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

Pull drm updates from Dave Airlie:
 "This is the main drm pull request for 4.5.  I don't think I've missed
  anything too major, I'm mostly back at work now but I'll probably get
  some sleep in 5 years time.

  Summary:

  New drivers:
   - etnaviv:

     GPU driver for the 3D core on the Vivante core used in numerous
     ARM boards.

  Highlights:

  Core:
   - Atomic suspend/resume helpers
   - Move the headers to using userspace friendlier types.
   - Documentation updates
   - Lots of struct_mutex removal.
   - Bunch of DP MST fixes from AMD.

  Panel:
   - More DSI helpers
   - Support for some new basic panels

  i915:
   - Basic Kabylake support
   - DP link training and detect code refactoring
   - fbc/psr fixes
   - FIFO underrun fixes
   - SDE interrupt handling fixes
   - dma-buf/fence support in pageflip path.
   - GPU side for MST audio support

  radeon/amdgpu:
   - Drop UMS support
   - GPUVM/Scheduler optimisations
   - Initial Powerplay support for Tonga/Fiji/CZ/ST
   - ACP audio prerequisites

  nouveau:
   - GK20a instmem improvements
   - PCIE link speed change support

  msm:
   - DSI support for msm8960/apq8064

  tegra:
   - Host1X support for Tegra210 SoC

  vc4:
   - 3D acceleration support

  armada:
   - Get rid of struct mutex

  tda998x:
   - Atomic modesetting support
   - TMDS clock limitations

  omapdrm:
   - Atomic modesetting support
   - improved TILER performance

  rockchip:
   - RK3036 VOP support
   - Atomic modesetting support
   - Synopsys DW MIPI DSI support

  exynos:
   - Runtime PM support
   - of_graph binding for DP panels
   - Cleanup of IPP code
   - Configurable plane support
   - Kernel panic fixes at release time"

* 'drm-next' of git://people.freedesktop.org/~airlied/linux: (711 commits)
  drm/fb_cma_helper: Remove implicit call to disable_unused_functions
  drm/amdgpu: add missing irq.h include
  drm/vmwgfx: Fix a width / pitch mismatch on framebuffer updates
  drm/vmwgfx: Fix an incorrect lock check
  drm: nouveau: fix nouveau_debugfs_init prototype
  drm/nouveau/pci: fix check in nvkm_pcie_set_link
  drm/amdgpu: validate duplicates first
  drm/amdgpu: move VM page tables to the LRU end on CS v2
  drm/ttm: add ttm_bo_move_to_lru_tail function v2
  drm/ttm: fix adding foreign BOs to the swap LRU
  drm/ttm: fix adding foreign BOs to the LRU during init v2
  drm/radeon: use kobj_to_dev()
  drm/amdgpu: use kobj_to_dev()
  drm/amdgpu/cz: force vce clocks when sclks are forced
  drm/amdgpu/cz: force uvd clocks when sclks are forced
  drm/amdgpu/cz: add code to enable forcing VCE clocks
  drm/amdgpu/cz: add code to enable forcing UVD clocks
  drm/amdgpu: fix lost sync_to if scheduler is enabled.
  drm/amd/powerplay: fix static checker warning for return meaningless value.
  drm/sysfs: use kobj_to_dev()
  ...

24 files changed, 1018 insertions, 518 deletions

diff --git a/include/uapi/drm/Kbuild b/include/uapi/drm/Kbuild
index 38d437096c35..9355dd8eff3b 100644
--- a/include/uapi/drm/Kbuild
+++ b/include/uapi/drm/Kbuild
@@ -3,6 +3,7 @@ header-y += drm.h
 header-y += drm_fourcc.h
 header-y += drm_mode.h
 header-y += drm_sarea.h
+header-y += amdgpu_drm.h
 header-y += exynos_drm.h
 header-y += i810_drm.h
 header-y += i915_drm.h
@@ -17,4 +18,5 @@ header-y += tegra_drm.h
 header-y += via_drm.h
 header-y += vmwgfx_drm.h
 header-y += msm_drm.h
+header-y += vc4_drm.h
 header-y += virtgpu_drm.h
diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
index e52933a73580..453a76af123c 100644
--- a/include/uapi/drm/amdgpu_drm.h
+++ b/include/uapi/drm/amdgpu_drm.h
@@ -76,19 +76,19 @@
 
 struct drm_amdgpu_gem_create_in  {
 	/** the requested memory size */
-	uint64_t bo_size;
+	__u64 bo_size;
 	/** physical start_addr alignment in bytes for some HW requirements */
-	uint64_t alignment;
+	__u64 alignment;
 	/** the requested memory domains */
-	uint64_t domains;
+	__u64 domains;
 	/** allocation flags */
-	uint64_t domain_flags;
+	__u64 domain_flags;
 };
 
 struct drm_amdgpu_gem_create_out  {
 	/** returned GEM object handle */
-	uint32_t handle;
-	uint32_t _pad;
+	__u32 handle;
+	__u32 _pad;
 };
 
 union drm_amdgpu_gem_create {
@@ -105,28 +105,28 @@ union drm_amdgpu_gem_create {
 
 struct drm_amdgpu_bo_list_in {
 	/** Type of operation */
-	uint32_t operation;
+	__u32 operation;
 	/** Handle of list or 0 if we want to create one */
-	uint32_t list_handle;
+	__u32 list_handle;
 	/** Number of BOs in list  */
-	uint32_t bo_number;
+	__u32 bo_number;
 	/** Size of each element describing BO */
-	uint32_t bo_info_size;
+	__u32 bo_info_size;
 	/** Pointer to array describing BOs */
-	uint64_t bo_info_ptr;
+	__u64 bo_info_ptr;
 };
 
 struct drm_amdgpu_bo_list_entry {
 	/** Handle of BO */
-	uint32_t bo_handle;
+	__u32 bo_handle;
 	/** New (if specified) BO priority to be used during migration */
-	uint32_t bo_priority;
+	__u32 bo_priority;
 };
 
 struct drm_amdgpu_bo_list_out {
 	/** Handle of resource list  */
-	uint32_t list_handle;
-	uint32_t _pad;
+	__u32 list_handle;
+	__u32 _pad;
 };
 
 union drm_amdgpu_bo_list {
@@ -150,26 +150,26 @@ union drm_amdgpu_bo_list {
 
 struct drm_amdgpu_ctx_in {
 	/** AMDGPU_CTX_OP_* */
-	uint32_t	op;
+	__u32	op;
 	/** For future use, no flags defined so far */
-	uint32_t	flags;
-	uint32_t	ctx_id;
-	uint32_t	_pad;
+	__u32	flags;
+	__u32	ctx_id;
+	__u32	_pad;
 };
 
 union drm_amdgpu_ctx_out {
 		struct {
-			uint32_t	ctx_id;
-			uint32_t	_pad;
+			__u32	ctx_id;
+			__u32	_pad;
 		} alloc;
 
 		struct {
 			/** For future use, no flags defined so far */
-			uint64_t	flags;
+			__u64	flags;
 			/** Number of resets caused by this context so far. */
-			uint32_t	hangs;
+			__u32	hangs;
 			/** Reset status since the last call of the ioctl. */
-			uint32_t	reset_status;
+			__u32	reset_status;
 		} state;
 };
 
