diff options
| author | Dave Airlie <airlied@redhat.com> | 2023-06-15 14:11:22 +1000 |
|---|---|---|
| committer | Dave Airlie <airlied@redhat.com> | 2023-06-15 14:11:22 +1000 |
| commit | 901bdf5ea1a836400ee69aa32b04e9c209271ec7 (patch) | |
| tree | ccb1851c8a71e776dbccf1ccae132dc9b5f093c6 /drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c | |
| parent | ba57b9b11f78530146f02b776854b2b6b6d344a4 (diff) | |
| parent | 3b718dcaf163d17fe907ea098c8449e0cd6bc271 (diff) | |
| download | lwn-901bdf5ea1a836400ee69aa32b04e9c209271ec7.tar.gz lwn-901bdf5ea1a836400ee69aa32b04e9c209271ec7.zip | |
Merge tag 'amd-drm-next-6.5-2023-06-09' of https://gitlab.freedesktop.org/agd5f/linux into drm-next
amd-drm-next-6.5-2023-06-02:
amdgpu:
- SR-IOV fixes
- Warning fixes
- Misc code cleanups and spelling fixes
- DCN 3.2 updates
- Improved DC FAMS support for better power management
- Improved DC SubVP support for better power management
- DCN 3.1.x fixes
- Max IB size query
- DC GPU reset fixes
- RAS updates
- DCN 3.0.x fixes
- S/G display fixes
- CP shadow buffer support
- Implement connector force callback
- Z8 power improvements
- PSP 13.0.10 vbflash support
- Mode2 reset fixes
- Store MQDs in VRAM to improve queue switch latency
- VCN 3.x fixes
- JPEG 3.x fixes
- Enable DC_FP on LoongArch
- GFXOFF fixes
- GC 9.4.3 partition support
- SDMA 4.4.2 partition support
- VCN/JPEG 4.0.3 partition support
- VCN 4.0.3 updates
- NBIO 7.9 updates
- GC 9.4.3 updates
- Take NUMA into account when allocating memory
- Handle NUMA for partitions
- SMU 13.0.6 updates
- GC 9.4.3 RAS updates
- Stop including unused swiotlb.h
- SMU 13.0.7 fixes
- Fix clock output ordering on some APUs
- Clean up DC FPGA code
- GFX9 preemption fixes
- Misc irq fixes
- S0ix fixes
- Add new DRM_AMDGPU_WERROR config parameter to help with CI
- PCIe fix for RDNA2
- kdoc fixes
- Documentation updates
amdkfd:
- Query TTM mem limit rather than hardcoding it
- GC 9.4.3 partition support
- Handle NUMA for partitions
radeon:
- Fix possible double free
- Stop including unused swiotlb.h
- Fix possible division by zero
ttm:
- Add query for TTM mem limit
- Add NUMA awareness to pools
- Export ttm_pool_fini()
UAPI:
- Add new ctx query flag to better handle GPU resets
Mesa MR: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22290
- Add new interface to query and set shadow buffer for RDNA3
Mesa MR: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21986
- Add new INFO query for max IB size
Proposed userspace: https://gitlab.freedesktop.org/bnieuwenhuizen/mesa/-/commits/ib-rejection-v3
amd-drm-next-6.5-2023-06-09:
amdgpu:
- S0ix fixes
- Initial SMU13 Overdrive support
- kdoc fixes
- Misc clode cleanups
- Flexible array fixes
- Display OTG fixes
- SMU 13.0.6 updates
- Revert some broken clock counter updates
- Misc display fixes
- GFX9 preemption fixes
- Add support for newer EEPROM bad page table format
- Add missing radeon secondary id
- Add support for new colorspace KMS API
- CSA fix
- Stable pstate fixes for APUs
- make vbl interface admin only
- Handle PCI accelerator class
amdkfd:
- Add debugger support for gdb
radeon:
- Fix possible UAF
drm:
- Add Colorspace functionality
UAPI:
- Add debugger interface for enabling gdb
Proposed userspace: https://github.com/ROCm-Developer-Tools/ROCdbgapi/tree/wip-dbgapi
- Add KMS colorspace API
Discussion: https://lists.freedesktop.org/archives/dri-devel/2023-June/408128.html
