc(hwmgr, enable ? PPSMC_MSG_UVDDPM_Enable : - PPSMC_MSG_UVDDPM_Disable); + PPSMC_MSG_UVDDPM_Disable, + NULL); } static int smu7_enable_disable_vce_dpm(struct pp_hwmgr *hwmgr, bool enable) { return smum_send_msg_to_smc(hwmgr, enable ? PPSMC_MSG_VCEDPM_Enable : - PPSMC_MSG_VCEDPM_Disable); + PPSMC_MSG_VCEDPM_Disable, + NULL); } static int smu7_update_uvd_dpm(struct pp_hwmgr *hwmgr, bool bgate) @@ -57,7 +59,8 @@ int smu7_powerdown_uvd(struct pp_hwmgr *hwmgr) { if (phm_cf_want_uvd_power_gating(hwmgr)) return smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_UVDPowerOFF); + PPSMC_MSG_UVDPowerOFF, + NULL); return 0; } @@ -67,10 +70,10 @@ static int smu7_powerup_uvd(struct pp_hwmgr *hwmgr) if (phm_cap_enabled(hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_UVDDynamicPowerGating)) { return smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_UVDPowerON, 1); + PPSMC_MSG_UVDPowerON, 1, NULL); } else { return smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_UVDPowerON, 0); + PPSMC_MSG_UVDPowerON, 0, NULL); } } @@ -81,7 +84,8 @@ static int smu7_powerdown_vce(struct pp_hwmgr *hwmgr) { if (phm_cf_want_vce_power_gating(hwmgr)) return smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_VCEPowerOFF); + PPSMC_MSG_VCEPowerOFF, + NULL); return 0; } @@ -89,7 +93,8 @@ static int smu7_powerup_vce(struct pp_hwmgr *hwmgr) { if (phm_cf_want_vce_power_gating(hwmgr)) return smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_VCEPowerON); + PPSMC_MSG_VCEPowerON, + NULL); return 0; } @@ -181,7 +186,7 @@ int smu7_update_clock_gatings(struct pp_hwmgr *hwmgr, value = CG_GFX_CGCG_MASK; if (smum_send_msg_to_smc_with_parameter( - hwmgr, msg, value)) + hwmgr, msg, value, NULL)) return -EINVAL; } if (PP_STATE_SUPPORT_LS & *msg_id) { @@ -191,7 +196,7 @@ int smu7_update_clock_gatings(struct pp_hwmgr *hwmgr, value = CG_GFX_CGLS_MASK; if (smum_send_msg_to_smc_with_parameter( - hwmgr, msg, value)) + hwmgr, msg, value, NULL)) return -EINVAL; } break; @@ -204,7 +209,7 @@ int smu7_update_clock_gatings(struct pp_hwmgr *hwmgr, value = CG_GFX_3DCG_MASK; if (smum_send_msg_to_smc_with_parameter( - hwmgr, msg, value)) + hwmgr, msg, value, NULL)) return -EINVAL; } @@ -215,7 +220,7 @@ int smu7_update_clock_gatings(struct pp_hwmgr *hwmgr, value = CG_GFX_3DLS_MASK; if (smum_send_msg_to_smc_with_parameter( - hwmgr, msg, value)) + hwmgr, msg, value, NULL)) return -EINVAL; } break; @@ -228,7 +233,7 @@ int smu7_update_clock_gatings(struct pp_hwmgr *hwmgr, value = CG_GFX_RLC_LS_MASK; if (smum_send_msg_to_smc_with_parameter( - hwmgr, msg, value)) + hwmgr, msg, value, NULL)) return -EINVAL; } break; @@ -241,7 +246,7 @@ int smu7_update_clock_gatings(struct pp_hwmgr *hwmgr, value = CG_GFX_CP_LS_MASK; if (smum_send_msg_to_smc_with_parameter( - hwmgr, msg, value)) + hwmgr, msg, value, NULL)) return -EINVAL; } break; @@ -255,7 +260,7 @@ int smu7_update_clock_gatings(struct pp_hwmgr *hwmgr, CG_GFX_OTHERS_MGCG_MASK); if (smum_send_msg_to_smc_with_parameter( - hwmgr, msg, value)) + hwmgr, msg, value, NULL)) return -EINVAL; } break; @@ -275,7 +280,7 @@ int smu7_update_clock_gatings(struct pp_hwmgr *hwmgr, value = CG_SYS_BIF_MGCG_MASK; if (smum_send_msg_to_smc_with_parameter( - hwmgr, msg, value)) + hwmgr, msg, value, NULL)) return -EINVAL; } if (PP_STATE_SUPPORT_LS & *msg_id) { @@ -285,7 +290,7 @@ int smu7_update_clock_gatings(struct pp_hwmgr *hwmgr, value = CG_SYS_BIF_MGLS_MASK; if (smum_send_msg_to_smc_with_parameter( - hwmgr, msg, value)) + hwmgr, msg, value, NULL)) return -EINVAL; } break; @@ -298,7 +303,7 @@ int smu7_update_clock_gatings(struct pp_hwmgr *hwmgr, value = CG_SYS_MC_MGCG_MASK; if (smum_send_msg_to_smc_with_parameter( - hwmgr, msg, value)) + hwmgr, msg, value, NULL)) return -EINVAL; } @@ -309,7 +314,7 @@ int smu7_update_clock_gatings(struct pp_hwmgr *hwmgr, value = CG_SYS_MC_MGLS_MASK; if (smum_send_msg_to_smc_with_parameter( - hwmgr, msg, value)) + hwmgr, msg, value, NULL)) return -EINVAL; } break; @@ -322,7 +327,7 @@ int smu7_update_clock_gatings(struct pp_hwmgr *hwmgr, value = CG_SYS_DRM_MGCG_MASK; if (smum_send_msg_to_smc_with_parameter( - hwmgr, msg, value)) + hwmgr, msg, value, NULL)) return -EINVAL; } if (PP_STATE_SUPPORT_LS & *msg_id) { @@ -332,7 +337,7 @@ int smu7_update_clock_gatings(struct pp_hwmgr *hwmgr, value = CG_SYS_DRM_MGLS_MASK; if (smum_send_msg_to_smc_with_parameter( - hwmgr, msg, value)) + hwmgr, msg, value, NULL)) return -EINVAL; } break; @@ -345,7 +350,7 @@ int smu7_update_clock_gatings(struct pp_hwmgr *hwmgr, value = CG_SYS_HDP_MGCG_MASK; if (smum_send_msg_to_smc_with_parameter( - hwmgr, msg, value)) + hwmgr, msg, value, NULL)) return -EINVAL; } @@ -356,7 +361,7 @@ int smu7_update_clock_gatings(struct pp_hwmgr *hwmgr, value = CG_SYS_HDP_MGLS_MASK; if (smum_send_msg_to_smc_with_parameter( - hwmgr, msg, value)) + hwmgr, msg, value, NULL)) return -EINVAL; } break; @@ -369,7 +374,7 @@ int smu7_update_clock_gatings(struct pp_hwmgr *hwmgr, value = CG_SYS_SDMA_MGCG_MASK; if (smum_send_msg_to_smc_with_parameter( - hwmgr, msg, value)) + hwmgr, msg, value, NULL)) return -EINVAL; } @@ -380,7 +385,7 @@ int smu7_update_clock_gatings(struct pp_hwmgr *hwmgr, value = CG_SYS_SDMA_MGLS_MASK; if (smum_send_msg_to_smc_with_parameter( - hwmgr, msg, value)) + hwmgr, msg, value, NULL)) return -EINVAL; } break; @@ -393,7 +398,7 @@ int smu7_update_clock_gatings(struct pp_hwmgr *hwmgr, value = CG_SYS_ROM_MASK; if (smum_send_msg_to_smc_with_parameter( - hwmgr, msg, value)) + hwmgr, msg, value, NULL)) return -EINVAL; } break; @@ -423,8 +428,10 @@ int smu7_powergate_gfx(struct pp_hwmgr *hwmgr, bool enable) if (enable) return smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GFX_CU_PG_ENABLE, - adev->gfx.cu_info.number); + adev->gfx.cu_info.number, + NULL); else return smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_GFX_CU_PG_DISABLE); + PPSMC_MSG_GFX_CU_PG_DISABLE, + NULL); } diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c index 7740488999df..753cb2cf6b77 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c @@ -186,7 +186,7 @@ static int smu7_enable_smc_voltage_controller(struct pp_hwmgr *hwmgr) } if (hwmgr->feature_mask & PP_SMC_VOLTAGE_CONTROL_MASK) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_Voltage_Cntl_Enable); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_Voltage_Cntl_Enable, NULL); return 0; } @@ -493,7 +493,7 @@ static int smu7_copy_and_switch_arb_sets(struct pp_hwmgr *hwmgr, static int smu7_reset_to_default(struct pp_hwmgr *hwmgr) { - return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ResetToDefaults); + return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ResetToDefaults, NULL); } /** @@ -979,7 +979,8 @@ static int smu7_enable_vrhot_gpio_interrupt(struct pp_hwmgr *hwmgr) if (phm_cap_enabled(hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_RegulatorHot)) return smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_EnableVRHotGPIOInterrupt); + PPSMC_MSG_EnableVRHotGPIOInterrupt, + NULL); return 0; } @@ -996,7 +997,7 @@ static int smu7_enable_ulv(struct pp_hwmgr *hwmgr) struct smu7_hwmgr *data = (struct smu7_hwmgr *)(hwmgr->backend); if (data->ulv_supported) - return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnableULV); + return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnableULV, NULL); return 0; } @@ -1006,7 +1007,7 @@ static int smu7_disable_ulv(struct pp_hwmgr *hwmgr) struct smu7_hwmgr *data = (struct smu7_hwmgr *)(hwmgr->backend); if (data->ulv_supported) - return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_DisableULV); + return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_DisableULV, NULL); return 0; } @@ -1015,13 +1016,14 @@ static int smu7_enable_deep_sleep_master_switch(struct pp_hwmgr *hwmgr) { if (phm_cap_enabled(hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_SclkDeepSleep)) { - if (smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MASTER_DeepSleep_ON)) + if (smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MASTER_DeepSleep_ON, NULL)) PP_ASSERT_WITH_CODE(false, "Attempt to enable Master Deep Sleep switch failed!", return -EINVAL); } else { if (smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_MASTER_DeepSleep_OFF)) { + PPSMC_MSG_MASTER_DeepSleep_OFF, + NULL)) { PP_ASSERT_WITH_CODE(false, "Attempt to disable Master Deep Sleep switch failed!", return -EINVAL); @@ -1036,7 +1038,8 @@ static int smu7_disable_deep_sleep_master_switch(struct pp_hwmgr *hwmgr) if (phm_cap_enabled(hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_SclkDeepSleep)) { if (smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_MASTER_DeepSleep_OFF)) { + PPSMC_MSG_MASTER_DeepSleep_OFF, + NULL)) { PP_ASSERT_WITH_CODE(false, "Attempt to disable Master Deep Sleep switch failed!", return -EINVAL); @@ -1089,7 +1092,7 @@ static int smu7_enable_sclk_mclk_dpm(struct pp_hwmgr *hwmgr) smu7_disable_sclk_vce_handshake(hwmgr); PP_ASSERT_WITH_CODE( - (0 == smum_send_msg_to_smc(hwmgr, PPSMC_MSG_DPM_Enable)), + (0 == smum_send_msg_to_smc(hwmgr, PPSMC_MSG_DPM_Enable, NULL)), "Failed to enable SCLK DPM during DPM Start Function!", return -EINVAL); } @@ -1101,7 +1104,8 @@ static int smu7_enable_sclk_mclk_dpm(struct pp_hwmgr *hwmgr) PP_ASSERT_WITH_CODE( (0 == smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_MCLKDPM_Enable)), + PPSMC_MSG_MCLKDPM_Enable, + NULL)), "Failed to enable MCLK DPM during DPM Start Function!", return -EINVAL); @@ -1172,7 +1176,8 @@ static int smu7_start_dpm(struct pp_hwmgr *hwmgr) if (0 == data->pcie_dpm_key_disabled) { PP_ASSERT_WITH_CODE( (0 == smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_PCIeDPM_Enable)), + PPSMC_MSG_PCIeDPM_Enable, + NULL)), "Failed to enable pcie DPM during DPM Start Function!", return -EINVAL); } @@ -1180,7 +1185,8 @@ static int smu7_start_dpm(struct pp_hwmgr *hwmgr) if (phm_cap_enabled(hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_Falcon_QuickTransition)) { PP_ASSERT_WITH_CODE((0 == smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_EnableACDCGPIOInterrupt)), + PPSMC_MSG_EnableACDCGPIOInterrupt, + NULL)), "Failed to enable AC DC GPIO Interrupt!", ); } @@ -1197,7 +1203,7 @@ static int smu7_disable_sclk_mclk_dpm(struct pp_hwmgr *hwmgr) PP_ASSERT_WITH_CODE(true == smum_is_dpm_running(hwmgr), "Trying to disable SCLK DPM when DPM is disabled", return 0); - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_DPM_Disable); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_DPM_Disable, NULL); } /* disable MCLK dpm */ @@ -1205,7 +1211,7 @@ static int smu7_disable_sclk_mclk_dpm(struct pp_hwmgr *hwmgr) PP_ASSERT_WITH_CODE(true == smum_is_dpm_running(hwmgr), "Trying to disable MCLK DPM when DPM is disabled", return 0); - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MCLKDPM_Disable); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MCLKDPM_Disable, NULL); } return 0; @@ -1226,7 +1232,8 @@ static int smu7_stop_dpm(struct pp_hwmgr *hwmgr) if (!data->pcie_dpm_key_disabled) { PP_ASSERT_WITH_CODE( (smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_PCIeDPM_Disable) == 0), + PPSMC_MSG_PCIeDPM_Disable, + NULL) == 0), "Failed to disable pcie DPM during DPM Stop Function!", return -EINVAL); } @@ -1237,7 +1244,7 @@ static int smu7_stop_dpm(struct pp_hwmgr *hwmgr) "Trying to disable voltage DPM when DPM is disabled", return 0); - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_Voltage_Cntl_Disable); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_Voltage_Cntl_Disable, NULL); return 0; } @@ -1388,7 +1395,7 @@ static int smu7_enable_dpm_tasks(struct pp_hwmgr *hwmgr) PP_ASSERT_WITH_CODE((0 == tmp_result), "Failed to enable VR hot GPIO interrupt!", result = tmp_result); - smum_send_msg_to_smc(hwmgr, (PPSMC_Msg)PPSMC_NoDisplay); + smum_send_msg_to_smc(hwmgr, (PPSMC_Msg)PPSMC_NoDisplay, NULL); tmp_result = smu7_enable_sclk_control(hwmgr); PP_ASSERT_WITH_CODE((0 == tmp_result), @@ -1446,14 +1453,14 @@ static int smu7_avfs_control(struct pp_hwmgr *hwmgr, bool enable) if (!PHM_READ_VFPF_INDIRECT_FIELD(hwmgr->device, CGS_IND_REG__SMC, FEATURE_STATUS, AVS_ON)) { PP_ASSERT_WITH_CODE(!smum_send_msg_to_smc( - hwmgr, PPSMC_MSG_EnableAvfs), + hwmgr, PPSMC_MSG_EnableAvfs, NULL), "Failed to enable AVFS!", return -EINVAL); } } else if (PHM_READ_VFPF_INDIRECT_FIELD(hwmgr->device, CGS_IND_REG__SMC, FEATURE_STATUS, AVS_ON)) { PP_ASSERT_WITH_CODE(!smum_send_msg_to_smc( - hwmgr, PPSMC_MSG_DisableAvfs), + hwmgr, PPSMC_MSG_DisableAvfs, NULL), "Failed to disable AVFS!", return -EINVAL); } @@ -2609,7 +2616,8 @@ static int smu7_force_dpm_highest(struct pp_hwmgr *hwmgr) if (level) smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_PCIeDPM_ForceLevel, level); + PPSMC_MSG_PCIeDPM_ForceLevel, level, + NULL); } } @@ -2623,7 +2631,8 @@ static int smu7_force_dpm_highest(struct pp_hwmgr *hwmgr) if (level) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SCLKDPM_SetEnabledMask, - (1 << level)); + (1 << level), + NULL); } } @@ -2637,7 +2646,8 @@ static int smu7_force_dpm_highest(struct pp_hwmgr *hwmgr) if (level) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_MCLKDPM_SetEnabledMask, - (1 << level)); + (1 << level), + NULL); } } @@ -2656,14 +2666,16 @@ static int smu7_upload_dpm_level_enable_mask(struct pp_hwmgr *hwmgr) if (data->dpm_level_enable_mask.sclk_dpm_enable_mask) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SCLKDPM_SetEnabledMask, - data->dpm_level_enable_mask.sclk_dpm_enable_mask); + data->dpm_level_enable_mask.sclk_dpm_enable_mask, + NULL); } if (!data->mclk_dpm_key_disabled) { if (data->dpm_level_enable_mask.mclk_dpm_enable_mask) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_MCLKDPM_SetEnabledMask, - data->dpm_level_enable_mask.mclk_dpm_enable_mask); + data->dpm_level_enable_mask.mclk_dpm_enable_mask, + NULL); } return 0; @@ -2678,7 +2690,8 @@ static int smu7_unforce_dpm_levels(struct pp_hwmgr *hwmgr) if (!data->pcie_dpm_key_disabled) { smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_PCIeDPM_UnForceLevel); + PPSMC_MSG_PCIeDPM_UnForceLevel, + NULL); } return smu7_upload_dpm_level_enable_mask(hwmgr); @@ -2696,7 +2709,8 @@ static int smu7_force_dpm_lowest(struct pp_hwmgr *hwmgr) data->dpm_level_enable_mask.sclk_dpm_enable_mask); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SCLKDPM_SetEnabledMask, - (1 << level)); + (1 << level), + NULL); } @@ -2706,7 +2720,8 @@ static int smu7_force_dpm_lowest(struct pp_hwmgr *hwmgr) data->dpm_level_enable_mask.mclk_dpm_enable_mask); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_MCLKDPM_SetEnabledMask, - (1 << level)); + (1 << level), + NULL); } } @@ -2716,7 +2731,8 @@ static int smu7_force_dpm_lowest(struct pp_hwmgr *hwmgr) data->dpm_level_enable_mask.pcie_dpm_enable_mask); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_PCIeDPM_ForceLevel, - (level)); + (level), + NULL); } } @@ -3495,21 +3511,20 @@ static int smu7_get_gpu_power(struct pp_hwmgr *hwmgr, u32 *query) (adev->asic_type != CHIP_BONAIRE) && (adev->asic_type != CHIP_FIJI) && (adev->asic_type != CHIP_TONGA)) { - smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetCurrPkgPwr, 0); - tmp = cgs_read_register(hwmgr->device, mmSMC_MSG_ARG_0); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetCurrPkgPwr, 0, &tmp); *query = tmp; if (tmp != 0) return 0; } - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_PmStatusLogStart); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_PmStatusLogStart, NULL); cgs_write_ind_register(hwmgr->device, CGS_IND_REG__SMC, ixSMU_PM_STATUS_95, 0); for (i = 0; i < 10; i++) { msleep(500); - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_PmStatusLogSample); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_PmStatusLogSample, NULL); tmp = cgs_read_ind_register(hwmgr->device, CGS_IND_REG__SMC, ixSMU_PM_STATUS_95); @@ -3534,14 +3549,12 @@ static int smu7_read_sensor(struct pp_hwmgr *hwmgr, int idx, switch (idx) { case AMDGPU_PP_SENSOR_GFX_SCLK: - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_API_GetSclkFrequency); - sclk = cgs_read_register(hwmgr->device, mmSMC_MSG_ARG_0); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_API_GetSclkFrequency, &sclk); *((uint32_t *)value) = sclk; *size = 4; return 0; case AMDGPU_PP_SENSOR_GFX_MCLK: - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_API_GetMclkFrequency); - mclk = cgs_read_register(hwmgr->device, mmSMC_MSG_ARG_0); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_API_GetMclkFrequency, &mclk); *((uint32_t *)value) = mclk; *size = 4; return 0; @@ -3730,7 +3743,8 @@ static int smu7_freeze_sclk_mclk_dpm(struct pp_hwmgr *hwmgr) "Trying to freeze SCLK DPM when DPM is disabled", ); PP_ASSERT_WITH_CODE(0 == smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_SCLKDPM_FreezeLevel), + PPSMC_MSG_SCLKDPM_FreezeLevel, + NULL), "Failed to freeze SCLK DPM during FreezeSclkMclkDPM Function!", return -EINVAL); } @@ -3742,7 +3756,8 @@ static int smu7_freeze_sclk_mclk_dpm(struct pp_hwmgr *hwmgr) "Trying to freeze MCLK DPM when DPM is disabled", ); PP_ASSERT_WITH_CODE(0 == smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_MCLKDPM_FreezeLevel), + PPSMC_MSG_MCLKDPM_FreezeLevel, + NULL), "Failed to freeze MCLK DPM during FreezeSclkMclkDPM Function!", return -EINVAL); } @@ -3804,9 +3819,12 @@ static int smu7_trim_single_dpm_states(struct pp_hwmgr *hwmgr, { uint32_t i; + /* force the trim if mclk_switching is disabled to prevent flicker */ + bool force_trim = (low_limit == high_limit); for (i = 0; i < dpm_table->count; i++) { /*skip the trim if od is enabled*/ - if (!hwmgr->od_enabled && (dpm_table->dpm_levels[i].value < low_limit + if ((!hwmgr->od_enabled || force_trim) + && (dpm_table->dpm_levels[i].value < low_limit || dpm_table->dpm_levels[i].value > high_limit)) dpm_table->dpm_levels[i].enabled = false; else @@ -3881,7 +3899,8 @@ static int smu7_unfreeze_sclk_mclk_dpm(struct pp_hwmgr *hwmgr) "Trying to Unfreeze SCLK DPM when DPM is disabled", ); PP_ASSERT_WITH_CODE(0 == smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_SCLKDPM_UnfreezeLevel), + PPSMC_MSG_SCLKDPM_UnfreezeLevel, + NULL), "Failed to unfreeze SCLK DPM during UnFreezeSclkMclkDPM Function!", return -EINVAL); } @@ -3893,7 +3912,8 @@ static int smu7_unfreeze_sclk_mclk_dpm(struct pp_hwmgr *hwmgr) "Trying to Unfreeze MCLK DPM when DPM is disabled", ); PP_ASSERT_WITH_CODE(0 == smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_MCLKDPM_UnfreezeLevel), + PPSMC_MSG_MCLKDPM_UnfreezeLevel, + NULL), "Failed to unfreeze MCLK DPM during UnFreezeSclkMclkDPM Function!", return -EINVAL); } @@ -3946,12 +3966,14 @@ static int smu7_notify_smc_display(struct pp_hwmgr *hwmgr) if (hwmgr->feature_mask & PP_VBI_TIME_SUPPORT_MASK) { if (hwmgr->chip_id == CHIP_VEGAM) smum_send_msg_to_smc_with_parameter(hwmgr, - (PPSMC_Msg)PPSMC_MSG_SetVBITimeout_VEGAM, data->frame_time_x2); + (PPSMC_Msg)PPSMC_MSG_SetVBITimeout_VEGAM, data->frame_time_x2, + NULL); else smum_send_msg_to_smc_with_parameter(hwmgr, - (PPSMC_Msg)PPSMC_MSG_SetVBITimeout, data->frame_time_x2); + (PPSMC_Msg)PPSMC_MSG_SetVBITimeout, data->frame_time_x2, + NULL); } - return (smum_send_msg_to_smc(hwmgr, (PPSMC_Msg)PPSMC_HasDisplay) == 0) ? 0 : -EINVAL; + return (smum_send_msg_to_smc(hwmgr, (PPSMC_Msg)PPSMC_HasDisplay, NULL) == 0) ? 0 : -EINVAL; } static int smu7_set_power_state_tasks(struct pp_hwmgr *hwmgr, const void *input) @@ -4037,7 +4059,8 @@ static int smu7_set_max_fan_pwm_output(struct pp_hwmgr *hwmgr, uint16_t us_max_f advanceFanControlParameters.usMaxFanPWM = us_max_fan_pwm; return smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_SetFanPwmMax, us_max_fan_pwm); + PPSMC_MSG_SetFanPwmMax, us_max_fan_pwm, + NULL); } static int @@ -4045,7 +4068,7 @@ smu7_notify_smc_display_change(struct pp_hwmgr *hwmgr, bool has_display) { PPSMC_Msg msg = has_display ? (PPSMC_Msg)PPSMC_HasDisplay : (PPSMC_Msg)PPSMC_NoDisplay; - return (smum_send_msg_to_smc(hwmgr, msg) == 0) ? 0 : -1; + return (smum_send_msg_to_smc(hwmgr, msg, NULL) == 0) ? 0 : -1; } static int @@ -4129,7 +4152,8 @@ static int smu7_set_max_fan_rpm_output(struct pp_hwmgr *hwmgr, uint16_t us_max_f advanceFanControlParameters.usMaxFanRPM = us_max_fan_rpm; return smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_SetFanRpmMax, us_max_fan_rpm); + PPSMC_MSG_SetFanRpmMax, us_max_fan_rpm, + NULL); } static const struct amdgpu_irq_src_funcs smu7_irq_funcs = { @@ -4259,14 +4283,14 @@ static int smu7_check_mc_firmware(struct pp_hwmgr *hwmgr) if ((hwmgr->chip_id == CHIP_POLARIS10) || (hwmgr->chip_id == CHIP_POLARIS11) || (hwmgr->chip_id == CHIP_POLARIS12)) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnableFFC); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnableFFC, NULL); } else { data->mem_latency_high = 330; data->mem_latency_low = 330; if ((hwmgr->chip_id == CHIP_POLARIS10) || (hwmgr->chip_id == CHIP_POLARIS11) || (hwmgr->chip_id == CHIP_POLARIS12)) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_DisableFFC); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_DisableFFC, NULL); } return 0; @@ -4410,13 +4434,15 @@ static int smu7_force_clock_level(struct pp_hwmgr *hwmgr, if (!data->sclk_dpm_key_disabled) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SCLKDPM_SetEnabledMask, - data->dpm_level_enable_mask.sclk_dpm_enable_mask & mask); + data->dpm_level_enable_mask.sclk_dpm_enable_mask & mask, + NULL); break; case PP_MCLK: if (!data->mclk_dpm_key_disabled) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_MCLKDPM_SetEnabledMask, - data->dpm_level_enable_mask.mclk_dpm_enable_mask & mask); + data->dpm_level_enable_mask.mclk_dpm_enable_mask & mask, + NULL); break; case PP_PCIE: { @@ -4424,11 +4450,13 @@ static int smu7_force_clock_level(struct pp_hwmgr *hwmgr, if (!data->pcie_dpm_key_disabled) { if (fls(tmp) != ffs(tmp)) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_PCIeDPM_UnForceLevel); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_PCIeDPM_UnForceLevel, + NULL); else smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_PCIeDPM_ForceLevel, - fls(tmp) - 1); + fls(tmp) - 1, + NULL); } break; } @@ -4454,8 +4482,7 @@ static int smu7_print_clock_levels(struct pp_hwmgr *hwmgr, switch (type) { case PP_SCLK: - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_API_GetSclkFrequency); - clock = cgs_read_register(hwmgr->device, mmSMC_MSG_ARG_0); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_API_GetSclkFrequency, &clock); for (i = 0; i < sclk_table->count; i++) { if (clock > sclk_table->dpm_levels[i].value) @@ -4470,8 +4497,7 @@ static int smu7_print_clock_levels(struct pp_hwmgr *hwmgr, (i == now) ? "*" : ""); break; case PP_MCLK: - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_API_GetMclkFrequency); - clock = cgs_read_register(hwmgr->device, mmSMC_MSG_ARG_0); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_API_GetMclkFrequency, &clock); for (i = 0; i < mclk_table->count; i++) { if (clock > mclk_table->dpm_levels[i].value) diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_powertune.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_powertune.c index 58f5589aaf12..5d4971576111 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_powertune.