@@ -189,12 +189,12 @@ union drm_amdgpu_ctx {
 #define AMDGPU_GEM_USERPTR_REGISTER	(1 << 3)
 
 struct drm_amdgpu_gem_userptr {
-	uint64_t		addr;
-	uint64_t		size;
+	__u64		addr;
+	__u64		size;
 	/* AMDGPU_GEM_USERPTR_* */
-	uint32_t		flags;
+	__u32		flags;
 	/* Resulting GEM handle */
-	uint32_t		handle;
+	__u32		handle;
 };
 
 /* same meaning as the GB_TILE_MODE and GL_MACRO_TILE_MODE fields */
@@ -226,28 +226,28 @@ struct drm_amdgpu_gem_userptr {
 /** The same structure is shared for input/output */
 struct drm_amdgpu_gem_metadata {
 	/** GEM Object handle */
-	uint32_t	handle;
+	__u32	handle;
 	/** Do we want get or set metadata */
-	uint32_t	op;
+	__u32	op;
 	struct {
 		/** For future use, no flags defined so far */
-		uint64_t	flags;
+		__u64	flags;
 		/** family specific tiling info */
-		uint64_t	tiling_info;
-		uint32_t	data_size_bytes;
-		uint32_t	data[64];
+		__u64	tiling_info;
+		__u32	data_size_bytes;
+		__u32	data[64];
 	} data;
 };
 
 struct drm_amdgpu_gem_mmap_in {
 	/** the GEM object handle */
-	uint32_t handle;
-	uint32_t _pad;
+	__u32 handle;
+	__u32 _pad;
 };
 
 struct drm_amdgpu_gem_mmap_out {
 	/** mmap offset from the vma offset manager */
-	uint64_t addr_ptr;
+	__u64 addr_ptr;
 };
 
 union drm_amdgpu_gem_mmap {
@@ -257,18 +257,18 @@ union drm_amdgpu_gem_mmap {
 
 struct drm_amdgpu_gem_wait_idle_in {
 	/** GEM object handle */
-	uint32_t handle;
+	__u32 handle;
 	/** For future use, no flags defined so far */
-	uint32_t flags;
+	__u32 flags;
 	/** Absolute timeout to wait */
-	uint64_t timeout;
+	__u64 timeout;
 };
 
 struct drm_amdgpu_gem_wait_idle_out {
 	/** BO status:  0 - BO is idle, 1 - BO is busy */
-	uint32_t status;
+	__u32 status;
 	/** Returned current memory domain */
-	uint32_t domain;
+	__u32 domain;
 };
 
 union drm_amdgpu_gem_wait_idle {
@@ -278,18 +278,18 @@ union drm_amdgpu_gem_wait_idle {
 
 struct drm_amdgpu_wait_cs_in {
 	/** Command submission handle */
-	uint64_t handle;
+	__u64 handle;
 	/** Absolute timeout to wait */
-	uint64_t timeout;
-	uint32_t ip_type;
-	uint32_t ip_instance;
-	uint32_t ring;
-	uint32_t ctx_id;
+	__u64 timeout;
+	__u32 ip_type;
+	__u32 ip_instance;
+	__u32 ring;
+	__u32 ctx_id;
 };
 
 struct drm_amdgpu_wait_cs_out {
 	/** CS status:  0 - CS completed, 1 - CS still busy */
-	uint64_t status;
+	__u64 status;
 };
 
 union drm_amdgpu_wait_cs {
@@ -303,11 +303,11 @@ union drm_amdgpu_wait_cs {
 /* Sets or returns a value associated with a buffer. */
 struct drm_amdgpu_gem_op {
 	/** GEM object handle */
-	uint32_t	handle;
+	__u32	handle;
 	/** AMDGPU_GEM_OP_* */
-	uint32_t	op;
+	__u32	op;
 	/** Input or return value */
-	uint64_t	value;
+	__u64	value;
 };
 
 #define AMDGPU_VA_OP_MAP			1
@@ -326,18 +326,18 @@ struct drm_amdgpu_gem_op {
 
 struct drm_amdgpu_gem_va {
 	/** GEM object handle */
-	uint32_t handle;
-	uint32_t _pad;
+	__u32 handle;
+	__u32 _pad;
 	/** AMDGPU_VA_OP_* */
-	uint32_t operation;
+	__u32 operation;
 	/** AMDGPU_VM_PAGE_* */
-	uint32_t flags;
+	__u32 flags;
 	/** va address to assign . Must be correctly aligned.*/
-	uint64_t va_address;
+	__u64 va_address;
 	/** Specify offset inside of BO to assign. Must be correctly aligned.*/
-	uint64_t offset_in_bo;
+	__u64 offset_in_bo;
 	/** Specify mapping size. Must be correctly aligned. */
-	uint64_t map_size;
+	__u64 map_size;
 };
 
 #define AMDGPU_HW_IP_GFX          0
@@ -354,24 +354,24 @@ struct drm_amdgpu_gem_va {
 #define AMDGPU_CHUNK_ID_DEPENDENCIES	0x03
 
 struct drm_amdgpu_cs_chunk {
-	uint32_t		chunk_id;
-	uint32_t		length_dw;
-	uint64_t		chunk_data;
+	__u32		chunk_id;
+	__u32		length_dw;
+	__u64		chunk_data;
 };
 
 struct drm_amdgpu_cs_in {
 	/** Rendering context id */
-	uint32_t		ctx_id;
+	__u32		ctx_id;
 	/**  Handle of resource list associated with CS */
-	uint32_t		bo_list_handle;
-	uint32_t		num_chunks;
-	uint32_t		_pad;
-	/** this points to uint64_t * which point to cs chunks */
-	uint64_t		chunks;
+	__u32		bo_list_handle;
+	__u32		num_chunks;
+	__u32		_pad;
+	/** this points to __u64 * which point to cs chunks */
+	__u64		chunks;
 };
 
 struct drm_amdgpu_cs_out {
-	uint64_t handle;
+	__u64 handle;
 };
 
 union drm_amdgpu_cs {
@@ -388,32 +388,32 @@ union drm_amdgpu_cs {
 #define AMDGPU_IB_FLAG_PREAMBLE (1<<1)
 
 struct drm_amdgpu_cs_chunk_ib {
-	uint32_t _pad;
+	__u32 _pad;
 	/** AMDGPU_IB_FLAG_* */
-	uint32_t flags;
+	__u32 flags;
 	/** Virtual address to begin IB execution */
-	uint64_t va_start;
+	__u64 va_start;
 	/** Size of submission */
-	uint32_t ib_bytes;
+	__u32 ib_bytes;
 	/** HW IP to submit to */
-	uint32_t ip_type;
+	__u32 ip_type;
 	/** HW IP index of the same type to submit to  */
-	uint32_t ip_instance;
+	__u32 ip_instance;
 	/** Ring index to submit to */
-	uint32_t ring;
+	__u32 ring;
 };
 