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230609174817.7764-1-alexander.deucher@amd.com
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c')
| -rw-r--r-- | drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c | 210 |
1 files changed, 198 insertions, 12 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c index 7e80caa05060..91c3574ebed3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c @@ -30,6 +30,7 @@ #include "soc15d.h" #include "v11_structs.h" #include "soc21.h" +#include <uapi/linux/kfd_ioctl.h> enum hqd_dequeue_request_type { NO_ACTION = 0, @@ -78,7 +79,7 @@ static void program_sh_mem_settings_v11(struct amdgpu_device *adev, uint32_t vmi uint32_t sh_mem_config, uint32_t sh_mem_ape1_base, uint32_t sh_mem_ape1_limit, - uint32_t sh_mem_bases) + uint32_t sh_mem_bases, uint32_t inst) { lock_srbm(adev, 0, 0, 0, vmid); @@ -89,7 +90,7 @@ static void program_sh_mem_settings_v11(struct amdgpu_device *adev, uint32_t vmi } static int set_pasid_vmid_mapping_v11(struct amdgpu_device *adev, unsigned int pasid, - unsigned int vmid) + unsigned int vmid, uint32_t inst) { uint32_t value = pasid << IH_VMID_0_LUT__PASID__SHIFT; @@ -101,7 +102,8 @@ static int set_pasid_vmid_mapping_v11(struct amdgpu_device *adev, unsigned int p return 0; } -static int init_interrupts_v11(struct amdgpu_device *adev, uint32_t pipe_id) +static int init_interrupts_v11(struct amdgpu_device *adev, uint32_t pipe_id, + uint32_t inst) { uint32_t mec; uint32_t pipe; @@ -162,7 +164,7 @@ static inline struct v11_sdma_mqd *get_sdma_mqd(void *mqd) static int hqd_load_v11(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id, uint32_t queue_id, uint32_t __user *wptr, uint32_t wptr_shift, uint32_t wptr_mask, - struct mm_struct *mm) + struct mm_struct *mm, uint32_t inst) { struct v11_compute_mqd *m; uint32_t *mqd_hqd; @@ -258,9 +260,9 @@ static int hqd_load_v11(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id, static int hiq_mqd_load_v11(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id, uint32_t queue_id, - uint32_t doorbell_off) + uint32_t doorbell_off, uint32_t inst) { - struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring; + struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring; struct v11_compute_mqd *m; uint32_t mec, pipe; int r; @@ -275,7 +277,7 @@ static int hiq_mqd_load_v11(struct amdgpu_device *adev, void *mqd, pr_debug("kfd: set HIQ, mec:%d, pipe:%d, queue:%d.\n", mec, pipe, queue_id); - spin_lock(&adev->gfx.kiq.ring_lock); + spin_lock(&adev->gfx.kiq[0].ring_lock); r = amdgpu_ring_alloc(kiq_ring, 7); if (r) { pr_err("Failed to alloc KIQ (%d).\n", r); @@ -302,7 +304,7 @@ static int hiq_mqd_load_v11(struct amdgpu_device *adev, void *mqd, amdgpu_ring_commit(kiq_ring); out_unlock: - spin_unlock(&adev->gfx.kiq.ring_lock); + spin_unlock(&adev->gfx.kiq[0].ring_lock); release_queue(adev); return r; @@ -310,7 +312,7 @@ out_unlock: static int hqd_dump_v11(struct amdgpu_device *adev, uint32_t pipe_id, uint32_t queue_id, - uint32_t (**dump)[2], uint32_t *n_regs) + uint32_t (**dump)[2], uint32_t *n_regs, uint32_t inst) { uint32_t i = 0, reg; #define HQD_N_REGS 56 @@ -445,7 +447,7 @@ static int hqd_sdma_dump_v11(struct amdgpu_device *adev, } static bool hqd_is_occupied_v11(struct amdgpu_device *adev, uint64_t queue_address, - uint32_t pipe_id, uint32_t queue_id) + uint32_t pipe_id, uint32_t queue_id, uint32_t inst) { uint32_t act; bool retval = false; @@ -486,7 +488,7 @@ static bool