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_powertune.c @@ -887,7 +887,10 @@ static int smu7_enable_didt(struct pp_hwmgr *hwmgr, const bool enable) didt_block |= block_en << TCP_Enable_SHIFT; if (enable) - result = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_Didt_Block_Function, didt_block); + result = smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_Didt_Block_Function, + didt_block, + NULL); return result; } @@ -1009,7 +1012,8 @@ int smu7_enable_didt_config(struct pp_hwmgr *hwmgr) if (hwmgr->chip_id == CHIP_POLARIS11) { result = smum_send_msg_to_smc(hwmgr, - (uint16_t)(PPSMC_MSG_EnableDpmDidt)); + (uint16_t)(PPSMC_MSG_EnableDpmDidt), + NULL); PP_ASSERT_WITH_CODE((0 == result), "Failed to enable DPM DIDT.", goto error); } @@ -1042,7 +1046,8 @@ int smu7_disable_didt_config(struct pp_hwmgr *hwmgr) goto error); if (hwmgr->chip_id == CHIP_POLARIS11) { result = smum_send_msg_to_smc(hwmgr, - (uint16_t)(PPSMC_MSG_DisableDpmDidt)); + (uint16_t)(PPSMC_MSG_DisableDpmDidt), + NULL); PP_ASSERT_WITH_CODE((0 == result), "Failed to disable DPM DIDT.", goto error); } @@ -1063,7 +1068,8 @@ int smu7_enable_smc_cac(struct pp_hwmgr *hwmgr) if (PP_CAP(PHM_PlatformCaps_CAC)) { int smc_result; smc_result = smum_send_msg_to_smc(hwmgr, - (uint16_t)(PPSMC_MSG_EnableCac)); + (uint16_t)(PPSMC_MSG_EnableCac), + NULL); PP_ASSERT_WITH_CODE((0 == smc_result), "Failed to enable CAC in SMC.", result = -1); @@ -1079,7 +1085,8 @@ int smu7_disable_smc_cac(struct pp_hwmgr *hwmgr) if (PP_CAP(PHM_PlatformCaps_CAC) && data->cac_enabled) { int smc_result = smum_send_msg_to_smc(hwmgr, - (uint16_t)(PPSMC_MSG_DisableCac)); + (uint16_t)(PPSMC_MSG_DisableCac), + NULL); PP_ASSERT_WITH_CODE((smc_result == 0), "Failed to disable CAC in SMC.", result = -1); @@ -1095,7 +1102,9 @@ int smu7_set_power_limit(struct pp_hwmgr *hwmgr, uint32_t n) if (data->power_containment_features & POWERCONTAINMENT_FEATURE_PkgPwrLimit) return smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_PkgPwrSetLimit, n<<8); + PPSMC_MSG_PkgPwrSetLimit, + n<<8, + NULL); return 0; } @@ -1103,7 +1112,9 @@ static int smu7_set_overdriver_target_tdp(struct pp_hwmgr *hwmgr, uint32_t target_tdp) { return smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_OverDriveSetTargetTdp, target_tdp); + PPSMC_MSG_OverDriveSetTargetTdp, + target_tdp, + NULL); } int smu7_enable_power_containment(struct pp_hwmgr *hwmgr) @@ -1124,7 +1135,8 @@ int smu7_enable_power_containment(struct pp_hwmgr *hwmgr) if (PP_CAP(PHM_PlatformCaps_PowerContainment)) { if (data->enable_tdc_limit_feature) { smc_result = smum_send_msg_to_smc(hwmgr, - (uint16_t)(PPSMC_MSG_TDCLimitEnable)); + (uint16_t)(PPSMC_MSG_TDCLimitEnable), + NULL); PP_ASSERT_WITH_CODE((0 == smc_result), "Failed to enable TDCLimit in SMC.", result = -1;); if (0 == smc_result) @@ -1134,7 +1146,8 @@ int smu7_enable_power_containment(struct pp_hwmgr *hwmgr) if (data->enable_pkg_pwr_tracking_feature) { smc_result = smum_send_msg_to_smc(hwmgr, - (uint16_t)(PPSMC_MSG_PkgPwrLimitEnable)); + (uint16_t)(PPSMC_MSG_PkgPwrLimitEnable), + NULL); PP_ASSERT_WITH_CODE((0 == smc_result), "Failed to enable PkgPwrTracking in SMC.", result = -1;); if (0 == smc_result) { @@ -1163,7 +1176,8 @@ int smu7_disable_power_containment(struct pp_hwmgr *hwmgr) if (data->power_containment_features & POWERCONTAINMENT_FEATURE_TDCLimit) { smc_result = smum_send_msg_to_smc(hwmgr, - (uint16_t)(PPSMC_MSG_TDCLimitDisable)); + (uint16_t)(PPSMC_MSG_TDCLimitDisable), + NULL); PP_ASSERT_WITH_CODE((smc_result == 0), "Failed to disable TDCLimit in SMC.", result = smc_result); @@ -1172,7 +1186,8 @@ int smu7_disable_power_containment(struct pp_hwmgr *hwmgr) if (data->power_containment_features & POWERCONTAINMENT_FEATURE_DTE) { smc_result = smum_send_msg_to_smc(hwmgr, - (uint16_t)(PPSMC_MSG_DisableDTE)); + (uint16_t)(PPSMC_MSG_DisableDTE), + NULL); PP_ASSERT_WITH_CODE((smc_result == 0), "Failed to disable DTE in SMC.", result = smc_result); @@ -1181,7 +1196,8 @@ int smu7_disable_power_containment(struct pp_hwmgr *hwmgr) if (data->power_containment_features & POWERCONTAINMENT_FEATURE_PkgPwrLimit) { smc_result = smum_send_msg_to_smc(hwmgr, - (uint16_t)(PPSMC_MSG_PkgPwrLimitDisable)); + (uint16_t)(PPSMC_MSG_PkgPwrLimitDisable), + NULL); PP_ASSERT_WITH_CODE((smc_result == 0), "Failed to disable PkgPwrTracking in SMC.", result = smc_result); diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_thermal.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_thermal.c index 5bdc0df5a9f4..0b30f73649a8 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_thermal.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_thermal.c @@ -151,8 +151,8 @@ int smu7_fan_ctrl_start_smc_fan_control(struct pp_hwmgr *hwmgr) int result; if (PP_CAP(PHM_PlatformCaps_ODFuzzyFanControlSupport)) { - cgs_write_register(hwmgr->device, mmSMC_MSG_ARG_0, FAN_CONTROL_FUZZY); - result = smum_send_msg_to_smc(hwmgr, PPSMC_StartFanControl); + result = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_StartFanControl, + FAN_CONTROL_FUZZY, NULL); if (PP_CAP(PHM_PlatformCaps_FanSpeedInTableIsRPM)) hwmgr->hwmgr_func->set_max_fan_rpm_output(hwmgr, @@ -164,8 +164,8 @@ int smu7_fan_ctrl_start_smc_fan_control(struct pp_hwmgr *hwmgr) advanceFanControlParameters.usMaxFanPWM); } else { - cgs_write_register(hwmgr->device, mmSMC_MSG_ARG_0, FAN_CONTROL_TABLE); - result = smum_send_msg_to_smc(hwmgr, PPSMC_StartFanControl); + result = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_StartFanControl, + FAN_CONTROL_TABLE, NULL); } if (!result && hwmgr->thermal_controller. @@ -173,7 +173,8 @@ int smu7_fan_ctrl_start_smc_fan_control(struct pp_hwmgr *hwmgr) result = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetFanTemperatureTarget, hwmgr->thermal_controller. - advanceFanControlParameters.ucTargetTemperature); + advanceFanControlParameters.ucTargetTemperature, + NULL); hwmgr->fan_ctrl_enabled = true; return result; @@ -183,7 +184,7 @@ int smu7_fan_ctrl_start_smc_fan_control(struct pp_hwmgr *hwmgr) int smu7_fan_ctrl_stop_smc_fan_control(struct pp_hwmgr *hwmgr) { hwmgr->fan_ctrl_enabled = false; - return smum_send_msg_to_smc(hwmgr, PPSMC_StopFanControl); + return smum_send_msg_to_smc(hwmgr, PPSMC_StopFanControl, NULL); } /** @@ -372,7 +373,7 @@ static void smu7_thermal_enable_alert(struct pp_hwmgr *hwmgr) CG_THERMAL_INT, THERM_INT_MASK, alert); /* send message to SMU to enable internal thermal interrupts */ - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_Thermal_Cntl_Enable); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_Thermal_Cntl_Enable, NULL); } /** @@ -390,7 +391,7 @@ int smu7_thermal_disable_alert(struct pp_hwmgr *hwmgr) CG_THERMAL_INT, THERM_INT_MASK, alert); /* send message to SMU to disable internal thermal interrupts */ - return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_Thermal_Cntl_Disable); + return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_Thermal_Cntl_Disable, NULL); } /** diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu8_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu8_hwmgr.c index 019d6a206492..a6c6a793e98e 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu8_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu8_hwmgr.c @@ -162,8 +162,10 @@ static uint32_t smu8_get_max_sclk_level(struct pp_hwmgr *hwmgr) struct smu8_hwmgr *data = hwmgr->backend; if (data->max_sclk_level == 0) { - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetMaxSclkLevel); - data->max_sclk_level = smum_get_argument(hwmgr) + 1; + smum_send_msg_to_smc(hwmgr, + PPSMC_MSG_GetMaxSclkLevel, + &data->max_sclk_level); + data->max_sclk_level += 1; } return data->max_sclk_level; @@ -580,7 +582,8 @@ static int smu8_init_uvd_limit(struct pp_hwmgr *hwmgr) struct smu8_hwmgr *data = hwmgr->backend; struct phm_uvd_clock_voltage_dependency_table *table = hwmgr->dyn_state.uvd_clock_voltage_dependency_table; - unsigned long clock = 0, level; + unsigned long clock = 0; + uint32_t level; if (NULL == table || table->count <= 0) return -EINVAL; @@ -588,8 +591,7 @@ static int smu8_init_uvd_limit(struct pp_hwmgr *hwmgr) data->uvd_dpm.soft_min_clk = 0; data->uvd_dpm.hard_min_clk = 0; - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetMaxUvdLevel); - level = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetMaxUvdLevel, &level); if (level < table->count) clock = table->entries[level].vclk; @@ -607,7 +609,8 @@ static int smu8_init_vce_limit(struct pp_hwmgr *hwmgr) struct smu8_hwmgr *data = hwmgr->backend; struct phm_vce_clock_voltage_dependency_table *table = hwmgr->dyn_state.vce_clock_voltage_dependency_table; - unsigned long clock = 0, level; + unsigned long clock = 0; + uint32_t level; if (NULL == table || table->count <= 0) return -EINVAL; @@ -615,8 +618,7 @@ static int smu8_init_vce_limit(struct pp_hwmgr *hwmgr) data->vce_dpm.soft_min_clk = 0; data->vce_dpm.hard_min_clk = 0; - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetMaxEclkLevel); - level = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetMaxEclkLevel, &level); if (level < table->count) clock = table->entries[level].ecclk; @@ -634,7 +636,8 @@ static int smu8_init_acp_limit(struct pp_hwmgr *hwmgr) struct smu8_hwmgr *data = hwmgr->backend; struct phm_acp_clock_voltage_dependency_table *table = hwmgr->dyn_state.acp_clock_voltage_dependency_table; - unsigned long clock = 0, level; + unsigned long clock = 0; + uint32_t level; if (NULL == table || table->count <= 0) return -EINVAL; @@ -642,8 +645,7 @@ static int smu8_init_acp_limit(struct pp_hwmgr *hwmgr) data->acp_dpm.soft_min_clk = 0; data->acp_dpm.hard_min_clk = 0; - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetMaxAclkLevel); - level = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetMaxAclkLevel, &level); if (level < table->count) clock = table->entries[level].acpclk; @@ -665,7 +667,7 @@ static void smu8_init_power_gate_state(struct pp_hwmgr *hwmgr) #ifdef CONFIG_DRM_AMD_ACP data->acp_power_gated = false; #else - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ACPPowerOFF); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ACPPowerOFF, NULL); data->acp_power_gated = true; #endif @@ -708,7 +710,8 @@ static int smu8_update_sclk_limit(struct pp_hwmgr *hwmgr) PPSMC_MSG_SetSclkHardMin, smu8_get_sclk_level(hwmgr, data->sclk_dpm.hard_min_clk, - PPSMC_MSG_SetSclkHardMin)); + PPSMC_MSG_SetSclkHardMin), + NULL); } clock = data->sclk_dpm.soft_min_clk; @@ -731,7 +734,8 @@ static int smu8_update_sclk_limit(struct pp_hwmgr *hwmgr) PPSMC_MSG_SetSclkSoftMin, smu8_get_sclk_level(hwmgr, data->sclk_dpm.soft_min_clk, - PPSMC_MSG_SetSclkSoftMin)); + PPSMC_MSG_SetSclkSoftMin), + NULL); } if (phm_cap_enabled(hwmgr->platform_descriptor.platformCaps, @@ -742,7 +746,8 @@ static int smu8_update_sclk_limit(struct pp_hwmgr *hwmgr) PPSMC_MSG_SetSclkSoftMax, smu8_get_sclk_level(hwmgr, data->sclk_dpm.soft_max_clk, - PPSMC_MSG_SetSclkSoftMax)); + PPSMC_MSG_SetSclkSoftMax), + NULL); } return 0; @@ -760,7 +765,8 @@ static int smu8_set_deep_sleep_sclk_threshold(struct pp_hwmgr *hwmgr) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetMinDeepSleepSclk, - clks); + clks, + NULL); } return 0; @@ -773,7 +779,8 @@ static int smu8_set_watermark_threshold(struct pp_hwmgr *hwmgr) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetWatermarkFrequency, - data->sclk_dpm.soft_max_clk); + data->sclk_dpm.soft_max_clk, + NULL); return 0; } @@ -788,13 +795,15 @@ static int smu8_nbdpm_pstate_enable_disable(struct pp_hwmgr *hwmgr, bool enable, return smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_EnableLowMemoryPstate, - (lock ? 1 : 0)); + (lock ? 1 : 0), + NULL); } else { PP_DBG_LOG("disable Low Memory PState.\n"); return smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DisableLowMemoryPstate, - (lock ? 1 : 0)); + (lock ? 1 : 0), + NULL); } } @@ -814,7 +823,8 @@ static int smu8_disable_nb_dpm(struct pp_hwmgr *hwmgr) ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_DisableAllSmuFeatures, - dpm_features); + dpm_features, + NULL); if (ret == 0) data->is_nb_dpm_enabled = false; } @@ -835,7 +845,8 @@ static int smu8_enable_nb_dpm(struct pp_hwmgr *hwmgr) ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_EnableAllSmuFeatures, - dpm_features); + dpm_features, + NULL); if (ret == 0) data->is_nb_dpm_enabled = true; } @@ -953,7 +964,8 @@ static int smu8_start_dpm(struct pp_hwmgr *hwmgr) return smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_EnableAllSmuFeatures, - SCLK_DPM_MASK); + SCLK_DPM_MASK, + NULL); } static int smu8_stop_dpm(struct pp_hwmgr *hwmgr) @@ -967,7 +979,8 @@ static int smu8_stop_dpm(struct pp_hwmgr *hwmgr) data->dpm_flags &= ~DPMFlags_SCLK_Enabled; ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DisableAllSmuFeatures, - dpm_features); + dpm_features, + NULL); } return ret; } @@ -983,13 +996,15 @@ static int smu8_program_bootup_state(struct pp_hwmgr *hwmgr) PPSMC_MSG_SetSclkSoftMin, smu8_get_sclk_level(hwmgr, data->sclk_dpm.soft_min_clk, - PPSMC_MSG_SetSclkSoftMin)); + PPSMC_MSG_SetSclkSoftMin), + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSclkSoftMax, smu8_get_sclk_level(hwmgr, data->sclk_dpm.soft_max_clk, - PPSMC_MSG_SetSclkSoftMax)); + PPSMC_MSG_SetSclkSoftMax), + NULL); return 0; } @@ -1127,13 +1142,15 @@ static int smu8_phm_force_dpm_highest(struct pp_hwmgr *hwmgr) PPSMC_MSG_SetSclkSoftMin, smu8_get_sclk_level(hwmgr, data->sclk_dpm.soft_max_clk, - PPSMC_MSG_SetSclkSoftMin)); + PPSMC_MSG_SetSclkSoftMin), + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSclkSoftMax, smu8_get_sclk_level(hwmgr, data->sclk_dpm.soft_max_clk, - PPSMC_MSG_SetSclkSoftMax)); + PPSMC_MSG_SetSclkSoftMax), + NULL); return 0; } @@ -1167,13 +1184,15 @@ static int smu8_phm_unforce_dpm_levels(struct pp_hwmgr *hwmgr) PPSMC_MSG_SetSclkSoftMin, smu8_get_sclk_level(hwmgr, data->sclk_dpm.soft_min_clk, - PPSMC_MSG_SetSclkSoftMin)); + PPSMC_MSG_SetSclkSoftMin), + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSclkSoftMax, smu8_get_sclk_level(hwmgr, data->sclk_dpm.soft_max_clk, - PPSMC_MSG_SetSclkSoftMax)); + PPSMC_MSG_SetSclkSoftMax), + NULL); return 0; } @@ -1186,13 +1205,15 @@ static int smu8_phm_force_dpm_lowest(struct pp_hwmgr *hwmgr) PPSMC_MSG_SetSclkSoftMax, smu8_get_sclk_level(hwmgr, data->sclk_dpm.soft_min_clk, - PPSMC_MSG_SetSclkSoftMax)); + PPSMC_MSG_SetSclkSoftMax), + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSclkSoftMin, smu8_get_sclk_level(hwmgr, data->sclk_dpm.soft_min_clk, - PPSMC_MSG_SetSclkSoftMin)); + PPSMC_MSG_SetSclkSoftMin), + NULL); return 0; } @@ -1227,7 +1248,7 @@ static int smu8_dpm_force_dpm_level(struct pp_hwmgr *hwmgr, static int smu8_dpm_powerdown_uvd(struct pp_hwmgr *hwmgr) { if (PP_CAP(PHM_PlatformCaps_UVDPowerGating)) - return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_UVDPowerOFF); + return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_UVDPowerOFF, NULL); return 0; } @@ -1237,7 +1258,8 @@ static int smu8_dpm_powerup_uvd(struct pp_hwmgr *hwmgr) return smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_UVDPowerON, - PP_CAP(PHM_PlatformCaps_UVDDynamicPowerGating) ? 1 : 0); + PP_CAP(PHM_PlatformCaps_UVDDynamicPowerGating) ? 1 : 0, + NULL); } return 0; @@ -1259,15 +1281,20 @@ static int smu8_dpm_update_vce_dpm(struct pp_hwmgr *hwmgr) PPSMC_MSG_SetEclkHardMin, smu8_get_eclk_level(hwmgr, data->vce_dpm.hard_min_clk, - PPSMC_MSG_SetEclkHardMin)); + PPSMC_MSG_SetEclkHardMin), + NULL); } else { smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_SetEclkHardMin, 0); + PPSMC_MSG_SetEclkHardMin, + 0, + NULL); /* disable ECLK DPM 0. Otherwise VCE could hang if * switching SCLK from DPM 0 to 6/7 */ smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_SetEclkSoftMin, 1); + PPSMC_MSG_SetEclkSoftMin, + 1, + NULL); } return 0; } @@ -1276,7 +1303,8 @@ static int smu8_dpm_powerdown_vce(struct pp_hwmgr *hwmgr) { if (PP_CAP(PHM_PlatformCaps_VCEPowerGating)) return smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_VCEPowerOFF); + PPSMC_MSG_VCEPowerOFF, + NULL); return 0; } @@ -1284,7 +1312,8 @@ static int smu8_dpm_powerup_vce(struct pp_hwmgr *hwmgr) { if (PP_CAP(PHM_PlatformCaps_VCEPowerGating)) return smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_VCEPowerON); + PPSMC_MSG_VCEPowerON, + NULL); return 0; } @@ -1435,7 +1464,8 @@ static void smu8_hw_print_display_cfg( smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDisplaySizePowerParams, - data); + data, + NULL); } return 0; @@ -1497,10 +1527,12 @@ static int smu8_force_clock_level(struct pp_hwmgr *hwmgr, case PP_SCLK: smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSclkSoftMin, - mask); + mask, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSclkSoftMax, - mask); + mask, + NULL); break; default: break; @@ -1753,9 +1785,10 @@ static int smu8_read_sensor(struct pp_hwmgr *hwmgr, int idx, *((uint32_t *)value) = 0; return 0; case AMDGPU_PP_SENSOR_GPU_LOAD: - result = smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetAverageGraphicsActivity); + result = smum_send_msg_to_smc(hwmgr, + PPSMC_MSG_GetAverageGraphicsActivity, + &activity_percent); if (0 == result) { - activity_percent = cgs_read_register(hwmgr->device, mmSMU_MP1_SRBM2P_ARG_0); activity_percent = activity_percent > 100 ? 100 : activity_percent; } else { activity_percent = 50; @@ -1785,20 +1818,25 @@ static int smu8_notify_cac_buffer_info(struct pp_hwmgr *hwmgr, { smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DramAddrHiVirtual, - mc_addr_hi); + mc_addr_hi, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DramAddrLoVirtual, - mc_addr_low); + mc_addr_low, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DramAddrHiPhysical, - virtual_addr_hi); + virtual_addr_hi, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DramAddrLoPhysical, - virtual_addr_low); + virtual_addr_low, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DramBufferSize, - size); + size, + NULL); return 0; } @@ -1827,12 +1865,16 @@ static int smu8_enable_disable_uvd_dpm(struct pp_hwmgr *hwmgr, bool enable) data->dpm_flags |= DPMFlags_UVD_Enabled; dpm_features |= UVD_DPM_MASK; smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_EnableAllSmuFeatures, dpm_features); + PPSMC_MSG_EnableAllSmuFeatures, + dpm_features, + NULL); } else { dpm_features |= UVD_DPM_MASK; data->dpm_flags &= ~DPMFlags_UVD_Enabled; smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_DisableAllSmuFeatures, dpm_features); + PPSMC_MSG_DisableAllSmuFeatures, + dpm_features, + NULL); } return 0; } @@ -1854,7 +1896,8 @@ int smu8_dpm_update_uvd_dpm(struct pp_hwmgr *hwmgr, bool bgate) PPSMC_MSG_SetUvdHardMin, smu8_get_uvd_level(hwmgr, data->uvd_dpm.hard_min_clk, - PPSMC_MSG_SetUvdHardMin)); + PPSMC_MSG_SetUvdHardMin), + NULL); smu8_enable_disable_uvd_dpm(hwmgr, true); } else { @@ -1878,12 +1921,16 @@ static int smu8_enable_disable_vce_dpm(struct pp_hwmgr *hwmgr, bool enable) data->dpm_flags |= DPMFlags_VCE_Enabled; dpm_features |= VCE_DPM_MASK; smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_EnableAllSmuFeatures, dpm_features); + PPSMC_MSG_EnableAllSmuFeatures, + dpm_features, + NULL); } else { dpm_features |= VCE_DPM_MASK; data->dpm_flags &= ~DPMFlags_VCE_Enabled; smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_DisableAllSmuFeatures, dpm_features); + PPSMC_MSG_DisableAllSmuFeatures, + dpm_features, + NULL); } return 0; @@ -1898,9 +1945,9 @@ static void smu8_dpm_powergate_acp(struct pp_hwmgr *hwmgr, bool bgate) return; if (bgate) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ACPPowerOFF); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ACPPowerOFF, NULL); else - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ACPPowerON); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ACPPowerON, NULL); } static void smu8_dpm_powergate_uvd(struct pp_hwmgr *hwmgr, bool bgate) diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.c index d09690fca452..60b5ca974356 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.c @@ -22,6 +22,7 @@ */ #include <linux/pci.h> +#include <linux/reboot.h> #include "hwmgr.h" #include "pp_debug.h" @@ -557,7 +558,9 @@ void phm_apply_dal_min_voltage_request(struct pp_hwmgr *hwmgr) if (req_vddc <= vddc_table->entries[i].vddc) { req_volt = (((uint32_t)vddc_table->entries[i].vddc) * VOLTAGE_SCALE); smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_VddC_Request, req_volt); + PPSMC_MSG_VddC_Request, + req_volt, + NULL); return; } } @@ -593,37 +596,43 @@ int phm_irq_process(struct amdgpu_device *adev, uint32_t src_id = entry->src_id; if (client_id == AMDGPU_IRQ_CLIENTID_LEGACY) { - if (src_id == VISLANDS30_IV_SRCID_CG_TSS_THERMAL_LOW_TO_HIGH) - pr_warn("GPU over temperature range detected on PCIe %d:%d.%d!\n", - PCI_BUS_NUM(adev->pdev->devfn), - PCI_SLOT(adev->pdev->devfn), - PCI_FUNC(adev->pdev->devfn)); - else if (src_id == VISLANDS30_IV_SRCID_CG_TSS_THERMAL_HIGH_TO_LOW) - pr_warn("GPU under temperature range detected on PCIe %d:%d.%d!\n", - PCI_BUS_NUM(adev->pdev->devfn), - PCI_SLOT(adev->pdev->devfn), - PCI_FUNC(adev->pdev->devfn)); - else if (src_id == VISLANDS30_IV_SRCID_GPIO_19) - pr_warn("GPU Critical Temperature Fault detected on PCIe %d:%d.%d!\n", - PCI_BUS_NUM(adev->pdev->devfn), - PCI_SLOT(adev->pdev->devfn), - PCI_FUNC(adev->pdev->devfn)); + if (src_id == VISLANDS30_IV_SRCID_CG_TSS_THERMAL_LOW_TO_HIGH) { + dev_emerg(adev->dev, "ERROR: GPU over temperature range(SW CTF) detected!\n"); + /* + * SW CTF just occurred. + * Try to do a graceful shutdown to prevent further damage. + */ + dev_emerg(adev->dev, "ERROR: System is going to shutdown due to GPU SW CTF!\n"); + orderly_poweroff(true); + } else if (src_id == VISLANDS30_IV_SRCID_CG_TSS_THERMAL_HIGH_TO_LOW) + dev_emerg(adev->dev, "ERROR: GPU under temperature range detected!\n"); + else if (src_id == VISLANDS30_IV_SRCID_GPIO_19) { + dev_emerg(adev->dev, "ERROR: GPU HW Critical Temperature Fault(aka CTF) detected!\n"); + /* + * HW CTF just occurred. Shutdown to prevent further damage. + */ + dev_emerg(adev->dev, "ERROR: System is going to shutdown due to GPU HW CTF!\n"); + orderly_poweroff(true); + } } else if (client_id == SOC15_IH_CLIENTID_THM) { - if (src_id == 0) - pr_warn("GPU over temperature range detected on PCIe %d:%d.%d!\n", - PCI_BUS_NUM(adev->pdev->devfn), - PCI_SLOT(adev->pdev->devfn), - PCI_FUNC(adev->pdev->devfn)); - else - pr_warn("GPU under temperature range detected on PCIe %d:%d.%d!\n", - PCI_BUS_NUM(adev->pdev->devfn), - PCI_SLOT(adev->pdev->devfn), - PCI_FUNC(adev->pdev->devfn)); - } else if (client_id == SOC15_IH_CLIENTID_ROM_SMUIO) - pr_warn("GPU Critical Temperature Fault detected on PCIe %d:%d.%d!\n", - PCI_BUS_NUM(adev->pdev->devfn), - PCI_SLOT(adev->pdev->devfn), - PCI_FUNC(adev->pdev->devfn)); + if (src_id == 0) { + dev_emerg(adev->dev, "ERROR: GPU over temperature range(SW CTF) detected!\n"); + /* + * SW CTF just occurred. + * Try to do a graceful shutdown to prevent further damage. + */ + dev_emerg(adev->dev, "ERROR: System is going to shutdown due to GPU SW CTF!\n"); + orderly_poweroff(true); + } else + dev_emerg(adev->dev, "ERROR: GPU under temperature range detected!\n"); + } else if (client_id == SOC15_IH_CLIENTID_ROM_SMUIO) { + dev_emerg(adev->dev, "ERROR: GPU HW Critical Temperature Fault(aka CTF) detected!\n"); + /* + * HW CTF just occurred. Shutdown to prevent further damage. + */ + dev_emerg(adev->dev, "ERROR: System is going to shutdown due to GPU HW CTF!