 struct drm_amdgpu_cs_chunk_dep {
-	uint32_t ip_type;
-	uint32_t ip_instance;
-	uint32_t ring;
-	uint32_t ctx_id;
-	uint64_t handle;
+	__u32 ip_type;
+	__u32 ip_instance;
+	__u32 ring;
+	__u32 ctx_id;
+	__u64 handle;
 };
 
 struct drm_amdgpu_cs_chunk_fence {
-	uint32_t handle;
-	uint32_t offset;
+	__u32 handle;
+	__u32 offset;
 };
 
 struct drm_amdgpu_cs_chunk_data {
@@ -486,83 +486,83 @@ struct drm_amdgpu_cs_chunk_data {
 /* Input structure for the INFO ioctl */
 struct drm_amdgpu_info {
 	/* Where the return value will be stored */
-	uint64_t return_pointer;
+	__u64 return_pointer;
 	/* The size of the return value. Just like "size" in "snprintf",
 	 * it limits how many bytes the kernel can write. */
-	uint32_t return_size;
+	__u32 return_size;
 	/* The query request id. */
-	uint32_t query;
+	__u32 query;
 
 	union {
 		struct {
-			uint32_t id;
-			uint32_t _pad;
+			__u32 id;
+			__u32 _pad;
 		} mode_crtc;
 
 		struct {
 			/** AMDGPU_HW_IP_* */
-			uint32_t type;
+			__u32 type;
 			/**
 			 * Index of the IP if there are more IPs of the same
 			 * type. Ignored by AMDGPU_INFO_HW_IP_COUNT.
 			 */
-			uint32_t ip_instance;
+			__u32 ip_instance;
 		} query_hw_ip;
 
 		struct {
-			uint32_t dword_offset;
+			__u32 dword_offset;
 			/** number of registers to read */
-			uint32_t count;
-			uint32_t instance;
+			__u32 count;
+			__u32 instance;
 			/** For future use, no flags defined so far */
-			uint32_t flags;
+			__u32 flags;
 		} read_mmr_reg;
 
 		struct {
 			/** AMDGPU_INFO_FW_* */
-			uint32_t fw_type;
+			__u32 fw_type;
 			/**
 			 * Index of the IP if there are more IPs of
 			 * the same type.
 			 */
-			uint32_t ip_instance;
+			__u32 ip_instance;
 			/**
 			 * Index of the engine. Whether this is used depends
 			 * on the firmware type. (e.g. MEC, SDMA)
 			 */
-			uint32_t index;
-			uint32_t _pad;
+			__u32 index;
+			__u32 _pad;
 		} query_fw;
 	};
 };
 
 struct drm_amdgpu_info_gds {
 	/** GDS GFX partition size */
-	uint32_t gds_gfx_partition_size;
+	__u32 gds_gfx_partition_size;
 	/** GDS compute partition size */
-	uint32_t compute_partition_size;
+	__u32 compute_partition_size;
 	/** total GDS memory size */
-	uint32_t gds_total_size;
+	__u32 gds_total_size;
 	/** GWS size per GFX partition */
-	uint32_t gws_per_gfx_partition;
+	__u32 gws_per_gfx_partition;
 	/** GSW size per compute partition */
-	uint32_t gws_per_compute_partition;
+	__u32 gws_per_compute_partition;
 	/** OA size per GFX partition */
-	uint32_t oa_per_gfx_partition;
+	__u32 oa_per_gfx_partition;
 	/** OA size per compute partition */
-	uint32_t oa_per_compute_partition;
-	uint32_t _pad;
+	__u32 oa_per_compute_partition;
+	__u32 _pad;
 };
 
 struct drm_amdgpu_info_vram_gtt {
-	uint64_t vram_size;
-	uint64_t vram_cpu_accessible_size;
-	uint64_t gtt_size;
+	__u64 vram_size;
+	__u64 vram_cpu_accessible_size;
+	__u64 gtt_size;
 };
 
 struct drm_amdgpu_info_firmware {
-	uint32_t ver;
-	uint32_t feature;
+	__u32 ver;
+	__u32 feature;
 };
 
 #define AMDGPU_VRAM_TYPE_UNKNOWN 0
@@ -576,61 +576,61 @@ struct drm_amdgpu_info_firmware {
 
 struct drm_amdgpu_info_device {
 	/** PCI Device ID */
-	uint32_t device_id;
+	__u32 device_id;
 	/** Internal chip revision: A0, A1, etc.) */
-	uint32_t chip_rev;
-	uint32_t external_rev;
+	__u32 chip_rev;
+	__u32 external_rev;
 	/** Revision id in PCI Config space */
-	uint32_t pci_rev;
-	uint32_t family;
-	uint32_t num_shader_engines;
-	uint32_t num_shader_arrays_per_engine;
+	__u32 pci_rev;
+	__u32 family;
+	__u32 num_shader_engines;
+	__u32 num_shader_arrays_per_engine;
 	/* in KHz */
-	uint32_t gpu_counter_freq;
-	uint64_t max_engine_clock;
-	uint64_t max_memory_clock;
+	__u32 gpu_counter_freq;
+	__u64 max_engine_clock;
+	__u64 max_memory_clock;
 	/* cu information */
-	uint32_t cu_active_number;
-	uint32_t cu_ao_mask;
-	uint32_t cu_bitmap[4][4];
+	__u32 cu_active_number;
+	__u32 cu_ao_mask;
+	__u32 cu_bitmap[4][4];
 	/** Render backend pipe mask. One render backend is CB+DB. */
-	uint32_t enabled_rb_pipes_mask;
-	uint32_t num_rb_pipes;
-	uint32_t num_hw_gfx_contexts;
-	uint32_t _pad;
-	uint64_t ids_flags;
+	__u32 enabled_rb_pipes_mask;
+	__u32 num_rb_pipes;
+	__u32 num_hw_gfx_contexts;
+	__u32 _pad;
+	__u64 ids_flags;
 	/** Starting virtual address for UMDs. */
-	uint64_t virtual_address_offset;
+	__u64 virtual_address_offset;
 	/** The maximum virtual address */
-	uint64_t virtual_address_max;
+	__u64 virtual_address_max;
 	/** Required alignment of virtual addresses. */
-	uint32_t virtual_address_alignment;
+	__u32 virtual_address_alignment;
 	/** Page table entry - fragment size */
-	uint32_t pte_fragment_size;
-	uint32_t gart_page_size;
+	__u32 pte_fragment_size;
+	__u32 gart_page_size;
 	/** constant engine ram size*/
-	uint32_t ce_ram_size;
+	__u32 ce_ram_size;
 	/** video memory type info*/
-	uint32_t vram_type;
+	__u32 vram_type;
 	/** video memory bit width*/
-	uint32_t vram_bit_width;
+	__u32 vram_bit_width;
 	/* vce harvesting instance */
-	uint32_t vce_harvest_config;
+	__u32 vce_harvest_config;
 };
 
 struct drm_amdgpu_info_hw_ip {
 	/** Version of h/w IP */
-	uint32_t  hw_ip_version_major;
-	uint32_t  hw_ip_version_minor;
+	__u32  hw_ip_version_major;
+	__u32  hw_ip_version_minor;
 	/** Capabilities */
-	uint64_t  capabilities_flags;
+	__u64  capabilities_flags;
 	/** command buffer address start alignment*/
-	uint32_t  ib_start_alignment;
+	__u32  ib_start_alignment;
 	/** command buffer size alignment*/
-	uint32_t  ib_size_alignment;
+	__u32  ib_size_alignment;
 	/** Bitmask of available rings. Bit 0 means ring 0, etc. */
-	uint32_t  available_rings;
-	uint32_t  _pad;
+	__u32  available_rings;
+	__u32  _pad;
 };
 