hqd_sdma_is_occupied_v11(struct amdgpu_device *adev, void *mqd) static int hqd_destroy_v11(struct amdgpu_device *adev, void *mqd, enum kfd_preempt_type reset_type, unsigned int utimeout, uint32_t pipe_id, - uint32_t queue_id) + uint32_t queue_id, uint32_t inst) { enum hqd_dequeue_request_type type; unsigned long end_jiffies; @@ -571,7 +573,7 @@ static int hqd_sdma_destroy_v11(struct amdgpu_device *adev, void *mqd, static int wave_control_execute_v11(struct amdgpu_device *adev, uint32_t gfx_index_val, - uint32_t sq_cmd) + uint32_t sq_cmd, uint32_t inst) { uint32_t data = 0; @@ -606,6 +608,183 @@ static void set_vm_context_page_table_base_v11(struct amdgpu_device *adev, adev->gfxhub.funcs->setup_vm_pt_regs(adev, vmid, page_table_base); } +/* + * Returns TRAP_EN, EXCP_EN and EXCP_REPLACE. + * + * restore_dbg_registers is ignored here but is a general interface requirement + * for devices that support GFXOFF and where the RLC save/restore list + * does not support hw registers for debugging i.e. the driver has to manually + * initialize the debug mode registers after it has disabled GFX off during the + * debug session. + */ +static uint32_t kgd_gfx_v11_enable_debug_trap(struct amdgpu_device *adev, + bool restore_dbg_registers, + uint32_t vmid) +{ + uint32_t data = 0; + + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, 0); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE, 0); + + return data; +} + +/* Returns TRAP_EN, EXCP_EN and EXCP_REPLACE. */ +static uint32_t kgd_gfx_v11_disable_debug_trap(struct amdgpu_device *adev, + bool keep_trap_enabled, + uint32_t vmid) +{ + uint32_t data = 0; + + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, keep_trap_enabled); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, 0); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE, 0); + + return data; +} + +static int kgd_gfx_v11_validate_trap_override_request(struct amdgpu_device *adev, + uint32_t trap_override, + uint32_t *trap_mask_supported) +{ + *trap_mask_supported &= KFD_DBG_TRAP_MASK_FP_INVALID | + KFD_DBG_TRAP_MASK_FP_INPUT_DENORMAL | + KFD_DBG_TRAP_MASK_FP_DIVIDE_BY_ZERO | + KFD_DBG_TRAP_MASK_FP_OVERFLOW | + KFD_DBG_TRAP_MASK_FP_UNDERFLOW | + KFD_DBG_TRAP_MASK_FP_INEXACT | + KFD_DBG_TRAP_MASK_INT_DIVIDE_BY_ZERO | + KFD_DBG_TRAP_MASK_DBG_ADDRESS_WATCH | + KFD_DBG_TRAP_MASK_DBG_MEMORY_VIOLATION; + + if (adev->ip_versions[GC_HWIP][0] >= IP_VERSION(11, 0, 4)) + *trap_mask_supported |= KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_START | + KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_END; + + if (trap_override != KFD_DBG_TRAP_OVERRIDE_OR && + trap_override != KFD_DBG_TRAP_OVERRIDE_REPLACE) + return -EPERM; + + return 0; +} + +static uint32_t trap_mask_map_sw_to_hw(uint32_t mask) +{ + uint32_t trap_on_start = (mask & KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_START) ? 1 : 0; + uint32_t trap_on_end = (mask & KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_END) ? 