\n"); + orderly_poweroff(true); + } return 0; } diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.c index d168af4a4d78..46bb16c29cf6 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_baco.c @@ -98,7 +98,7 @@ int vega10_baco_set_state(struct pp_hwmgr *hwmgr, enum BACO_STATE state) if (state == BACO_STATE_IN) { if (soc15_baco_program_registers(hwmgr, pre_baco_tbl, ARRAY_SIZE(pre_baco_tbl))) { - if (smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnterBaco)) + if (smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnterBaco, NULL)) return -EINVAL; if (soc15_baco_program_registers(hwmgr, enter_baco_tbl, diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c index f29f95be1e56..675c7cab7cfc 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c @@ -484,8 +484,9 @@ static void vega10_init_dpm_defaults(struct pp_hwmgr *hwmgr) if (data->registry_data.vr0hot_enabled) data->smu_features[GNLD_VR0HOT].supported = true; - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetSmuVersion); - hwmgr->smu_version = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, + PPSMC_MSG_GetSmuVersion, + &hwmgr->smu_version); /* ACG firmware has major version 5 */ if ((hwmgr->smu_version & 0xff000000) == 0x5000000) data->smu_features[GNLD_ACG].supported = true; @@ -503,10 +504,8 @@ static void vega10_init_dpm_defaults(struct pp_hwmgr *hwmgr) data->smu_features[GNLD_PCC_LIMIT].supported = true; /* Get the SN to turn into a Unique ID */ - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ReadSerialNumTop32); - top32 = smum_get_argument(hwmgr); - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ReadSerialNumBottom32); - bottom32 = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ReadSerialNumTop32, &top32); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ReadSerialNumBottom32, &bottom32); adev->unique_id = ((uint64_t)bottom32 << 32) | top32; } @@ -993,7 +992,10 @@ static int vega10_setup_asic_task(struct pp_hwmgr *hwmgr) "Failed to set up led dpm config!", return -EINVAL); - smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_NumOfDisplays, 0); + smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_NumOfDisplays, + 0, + NULL); return 0; } @@ -2303,16 +2305,15 @@ static int vega10_acg_enable(struct pp_hwmgr *hwmgr) data->smu_features[GNLD_DPM_PREFETCHER].smu_feature_bitmap)) data->smu_features[GNLD_DPM_PREFETCHER].enabled = true; - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_InitializeAcg); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_InitializeAcg, NULL); - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_RunAcgBtc); - agc_btc_response = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_RunAcgBtc, &agc_btc_response); if (1 == agc_btc_response) { if (1 == data->acg_loop_state) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_RunAcgInClosedLoop); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_RunAcgInClosedLoop, NULL); else if (2 == data->acg_loop_state) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_RunAcgInOpenLoop); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_RunAcgInOpenLoop, NULL); if (0 == vega10_enable_smc_features(hwmgr, true, data->smu_features[GNLD_ACG].smu_feature_bitmap)) data->smu_features[GNLD_ACG].enabled = true; @@ -2429,11 +2430,9 @@ static int vega10_populate_and_upload_avfs_fuse_override(struct pp_hwmgr *hwmgr) struct vega10_hwmgr *data = hwmgr->backend; AvfsFuseOverride_t *avfs_fuse_table = &(data->smc_state_table.avfs_fuse_override_table); - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ReadSerialNumTop32); - top32 = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ReadSerialNumTop32, &top32); - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ReadSerialNumBottom32); - bottom32 = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ReadSerialNumBottom32, &bottom32); serial_number = ((uint64_t)bottom32 << 32) | top32; @@ -2610,14 +2609,16 @@ static int vega10_init_smc_table(struct pp_hwmgr *hwmgr) if (0 != boot_up_values.usVddc) { smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetFloorSocVoltage, - (boot_up_values.usVddc * 4)); + (boot_up_values.usVddc * 4), + NULL); data->vbios_boot_state.bsoc_vddc_lock = true; } else { data->vbios_boot_state.bsoc_vddc_lock = false; } smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetMinDeepSleepDcefclk, - (uint32_t)(data->vbios_boot_state.dcef_clock / 100)); + (uint32_t)(data->vbios_boot_state.dcef_clock / 100), + NULL); } result = vega10_populate_avfs_parameters(hwmgr); @@ -2904,7 +2905,8 @@ static int vega10_start_dpm(struct pp_hwmgr *hwmgr, uint32_t bitmap) if (data->vbios_boot_state.bsoc_vddc_lock) { smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_SetFloorSocVoltage, 0); + PPSMC_MSG_SetFloorSocVoltage, 0, + NULL); data->vbios_boot_state.bsoc_vddc_lock = false; } @@ -2947,7 +2949,8 @@ static int vega10_enable_dpm_tasks(struct pp_hwmgr *hwmgr) vega10_enable_disable_PCC_limit_feature(hwmgr, true); smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_ConfigureTelemetry, data->config_telemetry); + PPSMC_MSG_ConfigureTelemetry, data->config_telemetry, + NULL); tmp_result = vega10_construct_voltage_tables(hwmgr); PP_ASSERT_WITH_CODE(!tmp_result, @@ -3528,7 +3531,8 @@ static int vega10_upload_dpm_bootup_level(struct pp_hwmgr *hwmgr) data->dpm_table.gfx_table.dpm_state.soft_min_level) { smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMinGfxclkByIndex, - data->smc_state_table.gfx_boot_level); + data->smc_state_table.gfx_boot_level, + NULL); data->dpm_table.gfx_table.dpm_state.soft_min_level = data->smc_state_table.gfx_boot_level; @@ -3543,11 +3547,13 @@ static int vega10_upload_dpm_bootup_level(struct pp_hwmgr *hwmgr) socclk_idx = vega10_get_soc_index_for_max_uclk(hwmgr); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMinSocclkByIndex, - socclk_idx); + socclk_idx, + NULL); } else { smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMinUclkByIndex, - data->smc_state_table.mem_boot_level); + data->smc_state_table.mem_boot_level, + NULL); } data->dpm_table.mem_table.dpm_state.soft_min_level = data->smc_state_table.mem_boot_level; @@ -3562,7 +3568,8 @@ static int vega10_upload_dpm_bootup_level(struct pp_hwmgr *hwmgr) data->dpm_table.soc_table.dpm_state.soft_min_level) { smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMinSocclkByIndex, - data->smc_state_table.soc_boot_level); + data->smc_state_table.soc_boot_level, + NULL); data->dpm_table.soc_table.dpm_state.soft_min_level = data->smc_state_table.soc_boot_level; } @@ -3582,7 +3589,8 @@ static int vega10_upload_dpm_max_level(struct pp_hwmgr *hwmgr) data->dpm_table.gfx_table.dpm_state.soft_max_level) { smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxGfxclkByIndex, - data->smc_state_table.gfx_max_level); + data->smc_state_table.gfx_max_level, + NULL); data->dpm_table.gfx_table.dpm_state.soft_max_level = data->smc_state_table.gfx_max_level; } @@ -3593,7 +3601,8 @@ static int vega10_upload_dpm_max_level(struct pp_hwmgr *hwmgr) data->dpm_table.mem_table.dpm_state.soft_max_level) { smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxUclkByIndex, - data->smc_state_table.mem_max_level); + data->smc_state_table.mem_max_level, + NULL); data->dpm_table.mem_table.dpm_state.soft_max_level = data->smc_state_table.mem_max_level; } @@ -3607,7 +3616,8 @@ static int vega10_upload_dpm_max_level(struct pp_hwmgr *hwmgr) data->dpm_table.soc_table.dpm_state.soft_max_level) { smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMaxSocclkByIndex, - data->smc_state_table.soc_max_level); + data->smc_state_table.soc_max_level, + NULL); data->dpm_table.soc_table.dpm_state.soft_max_level = data->smc_state_table.soc_max_level; } @@ -3694,7 +3704,8 @@ static int vega10_update_sclk_threshold(struct pp_hwmgr *hwmgr) /* This message will also enable SmcToHost Interrupt */ smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetLowGfxclkInterruptThreshold, - (uint32_t)low_sclk_interrupt_threshold); + (uint32_t)low_sclk_interrupt_threshold, + NULL); } return 0; @@ -3801,8 +3812,7 @@ static int vega10_get_gpu_power(struct pp_hwmgr *hwmgr, if (!query) return -EINVAL; - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetCurrPkgPwr); - value = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetCurrPkgPwr, &value); /* SMC returning actual watts, keep consistent with legacy asics, low 8 bit as 8 fractional bits */ *query = value << 8; @@ -3822,13 +3832,11 @@ static int vega10_read_sensor(struct pp_hwmgr *hwmgr, int idx, switch (idx) { case AMDGPU_PP_SENSOR_GFX_SCLK: - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetAverageGfxclkActualFrequency); - sclk_mhz = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetAverageGfxclkActualFrequency, &sclk_mhz); *((uint32_t *)value) = sclk_mhz * 100; break; case AMDGPU_PP_SENSOR_GFX_MCLK: - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetCurrentUclkIndex); - mclk_idx = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetCurrentUclkIndex, &mclk_idx); if (mclk_idx < dpm_table->mem_table.count) { *((uint32_t *)value) = dpm_table->mem_table.dpm_levels[mclk_idx].value; *size = 4; @@ -3837,8 +3845,8 @@ static int vega10_read_sensor(struct pp_hwmgr *hwmgr, int idx, } break; case AMDGPU_PP_SENSOR_GPU_LOAD: - smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetAverageGfxActivity, 0); - activity_percent = smum_get_argument(hwmgr); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetAverageGfxActivity, 0, + &activity_percent); *((uint32_t *)value) = activity_percent > 100 ? 100 : activity_percent; *size = 4; break; @@ -3847,14 +3855,14 @@ static int vega10_read_sensor(struct pp_hwmgr *hwmgr, int idx, *size = 4; break; case AMDGPU_PP_SENSOR_HOTSPOT_TEMP: - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetTemperatureHotspot); - *((uint32_t *)value) = smum_get_argument(hwmgr) * + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetTemperatureHotspot, (uint32_t *)value); + *((uint32_t *)value) = *((uint32_t *)value) * PP_TEMPERATURE_UNITS_PER_CENTIGRADES; *size = 4; break; case AMDGPU_PP_SENSOR_MEM_TEMP: - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetTemperatureHBM); - *((uint32_t *)value) = smum_get_argument(hwmgr) * + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetTemperatureHBM, (uint32_t *)value); + *((uint32_t *)value) = *((uint32_t *)value) * PP_TEMPERATURE_UNITS_PER_CENTIGRADES; *size = 4; break; @@ -3893,7 +3901,8 @@ static void vega10_notify_smc_display_change(struct pp_hwmgr *hwmgr, { smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetUclkFastSwitch, - has_disp ? 1 : 0); + has_disp ? 1 : 0, + NULL); } int vega10_display_clock_voltage_request(struct pp_hwmgr *hwmgr, @@ -3928,7 +3937,8 @@ int vega10_display_clock_voltage_request(struct pp_hwmgr *hwmgr, clk_request = (clk_freq << 16) | clk_select; smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_RequestDisplayClockByFreq, - clk_request); + clk_request, + NULL); } return result; @@ -3990,7 +4000,8 @@ static int vega10_notify_smc_display_config_after_ps_adjustment( if (!vega10_display_clock_voltage_request(hwmgr, &clock_req)) { smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetMinDeepSleepDcefclk, - min_clocks.dcefClockInSR / 100); + min_clocks.dcefClockInSR / 100, + NULL); } else { pr_info("Attempt to set Hard Min for DCEFCLK Failed!"); } @@ -4000,7 +4011,8 @@ static int vega10_notify_smc_display_config_after_ps_adjustment( if (min_clocks.memoryClock != 0) { idx = vega10_get_uclk_index(hwmgr, mclk_table, min_clocks.memoryClock); - smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMinUclkByIndex, idx); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMinUclkByIndex, idx, + NULL); data->dpm_table.mem_table.dpm_state.soft_min_level= idx; } @@ -4541,8 +4553,7 @@ static int vega10_print_clock_levels(struct pp_hwmgr *hwmgr, if (data->registry_data.sclk_dpm_key_disabled) break; - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetCurrentGfxclkIndex); - now = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetCurrentGfxclkIndex, &now); if (hwmgr->pp_one_vf && (hwmgr->dpm_level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK)) @@ -4558,8 +4569,7 @@ static int vega10_print_clock_levels(struct pp_hwmgr *hwmgr, if (data->registry_data.mclk_dpm_key_disabled) break; - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetCurrentUclkIndex); - now = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetCurrentUclkIndex, &now); for (i = 0; i < mclk_table->count; i++) size += sprintf(buf + size, "%d: %uMhz %s\n", @@ -4570,8 +4580,7 @@ static int vega10_print_clock_levels(struct pp_hwmgr *hwmgr, if (data->registry_data.socclk_dpm_key_disabled) break; - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetCurrentSocclkIndex); - now = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetCurrentSocclkIndex, &now); for (i = 0; i < soc_table->count; i++) size += sprintf(buf + size, "%d: %uMhz %s\n", @@ -4583,8 +4592,7 @@ static int vega10_print_clock_levels(struct pp_hwmgr *hwmgr, break; smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_GetClockFreqMHz, CLK_DCEFCLK); - now = smum_get_argument(hwmgr); + PPSMC_MSG_GetClockFreqMHz, CLK_DCEFCLK, &now); for (i = 0; i < dcef_table->count; i++) size += sprintf(buf + size, "%d: %uMhz %s\n", @@ -4593,8 +4601,7 @@ static int vega10_print_clock_levels(struct pp_hwmgr *hwmgr, "*" : ""); break; case PP_PCIE: - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetCurrentLinkIndex); - now = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetCurrentLinkIndex, &now); for (i = 0; i < pcie_table->count; i++) size += sprintf(buf + size, "%d: %s %s\n", i, @@ -4658,7 +4665,8 @@ static int vega10_display_configuration_changed_task(struct pp_hwmgr *hwmgr) if (data->water_marks_bitmap & WaterMarksLoaded) { smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_NumOfDisplays, hwmgr->display_config->num_display); + PPSMC_MSG_NumOfDisplays, hwmgr->display_config->num_display, + NULL); } return result; @@ -4924,21 +4932,26 @@ static int vega10_notify_cac_buffer_info(struct pp_hwmgr *hwmgr, { smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSystemVirtualDramAddrHigh, - virtual_addr_hi); + virtual_addr_hi, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSystemVirtualDramAddrLow, - virtual_addr_low); + virtual_addr_low, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DramLogSetDramAddrHigh, - mc_addr_hi); + mc_addr_hi, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DramLogSetDramAddrLow, - mc_addr_low); + mc_addr_low, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DramLogSetDramSize, - size); + size, + NULL); return 0; } @@ -5040,12 +5053,14 @@ static int vega10_set_power_profile_mode(struct pp_hwmgr *hwmgr, long *input, ui smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetCustomGfxDpmParameters, busy_set_point | FPS<<8 | - use_rlc_busy << 16 | min_active_level<<24); + use_rlc_busy << 16 | min_active_level<<24, + NULL); } out: smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetWorkloadMask, - 1 << power_profile_mode); + 1 << power_profile_mode, + NULL); hwmgr->power_profile_mode = power_profile_mode; return 0; @@ -5302,7 +5317,7 @@ static int vega10_set_mp1_state(struct pp_hwmgr *hwmgr, return 0; } - PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc(hwmgr, msg)) == 0, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc(hwmgr, msg, NULL)) == 0, "[PrepareMp1] Failed!", return ret); diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_powertune.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_powertune.c index 0a677d4bc87b..9757d47dd6b8 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_powertune.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_powertune.c @@ -651,18 +651,6 @@ static const struct vega10_didt_config_reg PSMSEEDCStallDelayConfig_Vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg PSMSEEDCThresholdConfig_Vega10[] = -{ -/* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- - * Offset Mask Shift Value - * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- - */ - /* SQ EDC THRESHOLD */ - { ixDIDT_SQ_EDC_THRESHOLD, DIDT_SQ_EDC_THRESHOLD__EDC_THRESHOLD_MASK, DIDT_SQ_EDC_THRESHOLD__EDC_THRESHOLD__SHIFT, 0x0000 }, - - { 0xFFFFFFFF } /* End of list */ -}; - static const struct vega10_didt_config_reg PSMSEEDCCtrlResetConfig_Vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -707,17 +695,6 @@ static const struct vega10_didt_config_reg PSMSEEDCCtrlConfig_Vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg PSMGCEDCThresholdConfig_vega10[] = -{ -/* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- - * Offset Mask Shift Value - * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- - */ - { mmGC_EDC_THRESHOLD, GC_EDC_THRESHOLD__EDC_THRESHOLD_MASK, GC_EDC_THRESHOLD__EDC_THRESHOLD__SHIFT, 0x0000000 }, - - { 0xFFFFFFFF } /* End of list */ -}; - static const struct vega10_didt_config_reg PSMGCEDCDroopCtrlConfig_vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -925,7 +902,8 @@ static void vega10_didt_set_mask(struct pp_hwmgr *hwmgr, const bool enable) /* For Vega10, SMC does not support any mask yet. */ if (enable) - smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_ConfigureGfxDidt, didt_block_info); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_ConfigureGfxDidt, didt_block_info, + NULL); } @@ -1327,7 +1305,8 @@ int vega10_set_power_limit(struct pp_hwmgr *hwmgr, uint32_t n) if (data->registry_data.enable_pkg_pwr_tracking_feature) smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_SetPptLimit, n); + PPSMC_MSG_SetPptLimit, n, + NULL); return 0; } @@ -1393,7 +1372,8 @@ static void vega10_set_overdrive_target_percentage(struct pp_hwmgr *hwmgr, uint32_t adjust_percent) { smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_OverDriveSetPercentage, adjust_percent); + PPSMC_MSG_OverDriveSetPercentage, adjust_percent, + NULL); } int vega10_power_control_set_level(struct pp_hwmgr *hwmgr) diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.c index ba8763daa380..7783c7fd7ccb 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_thermal.c @@ -31,8 +31,7 @@ static int vega10_get_current_rpm(struct pp_hwmgr *hwmgr, uint32_t *current_rpm) { - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetCurrentRpm); - *current_rpm = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetCurrentRpm, current_rpm); return 0; } @@ -520,7 +519,8 @@ int vega10_thermal_setup_fan_table(struct pp_hwmgr *hwmgr) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetFanTemperatureTarget, - (uint32_t)table->FanTargetTemperature); + (uint32_t)table->FanTargetTemperature, + NULL); table->FanPwmMin = hwmgr->thermal_controller. advanceFanControlParameters.usPWMMin * 255 / 100; diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_baco.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_baco.c index 9d8ca94a8f0c..bc53cce4f32d 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_baco.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_baco.c @@ -96,7 +96,7 @@ int vega12_baco_set_state(struct pp_hwmgr *hwmgr, enum BACO_STATE state) if (state == BACO_STATE_IN) { if (soc15_baco_program_registers(hwmgr, pre_baco_tbl, ARRAY_SIZE(pre_baco_tbl))) { - if (smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_EnterBaco, 0)) + if (smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_EnterBaco, 0, NULL)) return -EINVAL; if (soc15_baco_program_registers(hwmgr, enter_baco_tbl, diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c index aca61d1ff3c2..f4d1692cccf3 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c @@ -357,10 +357,8 @@ static void vega12_init_dpm_defaults(struct pp_hwmgr *hwmgr) } /* Get the SN to turn into a Unique ID */ - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ReadSerialNumTop32); - top32 = smum_get_argument(hwmgr); - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ReadSerialNumBottom32); - bottom32 = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ReadSerialNumTop32, &top32); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ReadSerialNumBottom32, &bottom32); adev->unique_id = ((uint64_t)bottom32 << 32) | top32; } @@ -483,16 +481,12 @@ static int vega12_get_number_of_dpm_level(struct pp_hwmgr *hwmgr, ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetDpmFreqByIndex, - (clk_id << 16 | 0xFF)); + (clk_id << 16 | 0xFF), + num_of_levels); PP_ASSERT_WITH_CODE(!ret, "[GetNumOfDpmLevel] failed to get dpm levels!", return ret); - *num_of_levels = smum_get_argument(hwmgr); - PP_ASSERT_WITH_CODE(*num_of_levels > 0, - "[GetNumOfDpmLevel] number of clk levels is invalid!", - return -EINVAL); - return ret; } @@ -504,12 +498,11 @@ static int vega12_get_dpm_frequency_by_index(struct pp_hwmgr *hwmgr, *Lower 16 bits specify the level */ PP_ASSERT_WITH_CODE(smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_GetDpmFreqByIndex, (clkID << 16 | index)) == 0, + PPSMC_MSG_GetDpmFreqByIndex, (clkID << 16 | index), + clock) == 0, "[GetDpmFrequencyByIndex] Failed to get dpm frequency from SMU!", return -EINVAL); - *clock = smum_get_argument(hwmgr); - return 0; } @@ -749,7 +742,8 @@ static int vega12_init_smc_table(struct pp_hwmgr *hwmgr) data->vbios_boot_state.vclock = boot_up_values.ulVClk; smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetMinDeepSleepDcefclk, - (uint32_t)(data->vbios_boot_state.dcef_clock / 100)); + (uint32_t)(data->vbios_boot_state.dcef_clock / 100), + NULL); } memcpy(pp_table, pptable_information->smc_pptable, sizeof(PPTable_t)); @@ -767,11 +761,10 @@ static int vega12_run_acg_btc(struct pp_hwmgr *hwmgr) uint32_t result; PP_ASSERT_WITH_CODE( - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_RunAcgBtc) == 0, + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_RunAcgBtc, &result) == 0, "[Run_ACG_BTC] Attempt to run ACG BTC failed!", return -EINVAL); - result = smum_get_argument(hwmgr); PP_ASSERT_WITH_CODE(result == 1, "Failed to run ACG BTC!", return -EINVAL); @@ -792,12 +785,14 @@ static int vega12_set_allowed_featuresmask(struct pp_hwmgr *hwmgr) (allowed_features_low |= ((data->smu_features[i].smu_feature_bitmap >> SMU_FEATURES_LOW_SHIFT) & 0xFFFFFFFF)); PP_ASSERT_WITH_CODE( - smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetAllowedFeaturesMaskHigh, allowed_features_high) == 0, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetAllowedFeaturesMaskHigh, allowed_features_high, + NULL) == 0, "[SetAllowedFeaturesMask] Attempt to set allowed features mask (high) failed!", return -1); PP_ASSERT_WITH_CODE( - smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetAllowedFeaturesMaskLow, allowed_features_low) == 0, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetAllowedFeaturesMaskLow, allowed_features_low, + NULL) == 0, "[SetAllowedFeaturesMask] Attempt to set allowed features mask (low) failed!", return -1); @@ -828,7 +823,7 @@ static int vega12_enable_all_smu_features(struct pp_hwmgr *hwmgr) bool enabled; PP_ASSERT_WITH_CODE( - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnableAllSmuFeatures) == 0, + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnableAllSmuFeatures, NULL) == 0, "[EnableAllSMUFeatures] Failed to enable all smu features!", return -1); @@ -854,7 +849,7 @@ static int