 /*
diff --git a/include/uapi/drm/armada_drm.h b/include/uapi/drm/armada_drm.h
index 8dec3fdc99c7..6de7f0196ca0 100644
--- a/include/uapi/drm/armada_drm.h
+++ b/include/uapi/drm/armada_drm.h
@@ -9,6 +9,8 @@
 #ifndef DRM_ARMADA_IOCTL_H
 #define DRM_ARMADA_IOCTL_H
 
+#include "drm.h"
+
 #define DRM_ARMADA_GEM_CREATE		0x00
 #define DRM_ARMADA_GEM_MMAP		0x02
 #define DRM_ARMADA_GEM_PWRITE		0x03
diff --git a/include/uapi/drm/drm.h b/include/uapi/drm/drm.h
index 3801584a0c53..b4e92eb12044 100644
--- a/include/uapi/drm/drm.h
+++ b/include/uapi/drm/drm.h
@@ -54,6 +54,7 @@ typedef int32_t  __s32;
 typedef uint32_t __u32;
 typedef int64_t  __s64;
 typedef uint64_t __u64;
+typedef size_t   __kernel_size_t;
 typedef unsigned long drm_handle_t;
 
 #endif
@@ -129,11 +130,11 @@ struct drm_version {
 	int version_major;	  /**< Major version */
 	int version_minor;	  /**< Minor version */
 	int version_patchlevel;	  /**< Patch level */
-	size_t name_len;	  /**< Length of name buffer */
+	__kernel_size_t name_len;	  /**< Length of name buffer */
 	char __user *name;	  /**< Name of driver */
-	size_t date_len;	  /**< Length of date buffer */
+	__kernel_size_t date_len;	  /**< Length of date buffer */
 	char __user *date;	  /**< User-space buffer to hold date */
-	size_t desc_len;	  /**< Length of desc buffer */
+	__kernel_size_t desc_len;	  /**< Length of desc buffer */
 	char __user *desc;	  /**< User-space buffer to hold desc */
 };
 
@@ -143,7 +144,7 @@ struct drm_version {
  * \sa drmGetBusid() and drmSetBusId().
  */
 struct drm_unique {
-	size_t unique_len;	  /**< Length of unique */
+	__kernel_size_t unique_len;	  /**< Length of unique */
 	char __user *unique;	  /**< Unique name for driver instantiation */
 };
 
diff --git a/include/uapi/drm/drm_fourcc.h b/include/uapi/drm/drm_fourcc.h
index ee2d542c65f5..4d8da699a623 100644
--- a/include/uapi/drm/drm_fourcc.h
+++ b/include/uapi/drm/drm_fourcc.h
@@ -24,7 +24,7 @@
 #ifndef DRM_FOURCC_H
 #define DRM_FOURCC_H
 
-#include <linux/types.h>
+#include "drm.h"
 
 #define fourcc_code(a, b, c, d) ((__u32)(a) | ((__u32)(b) << 8) | \
 				 ((__u32)(c) << 16) | ((__u32)(d) << 24))
diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h
index 6c11ca401de8..50adb46204c2 100644
--- a/include/uapi/drm/drm_mode.h
+++ b/include/uapi/drm/drm_mode.h
@@ -27,7 +27,7 @@
 #ifndef _DRM_MODE_H
 #define _DRM_MODE_H
 
-#include <linux/types.h>
+#include "drm.h"
 
 #define DRM_DISPLAY_INFO_LEN	32
 #define DRM_CONNECTOR_NAME_LEN	32
@@ -526,14 +526,14 @@ struct drm_mode_crtc_page_flip {
 
 /* create a dumb scanout buffer */
 struct drm_mode_create_dumb {
-	uint32_t height;
-	uint32_t width;
-	uint32_t bpp;
-	uint32_t flags;
+	__u32 height;
+	__u32 width;
+	__u32 bpp;
+	__u32 flags;
 	/* handle, pitch, size will be returned */
-	uint32_t handle;
-	uint32_t pitch;
-	uint64_t size;
+	__u32 handle;
+	__u32 pitch;
+	__u64 size;
 };
 
 /* set up for mmap of a dumb scanout buffer */
@@ -550,7 +550,7 @@ struct drm_mode_map_dumb {
 };
 
 struct drm_mode_destroy_dumb {
-	uint32_t handle;
+	__u32 handle;
 };
 
 /* page-flip flags are valid, plus: */
diff --git a/include/uapi/drm/drm_sarea.h b/include/uapi/drm/drm_sarea.h
index 413a5642d49f..1d1a858a203d 100644
--- a/include/uapi/drm/drm_sarea.h
+++ b/include/uapi/drm/drm_sarea.h
@@ -32,7 +32,7 @@
 #ifndef _DRM_SAREA_H_
 #define _DRM_SAREA_H_
 
-#include <drm/drm.h>
+#include "drm.h"
 
