Merge branch 'drm-next' of git://people.freedesktop.org/~airlied/linux

Pull drm tree changes from Dave Airlie:
 "This is the main drm pull request, I have some overlap with sound and
  arm-soc, the sound patch is acked and may conflict based on -next
  reports but should be a trivial fixup, which I'll leave to you!

  Highlights:

   - new drivers:

     MSM driver from Rob Clark

   - non-drm:

     switcheroo and hdmi audio driver support for secondary GPU
     poweroff, so drivers can use runtime PM to poweroff the GPUs.  This
     can save 5 or 6W on some optimus laptops.

   - drm core:

     combined GEM and TTM VMA manager
     per-filp mmap permission tracking
     initial rendernode support (via a runtime enable for now, until we get api stable),
     remove old proc support,
     lots of cleanups of legacy code
     hdmi vendor infoframes and 4k modes
     lots of gem/prime locking and races fixes
     async pageflip scaffolding
     drm bridge objects

   - i915:

     Haswell PC8+ support and eLLC support, HDMI 4K support, initial
     per-process VMA pieces, watermark reworks, convert to generic hdmi
     infoframes, encoder reworking, fastboot support,

   - radeon:

     CIK PM support, remove 3d blit code in favour of DMA engines,
     Berlin GPU support, HDMI audio fixes

   - nouveau:

     secondary GPU power down support for optimus laptops, lots of
     fixes, use MSI, VP3 engine support

   - exynos:

     runtime pm support for g2d, DT support, remove non-DT,

   - tda998x i2c driver:

     lots of fixes for sync issues

   - gma500:

     lots of cleanups

   - rcar:

     add LVDS support, fbdev emulation,

   - tegra:

     just minor fixes"

* 'drm-next' of git://people.freedesktop.org/~airlied/linux: (684 commits)
  drm/exynos: Fix build error with exynos_drm_connector.c
  drm/exynos: Remove non-DT support in exynos_drm_fimd
  drm/exynos: Remove non-DT support in exynos_hdmi
  drm/exynos: Remove non-DT support in exynos_drm_g2d
  drm/exynos: Remove non-DT support in exynos_hdmiphy
  drm/exynos: Remove non-DT support in exynos_ddc
  drm/exynos: Make Exynos DRM drivers depend on OF
  drm/exynos: Consider fallback option to allocation fail
  drm/exynos: fimd: move platform data parsing to separate function
  drm/exynos: fimd: get signal polarities from device tree
  drm/exynos: fimd: replace struct fb_videomode with videomode
  drm/exynos: check a pixel format to a particular window layer
  drm/exynos: fix fimd pixel format setting
  drm/exynos: Add NULL pointer check
  drm/exynos: Remove redundant error messages
  drm/exynos: Add missing of.h header include
  drm/exynos: Remove redundant NULL check in exynos_drm_buf
  drm/exynos: add device tree support for rotator
  drm/exynos: Add missing includes
  drm/exynos: add runtime pm interfaces to g2d driver
  ...

6 files changed, 271 insertions, 7 deletions

diff --git a/include/uapi/drm/Kbuild b/include/uapi/drm/Kbuild
index 119487e05e65..2d9a25daab05 100644
--- a/include/uapi/drm/Kbuild
+++ b/include/uapi/drm/Kbuild
@@ -16,3 +16,4 @@ header-y += sis_drm.h
 header-y += tegra_drm.h
 header-y += via_drm.h
 header-y += vmwgfx_drm.h
+header-y += msm_drm.h
diff --git a/include/uapi/drm/drm.h b/include/uapi/drm/drm.h
index 238a166b9fe6..ece867889cc7 100644
--- a/include/uapi/drm/drm.h
+++ b/include/uapi/drm/drm.h
@@ -181,7 +181,7 @@ enum drm_map_type {
 	_DRM_AGP = 3,		  /**< AGP/GART */
 	_DRM_SCATTER_GATHER = 4,  /**< Scatter/gather memory for PCI DMA */
 	_DRM_CONSISTENT = 5,	  /**< Consistent memory for PCI DMA */
-	_DRM_GEM = 6,		  /**< GEM object */
+	_DRM_GEM = 6,		  /**< GEM object (obsolete) */
 };
 