vega12_disable_all_smu_features(struct pp_hwmgr *hwmgr) bool enabled; PP_ASSERT_WITH_CODE( - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_DisableAllSmuFeatures) == 0, + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_DisableAllSmuFeatures, NULL) == 0, "[DisableAllSMUFeatures] Failed to disable all smu features!", return -1); @@ -879,7 +874,8 @@ static int vega12_set_overdrive_target_percentage(struct pp_hwmgr *hwmgr, uint32_t adjust_percent) { return smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_OverDriveSetPercentage, adjust_percent); + PPSMC_MSG_OverDriveSetPercentage, adjust_percent, + NULL); } static int vega12_power_control_set_level(struct pp_hwmgr *hwmgr) @@ -902,24 +898,24 @@ static int vega12_get_all_clock_ranges_helper(struct pp_hwmgr *hwmgr, { /* AC Max */ PP_ASSERT_WITH_CODE( - smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetMaxDpmFreq, (clkid << 16)) == 0, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetMaxDpmFreq, (clkid << 16), + &(clock->ACMax)) == 0, "[GetClockRanges] Failed to get max ac clock from SMC!", return -EINVAL); - clock->ACMax = smum_get_argument(hwmgr); /* AC Min */ PP_ASSERT_WITH_CODE( - smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetMinDpmFreq, (clkid << 16)) == 0, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetMinDpmFreq, (clkid << 16), + &(clock->ACMin)) == 0, "[GetClockRanges] Failed to get min ac clock from SMC!", return -EINVAL); - clock->ACMin = smum_get_argument(hwmgr); /* DC Max */ PP_ASSERT_WITH_CODE( - smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetDcModeMaxDpmFreq, (clkid << 16)) == 0, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetDcModeMaxDpmFreq, (clkid << 16), + &(clock->DCMax)) == 0, "[GetClockRanges] Failed to get max dc clock from SMC!", return -EINVAL); - clock->DCMax = smum_get_argument(hwmgr); return 0; } @@ -944,7 +940,7 @@ static int vega12_enable_dpm_tasks(struct pp_hwmgr *hwmgr) int tmp_result, result = 0; smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_NumOfDisplays, 0); + PPSMC_MSG_NumOfDisplays, 0, NULL); result = vega12_set_allowed_featuresmask(hwmgr); PP_ASSERT_WITH_CODE(result == 0, @@ -1043,7 +1039,8 @@ static int vega12_upload_dpm_min_level(struct pp_hwmgr *hwmgr) min_freq = data->dpm_table.gfx_table.dpm_state.soft_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMinByFreq, - (PPCLK_GFXCLK << 16) | (min_freq & 0xffff))), + (PPCLK_GFXCLK << 16) | (min_freq & 0xffff), + NULL)), "Failed to set soft min gfxclk !", return ret); } @@ -1052,14 +1049,16 @@ static int vega12_upload_dpm_min_level(struct pp_hwmgr *hwmgr) min_freq = data->dpm_table.mem_table.dpm_state.soft_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMinByFreq, - (PPCLK_UCLK << 16) | (min_freq & 0xffff))), + (PPCLK_UCLK << 16) | (min_freq & 0xffff), + NULL)), "Failed to set soft min memclk !", return ret); min_freq = data->dpm_table.mem_table.dpm_state.hard_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetHardMinByFreq, - (PPCLK_UCLK << 16) | (min_freq & 0xffff))), + (PPCLK_UCLK << 16) | (min_freq & 0xffff), + NULL)), "Failed to set hard min memclk !", return ret); } @@ -1069,7 +1068,8 @@ static int vega12_upload_dpm_min_level(struct pp_hwmgr *hwmgr) PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMinByFreq, - (PPCLK_VCLK << 16) | (min_freq & 0xffff))), + (PPCLK_VCLK << 16) | (min_freq & 0xffff), + NULL)), "Failed to set soft min vclk!", return ret); @@ -1077,7 +1077,8 @@ static int vega12_upload_dpm_min_level(struct pp_hwmgr *hwmgr) PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMinByFreq, - (PPCLK_DCLK << 16) | (min_freq & 0xffff))), + (PPCLK_DCLK << 16) | (min_freq & 0xffff), + NULL)), "Failed to set soft min dclk!", return ret); } @@ -1087,7 +1088,8 @@ static int vega12_upload_dpm_min_level(struct pp_hwmgr *hwmgr) PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMinByFreq, - (PPCLK_ECLK << 16) | (min_freq & 0xffff))), + (PPCLK_ECLK << 16) | (min_freq & 0xffff), + NULL)), "Failed to set soft min eclk!", return ret); } @@ -1097,7 +1099,8 @@ static int vega12_upload_dpm_min_level(struct pp_hwmgr *hwmgr) PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMinByFreq, - (PPCLK_SOCCLK << 16) | (min_freq & 0xffff))), + (PPCLK_SOCCLK << 16) | (min_freq & 0xffff), + NULL)), "Failed to set soft min socclk!", return ret); } @@ -1107,7 +1110,8 @@ static int vega12_upload_dpm_min_level(struct pp_hwmgr *hwmgr) PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetHardMinByFreq, - (PPCLK_DCEFCLK << 16) | (min_freq & 0xffff))), + (PPCLK_DCEFCLK << 16) | (min_freq & 0xffff), + NULL)), "Failed to set hard min dcefclk!", return ret); } @@ -1127,7 +1131,8 @@ static int vega12_upload_dpm_max_level(struct pp_hwmgr *hwmgr) PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMaxByFreq, - (PPCLK_GFXCLK << 16) | (max_freq & 0xffff))), + (PPCLK_GFXCLK << 16) | (max_freq & 0xffff), + NULL)), "Failed to set soft max gfxclk!", return ret); } @@ -1137,7 +1142,8 @@ static int vega12_upload_dpm_max_level(struct pp_hwmgr *hwmgr) PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMaxByFreq, - (PPCLK_UCLK << 16) | (max_freq & 0xffff))), + (PPCLK_UCLK << 16) | (max_freq & 0xffff), + NULL)), "Failed to set soft max memclk!", return ret); } @@ -1147,14 +1153,16 @@ static int vega12_upload_dpm_max_level(struct pp_hwmgr *hwmgr) PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMaxByFreq, - (PPCLK_VCLK << 16) | (max_freq & 0xffff))), + (PPCLK_VCLK << 16) | (max_freq & 0xffff), + NULL)), "Failed to set soft max vclk!", return ret); max_freq = data->dpm_table.dclk_table.dpm_state.soft_max_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMaxByFreq, - (PPCLK_DCLK << 16) | (max_freq & 0xffff))), + (PPCLK_DCLK << 16) | (max_freq & 0xffff), + NULL)), "Failed to set soft max dclk!", return ret); } @@ -1164,7 +1172,8 @@ static int vega12_upload_dpm_max_level(struct pp_hwmgr *hwmgr) PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMaxByFreq, - (PPCLK_ECLK << 16) | (max_freq & 0xffff))), + (PPCLK_ECLK << 16) | (max_freq & 0xffff), + NULL)), "Failed to set soft max eclk!", return ret); } @@ -1174,7 +1183,8 @@ static int vega12_upload_dpm_max_level(struct pp_hwmgr *hwmgr) PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMaxByFreq, - (PPCLK_SOCCLK << 16) | (max_freq & 0xffff))), + (PPCLK_SOCCLK << 16) | (max_freq & 0xffff), + NULL)), "Failed to set soft max socclk!", return ret); } @@ -1287,10 +1297,10 @@ static int vega12_get_current_gfx_clk_freq(struct pp_hwmgr *hwmgr, uint32_t *gfx *gfx_freq = 0; PP_ASSERT_WITH_CODE(smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_GetDpmClockFreq, (PPCLK_GFXCLK << 16)) == 0, + PPSMC_MSG_GetDpmClockFreq, (PPCLK_GFXCLK << 16), + &gfx_clk) == 0, "[GetCurrentGfxClkFreq] Attempt to get Current GFXCLK Frequency Failed!", return -EINVAL); - gfx_clk = smum_get_argument(hwmgr); *gfx_freq = gfx_clk * 100; @@ -1304,10 +1314,10 @@ static int vega12_get_current_mclk_freq(struct pp_hwmgr *hwmgr, uint32_t *mclk_f *mclk_freq = 0; PP_ASSERT_WITH_CODE( - smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetDpmClockFreq, (PPCLK_UCLK << 16)) == 0, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetDpmClockFreq, (PPCLK_UCLK << 16), + &mem_clk) == 0, "[GetCurrentMClkFreq] Attempt to get Current MCLK Frequency Failed!", return -EINVAL); - mem_clk = smum_get_argument(hwmgr); *mclk_freq = mem_clk * 100; @@ -1420,7 +1430,8 @@ static int vega12_notify_smc_display_change(struct pp_hwmgr *hwmgr, if (data->smu_features[GNLD_DPM_UCLK].enabled) return smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetUclkFastSwitch, - has_disp ? 1 : 0); + has_disp ? 1 : 0, + NULL); return 0; } @@ -1459,7 +1470,8 @@ int vega12_display_clock_voltage_request(struct pp_hwmgr *hwmgr, clk_request = (clk_select << 16) | clk_freq; result = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinByFreq, - clk_request); + clk_request, + NULL); } } @@ -1493,7 +1505,8 @@ static int vega12_notify_smc_display_config_after_ps_adjustment( PP_ASSERT_WITH_CODE( !smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetMinDeepSleepDcefclk, - min_clocks.dcefClockInSR /100), + min_clocks.dcefClockInSR /100, + NULL), "Attempt to set divider for DCEFCLK Failed!", return -1); } else { @@ -2124,10 +2137,10 @@ static int vega12_print_clock_levels(struct pp_hwmgr *hwmgr, case PP_SOCCLK: PP_ASSERT_WITH_CODE( smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_GetDpmClockFreq, (PPCLK_SOCCLK << 16)) == 0, + PPSMC_MSG_GetDpmClockFreq, (PPCLK_SOCCLK << 16), + &now) == 0, "Attempt to get Current SOCCLK Frequency Failed!", return -EINVAL); - now = smum_get_argument(hwmgr); PP_ASSERT_WITH_CODE( vega12_get_socclocks(hwmgr, &clocks) == 0, @@ -2142,10 +2155,10 @@ static int vega12_print_clock_levels(struct pp_hwmgr *hwmgr, case PP_DCEFCLK: PP_ASSERT_WITH_CODE( smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_GetDpmClockFreq, (PPCLK_DCEFCLK << 16)) == 0, + PPSMC_MSG_GetDpmClockFreq, (PPCLK_DCEFCLK << 16), + &now) == 0, "Attempt to get Current DCEFCLK Frequency Failed!", return -EINVAL); - now = smum_get_argument(hwmgr); PP_ASSERT_WITH_CODE( vega12_get_dcefclocks(hwmgr, &clocks) == 0, @@ -2343,7 +2356,8 @@ static int vega12_set_uclk_to_highest_dpm_level(struct pp_hwmgr *hwmgr, dpm_table->dpm_state.hard_min_level = dpm_table->dpm_levels[dpm_table->count - 1].value; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinByFreq, - (PPCLK_UCLK << 16 ) | dpm_table->dpm_state.hard_min_level)), + (PPCLK_UCLK << 16 ) | dpm_table->dpm_state.hard_min_level, + NULL)), "[SetUclkToHightestDpmLevel] Set hard min uclk failed!", return ret); } @@ -2357,7 +2371,8 @@ static int vega12_pre_display_configuration_changed_task(struct pp_hwmgr *hwmgr) int ret = 0; smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_NumOfDisplays, 0); + PPSMC_MSG_NumOfDisplays, 0, + NULL); ret = vega12_set_uclk_to_highest_dpm_level(hwmgr, &data->dpm_table.mem_table); @@ -2383,7 +2398,8 @@ static int vega12_display_configuration_changed_task(struct pp_hwmgr *hwmgr) data->smu_features[GNLD_DPM_DCEFCLK].supported && data->smu_features[GNLD_DPM_SOCCLK].supported) smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_NumOfDisplays, hwmgr->display_config->num_display); + PPSMC_MSG_NumOfDisplays, hwmgr->display_config->num_display, + NULL); return result; } @@ -2555,21 +2571,26 @@ static int vega12_notify_cac_buffer_info(struct pp_hwmgr *hwmgr, { smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSystemVirtualDramAddrHigh, - virtual_addr_hi); + virtual_addr_hi, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSystemVirtualDramAddrLow, - virtual_addr_low); + virtual_addr_low, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DramLogSetDramAddrHigh, - mc_addr_hi); + mc_addr_hi, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DramLogSetDramAddrLow, - mc_addr_low); + mc_addr_low, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DramLogSetDramSize, - size); + size, + NULL); return 0; } @@ -2605,7 +2626,7 @@ static int vega12_enable_gfx_off(struct pp_hwmgr *hwmgr) int ret = 0; if (data->gfxoff_controlled_by_driver) - ret = smum_send_msg_to_smc(hwmgr, PPSMC_MSG_AllowGfxOff); + ret = smum_send_msg_to_smc(hwmgr, PPSMC_MSG_AllowGfxOff, NULL); return ret; } @@ -2617,7 +2638,7 @@ static int vega12_disable_gfx_off(struct pp_hwmgr *hwmgr) int ret = 0; if (data->gfxoff_controlled_by_driver) - ret = smum_send_msg_to_smc(hwmgr, PPSMC_MSG_DisallowGfxOff); + ret = smum_send_msg_to_smc(hwmgr, PPSMC_MSG_DisallowGfxOff, NULL); return ret; } @@ -2654,7 +2675,7 @@ static int vega12_set_mp1_state(struct pp_hwmgr *hwmgr, return 0; } - PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc(hwmgr, msg)) == 0, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc(hwmgr, msg, NULL)) == 0, "[PrepareMp1] Failed!", return ret); diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_thermal.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_thermal.c index 904eb2c9155b..c85806a6f62e 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_thermal.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_thermal.c @@ -32,10 +32,10 @@ static int vega12_get_current_rpm(struct pp_hwmgr *hwmgr, uint32_t *current_rpm) { PP_ASSERT_WITH_CODE(!smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_GetCurrentRpm), + PPSMC_MSG_GetCurrentRpm, + current_rpm), "Attempt to get current RPM from SMC Failed!", return -EINVAL); - *current_rpm = smum_get_argument(hwmgr); return 0; } @@ -259,7 +259,8 @@ int vega12_thermal_setup_fan_table(struct pp_hwmgr *hwmgr) ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetFanTemperatureTarget, - (uint32_t)table->FanTargetTemperature); + (uint32_t)table->FanTargetTemperature, + NULL); return ret; } diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.c index 9b5e72bdceca..2a28c9df15a0 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_baco.c @@ -91,16 +91,16 @@ int vega20_baco_set_state(struct pp_hwmgr *hwmgr, enum BACO_STATE state) WREG32_SOC15(THM, 0, mmTHM_BACO_CNTL, data); if(smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_EnterBaco, 0)) + PPSMC_MSG_EnterBaco, 0, NULL)) return -EINVAL; } else { if(smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_EnterBaco, 1)) + PPSMC_MSG_EnterBaco, 1, NULL)) return -EINVAL; } } else if (state == BACO_STATE_OUT) { - if (smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ExitBaco)) + if (smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ExitBaco, NULL)) return -EINVAL; if (!soc15_baco_program_registers(hwmgr, clean_baco_tbl, ARRAY_SIZE(clean_baco_tbl))) @@ -118,5 +118,5 @@ int vega20_baco_apply_vdci_flush_workaround(struct pp_hwmgr *hwmgr) if (ret) return ret; - return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_BacoWorkAroundFlushVDCI); + return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_BacoWorkAroundFlushVDCI, NULL); } diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c index 08b6ba39a6d7..9ff470f1b826 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c @@ -92,8 +92,7 @@ static void vega20_set_default_registry_data(struct pp_hwmgr *hwmgr) */ data->registry_data.disallowed_features = 0xE0041C00; /* ECC feature should be disabled on old SMUs */ - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetSmuVersion); - hwmgr->smu_version = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetSmuVersion, &hwmgr->smu_version); if (hwmgr->smu_version < 0x282100) data->registry_data.disallowed_features |= FEATURE_ECC_MASK; @@ -400,10 +399,8 @@ static void vega20_init_dpm_defaults(struct pp_hwmgr *hwmgr) } /* Get the SN to turn into a Unique ID */ - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ReadSerialNumTop32); - top32 = smum_get_argument(hwmgr); - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ReadSerialNumBottom32); - bottom32 = smum_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ReadSerialNumTop32, &top32); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ReadSerialNumBottom32, &bottom32); adev->unique_id = ((uint64_t)bottom32 << 32) | top32; } @@ -527,16 +524,12 @@ static int vega20_get_number_of_dpm_level(struct pp_hwmgr *hwmgr, ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetDpmFreqByIndex, - (clk_id << 16 | 0xFF)); + (clk_id << 16 | 0xFF), + num_of_levels); PP_ASSERT_WITH_CODE(!ret, "[GetNumOfDpmLevel] failed to get dpm levels!", return ret); - *num_of_levels = smum_get_argument(hwmgr); - PP_ASSERT_WITH_CODE(*num_of_levels > 0, - "[GetNumOfDpmLevel] number of clk levels is invalid!", - return -EINVAL); - return ret; } @@ -547,16 +540,12 @@ static int vega20_get_dpm_frequency_by_index(struct pp_hwmgr *hwmgr, ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetDpmFreqByIndex, - (clk_id << 16 | index)); + (clk_id << 16 | index), + clk); PP_ASSERT_WITH_CODE(!ret, "[GetDpmFreqByIndex] failed to get dpm freq by index!", return ret); - *clk = smum_get_argument(hwmgr); - PP_ASSERT_WITH_CODE(*clk, - "[GetDpmFreqByIndex] clk value is invalid!", - return -EINVAL); - return ret; } @@ -813,7 +802,8 @@ static int vega20_init_smc_table(struct pp_hwmgr *hwmgr) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetMinDeepSleepDcefclk, - (uint32_t)(data->vbios_boot_state.dcef_clock / 100)); + (uint32_t)(data->vbios_boot_state.dcef_clock / 100), + NULL); memcpy(pp_table, pptable_information->smc_pptable, sizeof(PPTable_t)); @@ -868,7 +858,8 @@ static int vega20_override_pcie_parameters(struct pp_hwmgr *hwmgr) */ smu_pcie_arg = (1 << 16) | (pcie_gen << 8) | pcie_width; ret = smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_OverridePcieParameters, smu_pcie_arg); + PPSMC_MSG_OverridePcieParameters, smu_pcie_arg, + NULL); PP_ASSERT_WITH_CODE(!ret, "[OverridePcieParameters] Attempt to override pcie params failed!", return ret); @@ -899,13 +890,13 @@ static int vega20_set_allowed_featuresmask(struct pp_hwmgr *hwmgr) & 0xFFFFFFFF)); ret = smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_SetAllowedFeaturesMaskHigh, allowed_features_high); + PPSMC_MSG_SetAllowedFeaturesMaskHigh, allowed_features_high, NULL); PP_ASSERT_WITH_CODE(!ret, "[SetAllowedFeaturesMask] Attempt to set allowed features mask(high) failed!", return ret); ret = smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_SetAllowedFeaturesMaskLow, allowed_features_low); + PPSMC_MSG_SetAllowedFeaturesMaskLow, allowed_features_low, NULL); PP_ASSERT_WITH_CODE(!ret, "[SetAllowedFeaturesMask] Attempt to set allowed features mask (low) failed!", return ret); @@ -915,12 +906,12 @@ static int vega20_set_allowed_featuresmask(struct pp_hwmgr *hwmgr) static int vega20_run_btc(struct pp_hwmgr *hwmgr) { - return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_RunBtc); + return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_RunBtc, NULL); } static int vega20_run_btc_afll(struct pp_hwmgr *hwmgr) { - return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_RunAfllBtc); + return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_RunAfllBtc, NULL); } static int vega20_enable_all_smu_features(struct pp_hwmgr *hwmgr) @@ -933,7 +924,8 @@ static int vega20_enable_all_smu_features(struct pp_hwmgr *hwmgr) int ret = 0; PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_EnableAllSmuFeatures)) == 0, + PPSMC_MSG_EnableAllSmuFeatures, + NULL)) == 0, "[EnableAllSMUFeatures] Failed to enable all smu features!", return ret); @@ -966,7 +958,8 @@ static int vega20_notify_smc_display_change(struct pp_hwmgr *hwmgr) if (data->smu_features[GNLD_DPM_UCLK].enabled) return smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetUclkFastSwitch, - 1); + 1, + NULL); return 0; } @@ -978,7 +971,8 @@ static int vega20_send_clock_ratio(struct pp_hwmgr *hwmgr) return smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetFclkGfxClkRatio, - data->registry_data.fclk_gfxclk_ratio); + data->registry_data.fclk_gfxclk_ratio, + NULL); } static int vega20_disable_all_smu_features(struct pp_hwmgr *hwmgr) @@ -991,7 +985,8 @@ static int vega20_disable_all_smu_features(struct pp_hwmgr *hwmgr) int ret = 0; PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_DisableAllSmuFeatures)) == 0, + PPSMC_MSG_DisableAllSmuFeatures, + NULL)) == 0, "[DisableAllSMUFeatures] Failed to disable all smu features!", return ret); @@ -1199,12 +1194,12 @@ static int vega20_od8_get_gfx_clock_base_voltage( ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetAVFSVoltageByDpm, - ((AVFS_CURVE << 24) | (OD8_HOTCURVE_TEMPERATURE << 16) | freq)); + ((AVFS_CURVE << 24) | (OD8_HOTCURVE_TEMPERATURE << 16) | freq), + voltage); PP_ASSERT_WITH_CODE(!ret, "[GetBaseVoltage] failed to get GFXCLK AVFS voltage from SMU!", return ret); - *voltage = smum_get_argument(hwmgr); *voltage = *voltage / VOLTAGE_SCALE; return 0; @@ -1560,19 +1555,19 @@ static int vega20_get_max_sustainable_clock(struct pp_hwmgr *hwmgr, PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetDcModeMaxDpmFreq, - (clock_select << 16))) == 0, + (clock_select << 16), + clock)) == 0, "[GetMaxSustainableClock] Failed to get max DC clock from SMC!", return ret); - *clock = smum_get_argument(hwmgr); /* if DC limit is zero, return AC limit */ if (*clock == 0) { PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetMaxDpmFreq, - (clock_select << 16))) == 0, + (clock_select << 16), + clock)) == 0, "[GetMaxSustainableClock] failed to get max AC clock from SMC!", return ret); - *clock = smum_get_argument(hwmgr); } return 0; @@ -1641,7 +1636,8 @@ static int vega20_enable_mgpu_fan_boost(struct pp_hwmgr *hwmgr) int result; result = smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_SetMGpuFanBoostLimitRpm); + PPSMC_MSG_SetMGpuFanBoostLimitRpm, + NULL); PP_ASSERT_WITH_CODE(!result, "[EnableMgpuFan] Failed to enable mgpu fan boost!", return result); @@ -1669,7 +1665,7 @@ static int vega20_enable_dpm_tasks(struct pp_hwmgr *hwmgr) int result = 0; smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_NumOfDisplays, 0); + PPSMC_MSG_NumOfDisplays, 0, NULL); result = vega20_set_allowed_featuresmask(hwmgr); PP_ASSERT_WITH_CODE(!result, @@ -1740,12 +1736,12 @@ static int vega20_enable_dpm_tasks(struct pp_hwmgr *hwmgr) return result); result = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetPptLimit, - POWER_SOURCE_AC << 16); + POWER_SOURCE_AC << 16, &hwmgr->default_power_limit); PP_ASSERT_WITH_CODE(!result, "[GetPptLimit] get default PPT limit failed!", return result); hwmgr->power_limit = - hwmgr->default_power_limit = smum_get_argument(hwmgr); + hwmgr->default_power_limit; return 0; } @@ -1806,7 +1802,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr, uint32_t feature_ min_freq = data->dpm_table.gfx_table.dpm_state.soft_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMinByFreq, - (PPCLK_GFXCLK << 16) | (min_freq & 0xffff))), + (PPCLK_GFXCLK << 16) | (min_freq & 0xffff), + NULL)), "Failed to set soft min gfxclk !", return ret); } @@ -1816,7 +1813,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr, uint32_t feature_ min_freq = data->dpm_table.mem_table.dpm_state.soft_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMinByFreq, - (PPCLK_UCLK << 16) | (min_freq & 0xffff))), + (PPCLK_UCLK << 16) | (min_freq & 0xffff), + NULL)), "Failed to set soft min memclk !", return ret); } @@ -1827,7 +1825,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr, uint32_t feature_ PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMinByFreq, - (PPCLK_VCLK << 16) | (min_freq & 0xffff))), + (PPCLK_VCLK << 16) | (min_freq & 0xffff), + NULL)), "Failed to set soft min vclk!", return ret); @@ -1835,7 +1834,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr, uint32_t feature_ PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMinByFreq, - (PPCLK_DCLK << 16) | (min_freq & 0xffff))), + (PPCLK_DCLK << 16) | (min_freq & 0xffff), + NULL)), "Failed to set soft min dclk!", return ret); } @@ -1846,7 +1846,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr, uint32_t feature_ PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMinByFreq, - (PPCLK_ECLK << 16) | (min_freq & 0xffff))), + (PPCLK_ECLK << 16) | (min_freq & 0xffff), + NULL)), "Failed to set soft min eclk!", return ret); } @@ -1857,7 +1858,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr, uint32_t feature_ PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMinByFreq, - (PPCLK_SOCCLK << 16) | (min_freq & 0xffff))), + (PPCLK_SOCCLK << 16) | (min_freq & 0xffff), + NULL)), "Failed to set soft min socclk!", return ret); } @@ -1868,7 +1870,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr, uint32_t feature_ PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMinByFreq, - (PPCLK_FCLK << 16) | (min_freq & 0xffff))), + (PPCLK_FCLK << 16) | (min_freq & 0xffff), + NULL)), "Failed to set soft min fclk!", return ret); } @@ -1879,7 +1882,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr, uint32_t feature_ PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetHardMinByFreq, - (PPCLK_DCEFCLK << 16) | (min_freq & 0xffff))), + (PPCLK_DCEFCLK << 16) | (min_freq & 0xffff), + NULL)), "Failed to set hard min dcefclk!", return ret); } @@ -1900,7 +1904,8 @@ static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr, uint32_t feature_ PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMaxByFreq, - (PPCLK_GFXCLK << 16) | (max_freq & 0xffff))), + (PPCLK_GFXCLK << 16) | (max_freq & 0xffff), + NULL)), "Failed to set soft max gfxclk!", return ret); } @@ -1911,7 +1916,8 @@ static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr, uint32_t feature_ PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMaxByFreq, - (PPCLK_UCLK << 16) | (max_freq & 0xffff))), + (PPCLK_UCLK << 16) | (max_freq & 0xffff), + NULL)), "Failed to set soft max memclk!", return ret); } @@ -1922,14 +1928,16 @@ static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr, uint32_t feature_ PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMaxByFreq, - (PPCLK_VCLK << 16) | (max_freq & 0xffff))), + (PPCLK_VCLK << 16) | (max_freq & 0xffff), + NULL)), "Failed to set soft max vclk!", return ret); max_freq = data->dpm_table.dclk_table.dpm_state.soft_max_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMaxByFreq, - (PPCLK_DCLK << 16) | (max_freq & 0xffff))), + (PPCLK_DCLK << 16) | (max_freq & 0xffff), + NULL)), "Failed to set soft max dclk!", return ret); } @@ -1940,7 +1948,8 @@ static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr, uint32_t feature_ PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMaxByFreq, - (PPCLK_ECLK << 16) | (max_freq & 0xffff))), + (PPCLK_ECLK << 16) | (max_freq & 0xffff), + NULL)), "Failed to set soft max eclk!", return ret); } @@ -1951,7 +1960,8 @@ static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr, uint32_t feature_ PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMaxByFreq, - (PPCLK_SOCCLK << 16) | (max_freq & 0xffff))), + (PPCLK_SOCCLK << 16) | (max_freq & 0xffff), + NULL)), "Failed to set soft max socclk!", return ret); } @@ -1962,7 +1972,8 @@ static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr, uint32_t feature_ PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMaxByFreq, - (PPCLK_FCLK << 16) | (max_freq & 0xffff))), + (PPCLK_FCLK << 16) | (max_freq & 0xffff), + NULL)), "Failed to set soft max fclk!", return ret); } @@ -2006,17 +2017,17 @@ static int vega20_get_clock_ranges(struct pp_hwmgr *hwmgr, if (max) { PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_GetMaxDpmFreq, (clock_select << 16))) == 0, + PPSMC_MSG_GetMaxDpmFreq, (clock_select << 16), + clock)) == 0, "[GetClockRanges] Failed to get max clock from SMC!", return ret); - *clock = smum_get_argument(hwmgr); } else { PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetMinDpmFreq, - (clock_select << 16))) == 0, + (clock_select << 16), + clock)) == 0, "[GetClockRanges] Failed to get min clock from SMC!", return ret); - *clock = smum_get_argument(hwmgr); } return 0; @@ -2122,10 +2133,10 @@ static int vega20_get_current_clk_freq(struct pp_hwmgr *hwmgr, *clk_freq = 0; PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_GetDpmClockFreq, (clk_id << 16))) == 0, + PPSMC_MSG_GetDpmClockFreq, (clk_id << 16), + clk_freq)) == 0, "[GetCurrentClkFreq] Attempt to get Current Frequency Failed!", return ret); - *clk_freq = smum_get_argument(hwmgr); *clk_freq = *clk_freq * 100; @@ -2276,7 +2287,8 @@ int vega20_display_clock_voltage_request(struct pp_hwmgr *hwmgr, clk_request = (clk_select << 16) | clk_freq; result = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinByFreq, - clk_request); + clk_request, + NULL); } } @@ -2312,7 +2324,8 @@ static int vega20_notify_smc_display_config_after_ps_adjustment( if (data->smu_features[GNLD_DS_DCEFCLK].supported) PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetMinDeepSleepDcefclk, - min_clocks.dcefClockInSR / 100)) == 0, + min_clocks.dcefClockInSR / 100, + NULL)) == 0, "Attempt to set divider for DCEFCLK Failed!", return ret); } else { @@ -2324,7 +2337,8 @@ static int vega20_notify_smc_display_config_after_ps_adjustment( dpm_table->dpm_state.hard_min_level = min_clocks.memoryClock / 100; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinByFreq, - (PPCLK_UCLK << 16 ) | dpm_table->dpm_state.hard_min_level)), + (PPCLK_UCLK << 16 ) | dpm_table->dpm_state.hard_min_level, + NULL)), "[SetHardMinFreq] Set hard min uclk failed!", return ret); } @@ -2656,7 +2670,8 @@ static int vega20_force_clock_level(struct pp_hwmgr *hwmgr, return -EINVAL; ret = smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_SetMinLinkDpmByIndex, soft_min_level); + PPSMC_MSG_SetMinLinkDpmByIndex, soft_min_level, + NULL); PP_ASSERT_WITH_CODE(!ret, "Failed to set min link dpm level!", return ret); @@ -3140,7 +3155,7 @@ static int vega20_set_mp1_state(struct pp_hwmgr *hwmgr, return 0; } - PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc(hwmgr, msg)) == 0, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc(hwmgr, msg, NULL)) == 0, "[PrepareMp1] Failed!", return ret); @@ -3495,7 +3510,8 @@ static int vega20_set_uclk_to_highest_dpm_level(struct pp_hwmgr *hwmgr, dpm_table->dpm_state.hard_min_level = dpm_table->dpm_levels[dpm_table->count - 1].value; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinByFreq, - (PPCLK_UCLK << 16 ) | dpm_table->dpm_state.hard_min_level)), + (PPCLK_UCLK << 16 ) | dpm_table->dpm_state.hard_min_level, + NULL)), "[SetUclkToHightestDpmLevel] Set hard min uclk failed!", return ret); } @@ -3520,7 +3536,8 @@ static int vega20_set_fclk_to_highest_dpm_level(struct pp_hwmgr *hwmgr) dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[dpm_table->count - 1].value; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMinByFreq, - (PPCLK_FCLK << 16 ) | dpm_table->dpm_state.soft_min_level)), + (PPCLK_FCLK << 16 ) | dpm_table->dpm_state.soft_min_level, + NULL)), "[SetFclkToHightestDpmLevel] Set soft min fclk failed!", return ret); } @@ -3534,7 +3551,7 @@ static int vega20_pre_display_configuration_changed_task(struct pp_hwmgr *hwmgr) int ret = 0; smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_NumOfDisplays, 0); + PPSMC_MSG_NumOfDisplays, 0, NULL); ret = vega20_set_uclk_to_highest_dpm_level(hwmgr, &data->dpm_table.mem_table); @@ -3565,7 +3582,8 @@ static int vega20_display_configuration_changed_task(struct pp_hwmgr *hwmgr) data->smu_features[GNLD_DPM_SOCCLK].supported) { result = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_NumOfDisplays, - hwmgr->display_config->num_display); + hwmgr->display_config->num_display, + NULL); } return result; @@ -4082,7 +4100,8 @@ out: workload_type = conv_power_profile_to_pplib_workload(power_profile_mode); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetWorkloadMask, - 1 << workload_type); + 1 << workload_type, + NULL); hwmgr->power_profile_mode = power_profile_mode; @@ -4098,21 +4117,26 @@ static int vega20_notify_cac_buffer_info(struct pp_hwmgr *hwmgr, { smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSystemVirtualDramAddrHigh, - virtual_addr_hi); + virtual_addr_hi, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSystemVirtualDramAddrLow, - virtual_addr_low); + virtual_addr_low, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DramLogSetDramAddrHigh, - mc_addr_hi); + mc_addr_hi, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DramLogSetDramAddrLow, - mc_addr_low); + mc_addr_low, + NULL); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DramLogSetDramSize, - size); + size, + NULL); return 0; } @@ -4153,7 +4177,8 @@ static int vega20_smu_i2c_bus_access(struct pp_hwmgr *hwmgr, bool acquire) (acquire ? PPSMC_MSG_RequestI2CBus : PPSMC_MSG_ReleaseI2CBus), - 0); + 0, + NULL); PP_ASSERT_WITH_CODE(!res, "[SmuI2CAccessBus] Failed to access bus!", return res); return res; @@ -4170,7 +4195,8 @@ static int vega20_set_df_cstate(struct pp_hwmgr *hwmgr, return -EINVAL; } - ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DFCstateControl, state); + ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DFCstateControl, state, + NULL); if (ret) pr_err("SetDfCstate failed!\n"); @@ -4184,7 +4210,8 @@ static int vega20_set_xgmi_pstate(struct pp_hwmgr *hwmgr, ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetXgmiMode, - pstate ? XGMI_MODE_PSTATE_D0 : XGMI_MODE_PSTATE_D3); + pstate ? XGMI_MODE_PSTATE_D0 : XGMI_MODE_PSTATE_D3, + NULL); if (ret) pr_err("SetXgmiPstate failed!\n"); diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_powertune.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_powertune.c index a0bfb65cc5d6..d7cc3d2d9e17 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_powertune.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_powertune.c @@ -36,7 +36,8 @@ int vega20_set_power_limit(struct pp_hwmgr *hwmgr, uint32_t n) if (data->smu_features[GNLD_PPT].enabled) return smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_SetPptLimit, n); + PPSMC_MSG_SetPptLimit, n, + NULL); return 0; } @@ -51,7 +52,8 @@ static int vega20_set_overdrive_target_percentage(struct pp_hwmgr *hwmgr, uint32_t adjust_percent) { return smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_OverDriveSetPercentage, adjust_percent); + PPSMC_MSG_OverDriveSetPercentage, adjust_percent, + NULL); } int vega20_power_control_set_level(struct pp_hwmgr *hwmgr) diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_thermal.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_thermal.c index ede54e87e287..7add2f60f49c 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_thermal.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_thermal.c @@ -106,10 +106,10 @@ static int vega20_get_current_rpm(struct pp_hwmgr *hwmgr, uint32_t *current_rpm) int ret = 0; PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc(hwmgr, - PPSMC_MSG_GetCurrentRpm)) == 0, + PPSMC_MSG_GetCurrentRpm, + current_rpm)) == 0, "Attempt to get current RPM from SMC Failed!", return ret); - *current_rpm = smum_get_argument(hwmgr); return 0; } @@ -329,7 +329,8 @@ static int vega20_thermal_setup_fan_table(struct pp_hwmgr *hwmgr) ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetFanTemperatureTarget, - (uint32_t)table->FanTargetTemperature); + (uint32_t)table->FanTargetTemperature, + NULL); return ret; } diff --git a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h index ae2c318dd6fa..4d1c2a44a8b6 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h +++ b/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h @@ -405,7 +405,9 @@ struct smu_context bool pm_enabled; bool is_apu; - uint32_t smc_if_version; + uint32_t smc_driver_if_version; + uint32_t smc_fw_if_version; + uint32_t smc_fw_version; bool uploading_custom_pp_table; bool dc_controlled_by_gpio; @@ -489,6 +491,7 @@ struct pptable_funcs { int (*get_dpm_clk_limited)(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t dpm_level, uint32_t *freq); int (*set_df_cstate)(struct smu_context *smu, enum pp_df_cstate state); + int (*allow_xgmi_power_down)(struct smu_context *smu, bool en); int (*update_pcie_parameters)(struct smu_context *smu, uint32_t pcie_gen_cap, uint32_t pcie_width_cap); int (*i2c_eeprom_init)(struct i2c_adapter *control); void (*i2c_eeprom_fini)(struct i2c_adapter *control); @@ -580,11 +583,6 @@ int smu_check_fw_status(struct smu_context *smu); int smu_set_gfx_cgpg(struct smu_context *smu, bool enabled); -#define smu_i2c_eeprom_init(smu, control) \ - ((smu)->ppt_funcs->i2c_eeprom_init ? (smu)->ppt_funcs->i2c_eeprom_init((control)) : -EINVAL) -#define smu_i2c_eeprom_fini(smu, control) \ - ((smu)->ppt_funcs->i2c_eeprom_fini ? (smu)->ppt_funcs->i2c_eeprom_fini((control)) : -EINVAL) - int smu_set_fan_speed_rpm(struct smu_context *smu, uint32_t speed); int smu_get_power_limit(struct smu_context *smu, @@ -734,6 +732,7 @@ int smu_set_mp1_state(struct smu_context *smu, enum pp_mp1_state mp1_state); int smu_set_df_cstate(struct smu_context *smu, enum pp_df_cstate state); +int smu_allow_xgmi_power_down(struct smu_context *smu, bool en); int smu_get_max_sustainable_clocks_by_dc(struct smu_context *smu, struct pp_smu_nv_clock_table *max_clocks); diff --git a/drivers/gpu/drm/amd/powerplay/inc/arcturus_ppsmc.h b/drivers/gpu/drm/amd/powerplay/inc/arcturus_ppsmc.h index f736d773f9d6..e07478b6ac04 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/arcturus_ppsmc.h +++ b/drivers/gpu/drm/amd/powerplay/inc/arcturus_ppsmc.h @@ -114,7 +114,8 @@ #define PPSMC_MSG_SetNumBadHbmPagesRetired 0x3A #define PPSMC_MSG_DFCstateControl 0x3B -#define PPSMC_Message_Count 0x3C +#define PPSMC_MSG_GmiPwrDnControl 0x3D +#define PPSMC_Message_Count 0x3E typedef uint32_t PPSMC_Result; typedef uint32_t PPSMC_Msg; diff --git a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h index 2ffb666b97e6..15ed6cbdf366 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h +++ b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h @@ -743,6 +743,7 @@ struct pp_hwmgr { bool pm_en; bool pp_one_vf; struct mutex smu_lock; + struct mutex msg_lock; uint32_t pp_table_version; void *device; diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_arcturus.h b/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_arcturus.h index ce5b5011c122..8b82059d97e7 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_arcturus.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smu11_driver_if_arcturus.h @@ -82,8 +82,8 @@ // Other #define FEATURE_OUT_OF_BAND_MONITOR_BIT 24 #define FEATURE_TEMP_DEPENDENT_VMIN_BIT 25 +#define FEATURE_PER_PART_VMIN_BIT 26 -#define FEATURE_SPARE_26_BIT 26 #define FEATURE_SPARE_27_BIT 27 #define FEATURE_SPARE_28_BIT 28 #define FEATURE_SPARE_29_BIT 29 @@ -154,6 +154,7 @@ #define FEATURE_OUT_OF_BAND_MONITOR_MASK (1 << FEATURE_OUT_OF_BAND_MONITOR_BIT ) #define FEATURE_TEMP_DEPENDENT_VMIN_MASK (1 << FEATURE_TEMP_DEPENDENT_VMIN_BIT ) +#define FEATURE_PER_PART_VMIN_MASK (1 << FEATURE_PER_PART_VMIN_BIT ) //FIXME need updating @@ -628,8 +629,14 @@ typedef struct { uint16_t BasePerformanceFrequencyCap; //In Mhz uint16_t MaxPerformanceFrequencyCap; //In Mhz + // Per-Part Vmin + uint16_t VDDGFX_VminLow; // mv Q2 + uint16_t VDDGFX_TVminLow; //Celcius + uint16_t VDDGFX_VminLow_HiTemp; // mv Q2 + uint16_t VDDGFX_VminLow_LoTemp; // mv Q2 + // SECTION: Reserved - uint32_t Reserved[9]; + uint32_t Reserved[7]; // SECTION: BOARD PARAMETERS @@ -869,6 +876,10 @@ typedef struct { uint8_t Mem_DownHystLimit; uint16_t Mem_Fps; + uint32_t BusyThreshold; // Q16 + uint32_t BusyHyst; + uint32_t IdleHyst; + uint32_t MmHubPadding[8]; // SMU internal use } DpmActivityMonitorCoeffInt_t; diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu12_driver_if.h b/drivers/gpu/drm/amd/powerplay/inc/smu12_driver_if.h index 2f85a34c0591..e9315eb5b48e 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smu12_driver_if.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smu12_driver_if.h @@ -27,7 +27,7 @@ // *** IMPORTANT *** // SMU TEAM: Always increment the interface version if // any structure is changed in this file -#define SMU12_DRIVER_IF_VERSION 11 +#define SMU12_DRIVER_IF_VERSION 14 typedef struct { int32_t value; @@ -154,15 +154,19 @@ typedef enum { } CLOCK_IDs_e; // Throttler Status Bitmask -#define THROTTLER_STATUS_BIT_SPL 0 -#define THROTTLER_STATUS_BIT_FPPT 1 -#define THROTTLER_STATUS_BIT_SPPT 2 -#define THROTTLER_STATUS_BIT_SPPT_APU 3 -#define THROTTLER_STATUS_BIT_THM_CORE 4 -#define THROTTLER_STATUS_BIT_THM_GFX 5 -#define THROTTLER_STATUS_BIT_THM_SOC 6 -#define THROTTLER_STATUS_BIT_TDC_VDD 7 -#define THROTTLER_STATUS_BIT_TDC_SOC 8 +#define THROTTLER_STATUS_BIT_SPL 0 +#define THROTTLER_STATUS_BIT_FPPT 1 +#define THROTTLER_STATUS_BIT_SPPT 2 +#define THROTTLER_STATUS_BIT_SPPT_APU 3 +#define THROTTLER_STATUS_BIT_THM_CORE 4 +#define THROTTLER_STATUS_BIT_THM_GFX 5 +#define THROTTLER_STATUS_BIT_THM_SOC 6 +#define THROTTLER_STATUS_BIT_TDC_VDD 7 +#define THROTTLER_STATUS_BIT_TDC_SOC 8 +#define THROTTLER_STATUS_BIT_PROCHOT_CPU 9 +#define THROTTLER_STATUS_BIT_PROCHOT_GFX 10 +#define THROTTLER_STATUS_BIT_EDC_CPU 11 +#define THROTTLER_STATUS_BIT_EDC_GFX 12 typedef struct { uint16_t ClockFrequency[CLOCK_COUNT]; //[MHz] @@ -180,7 +184,7 @@ typedef struct { uint16_t Power[2]; //[mW] indices: VDDCR_VDD, VDDCR_SOC uint16_t FanPwm; //[milli] - uint16_t CurrentSocketPower; //[mW] + uint16_t CurrentSocketPower; //[W] uint16_t CoreFrequency[8]; //[MHz] uint16_t CorePower[8]; //[mW] @@ -193,10 +197,16 @@ typedef struct { uint16_t ThrottlerStatus; uint16_t spare; - uint16_t StapmOriginalLimit; //[mW] - uint16_t StapmCurrentLimit; //[mW] - uint16_t ApuPower; //[mW] - uint16_t dGpuPower; //[mW] + uint16_t StapmOriginalLimit; //[W] + uint16_t StapmCurrentLimit; //[W] + uint16_t ApuPower; //[W] + uint16_t dGpuPower; //[W] + + uint16_t VddTdcValue; //[mA] + uint16_t SocTdcValue; //[mA] + uint16_t VddEdcValue; //[mA] + uint16_t SocEdcValue; //[mA] + uint16_t reserve[2]; } SmuMetrics_t; diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_types.h b/drivers/gpu/drm/amd/powerplay/inc/smu_types.h index a5b4df146713..ee7dac4693d4 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smu_types.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smu_types.h @@ -170,6 +170,7 @@ __SMU_DUMMY_MAP(SetSoftMinJpeg), \ __SMU_DUMMY_MAP(SetHardMinFclkByFreq), \ __SMU_DUMMY_MAP(DFCstateControl), \ + __SMU_DUMMY_MAP(GmiPwrDnControl), \ __SMU_DUMMY_MAP(DAL_DISABLE_DUMMY_PSTATE_CHANGE), \ __SMU_DUMMY_MAP(DAL_ENABLE_DUMMY_PSTATE_CHANGE), \ diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h b/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h index 674e426ed59b..6b3b451a8018 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smu_v11_0.h @@ -27,8 +27,8 @@ #define SMU11_DRIVER_IF_VERSION_INV 0xFFFFFFFF #define SMU11_DRIVER_IF_VERSION_VG20 0x13 -#define SMU11_DRIVER_IF_VERSION_ARCT 0x12 -#define SMU11_DRIVER_IF_VERSION_NV10 0x35 +#define SMU11_DRIVER_IF_VERSION_ARCT 0x14 +#define SMU11_DRIVER_IF_VERSION_NV10 0x36 #define SMU11_DRIVER_IF_VERSION_NV12 0x33 #define SMU11_DRIVER_IF_VERSION_NV14 0x36 @@ -37,7 +37,6 @@ #define MP0_SRAM 0x03900000 #define MP1_Public 0x03b00000 #define MP1_SRAM 0x03c00004 -#define MP1_SMC_SIZE 0x40000 /* address block */ #define smnMP1_FIRMWARE_FLAGS 0x3010024 diff --git a/drivers/gpu/drm/amd/powerplay/inc/smumgr.h b/drivers/gpu/drm/amd/powerplay/inc/smumgr.h index c5288831aa15..ad100b533d04 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smumgr.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smumgr.h @@ -81,16 +81,15 @@ enum SMU10_TABLE_ID { SMU10_CLOCKTABLE, }; -extern uint32_t smum_get_argument(struct pp_hwmgr *hwmgr); - extern int smum_download_powerplay_table(struct pp_hwmgr *hwmgr, void **table); extern int smum_upload_powerplay_table(struct pp_hwmgr *hwmgr); -extern int smum_send_msg_to_smc(struct pp_hwmgr *hwmgr, uint16_t msg); +extern int smum_send_msg_to_smc(struct pp_hwmgr *hwmgr, uint16_t msg, uint32_t *resp); extern int smum_send_msg_to_smc_with_parameter(struct pp_hwmgr *hwmgr, - uint16_t msg, uint32_t parameter); + uint16_t msg, uint32_t parameter, + uint32_t *resp); extern int smum_update_sclk_threshold(struct pp_hwmgr *hwmgr); diff --git a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c b/drivers/gpu/drm/amd/powerplay/navi10_ppt.c index 15030284b444..0c9be864d072 100644 --- a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c +++ b/drivers/gpu/drm/amd/powerplay/navi10_ppt.c @@ -423,6 +423,7 @@ static int navi10_append_powerplay_table(struct smu_context *smu) struct smu_table_context *table_context = &smu->smu_table; PPTable_t *smc_pptable = table_context->driver_pptable; struct atom_smc_dpm_info_v4_5 *smc_dpm_table; + struct atom_smc_dpm_info_v4_7 *smc_dpm_table_v4_7; int index, ret; index = get_index_into_master_table(atom_master_list_of_data_tables_v2_1, @@ -433,77 +434,33 @@ static int navi10_append_powerplay_table(struct smu_context *smu) if (ret) return ret; - memcpy(smc_pptable->I2cControllers, smc_dpm_table->I2cControllers, - sizeof(I2cControllerConfig_t) * NUM_I2C_CONTROLLERS); - - /* SVI2 Board Parameters */ - smc_pptable->MaxVoltageStepGfx = smc_dpm_table->MaxVoltageStepGfx; - smc_pptable->MaxVoltageStepSoc = smc_dpm_table->MaxVoltageStepSoc; - smc_pptable->VddGfxVrMapping = smc_dpm_table->VddGfxVrMapping; - smc_pptable->VddSocVrMapping = smc_dpm_table->VddSocVrMapping; - smc_pptable->VddMem0VrMapping = smc_dpm_table->VddMem0VrMapping; - smc_pptable->VddMem1VrMapping = smc_dpm_table->VddMem1VrMapping; - smc_pptable->GfxUlvPhaseSheddingMask = smc_dpm_table->GfxUlvPhaseSheddingMask; - smc_pptable->SocUlvPhaseSheddingMask = smc_dpm_table->SocUlvPhaseSheddingMask; - smc_pptable->ExternalSensorPresent = smc_dpm_table->ExternalSensorPresent; - smc_pptable->Padding8_V = smc_dpm_table->Padding8_V; - - /* Telemetry Settings */ - smc_pptable->GfxMaxCurrent = smc_dpm_table->GfxMaxCurrent; - smc_pptable->GfxOffset = smc_dpm_table->GfxOffset; - smc_pptable->Padding_TelemetryGfx = smc_dpm_table->Padding_TelemetryGfx; - smc_pptable->SocMaxCurrent = smc_dpm_table->SocMaxCurrent; - smc_pptable->SocOffset = smc_dpm_table->SocOffset; - smc_pptable->Padding_TelemetrySoc = smc_dpm_table->Padding_TelemetrySoc; - smc_pptable->Mem0MaxCurrent = smc_dpm_table->Mem0MaxCurrent; - smc_pptable->Mem0Offset = smc_dpm_table->Mem0Offset; - smc_pptable->Padding_TelemetryMem0 = smc_dpm_table->Padding_TelemetryMem0; - smc_pptable->Mem1MaxCurrent = smc_dpm_table->Mem1MaxCurrent; - smc_pptable->Mem1Offset = smc_dpm_table->Mem1Offset; - smc_pptable->Padding_TelemetryMem1 = smc_dpm_table->Padding_TelemetryMem1; - - /* GPIO Settings */ - smc_pptable->AcDcGpio = smc_dpm_table->AcDcGpio; - smc_pptable->AcDcPolarity = smc_dpm_table->AcDcPolarity; - smc_pptable->VR0HotGpio = smc_dpm_table->VR0HotGpio; - smc_pptable->VR0HotPolarity = smc_dpm_table->VR0HotPolarity; - smc_pptable->VR1HotGpio = smc_dpm_table->VR1HotGpio; - smc_pptable->VR1HotPolarity = smc_dpm_table->VR1HotPolarity; - smc_pptable->GthrGpio = smc_dpm_table->GthrGpio; - smc_pptable->GthrPolarity = smc_dpm_table->GthrPolarity; - - /* LED Display Settings */ - smc_pptable->LedPin0 = smc_dpm_table->LedPin0; - smc_pptable->LedPin1 = smc_dpm_table->LedPin1; - smc_pptable->LedPin2 = smc_dpm_table->LedPin2; - smc_pptable->padding8_4 = smc_dpm_table->padding8_4; - - /* GFXCLK PLL Spread Spectrum */ - smc_pptable->PllGfxclkSpreadEnabled = smc_dpm_table->PllGfxclkSpreadEnabled; - smc_pptable->PllGfxclkSpreadPercent = smc_dpm_table->PllGfxclkSpreadPercent; - smc_pptable->PllGfxclkSpreadFreq = smc_dpm_table->PllGfxclkSpreadFreq; - - /* GFXCLK DFLL Spread Spectrum */ - smc_pptable->DfllGfxclkSpreadEnabled = smc_dpm_table->DfllGfxclkSpreadEnabled; - smc_pptable->DfllGfxclkSpreadPercent = smc_dpm_table->DfllGfxclkSpreadPercent; - smc_pptable->DfllGfxclkSpreadFreq = smc_dpm_table->DfllGfxclkSpreadFreq; - - /* UCLK Spread Spectrum */ - smc_pptable->UclkSpreadEnabled = smc_dpm_table->UclkSpreadEnabled; - smc_pptable->UclkSpreadPercent = smc_dpm_table->UclkSpreadPercent; - smc_pptable->UclkSpreadFreq = smc_dpm_table->UclkSpreadFreq; - - /* SOCCLK Spread Spectrum */ - smc_pptable->SoclkSpreadEnabled = smc_dpm_table->SoclkSpreadEnabled; - smc_pptable->SocclkSpreadPercent = smc_dpm_table->SocclkSpreadPercent; - smc_pptable->SocclkSpreadFreq = smc_dpm_table->SocclkSpreadFreq; - - /* Total board power */ - smc_pptable->TotalBoardPower = smc_dpm_table->TotalBoardPower; - smc_pptable->BoardPadding = smc_dpm_table->BoardPadding; - - /* Mvdd Svi2 Div Ratio Setting */ - smc_pptable->MvddRatio = smc_dpm_table->MvddRatio; + pr_info("smc_dpm_info table revision(format.content): %d.