 /* SAREA area needs to be at least a page */
 #if defined(__alpha__)
diff --git a/include/uapi/drm/etnaviv_drm.h b/include/uapi/drm/etnaviv_drm.h
new file mode 100644
index 000000000000..4cc989ad6851
--- /dev/null
+++ b/include/uapi/drm/etnaviv_drm.h
@@ -0,0 +1,222 @@
+/*
+ * Copyright (C) 2015 Etnaviv Project
+ *
+ * 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 __ETNAVIV_DRM_H__
+#define __ETNAVIV_DRM_H__
+
+#include "drm.h"
+
+/* Please note that modifications to all structs defined here are
+ * subject to backwards-compatibility constraints:
+ *  1) Do not use pointers, use __u64 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.
+ */
+
+/* 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_etnaviv_timespec {
+	__s64 tv_sec;          /* seconds */
+	__s64 tv_nsec;         /* nanoseconds */
+};
+
+#define ETNAVIV_PARAM_GPU_MODEL                     0x01
+#define ETNAVIV_PARAM_GPU_REVISION                  0x02
+#define ETNAVIV_PARAM_GPU_FEATURES_0                0x03
+#define ETNAVIV_PARAM_GPU_FEATURES_1                0x04
+#define ETNAVIV_PARAM_GPU_FEATURES_2                0x05
+#define ETNAVIV_PARAM_GPU_FEATURES_3                0x06
+#define ETNAVIV_PARAM_GPU_FEATURES_4                0x07
+
+#define ETNAVIV_PARAM_GPU_STREAM_COUNT              0x10
+#define ETNAVIV_PARAM_GPU_REGISTER_MAX              0x11
+#define ETNAVIV_PARAM_GPU_THREAD_COUNT              0x12
+#define ETNAVIV_PARAM_GPU_VERTEX_CACHE_SIZE         0x13
+#define ETNAVIV_PARAM_GPU_SHADER_CORE_COUNT         0x14
+#define ETNAVIV_PARAM_GPU_PIXEL_PIPES               0x15
+#define ETNAVIV_PARAM_GPU_VERTEX_OUTPUT_BUFFER_SIZE 0x16
+#define ETNAVIV_PARAM_GPU_BUFFER_SIZE               0x17
+#define ETNAVIV_PARAM_GPU_INSTRUCTION_COUNT         0x18
+#define ETNAVIV_PARAM_GPU_NUM_CONSTANTS             0x19
+
+#define ETNA_MAX_PIPES 4
+
+struct drm_etnaviv_param {
+	__u32 pipe;           /* in */
+	__u32 param;          /* in, ETNAVIV_PARAM_x */
+	__u64 value;          /* out (get_param) or in (set_param) */
+};
+
+/*
+ * GEM buffers:
+ */
+
+#define ETNA_BO_CACHE_MASK   0x000f0000
+/* cache modes */
+#define ETNA_BO_CACHED       0x00010000
+#define ETNA_BO_WC           0x00020000
+#define ETNA_BO_UNCACHED     0x00040000
+/* map flags */
+#define ETNA_BO_FORCE_MMU    0x00100000
+
+struct drm_etnaviv_gem_new {
+	__u64 size;           /* in */
+	__u32 flags;          /* in, mask of ETNA_BO_x */
+	__u32 handle;         /* out */
+};
+
+struct drm_etnaviv_gem_info {
+	__u32 handle;         /* in */
+	__u32 pad;
+	__u64 offset;         /* out, offset to pass to mmap() */
+};
+
+#define ETNA_PREP_READ        0x01
+#define ETNA_PREP_WRITE       0x02
+#define ETNA_PREP_NOSYNC      0x04
+
+struct drm_etnaviv_gem_cpu_prep {
+	__u32 handle;         /* in */
+	__u32 op;             /* in, mask of ETNA_PREP_x */
+	struct drm_etnaviv_timespec timeout;   /* in */
+};
+
+struct drm_etnaviv_gem_cpu_fini {
+	__u32 handle;         /* in */
+	__u32 flags;          /* in, placeholder for now, no defined values */
+};
+
+/*
+ * Cmdstream Submission:
+ */
+
+/* The value written into the cmdstream is logically:
+ * relocbuf->gpuaddr + reloc_offset
+ *
+ * NOTE that reloc's must be sorted by order of increasing submit_offset,
+ * otherwise EINVAL.
+ */
+struct drm_etnaviv_gem_submit_reloc {
+	__u32 submit_offset;  /* in, offset from submit_bo */
+	__u32 reloc_idx;      /* in, index of reloc_bo buffer */
+	__u64 reloc_offset;   /* in, offset from start of reloc_bo */
+	__u32 flags;          /* in, placeholder for now, no defined values */
+};
+
+/* 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 ETNA_SUBMIT_BO_READ             0x0001
+#define ETNA_SUBMIT_BO_WRITE            0x0002
+struct drm_etnaviv_gem_submit_bo {
+	__u32 flags;          /* in, mask of ETNA_SUBMIT_BO_x */
+	__u32 handle;         /* in, GEM handle */
+	__u64 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.
+ */
+#define ETNA_PIPE_3D      0x00
+#define ETNA_PIPE_2D      0x01
+#define ETNA_PIPE_VG      0x02
+struct drm_etnaviv_gem_submit {
+	__u32 fence;          /* out */
+	__u32 pipe;           /* in */
+	__u32 exec_state;     /* in, initial execution state (ETNA_PIPE_x) */
+	__u32 nr_bos;         /* in, number of submit_bo's */
+	__u32 nr_relocs;      /* in, number of submit_reloc's */
+	__u32 stream_size;    /* in, cmdstream size */
+	__u64 bos;            /* in, ptr to array of submit_bo's */
+	__u64 relocs;         /* in, ptr to array of submit_reloc's */
+	__u64 stream;         /* in, ptr to cmdstream */
+};
+
+/* 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.
+ */
+#define ETNA_WAIT_NONBLOCK      0x01
+struct drm_etnaviv_wait_fence {
+	__u32 pipe;           /* in */
+	__u32 fence;          /* in */
+	__u32 flags;          /* in, mask of ETNA_WAIT_x */
+	__u32 pad;
+	struct drm_etnaviv_timespec timeout;   /* in */
+};
+
+#define ETNA_USERPTR_READ	0x01
+#define ETNA_USERPTR_WRITE	0x02
+struct drm_etnaviv_gem_userptr {
+	__u64 user_ptr;	/* in, page aligned user pointer */
+	__u64 user_size;	/* in, page aligned user size */
+	__u32 flags;		/* in, flags */
+	__u32 handle;	/* out, non-zero handle */
+};
+
+struct drm_etnaviv_gem_wait {
+	__u32 pipe;				/* in */
+	__u32 handle;				/* in, bo to be waited for */
+	__u32 flags;				/* in, mask of ETNA_WAIT_x  */
+	__u32 pad;
+	struct drm_etnaviv_timespec timeout;	/* in */
+};
+
+#define DRM_ETNAVIV_GET_PARAM          0x00
+/* placeholder:
+#define DRM_ETNAVIV_SET_PARAM          0x01
+ */
+#define DRM_ETNAVIV_GEM_NEW            0x02
+#define DRM_ETNAVIV_GEM_INFO           0x03
+#define DRM_ETNAVIV_GEM_CPU_PREP       0x04
+#define DRM_ETNAVIV_GEM_CPU_FINI       0x05
+#define DRM_ETNAVIV_GEM_SUBMIT         0x06
+#define DRM_ETNAVIV_WAIT_FENCE         0x07
+#define DRM_ETNAVIV_GEM_USERPTR        0x08
+#define DRM_ETNAVIV_GEM_WAIT           0x09
+#define DRM_ETNAVIV_NUM_IOCTLS         0x0a
+
+#define DRM_IOCTL_ETNAVIV_GET_PARAM    DRM_IOWR(DRM_COMMAND_BASE + DRM_ETNAVIV_GET_PARAM, struct drm_etnaviv_param)
+#define DRM_IOCTL_ETNAVIV_GEM_NEW      DRM_IOWR(DRM_COMMAND_BASE + DRM_ETNAVIV_GEM_NEW, struct drm_etnaviv_gem_new)
+#define DRM_IOCTL_ETNAVIV_GEM_INFO     DRM_IOWR(DRM_COMMAND_BASE + DRM_ETNAVIV_GEM_INFO, struct drm_etnaviv_gem_info)
+#define DRM_IOCTL_ETNAVIV_GEM_CPU_PREP DRM_IOW(DRM_COMMAND_BASE + DRM_ETNAVIV_GEM_CPU_PREP, struct drm_etnaviv_gem_cpu_prep)
+#define DRM_IOCTL_ETNAVIV_GEM_CPU_FINI DRM_IOW(DRM_COMMAND_BASE + DRM_ETNAVIV_GEM_CPU_FINI, struct drm_etnaviv_gem_cpu_fini)
+#define DRM_IOCTL_ETNAVIV_GEM_SUBMIT   DRM_IOWR(DRM_COMMAND_BASE + DRM_ETNAVIV_GEM_SUBMIT, struct drm_etnaviv_gem_submit)
+#define DRM_IOCTL_ETNAVIV_WAIT_FENCE   DRM_IOW(DRM_COMMAND_BASE + DRM_ETNAVIV_WAIT_FENCE, struct drm_etnaviv_wait_fence)
+#define DRM_IOCTL_ETNAVIV_GEM_USERPTR  DRM_IOWR(DRM_COMMAND_BASE + DRM_ETNAVIV_GEM_USERPTR, struct drm_etnaviv_gem_userptr)
+#define DRM_IOCTL_ETNAVIV_GEM_WAIT     DRM_IOW(DRM_COMMAND_BASE + DRM_ETNAVIV_GEM_WAIT, struct drm_etnaviv_gem_wait)
+
+#endif /* __ETNAVIV_DRM_H__ */
diff --git a/include/uapi/drm/exynos_drm.h b/include/uapi/drm/exynos_drm.h
index 5575ed1598bd..312c67d744ae 100644
--- a/include/uapi/drm/exynos_drm.h
+++ b/include/uapi/drm/exynos_drm.h
@@ -15,7 +15,7 @@
 #ifndef _UAPI_EXYNOS_DRM_H_
 #define _UAPI_EXYNOS_DRM_H_
 