 /**
@@ -780,6 +780,7 @@ struct drm_event_vblank {
 #define DRM_CAP_DUMB_PREFER_SHADOW 0x4
 #define DRM_CAP_PRIME 0x5
 #define DRM_CAP_TIMESTAMP_MONOTONIC 0x6
+#define DRM_CAP_ASYNC_PAGE_FLIP 0x7
 
 #define DRM_PRIME_CAP_IMPORT 0x1
 #define DRM_PRIME_CAP_EXPORT 0x2
diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h
index 53db7cea373b..550811712f78 100644
--- a/include/uapi/drm/drm_mode.h
+++ b/include/uapi/drm/drm_mode.h
@@ -412,7 +412,8 @@ struct drm_mode_crtc_lut {
 };
 
 #define DRM_MODE_PAGE_FLIP_EVENT 0x01
-#define DRM_MODE_PAGE_FLIP_FLAGS DRM_MODE_PAGE_FLIP_EVENT
+#define DRM_MODE_PAGE_FLIP_ASYNC 0x02
+#define DRM_MODE_PAGE_FLIP_FLAGS (DRM_MODE_PAGE_FLIP_EVENT|DRM_MODE_PAGE_FLIP_ASYNC)
 
 /*
  * Request a page flip on the specified crtc.
@@ -426,11 +427,14 @@ struct drm_mode_crtc_lut {
  * flip is already pending as the ioctl is called, EBUSY will be
  * returned.
  *
- * The ioctl supports one flag, DRM_MODE_PAGE_FLIP_EVENT, which will
- * request that drm sends back a vblank event (see drm.h: struct
- * drm_event_vblank) when the page flip is done.  The user_data field
- * passed in with this ioctl will be returned as the user_data field
- * in the vblank event struct.
+ * Flag DRM_MODE_PAGE_FLIP_EVENT requests that drm sends back a vblank
+ * event (see drm.h: struct drm_event_vblank) when the page flip is
+ * done.  The user_data field passed in with this ioctl will be
+ * returned as the user_data field in the vblank event struct.
+ *
+ * Flag DRM_MODE_PAGE_FLIP_ASYNC requests that the flip happen
+ * 'as soon as possible', meaning that it not delay waiting for vblank.
+ * This may cause tearing on the screen.
  *
  * The reserved field must be zero until we figure out something
  * clever to use it for.
diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h
index 923ed7fe5775..55bb5729bd78 100644
--- a/include/uapi/drm/i915_drm.h
+++ b/include/uapi/drm/i915_drm.h
@@ -33,6 +33,30 @@
  * subject to backwards-compatibility constraints.
  */
 
+/**
+ * DOC: uevents generated by i915 on it's device node
+ *
+ * I915_L3_PARITY_UEVENT - Generated when the driver receives a parity mismatch
+ *	event from the gpu l3 cache. Additional information supplied is ROW,
+ *	BANK, SUBBANK of the affected cacheline. Userspace should keep track of
+ *	these events and if a specific cache-line seems to have a persistent
+ *	error remap it with the l3 remapping tool supplied in intel-gpu-tools.
+ *	The value supplied with the event is always 1.
+ *
+ * I915_ERROR_UEVENT - Generated upon error detection, currently only via
+ *	hangcheck. The error detection event is a good indicator of when things
+ *	began to go badly. The value supplied with the event is a 1 upon error
+ *	detection, and a 0 upon reset completion, signifying no more error
+ *	exists. NOTE: Disabling hangcheck or reset via module parameter will
+ *	cause the related events to not be seen.
+ *
+ * I915_RESET_UEVENT - Event is generated just before an attempt to reset the
+ *	the GPU. The value supplied with the event is always 1. NOTE: Disable
+ *	reset via module parameter will cause this event to not be seen.
+ */
+#define I915_L3_PARITY_UEVENT		"L3_PARITY_ERROR"
+#define I915_ERROR_UEVENT		"ERROR"
+#define I915_RESET_UEVENT		"RESET"
 