1 : 0; + uint32_t excp_en = mask & (KFD_DBG_TRAP_MASK_FP_INVALID | + KFD_DBG_TRAP_MASK_FP_INPUT_DENORMAL | + KFD_DBG_TRAP_MASK_FP_DIVIDE_BY_ZERO | + KFD_DBG_TRAP_MASK_FP_OVERFLOW | + KFD_DBG_TRAP_MASK_FP_UNDERFLOW | + KFD_DBG_TRAP_MASK_FP_INEXACT | + KFD_DBG_TRAP_MASK_INT_DIVIDE_BY_ZERO | + KFD_DBG_TRAP_MASK_DBG_ADDRESS_WATCH | + KFD_DBG_TRAP_MASK_DBG_MEMORY_VIOLATION); + uint32_t ret; + + ret = REG_SET_FIELD(0, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, excp_en); + ret = REG_SET_FIELD(ret, SPI_GDBG_PER_VMID_CNTL, TRAP_ON_START, trap_on_start); + ret = REG_SET_FIELD(ret, SPI_GDBG_PER_VMID_CNTL, TRAP_ON_END, trap_on_end); + + return ret; +} + +static uint32_t trap_mask_map_hw_to_sw(uint32_t mask) +{ + uint32_t ret = REG_GET_FIELD(mask, SPI_GDBG_PER_VMID_CNTL, EXCP_EN); + + if (REG_GET_FIELD(mask, SPI_GDBG_PER_VMID_CNTL, TRAP_ON_START)) + ret |= KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_START; + + if (REG_GET_FIELD(mask, SPI_GDBG_PER_VMID_CNTL, TRAP_ON_END)) + ret |= KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_END; + + return ret; +} + +/* Returns TRAP_EN, EXCP_EN and EXCP_REPLACE. */ +static uint32_t kgd_gfx_v11_set_wave_launch_trap_override(struct amdgpu_device *adev, + uint32_t vmid, + uint32_t trap_override, + uint32_t trap_mask_bits, + uint32_t trap_mask_request, + uint32_t *trap_mask_prev, + uint32_t kfd_dbg_trap_cntl_prev) +{ + uint32_t data = 0; + + *trap_mask_prev = trap_mask_map_hw_to_sw(kfd_dbg_trap_cntl_prev); + + data = (trap_mask_bits & trap_mask_request) | (*trap_mask_prev & ~trap_mask_request); + data = trap_mask_map_sw_to_hw(data); + + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE, trap_override); + + return data; +} + +static uint32_t kgd_gfx_v11_set_wave_launch_mode(struct amdgpu_device *adev, + uint8_t wave_launch_mode, + uint32_t vmid) +{ + uint32_t data = 0; + + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, LAUNCH_MODE, wave_launch_mode); + + return data; +} + +#define TCP_WATCH_STRIDE (regTCP_WATCH1_ADDR_H - regTCP_WATCH0_ADDR_H) +static uint32_t kgd_gfx_v11_set_address_watch(struct amdgpu_device *adev, + uint64_t watch_address, + uint32_t watch_address_mask, + uint32_t watch_id, + uint32_t watch_mode, + uint32_t debug_vmid) +{ + uint32_t watch_address_high; + uint32_t watch_address_low; + uint32_t watch_address_cntl; + + watch_address_cntl = 0; + watch_address_low = lower_32_bits(watch_address); + watch_address_high = upper_32_bits(watch_address) & 0xffff; + + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + MODE, + watch_mode); + + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + MASK, + watch_address_mask >> 7); + + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + VALID, + 1); + + WREG32_RLC((SOC15_REG_OFFSET(GC, 0, regTCP_WATCH0_ADDR_H) + + (watch_id * TCP_WATCH_STRIDE)), + watch_address_high); + + WREG32_RLC((SOC15_REG_OFFSET(GC, 0, regTCP_WATCH0_ADDR_L) + + (watch_id * TCP_WATCH_STRIDE)), + watch_address_low); + + return watch_address_cntl; +} + +static uint32_t kgd_gfx_v11_clear_address_watch(struct amdgpu_device *adev, + uint32_t watch_id) +{ + return 0; +} + const struct kfd2kgd_calls gfx_v11_kfd2kgd = { .program_sh_mem_settings = program_sh_mem_settings_v11, .set_pasid_vmid_mapping = set_pasid_vmid_mapping_v11, @@ -622,4 +801,11 @@ const struct kfd2kgd_calls gfx_v11_kfd2kgd = { .wave_control_execute = wave_control_execute_v11, .get_atc_vmid_pasid_mapping_info = NULL, .set_vm_context_page_table_base = set_vm_context_page_table_base_v11, + .enable_debug_trap = kgd_gfx_v11_enable_debug_trap, + .disable_debug_trap = kgd_gfx_v11_disable_debug_trap, + .validate_trap_override_request = kgd_gfx_v11_validate_trap_override_request, + .set_wave_launch_trap_override = kgd_gfx_v11_set_wave_launch_trap_override, + .set_wave_launch_mode = kgd_gfx_v11_set_wave_launch_mode, + .set_address_watch = kgd_gfx_v11_set_address_watch, + .clear_address_watch = kgd_gfx_v11_clear_address_watch }; |