-#include <drm/drm.h>
+#include "drm.h"
 
 /**
  * User-desired buffer creation information structure.
@@ -27,7 +27,7 @@
  *	- this handle will be set by gem module of kernel side.
  */
 struct drm_exynos_gem_create {
-	uint64_t size;
+	__u64 size;
 	unsigned int flags;
 	unsigned int handle;
 };
@@ -44,7 +44,7 @@ struct drm_exynos_gem_create {
 struct drm_exynos_gem_info {
 	unsigned int handle;
 	unsigned int flags;
-	uint64_t size;
+	__u64 size;
 };
 
 /**
@@ -58,7 +58,7 @@ struct drm_exynos_gem_info {
 struct drm_exynos_vidi_connection {
 	unsigned int connection;
 	unsigned int extensions;
-	uint64_t edid;
+	__u64 edid;
 };
 
 /* memory type definitions. */
diff --git a/include/uapi/drm/i810_drm.h b/include/uapi/drm/i810_drm.h
index 34736efd5824..bdb028723ded 100644
--- a/include/uapi/drm/i810_drm.h
+++ b/include/uapi/drm/i810_drm.h
@@ -1,7 +1,7 @@
 #ifndef _I810_DRM_H_
 #define _I810_DRM_H_
 
-#include <drm/drm.h>
+#include "drm.h"
 
 /* WARNING: These defines must be the same as what the Xserver uses.
  * if you change them, you must change the defines in the Xserver.
diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h
index 67ef73a5d6eb..acf21026c78a 100644
--- a/include/uapi/drm/i915_drm.h
+++ b/include/uapi/drm/i915_drm.h
@@ -27,7 +27,7 @@
 #ifndef _UAPI_I915_DRM_H_
 #define _UAPI_I915_DRM_H_
 
-#include <drm/drm.h>
+#include "drm.h"
 
 /* Please note that modifications to all structs defined here are
  * subject to backwards-compatibility constraints.
@@ -356,6 +356,7 @@ typedef struct drm_i915_irq_wait {
 #define I915_PARAM_EU_TOTAL		 34
 #define I915_PARAM_HAS_GPU_RESET	 35
 #define I915_PARAM_HAS_RESOURCE_STREAMER 36
+#define I915_PARAM_HAS_EXEC_SOFTPIN	 37
 
 typedef struct drm_i915_getparam {
 	__s32 param;
@@ -682,8 +683,12 @@ struct drm_i915_gem_exec_object2 {
 	__u64 alignment;
 
 	/**
-	 * Returned value of the updated offset of the object, for future
-	 * presumed_offset writes.
+	 * When the EXEC_OBJECT_PINNED flag is specified this is populated by
+	 * the user with the GTT offset at which this object will be pinned.
+	 * When the I915_EXEC_NO_RELOC flag is specified this must contain the
+	 * presumed_offset of the object.
+	 * During execbuffer2 the kernel populates it with the value of the
+	 * current GTT offset of the object, for future presumed_offset writes.
 	 */
 	__u64 offset;
 
@@ -691,7 +696,8 @@ struct drm_i915_gem_exec_object2 {
 #define EXEC_OBJECT_NEEDS_GTT	(1<<1)
 #define EXEC_OBJECT_WRITE	(1<<2)
 #define EXEC_OBJECT_SUPPORTS_48B_ADDRESS (1<<3)
-#define __EXEC_OBJECT_UNKNOWN_FLAGS -(EXEC_OBJECT_SUPPORTS_48B_ADDRESS<<1)
+#define EXEC_OBJECT_PINNED	(1<<4)
+#define __EXEC_OBJECT_UNKNOWN_FLAGS -(EXEC_OBJECT_PINNED<<1)
 	__u64 flags;
 
 	__u64 rsvd1;
diff --git a/include/uapi/drm/mga_drm.h b/include/uapi/drm/mga_drm.h
index 2375bfd6e5e9..fca817009e13 100644
--- a/include/uapi/drm/mga_drm.h
+++ b/include/uapi/drm/mga_drm.h
@@ -35,7 +35,7 @@
 #ifndef __MGA_DRM_H__
 #define __MGA_DRM_H__
 
-#include <drm/drm.h>
+#include "drm.h"
 
 /* WARNING: If you change any of these defines, make sure to change the
  * defines in the Xserver file (mga_sarea.h)
diff --git a/include/uapi/drm/msm_drm.h b/include/uapi/drm/msm_drm.h
index 75a232b9a970..81e6e0d1d360 100644
--- a/include/uapi/drm/msm_drm.h
+++ b/include/uapi/drm/msm_drm.h
@@ -18,8 +18,7 @@
 #ifndef __MSM_DRM_H__
 #define __MSM_DRM_H__
 
-#include <stddef.h>
-#include <drm/drm.h>
+#include "drm.h"
 
 /* Please note that modifications to all structs defined here are
  * subject to backwards-compatibility constraints:
@@ -122,7 +121,7 @@ struct drm_msm_gem_cpu_fini {
 struct drm_msm_gem_submit_reloc {
 	__u32 submit_offset;  /* in, offset from submit_bo */
 	__u32 or;             /* in, value OR'd with result */
-	__s32  shift;          /* in, amount of left shift (can be negative) */
+	__s32 shift;          /* in, amount of left shift (can be negative) */
 	__u32 reloc_idx;      /* in, index of reloc_bo buffer */
 	__u64 reloc_offset;   /* in, offset from start of reloc_bo */
 };
diff --git a/include/uapi/drm/nouveau_drm.h b/include/uapi/drm/nouveau_drm.h
index fd594cc73cc0..500d82aecbe4 100644
--- a/include/uapi/drm/nouveau_drm.h
+++ b/include/uapi/drm/nouveau_drm.h
@@ -27,6 +27,8 @@
 