 /* Each region is a minimum of 16k, and there are at most 255 of them.
  */
@@ -310,6 +334,7 @@ typedef struct drm_i915_irq_wait {
 #define I915_PARAM_HAS_PINNED_BATCHES	 24
 #define I915_PARAM_HAS_EXEC_NO_RELOC	 25
 #define I915_PARAM_HAS_EXEC_HANDLE_LUT   26
+#define I915_PARAM_HAS_WT     	 	 27
 
 typedef struct drm_i915_getparam {
 	int param;
@@ -744,8 +769,32 @@ struct drm_i915_gem_busy {
 	__u32 busy;
 };
 
+/**
+ * I915_CACHING_NONE
+ *
+ * GPU access is not coherent with cpu caches. Default for machines without an
+ * LLC.
+ */
 #define I915_CACHING_NONE		0
+/**
+ * I915_CACHING_CACHED
+ *
+ * GPU access is coherent with cpu caches and furthermore the data is cached in
+ * last-level caches shared between cpu cores and the gpu GT. Default on
+ * machines with HAS_LLC.
+ */
 #define I915_CACHING_CACHED		1
+/**
+ * I915_CACHING_DISPLAY
+ *
+ * Special GPU caching mode which is coherent with the scanout engines.
+ * Transparently falls back to I915_CACHING_NONE on platforms where no special
+ * cache mode (like write-through or gfdt flushing) is available. The kernel
+ * automatically sets this mode when using a buffer as a scanout target.
+ * Userspace can manually set this mode to avoid a costly stall and clflush in
+ * the hotpath of drawing the first frame.
+ */
+#define I915_CACHING_DISPLAY		2
 