%d\n", + smc_dpm_table->table_header.format_revision, + smc_dpm_table->table_header.content_revision); + + if (smc_dpm_table->table_header.format_revision != 4) { + pr_err("smc_dpm_info table format revision is not 4!\n"); + return -EINVAL; + } + + switch (smc_dpm_table->table_header.content_revision) { + case 5: /* nv10 and nv14 */ + memcpy(smc_pptable->I2cControllers, smc_dpm_table->I2cControllers, + sizeof(*smc_dpm_table) - sizeof(smc_dpm_table->table_header)); + break; + case 7: /* nv12 */ + ret = smu_get_atom_data_table(smu, index, NULL, NULL, NULL, + (uint8_t **)&smc_dpm_table_v4_7); + if (ret) + return ret; + memcpy(smc_pptable->I2cControllers, smc_dpm_table_v4_7->I2cControllers, + sizeof(*smc_dpm_table_v4_7) - sizeof(smc_dpm_table_v4_7->table_header)); + break; + default: + pr_err("smc_dpm_info with unsupported content revision %d!\n", + smc_dpm_table->table_header.content_revision); + return -EINVAL; + } if (adev->pm.pp_feature & PP_GFXOFF_MASK) { /* TODO: remove it once SMU fw fix it */ @@ -1336,8 +1293,6 @@ static int navi10_set_power_profile_mode(struct smu_context *smu, long *input, u } if (smu->power_profile_mode == PP_SMC_POWER_PROFILE_CUSTOM) { - if (size < 0) - return -EINVAL; ret = smu_update_table(smu, SMU_TABLE_ACTIVITY_MONITOR_COEFF, WORKLOAD_PPLIB_CUSTOM_BIT, @@ -1860,7 +1815,8 @@ static int navi10_get_power_limit(struct smu_context *smu, int power_src; if (!smu->power_limit) { - if (smu_feature_is_enabled(smu, SMU_FEATURE_PPT_BIT)) { + if (smu_feature_is_enabled(smu, SMU_FEATURE_PPT_BIT) && + !amdgpu_sriov_vf(smu->adev)) { power_src = smu_power_get_index(smu, SMU_POWER_SOURCE_AC); if (power_src < 0) return -EINVAL; @@ -2003,6 +1959,9 @@ static int navi10_set_default_od_settings(struct smu_context *smu, bool initiali OverDriveTable_t *od_table, *boot_od_table; int ret = 0; + if (amdgpu_sriov_vf(smu->adev)) + return 0; + ret = smu_v11_0_set_default_od_settings(smu, initialize, sizeof(OverDriveTable_t)); if (ret) return ret; diff --git a/drivers/gpu/drm/amd/powerplay/renoir_ppt.c b/drivers/gpu/drm/amd/powerplay/renoir_ppt.c index ff73a735b888..67476047c067 100644 --- a/drivers/gpu/drm/amd/powerplay/renoir_ppt.c +++ b/drivers/gpu/drm/amd/powerplay/renoir_ppt.c @@ -296,6 +296,8 @@ static int renoir_print_clk_levels(struct smu_context *smu, for (i = 0; i < count; i++) { GET_DPM_CUR_FREQ(clk_table, clk_type, i, value); + if (!value) + continue; size += sprintf(buf + size, "%d: %uMhz %s\n", i, value, cur_value == value ? "*" : ""); if (cur_value == value) @@ -847,7 +849,7 @@ static int renoir_get_power_profile_mode(struct smu_context *smu, uint32_t i, size = 0; int16_t workload_type = 0; - if (!smu->pm_enabled || !buf) + if (!buf) return -EINVAL; for (i = 0; i <= PP_SMC_POWER_PROFILE_CUSTOM; i++) { @@ -895,12 +897,17 @@ static int renoir_read_sensor(struct smu_context *smu, static bool renoir_is_dpm_running(struct smu_context *smu) { + struct amdgpu_device *adev = smu->adev; + /* - * Util now, the pmfw hasn't exported the interface of SMU + * Until now, the pmfw hasn't exported the interface of SMU * feature mask to APU SKU so just force on all the feature * at early initial stage. */ - return true; + if (adev->in_suspend) + return false; + else + return true; } @@ -950,6 +957,6 @@ static const struct pptable_funcs renoir_ppt_funcs = { void renoir_set_ppt_funcs(struct smu_context *smu) { smu->ppt_funcs = &renoir_ppt_funcs; - smu->smc_if_version = SMU12_DRIVER_IF_VERSION; + smu->smc_driver_if_version = SMU12_DRIVER_IF_VERSION; smu->is_apu = true; } diff --git a/drivers/gpu/drm/amd/powerplay/smu_internal.h b/drivers/gpu/drm/amd/powerplay/smu_internal.h index 40c35bcc5a0a..c97444841abc 100644 --- a/drivers/gpu/drm/amd/powerplay/smu_internal.h +++ b/drivers/gpu/drm/amd/powerplay/smu_internal.h @@ -214,4 +214,9 @@ static inline int smu_send_smc_msg(struct smu_context *smu, enum smu_message_typ #define smu_set_power_source(smu, power_src) \ ((smu)->ppt_funcs->set_power_source ? (smu)->ppt_funcs->set_power_source((smu), (power_src)) : 0) +#define smu_i2c_eeprom_init(smu, control) \ + ((smu)->ppt_funcs->i2c_eeprom_init ? (smu)->ppt_funcs->i2c_eeprom_init((control)) : 0) +#define smu_i2c_eeprom_fini(smu, control) \ + ((smu)->ppt_funcs->i2c_eeprom_fini ? (smu)->ppt_funcs->i2c_eeprom_fini((control)) : 0) + #endif diff --git a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c index 541c932a6005..aa76c2cea747 100644 --- a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c +++ b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c @@ -23,6 +23,7 @@ #include <linux/firmware.h> #include <linux/module.h> #include <linux/pci.h> +#include <linux/reboot.h> #define SMU_11_0_PARTIAL_PPTABLE @@ -57,7 +58,7 @@ static int smu_v11_0_send_msg_without_waiting(struct smu_context *smu, uint16_t msg) { struct amdgpu_device *adev = smu->adev; - WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_66, msg); + WREG32_SOC15_NO_KIQ(MP1, 0, mmMP1_SMN_C2PMSG_66, msg); return 0; } @@ -65,7 +66,7 @@ static int smu_v11_0_read_arg(struct smu_context *smu, uint32_t *arg) { struct amdgpu_device *adev = smu->adev; - *arg = RREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_82); + *arg = RREG32_SOC15_NO_KIQ(MP1, 0, mmMP1_SMN_C2PMSG_82); return 0; } @@ -75,7 +76,7 @@ static int smu_v11_0_wait_for_response(struct smu_context *smu) uint32_t cur_value, i, timeout = adev->usec_timeout * 10; for (i = 0; i < timeout; i++) { - cur_value = RREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_90); + cur_value = RREG32_SOC15_NO_KIQ(MP1, 0, mmMP1_SMN_C2PMSG_90); if ((cur_value & MP1_C2PMSG_90__CONTENT_MASK) != 0) return cur_value == 0x1 ? 0 : -EIO; @@ -83,7 +84,10 @@ static int smu_v11_0_wait_for_response(struct smu_context *smu) } /* timeout means wrong logic */ - return -ETIME; + if (i == timeout) + return -ETIME; + + return RREG32_SOC15_NO_KIQ(MP1, 0, mmMP1_SMN_C2PMSG_90) == 0x1 ? 0 : -EIO; } int @@ -107,9 +111,9 @@ smu_v11_0_send_msg_with_param(struct smu_context *smu, goto out; } - WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_90, 0); + WREG32_SOC15_NO_KIQ(MP1, 0, mmMP1_SMN_C2PMSG_90, 0); - WREG32_SOC15(MP1, 0, mmMP1_SMN_C2PMSG_82, param); + WREG32_SOC15_NO_KIQ(MP1, 0, mmMP1_SMN_C2PMSG_82, param); smu_v11_0_send_msg_without_waiting(smu, (uint16_t)index); @@ -119,6 +123,7 @@ smu_v11_0_send_msg_with_param(struct smu_context *smu, smu_get_message_name(smu, msg), index, param, ret); goto out; } + if (read_arg) { ret = smu_v11_0_read_arg(smu, read_arg); if (ret) { @@ -201,13 +206,15 @@ int smu_v11_0_load_microcode(struct smu_context *smu) const struct smc_firmware_header_v1_0 *hdr; uint32_t addr_start = MP1_SRAM; uint32_t i; + uint32_t smc_fw_size; uint32_t mp1_fw_flags; hdr = (const struct smc_firmware_header_v1_0 *) adev->pm.fw->data; src = (const uint32_t *)(adev->pm.fw->data + le32_to_cpu(hdr->header.ucode_array_offset_bytes)); + smc_fw_size = hdr->header.ucode_size_bytes; - for (i = 1; i < MP1_SMC_SIZE/4 - 1; i++) { + for (i = 1; i < smc_fw_size/4 - 1; i++) { WREG32_PCIE(addr_start, src[i]); addr_start += 4; } @@ -264,23 +271,23 @@ int smu_v11_0_check_fw_version(struct smu_context *smu) switch (smu->adev->asic_type) { case CHIP_VEGA20: - smu->smc_if_version = SMU11_DRIVER_IF_VERSION_VG20; + smu->smc_driver_if_version = SMU11_DRIVER_IF_VERSION_VG20; break; case CHIP_ARCTURUS: - smu->smc_if_version = SMU11_DRIVER_IF_VERSION_ARCT; + smu->smc_driver_if_version = SMU11_DRIVER_IF_VERSION_ARCT; break; case CHIP_NAVI10: - smu->smc_if_version = SMU11_DRIVER_IF_VERSION_NV10; + smu->smc_driver_if_version = SMU11_DRIVER_IF_VERSION_NV10; break; case CHIP_NAVI12: - smu->smc_if_version = SMU11_DRIVER_IF_VERSION_NV12; + smu->smc_driver_if_version = SMU11_DRIVER_IF_VERSION_NV12; break; case CHIP_NAVI14: - smu->smc_if_version = SMU11_DRIVER_IF_VERSION_NV14; + smu->smc_driver_if_version = SMU11_DRIVER_IF_VERSION_NV14; break; default: pr_err("smu unsupported asic type:%d.\n", smu->adev->asic_type); - smu->smc_if_version = SMU11_DRIVER_IF_VERSION_INV; + smu->smc_driver_if_version = SMU11_DRIVER_IF_VERSION_INV; break; } @@ -292,10 +299,10 @@ int smu_v11_0_check_fw_version(struct smu_context *smu) * Considering above, we just leave user a warning message instead * of halt driver loading. */ - if (if_version != smu->smc_if_version) { + if (if_version != smu->smc_driver_if_version) { pr_info("smu driver if version = 0x%08x, smu fw if version = 0x%08x, " "smu fw version = 0x%08x (%d.%d.%d)\n", - smu->smc_if_version, if_version, + smu->smc_driver_if_version, if_version, smu_version, smu_major, smu_minor, smu_debug); pr_warn("SMU driver if version not matched\n"); } @@ -479,8 +486,6 @@ int smu_v11_0_init_power(struct smu_context *smu) { struct smu_power_context *smu_power = &smu->smu_power; - if (!smu->pm_enabled) - return 0; if (smu_power->power_context || smu_power->power_context_size != 0) return -EINVAL; @@ -497,8 +502,6 @@ int smu_v11_0_fini_power(struct smu_context *smu) { struct smu_power_context *smu_power = &smu->smu_power; - if (!smu->pm_enabled) - return 0; if (!smu_power->power_context || smu_power->power_context_size == 0) return -EINVAL; @@ -730,8 +733,9 @@ int smu_v11_0_parse_pptable(struct smu_context *smu) struct smu_table_context *table_context = &smu->smu_table; struct smu_table *table = &table_context->tables[SMU_TABLE_PPTABLE]; + /* during TDR we need to free and alloc the pptable */ if (table_context->driver_pptable) - return -EINVAL; + kfree(table_context->driver_pptable); table_context->driver_pptable = kzalloc(table->size, GFP_KERNEL); @@ -771,6 +775,9 @@ int smu_v11_0_set_deep_sleep_dcefclk(struct smu_context *smu, uint32_t clk) { int ret; + if (amdgpu_sriov_vf(smu->adev)) + return 0; + ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetMinDeepSleepDcefclk, clk, NULL); if (ret) @@ -783,8 +790,6 @@ int smu_v11_0_set_min_dcef_deep_sleep(struct smu_context *smu) { struct smu_table_context *table_context = &smu->smu_table; - if (!smu->pm_enabled) - return 0; if (!table_context) return -EINVAL; @@ -816,6 +821,9 @@ int smu_v11_0_set_tool_table_location(struct smu_context *smu) int ret = 0; struct smu_table *tool_table = &smu->smu_table.tables[SMU_TABLE_PMSTATUSLOG]; + if (amdgpu_sriov_vf(smu->adev)) + return 0; + if (tool_table->mc_address) { ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetToolsDramAddrHigh, @@ -835,6 +843,9 @@ int smu_v11_0_init_display_count(struct smu_context *smu, uint32_t count) { int ret = 0; + if (amdgpu_sriov_vf(smu->adev)) + return 0; + if (!smu->pm_enabled) return ret; @@ -849,6 +860,9 @@ int smu_v11_0_set_allowed_mask(struct smu_context *smu) int ret = 0; uint32_t feature_mask[2]; + if (amdgpu_sriov_vf(smu->adev)) + return 0; + mutex_lock(&feature->mutex); if (bitmap_empty(feature->allowed, SMU_FEATURE_MAX) || feature->feature_num < 64) goto failed; @@ -877,6 +891,9 @@ int smu_v11_0_get_enabled_mask(struct smu_context *smu, struct smu_feature *feature = &smu->smu_feature; int ret = 0; + if (amdgpu_sriov_vf(smu->adev) && !amdgpu_sriov_is_pp_one_vf(smu->adev)) + return 0; + if (!feature_mask || num < 2) return -EINVAL; @@ -932,8 +949,12 @@ int smu_v11_0_notify_display_change(struct smu_context *smu) { int ret = 0; + if (amdgpu_sriov_vf(smu->adev)) + return 0; + if (!smu->pm_enabled) return ret; + if (smu_feature_is_enabled(smu, SMU_FEATURE_DPM_UCLK_BIT) && smu->adev->gmc.vram_type == AMDGPU_VRAM_TYPE_HBM) ret = smu_send_smc_msg_with_param(smu, SMU_MSG_SetUclkFastSwitch, 1, NULL); @@ -948,9 +969,6 @@ smu_v11_0_get_max_sustainable_clock(struct smu_context *smu, uint32_t *clock, int ret = 0; int clk_id; - if (!smu->pm_enabled) - return ret; - if ((smu_msg_get_index(smu, SMU_MSG_GetDcModeMaxDpmFreq) < 0) || (smu_msg_get_index(smu, SMU_MSG_GetMaxDpmFreq) < 0)) return 0; @@ -1096,6 +1114,9 @@ int smu_v11_0_set_power_limit(struct smu_context *smu, uint32_t n) int ret = 0; uint32_t max_power_limit; + if (amdgpu_sriov_vf(smu->adev)) + return 0; + max_power_limit = smu_v11_0_get_max_power_limit(smu); if (n > max_power_limit) { @@ -1205,9 +1226,6 @@ int smu_v11_0_start_thermal_control(struct smu_context *smu) struct smu_temperature_range range; struct amdgpu_device *adev = smu->adev; - if (!smu->pm_enabled) - return ret; - memcpy(&range, &smu11_thermal_policy[0], sizeof(struct smu_temperature_range)); ret = smu_get_thermal_temperature_range(smu, &range); @@ -1321,9 +1339,6 @@ smu_v11_0_display_clock_voltage_request(struct smu_context *smu, enum smu_clk_type clk_select = 0; uint32_t clk_freq = clock_req->clock_freq_in_khz / 1000; - if (!smu->pm_enabled) - return -EINVAL; - if (smu_feature_is_enabled(smu, SMU_FEATURE_DPM_DCEFCLK_BIT) || smu_feature_is_enabled(smu, SMU_FEATURE_DPM_UCLK_BIT)) { switch (clk_type) { @@ -1533,39 +1548,65 @@ static int smu_v11_0_ack_ac_dc_interrupt(struct smu_context *smu) #define THM_11_0__SRCID__THM_DIG_THERM_L2H 0 /* ASIC_TEMP > CG_THERMAL_INT.DIG_THERM_INTH */ #define THM_11_0__SRCID__THM_DIG_THERM_H2L 1 /* ASIC_TEMP < CG_THERMAL_INT.DIG_THERM_INTL */ +#define SMUIO_11_0__SRCID__SMUIO_GPIO19 83 + static int smu_v11_0_irq_process(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) { uint32_t client_id = entry->client_id; uint32_t src_id = entry->src_id; + /* + * ctxid is used to distinguish different + * events for SMCToHost interrupt. + */ + uint32_t ctxid = entry->src_data[0]; + uint32_t data; if (client_id == SOC15_IH_CLIENTID_THM) { switch (src_id) { case THM_11_0__SRCID__THM_DIG_THERM_L2H: - pr_warn("GPU over temperature range detected on PCIe %d:%d.%d!\n", - PCI_BUS_NUM(adev->pdev->devfn), - PCI_SLOT(adev->pdev->devfn), - PCI_FUNC(adev->pdev->devfn)); + dev_emerg(adev->dev, "ERROR: GPU over temperature range(SW CTF) detected!\n"); + /* + * SW CTF just occurred. + * Try to do a graceful shutdown to prevent further damage. + */ + dev_emerg(adev->dev, "ERROR: System is going to shutdown due to GPU SW CTF!\n"); + orderly_poweroff(true); break; case THM_11_0__SRCID__THM_DIG_THERM_H2L: - pr_warn("GPU under temperature range detected on PCIe %d:%d.%d!\n", - PCI_BUS_NUM(adev->pdev->devfn), - PCI_SLOT(adev->pdev->devfn), - PCI_FUNC(adev->pdev->devfn)); + dev_emerg(adev->dev, "ERROR: GPU under temperature range detected\n"); break; default: - pr_warn("GPU under temperature range unknown src id (%d), detected on PCIe %d:%d.%d!\n", - src_id, - PCI_BUS_NUM(adev->pdev->devfn), - PCI_SLOT(adev->pdev->devfn), - PCI_FUNC(adev->pdev->devfn)); + dev_emerg(adev->dev, "ERROR: GPU under temperature range unknown src id (%d)\n", + src_id); break; - } + } else if (client_id == SOC15_IH_CLIENTID_ROM_SMUIO) { + dev_emerg(adev->dev, "ERROR: GPU HW Critical Temperature Fault(aka CTF) detected!\n"); + /* + * HW CTF just occurred. Shutdown to prevent further damage. + */ + dev_emerg(adev->dev, "ERROR: System is going to shutdown due to GPU HW CTF!\n"); + orderly_poweroff(true); } else if (client_id == SOC15_IH_CLIENTID_MP1) { - if (src_id == 0xfe) - smu_v11_0_ack_ac_dc_interrupt(&adev->smu); + if (src_id == 0xfe) { + /* ACK SMUToHost interrupt */ + data = RREG32_SOC15(MP1, 0, mmMP1_SMN_IH_SW_INT_CTRL); + data = REG_SET_FIELD(data, MP1_SMN_IH_SW_INT_CTRL, INT_ACK, 1); + WREG32_SOC15(MP1, 0, mmMP1_SMN_IH_SW_INT_CTRL, data); + + switch (ctxid) { + case 0x3: + dev_dbg(adev->dev, "Switched to AC mode!\n"); + smu_v11_0_ack_ac_dc_interrupt(&adev->smu); + break; + case 0x4: + dev_dbg(adev->dev, "Switched to DC mode!\n"); + smu_v11_0_ack_ac_dc_interrupt(&adev->smu); + break; + } + } } return 0; @@ -1605,6 +1646,13 @@ int smu_v11_0_register_irq_handler(struct smu_context *smu) if (ret) return ret; + /* Register CTF(GPIO_19) interrupt */ + ret = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_ROM_SMUIO, + SMUIO_11_0__SRCID__SMUIO_GPIO19, + irq_src); + if (ret) + return ret; + ret = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_MP1, 0xfe, irq_src); @@ -1718,6 +1766,12 @@ int smu_v11_0_baco_set_state(struct smu_context *smu, enum smu_baco_state state) if (ret) goto out; + if (ras && ras->supported) { + ret = smu_send_smc_msg(smu, SMU_MSG_PrepareMp1ForUnload, NULL); + if (ret) + goto out; + } + /* clear vbios scratch 6 and 7 for coming asic reinit */ WREG32(adev->bios_scratch_reg_offset + 6, 0); WREG32(adev->bios_scratch_reg_offset + 7, 0); @@ -1827,6 +1881,9 @@ int smu_v11_0_override_pcie_parameters(struct smu_context *smu) uint32_t pcie_gen = 0, pcie_width = 0; int ret; + if (amdgpu_sriov_vf(smu->adev)) + return 0; + if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN4) pcie_gen = 3; else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3) diff --git a/drivers/gpu/drm/amd/powerplay/smu_v12_0.c b/drivers/gpu/drm/amd/powerplay/smu_v12_0.c index 169ebdad87b8..4023d10fb49b 100644 --- a/drivers/gpu/drm/amd/powerplay/smu_v12_0.c +++ b/drivers/gpu/drm/amd/powerplay/smu_v12_0.c @@ -32,13 +32,15 @@ #include "asic_reg/mp/mp_12_0_0_offset.h" #include "asic_reg/mp/mp_12_0_0_sh_mask.h" +#include "asic_reg/smuio/smuio_12_0_0_offset.h" +#include "asic_reg/smuio/smuio_12_0_0_sh_mask.h" -#define smnMP1_FIRMWARE_FLAGS 0x3010024 +// because some SMU12 based ASICs use older ip offset tables +// we should undefine this register from the smuio12 header +// to prevent confusion down the road +#undef mmPWR_MISC_CNTL_STATUS -#define mmSMUIO_GFX_MISC_CNTL 0x00c8 -#define mmSMUIO_GFX_MISC_CNTL_BASE_IDX 0 -#define SMUIO_GFX_MISC_CNTL__PWR_GFXOFF_STATUS_MASK 0x00000006L -#define SMUIO_GFX_MISC_CNTL__PWR_GFXOFF_STATUS__SHIFT 0x1 +#define smnMP1_FIRMWARE_FLAGS 0x3010024 int smu_v12_0_send_msg_without_waiting(struct smu_context *smu, uint16_t msg) @@ -158,10 +160,10 @@ int smu_v12_0_check_fw_version(struct smu_context *smu) * Considering above, we just leave user a warning message instead * of halt driver loading. */ - if (if_version != smu->smc_if_version) { + if (if_version != smu->smc_driver_if_version) { pr_info("smu driver if version = 0x%08x, smu fw if version = 0x%08x, " "smu fw version = 0x%08x (%d.%d.%d)\n", - smu->smc_if_version, if_version, + smu->smc_driver_if_version, if_version, smu_version, smu_major, smu_minor, smu_debug); pr_warn("SMU driver if version not matched\n"); } diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/ci_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/ci_smumgr.c index 868e2d5f6e62..56923a96b450 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/ci_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/ci_smumgr.c @@ -239,7 +239,7 @@ static void ci_initialize_power_tune_defaults(struct pp_hwmgr *hwmgr) switch (dev_id) { case 0x67BA: - case 0x66B1: + case 0x67B1: smu_data->power_tune_defaults = &defaults_hawaii_pro; break; case 0x67B8: @@ -2780,7 +2780,7 @@ static int ci_update_dpm_settings(struct pp_hwmgr *hwmgr, if (setting->bupdate_sclk) { if (!data->sclk_dpm_key_disabled) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_SCLKDPM_FreezeLevel); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_SCLKDPM_FreezeLevel, NULL); for (i = 0; i < smu_data->smc_state_table.GraphicsDpmLevelCount; i++) { if (levels[i].ActivityLevel != cpu_to_be16(setting->sclk_activity)) { @@ -2810,12 +2810,12 @@ static int ci_update_dpm_settings(struct pp_hwmgr *hwmgr, } } if (!data->sclk_dpm_key_disabled) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_SCLKDPM_UnfreezeLevel); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_SCLKDPM_UnfreezeLevel, NULL); } if (setting->bupdate_mclk) { if (!data->mclk_dpm_key_disabled) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MCLKDPM_FreezeLevel); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MCLKDPM_FreezeLevel, NULL); for (i = 0; i < smu_data->smc_state_table.MemoryDpmLevelCount; i++) { if (mclk_levels[i].ActivityLevel != cpu_to_be16(setting->mclk_activity)) { @@ -2845,7 +2845,7 @@ static int ci_update_dpm_settings(struct pp_hwmgr *hwmgr, } } if (!data->mclk_dpm_key_disabled) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MCLKDPM_UnfreezeLevel); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MCLKDPM_UnfreezeLevel, NULL); } return 0; } @@ -2881,8 +2881,9 @@ static int ci_update_uvd_smc_table(struct pp_hwmgr *hwmgr) if (hwmgr->dpm_level & profile_mode_mask || !PP_CAP(PHM_PlatformCaps_UVDDPM)) break; } - ci_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_UVDDPM_SetEnabledMask, - data->dpm_level_enable_mask.uvd_dpm_enable_mask); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_UVDDPM_SetEnabledMask, + data->dpm_level_enable_mask.uvd_dpm_enable_mask, + NULL); return 0; } @@ -2912,8 +2913,9 @@ static int ci_update_vce_smc_table(struct pp_hwmgr *hwmgr) if (hwmgr->dpm_level & profile_mode_mask || !PP_CAP(PHM_PlatformCaps_VCEDPM)) break; } - ci_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_VCEDPM_SetEnabledMask, - data->dpm_level_enable_mask.vce_dpm_enable_mask); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_VCEDPM_SetEnabledMask, + data->dpm_level_enable_mask.vce_dpm_enable_mask, + NULL); return 0; } diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.c index 32ebb383c456..ecb9ee46d6b3 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.c @@ -137,9 +137,7 @@ static int fiji_start_smu_in_protection_mode(struct pp_hwmgr *hwmgr) PHM_WAIT_VFPF_INDIRECT_FIELD(hwmgr, SMC_IND, RCU_UC_EVENTS, INTERRUPTS_ENABLED, 1); - cgs_write_register(hwmgr->device, mmSMC_MSG_ARG_0, 0x20000); - cgs_write_register(hwmgr->device, mmSMC_MESSAGE_0, PPSMC_MSG_Test); - PHM_WAIT_FIELD_UNEQUAL(hwmgr, SMC_RESP_0, SMC_RESP, 0); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_Test, 0x20000, NULL); /* Wait for done bit to be set */ PHM_WAIT_VFPF_INDIRECT_FIELD_UNEQUAL(hwmgr, SMC_IND, @@ -203,8 +201,9 @@ static int fiji_start_avfs_btc(struct pp_hwmgr *hwmgr) struct smu7_smumgr *smu_data = (struct smu7_smumgr *)(hwmgr->smu_backend); if (0 != smu_data->avfs_btc_param) { - if (0 != smu7_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_PerformBtc, smu_data->avfs_btc_param)) { + if (0 != smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_PerformBtc, smu_data->avfs_btc_param, + NULL)) { pr_info("[AVFS][Fiji_PerformBtc] PerformBTC SMU msg failed"); result = -EINVAL; } @@ -1913,7 +1912,8 @@ static int fiji_setup_dpm_led_config(struct pp_hwmgr *hwmgr) if (mask) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_LedConfig, - mask); + mask, + NULL); return 0; } @@ -2220,14 +2220,16 @@ static int fiji_thermal_setup_fan_table(struct pp_hwmgr *hwmgr) res = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetFanMinPwm, hwmgr->thermal_controller. - advanceFanControlParameters.ucMinimumPWMLimit); + advanceFanControlParameters.ucMinimumPWMLimit, + NULL); if (!res && hwmgr->thermal_controller. advanceFanControlParameters.ulMinFanSCLKAcousticLimit) res = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetFanSclkTarget, hwmgr->thermal_controller. - advanceFanControlParameters.ulMinFanSCLKAcousticLimit); + advanceFanControlParameters.ulMinFanSCLKAcousticLimit, + NULL); if (res) phm_cap_unset(hwmgr->platform_descriptor.platformCaps, @@ -2242,7 +2244,7 @@ static int fiji_thermal_avfs_enable(struct pp_hwmgr *hwmgr) if (!hwmgr->avfs_supported) return 0; - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnableAvfs); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnableAvfs, NULL); return 0; } @@ -2390,7 +2392,8 @@ static int fiji_update_uvd_smc_table(struct pp_hwmgr *hwmgr) PHM_PlatformCaps_StablePState)) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_UVDDPM_SetEnabledMask, - (uint32_t)(1 << smu_data->smc_state_table.UvdBootLevel)); + (uint32_t)(1 << smu_data->smc_state_table.UvdBootLevel), + NULL); return 0; } @@ -2422,7 +2425,8 @@ static int fiji_update_vce_smc_table(struct pp_hwmgr *hwmgr) if (phm_cap_enabled(hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_StablePState)) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_VCEDPM_SetEnabledMask, - (uint32_t)1 << smu_data->smc_state_table.VceBootLevel); + (uint32_t)1 << smu_data->smc_state_table.VceBootLevel, + NULL); return 0; } @@ -2569,7 +2573,7 @@ static int fiji_update_dpm_settings(struct pp_hwmgr *hwmgr, if (setting->bupdate_sclk) { if (!data->sclk_dpm_key_disabled) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_SCLKDPM_FreezeLevel); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_SCLKDPM_FreezeLevel, NULL); for (i = 0; i < smu_data->smc_state_table.GraphicsDpmLevelCount; i++) { if (levels[i].ActivityLevel != cpu_to_be16(setting->sclk_activity)) { @@ -2599,12 +2603,12 @@ static int fiji_update_dpm_settings(struct pp_hwmgr *hwmgr, } } if (!data->sclk_dpm_key_disabled) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_SCLKDPM_UnfreezeLevel); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_SCLKDPM_UnfreezeLevel, NULL); } if (setting->bupdate_mclk) { if (!data->mclk_dpm_key_disabled) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MCLKDPM_FreezeLevel); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MCLKDPM_FreezeLevel, NULL); for (i = 0; i < smu_data->smc_state_table.MemoryDpmLevelCount; i++) { if (mclk_levels[i].ActivityLevel != cpu_to_be16(setting->mclk_activity)) { @@ -2634,7 +2638,7 @@ static int fiji_update_dpm_settings(struct pp_hwmgr *hwmgr, } } if (!data->mclk_dpm_key_disabled) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MCLKDPM_UnfreezeLevel); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MCLKDPM_UnfreezeLevel, NULL); } return 0; } @@ -2649,6 +2653,7 @@ const struct pp_smumgr_func fiji_smu_funcs = { .request_smu_load_specific_fw = NULL, .send_msg_to_smc = &smu7_send_msg_to_smc, .send_msg_to_smc_with_parameter = &smu7_send_msg_to_smc_with_parameter, + .get_argument = smu7_get_argument, .download_pptable_settings = NULL, .upload_pptable_settings = NULL, .update_smc_table = fiji_update_smc_table, diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/iceland_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/iceland_smumgr.c index 732005c03a82..431ad2fd38df 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/iceland_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/iceland_smumgr.c @@ -2669,6 +2669,7 @@ const struct pp_smumgr_func iceland_smu_funcs = { .request_smu_load_specific_fw = &iceland_request_smu_load_specific_fw, .send_msg_to_smc = &smu7_send_msg_to_smc, .send_msg_to_smc_with_parameter = &smu7_send_msg_to_smc_with_parameter, + .get_argument = smu7_get_argument, .download_pptable_settings = NULL, .upload_pptable_settings = NULL, .get_offsetof = iceland_get_offsetof, diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c index 23c12018dbc1..c3d2e6dcf62a 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c @@ -99,7 +99,8 @@ static int polaris10_perform_btc(struct pp_hwmgr *hwmgr) struct smu7_smumgr *smu_data = (struct smu7_smumgr *)(hwmgr->smu_backend); if (0 != smu_data->avfs_btc_param) { - if (0 != smu7_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_PerformBtc, smu_data->avfs_btc_param)) { + if (0 != smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_PerformBtc, smu_data->avfs_btc_param, + NULL)) { pr_info("[AVFS][SmuPolaris10_PerformBtc] PerformBTC SMU msg failed"); result = -1; } @@ -2049,15 +2050,16 @@ int polaris10_thermal_avfs_enable(struct pp_hwmgr *hwmgr) return 0; smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_SetGBDroopSettings, data->avfs_vdroop_override_setting); + PPSMC_MSG_SetGBDroopSettings, data->avfs_vdroop_override_setting, + NULL); - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnableAvfs); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnableAvfs, NULL); /* Apply avfs cks-off voltages to avoid the overshoot * when switching to the highest sclk frequency */ if (data->apply_avfs_cks_off_voltage) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ApplyAvfsCksOffVoltage); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ApplyAvfsCksOffVoltage, NULL); return 0; } @@ -2158,14 +2160,16 @@ static int polaris10_thermal_setup_fan_table(struct pp_hwmgr *hwmgr) res = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetFanMinPwm, hwmgr->thermal_controller. - advanceFanControlParameters.ucMinimumPWMLimit); + advanceFanControlParameters.ucMinimumPWMLimit, + NULL); if (!res && hwmgr->thermal_controller. advanceFanControlParameters.ulMinFanSCLKAcousticLimit) res = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetFanSclkTarget, hwmgr->thermal_controller. - advanceFanControlParameters.ulMinFanSCLKAcousticLimit); + advanceFanControlParameters.ulMinFanSCLKAcousticLimit, + NULL); if (res) phm_cap_unset(hwmgr->platform_descriptor.platformCaps, @@ -2202,7 +2206,8 @@ static int polaris10_update_uvd_smc_table(struct pp_hwmgr *hwmgr) PHM_PlatformCaps_StablePState)) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_UVDDPM_SetEnabledMask, - (uint32_t)(1 << smu_data->smc_state_table.UvdBootLevel)); + (uint32_t)(1 << smu_data->smc_state_table.UvdBootLevel), + NULL); return 0; } @@ -2234,7 +2239,8 @@ static int polaris10_update_vce_smc_table(struct pp_hwmgr *hwmgr) if (phm_cap_enabled(hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_StablePState)) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_VCEDPM_SetEnabledMask, - (uint32_t)1 << smu_data->smc_state_table.VceBootLevel); + (uint32_t)1 << smu_data->smc_state_table.VceBootLevel, + NULL); return 0; } @@ -2485,7 +2491,7 @@ static int polaris10_update_dpm_settings(struct pp_hwmgr *hwmgr, if (setting->bupdate_sclk) { if (!data->sclk_dpm_key_disabled) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_SCLKDPM_FreezeLevel); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_SCLKDPM_FreezeLevel, NULL); for (i = 0; i < smu_data->smc_state_table.GraphicsDpmLevelCount; i++) { if (levels[i].ActivityLevel != cpu_to_be16(setting->sclk_activity)) { @@ -2515,12 +2521,12 @@ static int polaris10_update_dpm_settings(struct pp_hwmgr *hwmgr, } } if (!data->sclk_dpm_key_disabled) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_SCLKDPM_UnfreezeLevel); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_SCLKDPM_UnfreezeLevel, NULL); } if (setting->bupdate_mclk) { if (!data->mclk_dpm_key_disabled) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MCLKDPM_FreezeLevel); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MCLKDPM_FreezeLevel, NULL); for (i = 0; i < smu_data->smc_state_table.MemoryDpmLevelCount; i++) { if (mclk_levels[i].ActivityLevel != cpu_to_be16(setting->mclk_activity)) { @@ -2550,7 +2556,7 @@ static int polaris10_update_dpm_settings(struct pp_hwmgr *hwmgr, } } if (!data->mclk_dpm_key_disabled) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MCLKDPM_UnfreezeLevel); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MCLKDPM_UnfreezeLevel, NULL); } return 0; } @@ -2565,6 +2571,7 @@ const struct pp_smumgr_func polaris10_smu_funcs = { .request_smu_load_specific_fw = NULL, .send_msg_to_smc = smu7_send_msg_to_smc, .send_msg_to_smc_with_parameter = smu7_send_msg_to_smc_with_parameter, + .get_argument = smu7_get_argument, .download_pptable_settings = NULL, .upload_pptable_settings = NULL, .update_smc_table = polaris10_update_smc_table, diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smu10_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/smu10_smumgr.c index 2319400a3fcb..ea2279bb8cbf 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/smu10_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/smu10_smumgr.c @@ -126,15 +126,18 @@ static int smu10_copy_table_from_smc(struct pp_hwmgr *hwmgr, "Invalid SMU Table version!", return -EINVAL;); PP_ASSERT_WITH_CODE(priv->smu_tables.entry[table_id].size != 0, "Invalid SMU Table Length!", return -EINVAL;); - smu10_send_msg_to_smc_with_parameter(hwmgr, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrHigh, - upper_32_bits(priv->smu_tables.entry[table_id].mc_addr)); - smu10_send_msg_to_smc_with_parameter(hwmgr, + upper_32_bits(priv->smu_tables.entry[table_id].mc_addr), + NULL); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrLow, - lower_32_bits(priv->smu_tables.entry[table_id].mc_addr)); - smu10_send_msg_to_smc_with_parameter(hwmgr, + lower_32_bits(priv->smu_tables.entry[table_id].mc_addr), + NULL); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_TransferTableSmu2Dram, - priv->smu_tables.entry[table_id].table_id); + priv->smu_tables.entry[table_id].table_id, + NULL); /* flush hdp cache */ amdgpu_asic_flush_hdp(adev, NULL); @@ -164,15 +167,18 @@ static int smu10_copy_table_to_smc(struct pp_hwmgr *hwmgr, amdgpu_asic_flush_hdp(adev, NULL); - smu10_send_msg_to_smc_with_parameter(hwmgr, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrHigh, - upper_32_bits(priv->smu_tables.entry[table_id].mc_addr)); - smu10_send_msg_to_smc_with_parameter(hwmgr, + upper_32_bits(priv->smu_tables.entry[table_id].mc_addr), + NULL); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrLow, - lower_32_bits(priv->smu_tables.entry[table_id].mc_addr)); - smu10_send_msg_to_smc_with_parameter(hwmgr, + lower_32_bits(priv->smu_tables.entry[table_id].mc_addr), + NULL); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_TransferTableDram2Smu, - priv->smu_tables.entry[table_id].table_id); + priv->smu_tables.entry[table_id].table_id, + NULL); return 0; } @@ -181,9 +187,9 @@ static int smu10_verify_smc_interface(struct pp_hwmgr *hwmgr) { uint32_t smc_driver_if_version; - smu10_send_msg_to_smc(hwmgr, - PPSMC_MSG_GetDriverIfVersion); - smc_driver_if_version = smu10_read_arg_from_smc(hwmgr); + smum_send_msg_to_smc(hwmgr, + PPSMC_MSG_GetDriverIfVersion, + &smc_driver_if_version); if ((smc_driver_if_version != SMU10_DRIVER_IF_VERSION) && (smc_driver_if_version != SMU10_DRIVER_IF_VERSION + 1)) { @@ -217,11 +223,11 @@ static int smu10_start_smu(struct pp_hwmgr *hwmgr) { struct amdgpu_device *adev = hwmgr->adev; - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetSmuVersion); - hwmgr->smu_version = smu10_read_arg_from_smc(hwmgr); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_GetSmuVersion, &hwmgr->smu_version); adev->pm.fw_version = hwmgr->smu_version >> 8; - if (adev->rev_id < 0x8 && adev->pdev->device != 0x15d8 && + if (!(adev->apu_flags & AMD_APU_IS_RAVEN2) && + (adev->apu_flags & AMD_APU_IS_RAVEN) && adev->pm.fw_version < 0x1e45) adev->pm.pp_feature &= ~PP_GFXOFF_MASK; diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c index 3f51d545e8ff..aae25243eb10 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c @@ -191,13 +191,6 @@ int smu7_send_msg_to_smc(struct pp_hwmgr *hwmgr, uint16_t msg) return 0; } -int smu7_send_msg_to_smc_without_waiting(struct pp_hwmgr *hwmgr, uint16_t msg) -{ - cgs_write_register(hwmgr->device, mmSMC_MESSAGE_0, msg); - - return 0; -} - int smu7_send_msg_to_smc_with_parameter(struct pp_hwmgr *hwmgr, uint16_t msg, uint32_t parameter) { PHM_WAIT_FIELD_UNEQUAL(hwmgr, SMC_RESP_0, SMC_RESP, 0); @@ -207,25 +200,14 @@ int smu7_send_msg_to_smc_with_parameter(struct pp_hwmgr *hwmgr, uint16_t msg, ui return smu7_send_msg_to_smc(hwmgr, msg); } -int smu7_send_msg_to_smc_with_parameter_without_waiting(struct pp_hwmgr *hwmgr, uint16_t msg, uint32_t parameter) +uint32_t smu7_get_argument(struct pp_hwmgr *hwmgr) { - cgs_write_register(hwmgr->device, mmSMC_MSG_ARG_0, parameter); - - return smu7_send_msg_to_smc_without_waiting(hwmgr, msg); + return cgs_read_register(hwmgr->device, mmSMC_MSG_ARG_0); } int smu7_send_msg_to_smc_offset(struct pp_hwmgr *hwmgr) { - cgs_write_register(hwmgr->device, mmSMC_MSG_ARG_0, 0x20000); - - cgs_write_register(hwmgr->device, mmSMC_MESSAGE_0, PPSMC_MSG_Test); - - PHM_WAIT_FIELD_UNEQUAL(hwmgr, SMC_RESP_0, SMC_RESP, 0); - - if (1 != PHM_READ_FIELD(hwmgr->device, SMC_RESP_0, SMC_RESP)) - pr_info("Failed to send Message.\n"); - - return 0; + return smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_Test, 0x20000, NULL); } enum cgs_ucode_id smu7_convert_fw_type_to_cgs(uint32_t fw_type) @@ -353,12 +335,14 @@ int smu7_request_smu_load_fw(struct pp_hwmgr *hwmgr) if (hwmgr->chip_id > CHIP_TOPAZ) { /* add support for Topaz */ if (hwmgr->not_vf) { - smu7_send_msg_to_smc_with_parameter(hwmgr, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SMU_DRAM_ADDR_HI, - upper_32_bits(smu_data->smu_buffer.mc_addr)); - smu7_send_msg_to_smc_with_parameter(hwmgr, + upper_32_bits(smu_data->smu_buffer.mc_addr), + NULL); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SMU_DRAM_ADDR_LO, - lower_32_bits(smu_data->smu_buffer.mc_addr)); + lower_32_bits(smu_data->smu_buffer.mc_addr), + NULL); } fw_to_load = UCODE_ID_RLC_G_MASK + UCODE_ID_SDMA0_MASK @@ -423,10 +407,16 @@ int smu7_request_smu_load_fw(struct pp_hwmgr *hwmgr) } memcpy_toio(smu_data->header_buffer.kaddr, smu_data->toc, sizeof(struct SMU_DRAMData_TOC)); - smu7_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DRV_DRAM_ADDR_HI, upper_32_bits(smu_data->header_buffer.mc_addr)); - smu7_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DRV_DRAM_ADDR_LO, lower_32_bits(smu_data->header_buffer.mc_addr)); - - smu7_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_LoadUcodes, fw_to_load); + smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_DRV_DRAM_ADDR_HI, + upper_32_bits(smu_data->header_buffer.mc_addr), + NULL); + smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_DRV_DRAM_ADDR_LO, + lower_32_bits(smu_data->header_buffer.mc_addr), + NULL); + + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_LoadUcodes, fw_to_load, NULL); r = smu7_check_fw_load_finish(hwmgr, fw_to_load); if (!r) diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.h b/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.h index 01f0538fba6b..e7303dc8c260 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.h +++ b/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.h @@ -60,11 +60,9 @@ int smu7_copy_bytes_to_smc(struct pp_hwmgr *hwmgr, uint32_t smc_start_address, int smu7_program_jump_on_start(struct pp_hwmgr *hwmgr); bool smu7_is_smc_ram_running(struct pp_hwmgr *hwmgr); int smu7_send_msg_to_smc(struct pp_hwmgr *hwmgr, uint16_t msg); -int smu7_send_msg_to_smc_without_waiting(struct pp_hwmgr *hwmgr, uint16_t msg); int smu7_send_msg_to_smc_with_parameter(struct pp_hwmgr *hwmgr, uint16_t msg, uint32_t parameter); -int smu7_send_msg_to_smc_with_parameter_without_waiting(struct pp_hwmgr *hwmgr, - uint16_t msg, uint32_t parameter); +uint32_t smu7_get_argument(struct pp_hwmgr *hwmgr); int smu7_send_msg_to_smc_offset(struct pp_hwmgr *hwmgr); enum cgs_ucode_id smu7_convert_fw_type_to_cgs(uint32_t fw_type); diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smu8_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/smu8_smumgr.c index 7dca04a89217..76d4f12ceedf 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/smu8_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/smu8_smumgr.c @@ -610,18 +610,21 @@ static int smu8_download_pptable_settings(struct pp_hwmgr *hwmgr, void **table) *table = (struct SMU8_Fusion_ClkTable *)smu8_smu->scratch_buffer[i].kaddr; - smu8_send_msg_to_smc_with_parameter(hwmgr, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetClkTableAddrHi, - upper_32_bits(smu8_smu->scratch_buffer[i].mc_addr)); + upper_32_bits(smu8_smu->scratch_buffer[i].mc_addr), + NULL); - smu8_send_msg_to_smc_with_parameter(hwmgr, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetClkTableAddrLo, - lower_32_bits(smu8_smu->scratch_buffer[i].mc_addr)); + lower_32_bits(smu8_smu->scratch_buffer[i].mc_addr), + NULL); - smu8_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_ExecuteJob, - smu8_smu->toc_entry_clock_table); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_ExecuteJob, + smu8_smu->toc_entry_clock_table, + NULL); - smu8_send_msg_to_smc(hwmgr, PPSMC_MSG_ClkTableXferToDram); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ClkTableXferToDram, NULL); return 0; } @@ -637,18 +640,21 @@ static int smu8_upload_pptable_settings(struct pp_hwmgr *hwmgr) break; } - smu8_send_msg_to_smc_with_parameter(hwmgr, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetClkTableAddrHi, - upper_32_bits(smu8_smu->scratch_buffer[i].mc_addr)); + upper_32_bits(smu8_smu->scratch_buffer[i].mc_addr), + NULL); - smu8_send_msg_to_smc_with_parameter(hwmgr, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetClkTableAddrLo, - lower_32_bits(smu8_smu->scratch_buffer[i].mc_addr)); + lower_32_bits(smu8_smu->scratch_buffer[i].mc_addr), + NULL); - smu8_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_ExecuteJob, - smu8_smu->toc_entry_clock_table); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_ExecuteJob, + smu8_smu->toc_entry_clock_table, + NULL); - smu8_send_msg_to_smc(hwmgr, PPSMC_MSG_ClkTableXferToSmu); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ClkTableXferToSmu, NULL); return 0; } @@ -671,25 +677,30 @@ static int smu8_request_smu_load_fw(struct pp_hwmgr *hwmgr) smu8_write_smc_sram_dword(hwmgr, smc_address, 0, smc_address+4); - smu8_send_msg_to_smc_with_parameter(hwmgr, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DriverDramAddrHi, - upper_32_bits(smu8_smu->toc_buffer.mc_addr)); + upper_32_bits(smu8_smu->toc_buffer.mc_addr), + NULL); - smu8_send_msg_to_smc_with_parameter(hwmgr, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_DriverDramAddrLo, - lower_32_bits(smu8_smu->toc_buffer.mc_addr)); + lower_32_bits(smu8_smu->toc_buffer.mc_addr), + NULL); - smu8_send_msg_to_smc(hwmgr, PPSMC_MSG_InitJobs); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_InitJobs, NULL); - smu8_send_msg_to_smc_with_parameter(hwmgr, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_ExecuteJob, - smu8_smu->toc_entry_aram); - smu8_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_ExecuteJob, - smu8_smu->toc_entry_power_profiling_index); + smu8_smu->toc_entry_aram, + NULL); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_ExecuteJob, + smu8_smu->toc_entry_power_profiling_index, + NULL); - smu8_send_msg_to_smc_with_parameter(hwmgr, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_ExecuteJob, - smu8_smu->toc_entry_initialize_index); + smu8_smu->toc_entry_initialize_index, + NULL); fw_to_check = UCODE_ID_RLC_G_MASK | UCODE_ID_SDMA0_MASK | @@ -860,11 +871,13 @@ static bool smu8_dpm_check_smu_features(struct pp_hwmgr *hwmgr, unsigned long check_feature) { int result; - unsigned long features; + uint32_t features; - result = smu8_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_GetFeatureStatus, 0); + result = smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_GetFeatureStatus, + 0, + &features); if (result == 0) { - features = smum_get_argument(hwmgr); if (features & check_feature) return true; } diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c index 4240aeec9000..b6fb48066841 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c @@ -103,14 +103,6 @@ int smum_process_firmware_header(struct pp_hwmgr *hwmgr) return 0; } -uint32_t smum_get_argument(struct pp_hwmgr *hwmgr) -{ - if (NULL != hwmgr->smumgr_funcs->get_argument) - return hwmgr->smumgr_funcs->get_argument(hwmgr); - - return 0; -} - uint32_t smum_get_mac_definition(struct pp_hwmgr *hwmgr, uint32_t value) { if (NULL != hwmgr->smumgr_funcs->get_mac_definition) @@ -135,22 +127,58 @@ int smum_upload_powerplay_table(struct pp_hwmgr *hwmgr) return 0; } -int smum_send_msg_to_smc(struct pp_hwmgr *hwmgr, uint16_t msg) +int smum_send_msg_to_smc(struct pp_hwmgr *hwmgr, uint16_t msg, uint32_t *resp) { - if (hwmgr == NULL || hwmgr->smumgr_funcs->send_msg_to_smc == NULL) + int ret = 0; + + if (hwmgr == NULL || + hwmgr->smumgr_funcs->send_msg_to_smc == NULL || + (resp && !hwmgr->smumgr_funcs->get_argument)) return -EINVAL; - return hwmgr->smumgr_funcs->send_msg_to_smc(hwmgr, msg); + mutex_lock(&hwmgr->msg_lock); + + ret = hwmgr->smumgr_funcs->send_msg_to_smc(hwmgr, msg); + if (ret) { + mutex_unlock(&hwmgr->msg_lock); + return ret; + } + + if (resp) + *resp = hwmgr->smumgr_funcs->get_argument(hwmgr); + + mutex_unlock(&hwmgr->msg_lock); + + return ret; } int smum_send_msg_to_smc_with_parameter(struct pp_hwmgr *hwmgr, - uint16_t msg, uint32_t parameter) + uint16_t msg, + uint32_t parameter, + uint32_t *resp) { + int ret = 0; + if (hwmgr == NULL || - hwmgr->smumgr_funcs->send_msg_to_smc_with_parameter == NULL) + hwmgr->smumgr_funcs->send_msg_to_smc_with_parameter == NULL || + (resp && !hwmgr->smumgr_funcs->get_argument)) return -EINVAL; - return hwmgr->smumgr_funcs->send_msg_to_smc_with_parameter( + + mutex_lock(&hwmgr->msg_lock); + + ret = hwmgr->smumgr_funcs->send_msg_to_smc_with_parameter( hwmgr, msg, parameter); + if (ret) { + mutex_unlock(&hwmgr->msg_lock); + return ret; + } + + if (resp) + *resp = hwmgr->smumgr_funcs->get_argument(hwmgr); + + mutex_unlock(&hwmgr->msg_lock); + + return ret; } int smum_init_smc_table(struct pp_hwmgr *hwmgr) diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/tonga_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/tonga_smumgr.c index f19bac7ef7ba..398e7e3587de 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/tonga_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/tonga_smumgr.c @@ -2702,7 +2702,8 @@ static int tonga_update_uvd_smc_table(struct pp_hwmgr *hwmgr) PHM_PlatformCaps_StablePState)) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_UVDDPM_SetEnabledMask, - (uint32_t)(1 << smu_data->smc_state_table.UvdBootLevel)); + (uint32_t)(1 << smu_data->smc_state_table.UvdBootLevel), + NULL); return 0; } @@ -2733,7 +2734,8 @@ static int tonga_update_vce_smc_table(struct pp_hwmgr *hwmgr) PHM_PlatformCaps_StablePState)) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_VCEDPM_SetEnabledMask, - (uint32_t)1 << smu_data->smc_state_table.VceBootLevel); + (uint32_t)1 << smu_data->smc_state_table.VceBootLevel, + NULL); return 0; } @@ -3168,7 +3170,7 @@ static int tonga_update_dpm_settings(struct pp_hwmgr *hwmgr, if (setting->bupdate_sclk) { if (!data->sclk_dpm_key_disabled) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_SCLKDPM_FreezeLevel); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_SCLKDPM_FreezeLevel, NULL); for (i = 0; i < smu_data->smc_state_table.GraphicsDpmLevelCount; i++) { if (levels[i].ActivityLevel != cpu_to_be16(setting->sclk_activity)) { @@ -3198,12 +3200,12 @@ static int tonga_update_dpm_settings(struct pp_hwmgr *hwmgr, } } if (!data->sclk_dpm_key_disabled) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_SCLKDPM_UnfreezeLevel); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_SCLKDPM_UnfreezeLevel, NULL); } if (setting->bupdate_mclk) { if (!data->mclk_dpm_key_disabled) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MCLKDPM_FreezeLevel); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MCLKDPM_FreezeLevel, NULL); for (i = 0; i < smu_data->smc_state_table.MemoryDpmLevelCount; i++) { if (mclk_levels[i].ActivityLevel != cpu_to_be16(setting->mclk_activity)) { @@ -3233,7 +3235,7 @@ static int tonga_update_dpm_settings(struct pp_hwmgr *hwmgr, } } if (!data->mclk_dpm_key_disabled) - smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MCLKDPM_UnfreezeLevel); + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_MCLKDPM_UnfreezeLevel, NULL); } return 0; } @@ -3248,6 +3250,7 @@ const struct pp_smumgr_func tonga_smu_funcs = { .request_smu_load_specific_fw = NULL, .send_msg_to_smc = &smu7_send_msg_to_smc, .send_msg_to_smc_with_parameter = &smu7_send_msg_to_smc_with_parameter, + .get_argument = smu7_get_argument, .download_pptable_settings = NULL, .upload_pptable_settings = NULL, .update_smc_table = tonga_update_smc_table, diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vega10_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/vega10_smumgr.c index 715564009089..1e222c5d91a4 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/vega10_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/vega10_smumgr.c @@ -47,15 +47,18 @@ static int vega10_copy_table_from_smc(struct pp_hwmgr *hwmgr, "Invalid SMU Table version!", return -EINVAL); PP_ASSERT_WITH_CODE(priv->smu_tables.entry[table_id].size != 0, "Invalid SMU Table Length!", return -EINVAL); - smu9_send_msg_to_smc_with_parameter(hwmgr, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrHigh, - upper_32_bits(priv->smu_tables.entry[table_id].mc_addr)); - smu9_send_msg_to_smc_with_parameter(hwmgr, + upper_32_bits(priv->smu_tables.entry[table_id].mc_addr), + NULL); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrLow, - lower_32_bits(priv->smu_tables.entry[table_id].mc_addr)); - smu9_send_msg_to_smc_with_parameter(hwmgr, + lower_32_bits(priv->smu_tables.entry[table_id].mc_addr), + NULL); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_TransferTableSmu2Dram, - priv->smu_tables.entry[table_id].table_id); + priv->smu_tables.entry[table_id].table_id, + NULL); /* flush hdp cache */ amdgpu_asic_flush_hdp(adev, NULL); @@ -90,15 +93,18 @@ static int vega10_copy_table_to_smc(struct pp_hwmgr *hwmgr, amdgpu_asic_flush_hdp(adev, NULL); - smu9_send_msg_to_smc_with_parameter(hwmgr, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrHigh, - upper_32_bits(priv->smu_tables.entry[table_id].mc_addr)); - smu9_send_msg_to_smc_with_parameter(hwmgr, + upper_32_bits(priv->smu_tables.entry[table_id].mc_addr), + NULL); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrLow, - lower_32_bits(priv->smu_tables.entry[table_id].mc_addr)); - smu9_send_msg_to_smc_with_parameter(hwmgr, + lower_32_bits(priv->smu_tables.entry[table_id].mc_addr), + NULL); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_TransferTableDram2Smu, - priv->smu_tables.entry[table_id].table_id); + priv->smu_tables.entry[table_id].table_id, + NULL); return 0; } @@ -118,17 +124,21 @@ int vega10_enable_smc_features(struct pp_hwmgr *hwmgr, return 0; return smum_send_msg_to_smc_with_parameter(hwmgr, - msg, feature_mask); + msg, feature_mask, NULL); } int vega10_get_enabled_smc_features(struct pp_hwmgr *hwmgr, uint64_t *features_enabled) { + uint32_t enabled_features; + if (features_enabled == NULL) return -EINVAL; - smu9_send_msg_to_smc(hwmgr, PPSMC_MSG_GetEnabledSmuFeatures); - *features_enabled = smu9_get_argument(hwmgr); + smum_send_msg_to_smc(hwmgr, + PPSMC_MSG_GetEnabledSmuFeatures, + &enabled_features); + *features_enabled = enabled_features; return 0; } @@ -150,12 +160,14 @@ static int vega10_set_tools_address(struct pp_hwmgr *hwmgr) struct vega10_smumgr *priv = hwmgr->smu_backend; if (priv->smu_tables.entry[TOOLSTABLE].mc_addr) { - smu9_send_msg_to_smc_with_parameter(hwmgr, + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetToolsDramAddrHigh, - upper_32_bits(priv->smu_tables.entry[TOOLSTABLE].mc_addr)); - smu9_send_msg_to_smc_with_parameter(hwmgr, + upper_32_bits(priv->smu_tables.entry[TOOLSTABLE].mc_addr), + NULL); + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetToolsDramAddrLow, - lower_32_bits(priv->smu_tables.entry[TOOLSTABLE].mc_addr)); + lower_32_bits(priv->smu_tables.entry[TOOLSTABLE].mc_addr), + NULL); } return 0; } @@ -167,11 +179,11 @@ static int vega10_verify_smc_interface(struct pp_hwmgr *hwmgr) uint32_t dev_id; uint32_t rev_id; - PP_ASSERT_WITH_CODE(!smu9_send_msg_to_smc(hwmgr, - PPSMC_MSG_GetDriverIfVersion), + PP_ASSERT_WITH_CODE(!smum_send_msg_to_smc(hwmgr, + PPSMC_MSG_GetDriverIfVersion, + &smc_driver_if_version), "Attempt to get SMC IF Version Number Failed!", return -EINVAL); - smc_driver_if_version = smu9_get_argument(hwmgr); dev_id = adev->pdev->device; rev_id = adev->pdev->revision; diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vega12_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/vega12_smumgr.c index 275dbf65f1a0..f54df76537e4 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/vega12_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/vega12_smumgr.c @@ -50,18 +50,21 @@ static int vega12_copy_table_from_smc(struct pp_hwmgr *hwmgr, "Invalid SMU Table version!", return -EINVAL); PP_ASSERT_WITH_CODE(priv->smu_tables.entry[table_id].size != 0, "Invalid SMU Table Length!", return -EINVAL); - PP_ASSERT_WITH_CODE(smu9_send_msg_to_smc_with_parameter(hwmgr, + PP_ASSERT_WITH_CODE(smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrHigh, - upper_32_bits(priv->smu_tables.entry[table_id].mc_addr)) == 0, + upper_32_bits(priv->smu_tables.entry[table_id].mc_addr), + NULL) == 0, "[CopyTableFromSMC] Attempt to Set Dram Addr High Failed!", return -EINVAL); - PP_ASSERT_WITH_CODE(smu9_send_msg_to_smc_with_parameter(hwmgr, + PP_ASSERT_WITH_CODE(smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrLow, - lower_32_bits(priv->smu_tables.entry[table_id].mc_addr)) == 0, + lower_32_bits(priv->smu_tables.entry[table_id].mc_addr), + NULL) == 0, "[CopyTableFromSMC] Attempt to Set Dram Addr Low Failed!", return -EINVAL); - PP_ASSERT_WITH_CODE(smu9_send_msg_to_smc_with_parameter(hwmgr, + PP_ASSERT_WITH_CODE(smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_TransferTableSmu2Dram, - table_id) == 0, + table_id, + NULL) == 0, "[CopyTableFromSMC] Attempt to Transfer Table From SMU Failed!", return -EINVAL); @@ -98,19 +101,22 @@ static int vega12_copy_table_to_smc(struct pp_hwmgr *hwmgr, amdgpu_asic_flush_hdp(adev, NULL); - PP_ASSERT_WITH_CODE(smu9_send_msg_to_smc_with_parameter(hwmgr, + PP_ASSERT_WITH_CODE(smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrHigh, - upper_32_bits(priv->smu_tables.entry[table_id].mc_addr)) == 0, + upper_32_bits(priv->smu_tables.entry[table_id].mc_addr), + NULL) == 0, "[CopyTableToSMC] Attempt to Set Dram Addr High Failed!", return -EINVAL;); - PP_ASSERT_WITH_CODE(smu9_send_msg_to_smc_with_parameter(hwmgr, + PP_ASSERT_WITH_CODE(smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrLow, - lower_32_bits(priv->smu_tables.entry[table_id].mc_addr)) == 0, + lower_32_bits(priv->smu_tables.entry[table_id].mc_addr), + NULL) == 0, "[CopyTableToSMC] Attempt to Set Dram Addr Low Failed!", return -EINVAL); - PP_ASSERT_WITH_CODE(smu9_send_msg_to_smc_with_parameter(hwmgr, + PP_ASSERT_WITH_CODE(smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_TransferTableDram2Smu, - table_id) == 0, + table_id, + NULL) == 0, "[CopyTableToSMC] Attempt to Transfer Table To SMU Failed!", return -EINVAL); @@ -126,21 +132,21 @@ int vega12_enable_smc_features(struct pp_hwmgr *hwmgr, smu_features_high = (uint32_t)((feature_mask & SMU_FEATURES_HIGH_MASK) >> SMU_FEATURES_HIGH_SHIFT); if (enable) { - PP_ASSERT_WITH_CODE(smu9_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_EnableSmuFeaturesLow, smu_features_low) == 0, + PP_ASSERT_WITH_CODE(smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_EnableSmuFeaturesLow, smu_features_low, NULL) == 0, "[EnableDisableSMCFeatures] Attempt to enable SMU features Low failed!", return -EINVAL); - PP_ASSERT_WITH_CODE(smu9_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_EnableSmuFeaturesHigh, smu_features_high) == 0, + PP_ASSERT_WITH_CODE(smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_EnableSmuFeaturesHigh, smu_features_high, NULL) == 0, "[EnableDisableSMCFeatures] Attempt to enable SMU features High failed!", return -EINVAL); } else { - PP_ASSERT_WITH_CODE(smu9_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_DisableSmuFeaturesLow, smu_features_low) == 0, + PP_ASSERT_WITH_CODE(smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_DisableSmuFeaturesLow, smu_features_low, NULL) == 0, "[EnableDisableSMCFeatures] Attempt to disable SMU features Low failed!", return -EINVAL); - PP_ASSERT_WITH_CODE(smu9_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_DisableSmuFeaturesHigh, smu_features_high) == 0, + PP_ASSERT_WITH_CODE(smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_DisableSmuFeaturesHigh, smu_features_high, NULL) == 0, "[EnableDisableSMCFeatures] Attempt to disable SMU features High failed!", return -EINVAL); } @@ -156,17 +162,17 @@ int vega12_get_enabled_smc_features(struct pp_hwmgr *hwmgr, if (features_enabled == NULL) return -EINVAL; - PP_ASSERT_WITH_CODE(smu9_send_msg_to_smc(hwmgr, - PPSMC_MSG_GetEnabledSmuFeaturesLow) == 0, + PP_ASSERT_WITH_CODE(smum_send_msg_to_smc(hwmgr, + PPSMC_MSG_GetEnabledSmuFeaturesLow, + &smc_features_low) == 0, "[GetEnabledSMCFeatures] Attempt to get SMU features Low failed!", return -EINVAL); - smc_features_low = smu9_get_argument(hwmgr); - PP_ASSERT_WITH_CODE(smu9_send_msg_to_smc(hwmgr, - PPSMC_MSG_GetEnabledSmuFeaturesHigh) == 0, + PP_ASSERT_WITH_CODE(smum_send_msg_to_smc(hwmgr, + PPSMC_MSG_GetEnabledSmuFeaturesHigh, + &smc_features_high) == 0, "[GetEnabledSMCFeatures] Attempt to get SMU features High failed!", return -EINVAL); - smc_features_high = smu9_get_argument(hwmgr); *features_enabled = ((((uint64_t)smc_features_low << SMU_FEATURES_LOW_SHIFT) & SMU_FEATURES_LOW_MASK) | (((uint64_t)smc_features_high << SMU_FEATURES_HIGH_SHIFT) & SMU_FEATURES_HIGH_MASK)); @@ -192,12 +198,14 @@ static int vega12_set_tools_address(struct pp_hwmgr *hwmgr) (struct vega12_smumgr *)(hwmgr->smu_backend); if (priv->smu_tables.entry[TABLE_PMSTATUSLOG].mc_addr) { - if (!smu9_send_msg_to_smc_with_parameter(hwmgr, + if (!smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetToolsDramAddrHigh, - upper_32_bits(priv->smu_tables.entry[TABLE_PMSTATUSLOG].mc_addr))) - smu9_send_msg_to_smc_with_parameter(hwmgr, + upper_32_bits(priv->smu_tables.entry[TABLE_PMSTATUSLOG].mc_addr), + NULL)) + smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetToolsDramAddrLow, - lower_32_bits(priv->smu_tables.entry[TABLE_PMSTATUSLOG].mc_addr)); + lower_32_bits(priv->smu_tables.entry[TABLE_PMSTATUSLOG].mc_addr), + NULL); } return 0; } diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.c index 16aa171971d3..2fb97554134f 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/vega20_smumgr.c @@ -175,18 +175,20 @@ static int vega20_copy_table_from_smc(struct pp_hwmgr *hwmgr, PP_ASSERT_WITH_CODE(priv->smu_tables.entry[table_id].size != 0, "Invalid SMU Table Length!", return -EINVAL); - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrHigh, - upper_32_bits(priv->smu_tables.entry[table_id].mc_addr))) == 0, + upper_32_bits(priv->smu_tables.entry[table_id].mc_addr), + NULL)) == 0, "[CopyTableFromSMC] Attempt to Set Dram Addr High Failed!", return ret); - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrLow, - lower_32_bits(priv->smu_tables.entry[table_id].mc_addr))) == 0, + lower_32_bits(priv->smu_tables.entry[table_id].mc_addr), + NULL)) == 0, "[CopyTableFromSMC] Attempt to Set Dram Addr Low Failed!", return ret); - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_TransferTableSmu2Dram, table_id)) == 0, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_TransferTableSmu2Dram, table_id, NULL)) == 0, "[CopyTableFromSMC] Attempt to Transfer Table From SMU Failed!", return ret); @@ -224,18 +226,20 @@ static int vega20_copy_table_to_smc(struct pp_hwmgr *hwmgr, amdgpu_asic_flush_hdp(adev, NULL); - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrHigh, - upper_32_bits(priv->smu_tables.entry[table_id].mc_addr))) == 0, + upper_32_bits(priv->smu_tables.entry[table_id].mc_addr), + NULL)) == 0, "[CopyTableToSMC] Attempt to Set Dram Addr High Failed!", return ret); - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrLow, - lower_32_bits(priv->smu_tables.entry[table_id].mc_addr))) == 0, + lower_32_bits(priv->smu_tables.entry[table_id].mc_addr), + NULL)) == 0, "[CopyTableToSMC] Attempt to Set Dram Addr Low Failed!", return ret); - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_TransferTableDram2Smu, table_id)) == 0, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_TransferTableDram2Smu, table_id, NULL)) == 0, "[CopyTableToSMC] Attempt to Transfer Table To SMU Failed!", return ret); @@ -255,18 +259,22 @@ int vega20_set_activity_monitor_coeff(struct pp_hwmgr *hwmgr, amdgpu_asic_flush_hdp(adev, NULL); - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrHigh, - upper_32_bits(priv->smu_tables.entry[TABLE_ACTIVITY_MONITOR_COEFF].mc_addr))) == 0, + upper_32_bits(priv->smu_tables.entry[TABLE_ACTIVITY_MONITOR_COEFF].mc_addr), + NULL)) == 0, "[SetActivityMonitor] Attempt to Set Dram Addr High Failed!", return ret); - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrLow, - lower_32_bits(priv->smu_tables.entry[TABLE_ACTIVITY_MONITOR_COEFF].mc_addr))) == 0, + lower_32_bits(priv->smu_tables.entry[TABLE_ACTIVITY_MONITOR_COEFF].mc_addr), + NULL)) == 0, "[SetActivityMonitor] Attempt to Set Dram Addr Low Failed!", return ret); - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_TransferTableDram2Smu, TABLE_ACTIVITY_MONITOR_COEFF | (workload_type << 16))) == 0, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_TransferTableDram2Smu, + TABLE_ACTIVITY_MONITOR_COEFF | (workload_type << 16), + NULL)) == 0, "[SetActivityMonitor] Attempt to Transfer Table To SMU Failed!", return ret); @@ -281,19 +289,21 @@ int vega20_get_activity_monitor_coeff(struct pp_hwmgr *hwmgr, struct amdgpu_device *adev = hwmgr->adev; int ret = 0; - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrHigh, - upper_32_bits(priv->smu_tables.entry[TABLE_ACTIVITY_MONITOR_COEFF].mc_addr))) == 0, + upper_32_bits(priv->smu_tables.entry[TABLE_ACTIVITY_MONITOR_COEFF].mc_addr), + NULL)) == 0, "[GetActivityMonitor] Attempt to Set Dram Addr High Failed!", return ret); - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrLow, - lower_32_bits(priv->smu_tables.entry[TABLE_ACTIVITY_MONITOR_COEFF].mc_addr))) == 0, + lower_32_bits(priv->smu_tables.entry[TABLE_ACTIVITY_MONITOR_COEFF].mc_addr), + NULL)) == 0, "[GetActivityMonitor] Attempt to Set Dram Addr Low Failed!", return ret); - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_TransferTableSmu2Dram, - TABLE_ACTIVITY_MONITOR_COEFF | (workload_type << 16))) == 0, + TABLE_ACTIVITY_MONITOR_COEFF | (workload_type << 16), NULL)) == 0, "[GetActivityMonitor] Attempt to Transfer Table From SMU Failed!", return ret); @@ -316,21 +326,21 @@ int vega20_enable_smc_features(struct pp_hwmgr *hwmgr, smu_features_high = (uint32_t)((feature_mask & SMU_FEATURES_HIGH_MASK) >> SMU_FEATURES_HIGH_SHIFT); if (enable) { - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_EnableSmuFeaturesLow, smu_features_low)) == 0, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_EnableSmuFeaturesLow, smu_features_low, NULL)) == 0, "[EnableDisableSMCFeatures] Attempt to enable SMU features Low failed!", return ret); - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_EnableSmuFeaturesHigh, smu_features_high)) == 0, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_EnableSmuFeaturesHigh, smu_features_high, NULL)) == 0, "[EnableDisableSMCFeatures] Attempt to enable SMU features High failed!", return ret); } else { - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_DisableSmuFeaturesLow, smu_features_low)) == 0, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_DisableSmuFeaturesLow, smu_features_low, NULL)) == 0, "[EnableDisableSMCFeatures] Attempt to disable SMU features Low failed!", return ret); - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_DisableSmuFeaturesHigh, smu_features_high)) == 0, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_DisableSmuFeaturesHigh, smu_features_high, NULL)) == 0, "[EnableDisableSMCFeatures] Attempt to disable SMU features High failed!", return ret); } @@ -347,16 +357,16 @@ int vega20_get_enabled_smc_features(struct pp_hwmgr *hwmgr, if (features_enabled == NULL) return -EINVAL; - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc(hwmgr, - PPSMC_MSG_GetEnabledSmuFeaturesLow)) == 0, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc(hwmgr, + PPSMC_MSG_GetEnabledSmuFeaturesLow, + &smc_features_low)) == 0, "[GetEnabledSMCFeatures] Attempt to get SMU features Low failed!", return ret); - smc_features_low = vega20_get_argument(hwmgr); - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc(hwmgr, - PPSMC_MSG_GetEnabledSmuFeaturesHigh)) == 0, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc(hwmgr, + PPSMC_MSG_GetEnabledSmuFeaturesHigh, + &smc_features_high)) == 0, "[GetEnabledSMCFeatures] Attempt to get SMU features High failed!", return ret); - smc_features_high = vega20_get_argument(hwmgr); *features_enabled = ((((uint64_t)smc_features_low << SMU_FEATURES_LOW_SHIFT) & SMU_FEATURES_LOW_MASK) | (((uint64_t)smc_features_high << SMU_FEATURES_HIGH_SHIFT) & SMU_FEATURES_HIGH_MASK)); @@ -371,13 +381,15 @@ static int vega20_set_tools_address(struct pp_hwmgr *hwmgr) int ret = 0; if (priv->smu_tables.entry[TABLE_PMSTATUSLOG].mc_addr) { - ret = vega20_send_msg_to_smc_with_parameter(hwmgr, + ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetToolsDramAddrHigh, - upper_32_bits(priv->smu_tables.entry[TABLE_PMSTATUSLOG].mc_addr)); + upper_32_bits(priv->smu_tables.entry[TABLE_PMSTATUSLOG].mc_addr), + NULL); if (!ret) - ret = vega20_send_msg_to_smc_with_parameter(hwmgr, + ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetToolsDramAddrLow, - lower_32_bits(priv->smu_tables.entry[TABLE_PMSTATUSLOG].mc_addr)); + lower_32_bits(priv->smu_tables.entry[TABLE_PMSTATUSLOG].mc_addr), + NULL); } return ret; @@ -389,14 +401,16 @@ int vega20_set_pptable_driver_address(struct pp_hwmgr *hwmgr) (struct vega20_smumgr *)(hwmgr->smu_backend); int ret = 0; - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrHigh, - upper_32_bits(priv->smu_tables.entry[TABLE_PPTABLE].mc_addr))) == 0, + upper_32_bits(priv->smu_tables.entry[TABLE_PPTABLE].mc_addr), + NULL)) == 0, "[SetPPtabeDriverAddress] Attempt to Set Dram Addr High Failed!", return ret); - PP_ASSERT_WITH_CODE((ret = vega20_send_msg_to_smc_with_parameter(hwmgr, + PP_ASSERT_WITH_CODE((ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetDriverDramAddrLow, - lower_32_bits(priv->smu_tables.entry[TABLE_PPTABLE].mc_addr))) == 0, + lower_32_bits(priv->smu_tables.entry[TABLE_PPTABLE].mc_addr), + NULL)) == 0, "[SetPPtabeDriverAddress] Attempt to Set Dram Addr Low Failed!", return ret); diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.c index b0e0d67cd54b..3da71a088b92 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.c @@ -356,7 +356,8 @@ static int vegam_update_uvd_smc_table(struct pp_hwmgr *hwmgr) PHM_PlatformCaps_StablePState)) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_UVDDPM_SetEnabledMask, - (uint32_t)(1 << smu_data->smc_state_table.UvdBootLevel)); + (uint32_t)(1 << smu_data->smc_state_table.UvdBootLevel), + NULL); return 0; } @@ -388,7 +389,8 @@ static int vegam_update_vce_smc_table(struct pp_hwmgr *hwmgr) if (phm_cap_enabled(hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_StablePState)) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_VCEDPM_SetEnabledMask, - (uint32_t)1 << smu_data->smc_state_table.VceBootLevel); + (uint32_t)1 << smu_data->smc_state_table.VceBootLevel, + NULL); return 0; } @@ -1906,7 +1908,8 @@ static int vegam_enable_reconfig_cus(struct pp_hwmgr *hwmgr) smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_EnableModeSwitchRLCNotification, - adev->gfx.cu_info.number); + adev->gfx.cu_info.number, + NULL); return 0; } @@ -2060,7 +2063,7 @@ static int vegam_init_smc_table(struct pp_hwmgr *hwmgr) table->AcDcGpio = gpio_pin.uc_gpio_pin_bit_shift; if (phm_cap_enabled(hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_AutomaticDCTransition) && - !smum_send_msg_to_smc(hwmgr, PPSMC_MSG_UseNewGPIOScheme)) + !smum_send_msg_to_smc(hwmgr, PPSMC_MSG_UseNewGPIOScheme, NULL)) phm_cap_set(hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_SMCtoPPLIBAcdcGpioScheme); } else { @@ -2250,10 +2253,12 @@ int vegam_thermal_avfs_enable(struct pp_hwmgr *hwmgr) if (!hwmgr->avfs_supported) return 0; - ret = smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnableAvfs); + ret = smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnableAvfs, NULL); if (!ret) { if (data->apply_avfs_cks_off_voltage) - ret = smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ApplyAvfsCksOffVoltage); + ret = smum_send_msg_to_smc(hwmgr, + PPSMC_MSG_ApplyAvfsCksOffVoltage, + NULL); } return ret; @@ -2279,6 +2284,7 @@ const struct pp_smumgr_func vegam_smu_funcs = { .request_smu_load_specific_fw = NULL, .send_msg_to_smc = smu7_send_msg_to_smc, .send_msg_to_smc_with_parameter = smu7_send_msg_to_smc_with_parameter, + .get_argument = smu7_get_argument, .process_firmware_header = vegam_process_firmware_header, .is_dpm_running = vegam_is_dpm_running, .get_mac_definition = vegam_get_mac_definition, diff --git a/drivers/gpu/drm/amd/powerplay/vega20_ppt.c b/drivers/gpu/drm/amd/powerplay/vega20_ppt.c index 3f1044326dcb..61923530b2e4 100644 --- a/drivers/gpu/drm/amd/powerplay/vega20_ppt.c +++ b/drivers/gpu/drm/amd/powerplay/vega20_ppt.c @@ -1796,7 +1796,7 @@ static int vega20_get_power_profile_mode(struct smu_context *smu, char *buf) "PD_Data_error_rate_coeff"}; int result = 0; - if (!smu->pm_enabled || !buf) + if (!buf) return -EINVAL; size += sprintf(buf + size, "%16s %s %s %s %s %s %s %s %s %s %s\n", @@ -1887,8 +1887,6 @@ static int vega20_set_power_profile_mode(struct smu_context *smu, long *input, u smu->power_profile_mode = input[size]; - if (!smu->pm_enabled) - return ret; if (smu->power_profile_mode > PP_SMC_POWER_PROFILE_CUSTOM) { pr_err("Invalid power profile mode %d\n", smu->power_profile_mode); return -EINVAL; |