 #define DRM_NOUVEAU_EVENT_NVIF                                       0x80000000
 
+#include <drm/drm.h>
+
 #define NOUVEAU_GEM_DOMAIN_CPU       (1 << 0)
 #define NOUVEAU_GEM_DOMAIN_VRAM      (1 << 1)
 #define NOUVEAU_GEM_DOMAIN_GART      (1 << 2)
@@ -41,34 +43,34 @@
 #define NOUVEAU_GEM_TILE_NONCONTIG   0x00000008
 
 struct drm_nouveau_gem_info {
-	uint32_t handle;
-	uint32_t domain;
-	uint64_t size;
-	uint64_t offset;
-	uint64_t map_handle;
-	uint32_t tile_mode;
-	uint32_t tile_flags;
+	__u32 handle;
+	__u32 domain;
+	__u64 size;
+	__u64 offset;
+	__u64 map_handle;
+	__u32 tile_mode;
+	__u32 tile_flags;
 };
 
 struct drm_nouveau_gem_new {
 	struct drm_nouveau_gem_info info;
-	uint32_t channel_hint;
-	uint32_t align;
+	__u32 channel_hint;
+	__u32 align;
 };
 
 #define NOUVEAU_GEM_MAX_BUFFERS 1024
 struct drm_nouveau_gem_pushbuf_bo_presumed {
-	uint32_t valid;
-	uint32_t domain;
-	uint64_t offset;
+	__u32 valid;
+	__u32 domain;
+	__u64 offset;
 };
 
 struct drm_nouveau_gem_pushbuf_bo {
-	uint64_t user_priv;
-	uint32_t handle;
-	uint32_t read_domains;
-	uint32_t write_domains;
-	uint32_t valid_domains;
+	__u64 user_priv;
+	__u32 handle;
+	__u32 read_domains;
+	__u32 write_domains;
+	__u32 valid_domains;
 	struct drm_nouveau_gem_pushbuf_bo_presumed presumed;
 };
 
@@ -77,46 +79,46 @@ struct drm_nouveau_gem_pushbuf_bo {
 #define NOUVEAU_GEM_RELOC_OR   (1 << 2)
 #define NOUVEAU_GEM_MAX_RELOCS 1024
 struct drm_nouveau_gem_pushbuf_reloc {
-	uint32_t reloc_bo_index;
-	uint32_t reloc_bo_offset;
-	uint32_t bo_index;
-	uint32_t flags;
-	uint32_t data;
-	uint32_t vor;
-	uint32_t tor;
+	__u32 reloc_bo_index;
+	__u32 reloc_bo_offset;
+	__u32 bo_index;
+	__u32 flags;
+	__u32 data;
+	__u32 vor;
+	__u32 tor;
 };
 
 #define NOUVEAU_GEM_MAX_PUSH 512
 struct drm_nouveau_gem_pushbuf_push {
-	uint32_t bo_index;
-	uint32_t pad;
-	uint64_t offset;
-	uint64_t length;
+	__u32 bo_index;
+	__u32 pad;
+	__u64 offset;
+	__u64 length;
 };
 
 struct drm_nouveau_gem_pushbuf {
-	uint32_t channel;
-	uint32_t nr_buffers;
-	uint64_t buffers;
-	uint32_t nr_relocs;
-	uint32_t nr_push;
-	uint64_t relocs;
-	uint64_t push;
-	uint32_t suffix0;
-	uint32_t suffix1;
-	uint64_t vram_available;
-	uint64_t gart_available;
+	__u32 channel;
+	__u32 nr_buffers;
+	__u64 buffers;
+	__u32 nr_relocs;
+	__u32 nr_push;
+	__u64 relocs;
+	__u64 push;
+	__u32 suffix0;
+	__u32 suffix1;
+	__u64 vram_available;
+	__u64 gart_available;
 };
 
 #define NOUVEAU_GEM_CPU_PREP_NOWAIT                                  0x00000001
 #define NOUVEAU_GEM_CPU_PREP_WRITE                                   0x00000004
 struct drm_nouveau_gem_cpu_prep {
-	uint32_t handle;
-	uint32_t flags;
+	__u32 handle;
+	__u32 flags;
 };
 
 struct drm_nouveau_gem_cpu_fini {
-	uint32_t handle;
+	__u32 handle;
 };
 
 #define DRM_NOUVEAU_GETPARAM           0x00 /* deprecated */
diff --git a/include/uapi/drm/omap_drm.h b/include/uapi/drm/omap_drm.h
index 1d0b1172664e..38a3bd847e15 100644
--- a/include/uapi/drm/omap_drm.h
+++ b/include/uapi/drm/omap_drm.h
@@ -20,7 +20,7 @@
 #ifndef __OMAP_DRM_H__
 #define __OMAP_DRM_H__
 
-#include <drm/drm.h>
+#include "drm.h"
 
 /* Please note that modifications to all structs defined here are
  * subject to backwards-compatibility constraints.
@@ -101,9 +101,6 @@ struct drm_omap_gem_info {
 
 #define DRM_OMAP_GET_PARAM		0x00
 #define DRM_OMAP_SET_PARAM		0x01
-/* placeholder for plugin-api
-#define DRM_OMAP_GET_BASE		0x02
-*/
 #define DRM_OMAP_GEM_NEW		0x03
 #define DRM_OMAP_GEM_CPU_PREP		0x04
 #define DRM_OMAP_GEM_CPU_FINI		0x05
@@ -112,9 +109,6 @@ struct drm_omap_gem_info {
 
 #define DRM_IOCTL_OMAP_GET_PARAM	DRM_IOWR(DRM_COMMAND_BASE + DRM_OMAP_GET_PARAM, struct drm_omap_param)
 #define DRM_IOCTL_OMAP_SET_PARAM	DRM_IOW (DRM_COMMAND_BASE + DRM_OMAP_SET_PARAM, struct drm_omap_param)
-/* placeholder for plugin-api
-#define DRM_IOCTL_OMAP_GET_BASE		DRM_IOWR(DRM_COMMAND_BASE + DRM_OMAP_GET_BASE, struct drm_omap_get_base)
-*/
 #define DRM_IOCTL_OMAP_GEM_NEW		DRM_IOWR(DRM_COMMAND_BASE + DRM_OMAP_GEM_NEW, struct drm_omap_gem_new)
 #define DRM_IOCTL_OMAP_GEM_CPU_PREP	DRM_IOW (DRM_COMMAND_BASE + DRM_OMAP_GEM_CPU_PREP, struct drm_omap_gem_cpu_prep)
 #define DRM_IOCTL_OMAP_GEM_CPU_FINI	DRM_IOW (DRM_COMMAND_BASE + DRM_OMAP_GEM_CPU_FINI, struct drm_omap_gem_cpu_fini)
diff --git a/include/uapi/drm/qxl_drm.h b/include/uapi/drm/qxl_drm.h
index ebebd36c4117..4d1e32640463 100644
--- a/include/uapi/drm/qxl_drm.h
+++ b/include/uapi/drm/qxl_drm.h
@@ -24,13 +24,12 @@
 #ifndef QXL_DRM_H
 #define QXL_DRM_H
 