 struct drm_i915_gem_caching {
 	/**
diff --git a/include/uapi/drm/msm_drm.h b/include/uapi/drm/msm_drm.h
new file mode 100644
index 000000000000..d3c62074016d
--- /dev/null
+++ b/include/uapi/drm/msm_drm.h
@@ -0,0 +1,207 @@
+/*
+ * Copyright (C) 2013 Red Hat
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef __MSM_DRM_H__
+#define __MSM_DRM_H__
+
+#include <stddef.h>
+#include <drm/drm.h>
+
+/* Please note that modifications to all structs defined here are
+ * subject to backwards-compatibility constraints:
+ *  1) Do not use pointers, use uint64_t instead for 32 bit / 64 bit
+ *     user/kernel compatibility
+ *  2) Keep fields aligned to their size
+ *  3) Because of how drm_ioctl() works, we can add new fields at
+ *     the end of an ioctl if some care is taken: drm_ioctl() will
+ *     zero out the new fields at the tail of the ioctl, so a zero
+ *     value should have a backwards compatible meaning.  And for
+ *     output params, userspace won't see the newly added output
+ *     fields.. so that has to be somehow ok.
+ */
+
+#define MSM_PIPE_NONE        0x00
+#define MSM_PIPE_2D0         0x01
+#define MSM_PIPE_2D1         0x02
+#define MSM_PIPE_3D0         0x10
+
+/* timeouts are specified in clock-monotonic absolute times (to simplify
+ * restarting interrupted ioctls).  The following struct is logically the
+ * same as 'struct timespec' but 32/64b ABI safe.
+ */
+struct drm_msm_timespec {
+	int64_t tv_sec;          /* seconds */
+	int64_t tv_nsec;         /* nanoseconds */
+};
+
+#define MSM_PARAM_GPU_ID     0x01
+#define MSM_PARAM_GMEM_SIZE  0x02
+
+struct drm_msm_param {
+	uint32_t pipe;           /* in, MSM_PIPE_x */
+	uint32_t param;          /* in, MSM_PARAM_x */
+	uint64_t value;          /* out (get_param) or in (set_param) */
+};
+
+/*
+ * GEM buffers:
+ */
+
+#define MSM_BO_SCANOUT       0x00000001     /* scanout capable */
+#define MSM_BO_GPU_READONLY  0x00000002
+#define MSM_BO_CACHE_MASK    0x000f0000
+/* cache modes */
+#define MSM_BO_CACHED        0x00010000
+#define MSM_BO_WC            0x00020000
+#define MSM_BO_UNCACHED      0x00040000
+
+struct drm_msm_gem_new {
+	uint64_t size;           /* in */
+	uint32_t flags;          /* in, mask of MSM_BO_x */
+	uint32_t handle;         /* out */
+};
+
+struct drm_msm_gem_info {
+	uint32_t handle;         /* in */
+	uint32_t pad;
+	uint64_t offset;         /* out, offset to pass to mmap() */
+};
+
+#define MSM_PREP_READ        0x01
+#define MSM_PREP_WRITE       0x02
+#define MSM_PREP_NOSYNC      0x04
+
+struct drm_msm_gem_cpu_prep {
+	uint32_t handle;         /* in */
+	uint32_t op;             /* in, mask of MSM_PREP_x */
+	struct drm_msm_timespec timeout;   /* in */
+};
+
+struct drm_msm_gem_cpu_fini {
+	uint32_t handle;         /* in */
+};
+
+/*
+ * Cmdstream Submission:
+ */
+
+/* The value written into the cmdstream is logically:
+ *
+ *   ((relocbuf->gpuaddr + reloc_offset) << shift) | or
+ *
+ * When we have GPU's w/ >32bit ptrs, it should be possible to deal
+ * with this by emit'ing two reloc entries with appropriate shift
+ * values.  Or a new MSM_SUBMIT_CMD_x type would also be an option.
+ *
+ * NOTE that reloc's must be sorted by order of increasing submit_offset,
+ * otherwise EINVAL.
+ */
+struct drm_msm_gem_submit_reloc {
+	uint32_t submit_offset;  /* in, offset from submit_bo */
+	uint32_t or;             /* in, value OR'd with result */
+	int32_t  shift;          /* in, amount of left shift (can be negative) */
+	uint32_t reloc_idx;      /* in, index of reloc_bo buffer */
+	uint64_t reloc_offset;   /* in, offset from start of reloc_bo */
+};
+
+/* submit-types:
+ *   BUF - this cmd buffer is executed normally.
+ *   IB_TARGET_BUF - this cmd buffer is an IB target.  Reloc's are
+ *      processed normally, but the kernel does not setup an IB to
+ *      this buffer in the first-level ringbuffer
+ *   CTX_RESTORE_BUF - only executed if there has been a GPU context
+ *      switch since the last SUBMIT ioctl
+ */
+#define MSM_SUBMIT_CMD_BUF             0x0001
+#define MSM_SUBMIT_CMD_IB_TARGET_BUF   0x0002
+#define MSM_SUBMIT_CMD_CTX_RESTORE_BUF 0x0003
+struct drm_msm_gem_submit_cmd {
+	uint32_t type;           /* in, one of MSM_SUBMIT_CMD_x */
+	uint32_t submit_idx;     /* in, index of submit_bo cmdstream buffer */
+	uint32_t submit_offset;  /* in, offset into submit_bo */
+	uint32_t size;           /* in, cmdstream size */
+	uint32_t pad;
+	uint32_t nr_relocs;      /* in, number of submit_reloc's */
+	uint64_t __user relocs;  /* in, ptr to array of submit_reloc's */
+};
+
+/* Each buffer referenced elsewhere in the cmdstream submit (ie. the
+ * cmdstream buffer(s) themselves or reloc entries) has one (and only
+ * one) entry in the submit->bos[] table.
+ *
+ * As a optimization, the current buffer (gpu virtual address) can be
+ * passed back through the 'presumed' field.  If on a subsequent reloc,
+ * userspace passes back a 'presumed' address that is still valid,
+ * then patching the cmdstream for this entry is skipped.  This can
+ * avoid kernel needing to map/access the cmdstream bo in the common
+ * case.
+ */
+#define MSM_SUBMIT_BO_READ             0x0001
+#define MSM_SUBMIT_BO_WRITE            0x0002
+struct drm_msm_gem_submit_bo {
+	uint32_t flags;          /* in, mask of MSM_SUBMIT_BO_x */
+	uint32_t handle;         /* in, GEM handle */
+	uint64_t presumed;       /* in/out, presumed buffer address */
+};
+
+/* Each cmdstream submit consists of a table of buffers involved, and
+ * one or more cmdstream buffers.  This allows for conditional execution
+ * (context-restore), and IB buffers needed for per tile/bin draw cmds.
+ */
+struct drm_msm_gem_submit {
+	uint32_t pipe;           /* in, MSM_PIPE_x */
+	uint32_t fence;          /* out */
+	uint32_t nr_bos;         /* in, number of submit_bo's */
+	uint32_t nr_cmds;        /* in, number of submit_cmd's */
+	uint64_t __user bos;     /* in, ptr to array of submit_bo's */
+	uint64_t __user cmds;    /* in, ptr to array of submit_cmd's */
+};
+
+/* The normal way to synchronize with the GPU is just to CPU_PREP on
+ * a buffer if you need to access it from the CPU (other cmdstream
+ * submission from same or other contexts, PAGE_FLIP ioctl, etc, all
+ * handle the required synchronization under the hood).  This ioctl
+ * mainly just exists as a way to implement the gallium pipe_fence
+ * APIs without requiring a dummy bo to synchronize on.
+ */
+struct drm_msm_wait_fence {
+	uint32_t fence;          /* in */
+	uint32_t pad;
+	struct drm_msm_timespec timeout;   /* in */
+};
+
+#define DRM_MSM_GET_PARAM              0x00
+/* placeholder:
+#define DRM_MSM_SET_PARAM              0x01
+ */
+#define DRM_MSM_GEM_NEW                0x02
+#define DRM_MSM_GEM_INFO               0x03
+#define DRM_MSM_GEM_CPU_PREP           0x04
+#define DRM_MSM_GEM_CPU_FINI           0x05
+#define DRM_MSM_GEM_SUBMIT             0x06
+#define DRM_MSM_WAIT_FENCE             0x07
+#define DRM_MSM_NUM_IOCTLS             0x08
+
+#define DRM_IOCTL_MSM_GET_PARAM        DRM_IOWR(DRM_COMMAND_BASE + DRM_MSM_GET_PARAM, struct drm_msm_param)
+#define DRM_IOCTL_MSM_GEM_NEW          DRM_IOWR(DRM_COMMAND_BASE + DRM_MSM_GEM_NEW, struct drm_msm_gem_new)
+#define DRM_IOCTL_MSM_GEM_INFO         DRM_IOWR(DRM_COMMAND_BASE + DRM_MSM_GEM_INFO, struct drm_msm_gem_info)
+#define DRM_IOCTL_MSM_GEM_CPU_PREP     DRM_IOW (DRM_COMMAND_BASE + DRM_MSM_GEM_CPU_PREP, struct drm_msm_gem_cpu_prep)
+#define DRM_IOCTL_MSM_GEM_CPU_FINI     DRM_IOW (DRM_COMMAND_BASE + DRM_MSM_GEM_CPU_FINI, struct drm_msm_gem_cpu_fini)
+#define DRM_IOCTL_MSM_GEM_SUBMIT       DRM_IOWR(DRM_COMMAND_BASE + DRM_MSM_GEM_SUBMIT, struct drm_msm_gem_submit)
+#define DRM_IOCTL_MSM_WAIT_FENCE       DRM_IOW (DRM_COMMAND_BASE + DRM_MSM_WAIT_FENCE, struct drm_msm_wait_fence)
+
+#endif /* __MSM_DRM_H__ */
diff --git a/include/uapi/drm/radeon_drm.h b/include/uapi/drm/radeon_drm.h
index 321d4ac5c512..fa8b3adf9ffb 100644
--- a/include/uapi/drm/radeon_drm.h
+++ b/include/uapi/drm/radeon_drm.h
@@ -979,6 +979,8 @@ struct drm_radeon_cs {
 #define RADEON_INFO_RING_WORKING	0x15
 /* SI tile mode array */
 #define RADEON_INFO_SI_TILE_MODE_ARRAY	0x16
+/* query if CP DMA is supported on the compute ring */
+#define RADEON_INFO_SI_CP_DMA_COMPUTE	0x17
 
 
 struct drm_radeon_info {