-#include <stddef.h>
-#include "drm/drm.h"
+#include "drm.h"
 
 /* Please note that modifications to all structs defined here are
  * subject to backwards-compatibility constraints.
  *
- * Do not use pointers, use uint64_t instead for 32 bit / 64 bit user/kernel
+ * Do not use pointers, use __u64 instead for 32 bit / 64 bit user/kernel
  * compatibility Keep fields aligned to their size
  */
 
@@ -48,14 +47,14 @@
 #define DRM_QXL_ALLOC_SURF  0x06
 
 struct drm_qxl_alloc {
-	uint32_t size;
-	uint32_t handle; /* 0 is an invalid handle */
+	__u32 size;
+	__u32 handle; /* 0 is an invalid handle */
 };
 
 struct drm_qxl_map {
-	uint64_t offset; /* use for mmap system call */
-	uint32_t handle;
-	uint32_t pad;
+	__u64 offset; /* use for mmap system call */
+	__u32 handle;
+	__u32 pad;
 };
 
 /*
@@ -68,59 +67,59 @@ struct drm_qxl_map {
 #define QXL_RELOC_TYPE_SURF 2
 
 struct drm_qxl_reloc {
-	uint64_t src_offset; /* offset into src_handle or src buffer */
-	uint64_t dst_offset; /* offset in dest handle */
-	uint32_t src_handle; /* dest handle to compute address from */
-	uint32_t dst_handle; /* 0 if to command buffer */
-	uint32_t reloc_type;
-	uint32_t pad;
+	__u64 src_offset; /* offset into src_handle or src buffer */
+	__u64 dst_offset; /* offset in dest handle */
+	__u32 src_handle; /* dest handle to compute address from */
+	__u32 dst_handle; /* 0 if to command buffer */
+	__u32 reloc_type;
+	__u32 pad;
 };
 
 struct drm_qxl_command {
-	uint64_t	 __user command; /* void* */
-	uint64_t	 __user relocs; /* struct drm_qxl_reloc* */
-	uint32_t		type;
-	uint32_t		command_size;
-	uint32_t		relocs_num;
-	uint32_t                pad;
+	__u64	 __user command; /* void* */
+	__u64	 __user relocs; /* struct drm_qxl_reloc* */
+	__u32		type;
+	__u32		command_size;
+	__u32		relocs_num;
+	__u32                pad;
 };
 
 /* XXX: call it drm_qxl_commands? */
 struct drm_qxl_execbuffer {
-	uint32_t		flags;		/* for future use */
-	uint32_t		commands_num;
-	uint64_t	 __user commands;	/* struct drm_qxl_command* */
+	__u32		flags;		/* for future use */
+	__u32		commands_num;
+	__u64	 __user commands;	/* struct drm_qxl_command* */
 };
 
 struct drm_qxl_update_area {
-	uint32_t handle;
-	uint32_t top;
-	uint32_t left;
-	uint32_t bottom;
-	uint32_t right;
-	uint32_t pad;
+	__u32 handle;
+	__u32 top;
+	__u32 left;
+	__u32 bottom;
+	__u32 right;
+	__u32 pad;
 };
 
 #define QXL_PARAM_NUM_SURFACES 1 /* rom->n_surfaces */
 #define QXL_PARAM_MAX_RELOCS 2
 struct drm_qxl_getparam {
-	uint64_t param;
-	uint64_t value;
+	__u64 param;
+	__u64 value;
 };
 
 /* these are one bit values */
 struct drm_qxl_clientcap {
-	uint32_t index;
-	uint32_t pad;
+	__u32 index;
+	__u32 pad;
 };
 
 struct drm_qxl_alloc_surf {
-	uint32_t format;
-	uint32_t width;
-	uint32_t height;
-	int32_t stride;
-	uint32_t handle;
-	uint32_t pad;
+	__u32 format;
+	__u32 width;
+	__u32 height;
+	__s32 stride;
+	__u32 handle;
+	__u32 pad;
 };
 
 #define DRM_IOCTL_QXL_ALLOC \
diff --git a/include/uapi/drm/r128_drm.h b/include/uapi/drm/r128_drm.h
index 76b0aa3e8210..7a44c6500a7e 100644
--- a/include/uapi/drm/r128_drm.h
+++ b/include/uapi/drm/r128_drm.h
@@ -33,7 +33,7 @@
 #ifndef __R128_DRM_H__
 #define __R128_DRM_H__
 
-#include <drm/drm.h>
+#include "drm.h"
 
 /* WARNING: If you change any of these defines, make sure to change the
  * defines in the X server file (r128_sarea.h)
diff --git a/include/uapi/drm/radeon_drm.h b/include/uapi/drm/radeon_drm.h
index 01aa2a8e3f8d..ccb9bcd82685 100644
--- a/include/uapi/drm/radeon_drm.h
+++ b/include/uapi/drm/radeon_drm.h
@@ -793,9 +793,9 @@ typedef struct drm_radeon_surface_free {
 #define RADEON_GEM_DOMAIN_VRAM		0x4
 
 struct drm_radeon_gem_info {
-	uint64_t	gart_size;
-	uint64_t	vram_size;
-	uint64_t	vram_visible;
+	__u64	gart_size;
+	__u64	vram_size;
+	__u64	vram_visible;
 };
 
 #define RADEON_GEM_NO_BACKING_STORE	(1 << 0)
@@ -807,11 +807,11 @@ struct drm_radeon_gem_info {
 #define RADEON_GEM_NO_CPU_ACCESS	(1 << 4)
 
 struct drm_radeon_gem_create {
-	uint64_t	size;
-	uint64_t	alignment;
-	uint32_t	handle;
-	uint32_t	initial_domain;
-	uint32_t	flags;
+	__u64	size;
+	__u64	alignment;
+	__u32	handle;
+	__u32	initial_domain;
+	__u32	flags;
 };
 
 /*
@@ -825,10 +825,10 @@ struct drm_radeon_gem_create {
 #define RADEON_GEM_USERPTR_REGISTER	(1 << 3)
 
 struct drm_radeon_gem_userptr {
-	uint64_t		addr;
-	uint64_t		size;
-	uint32_t		flags;
-	uint32_t		handle;
+	__u64		addr;
+	__u64		size;
+	__u32		flags;
+	__u32		handle;
 };
 
 #define RADEON_TILING_MACRO				0x1
@@ -850,72 +850,72 @@ struct drm_radeon_gem_userptr {
 #define RADEON_TILING_EG_STENCIL_TILE_SPLIT_MASK	0xf
 
 struct drm_radeon_gem_set_tiling {
-	uint32_t	handle;
-	uint32_t	tiling_flags;
-	uint32_t	pitch;
+	__u32	handle;
+	__u32	tiling_flags;
+	__u32	pitch;
 };
 
 struct drm_radeon_gem_get_tiling {
-	uint32_t	handle;
-	uint32_t	tiling_flags;
-	uint32_t	pitch;
+	__u32	handle;
+	__u32	tiling_flags;
+	__u32	pitch;
 };
 
 struct drm_radeon_gem_mmap {
-	uint32_t	handle;
-	uint32_t	pad;
-	uint64_t	offset;
-	uint64_t	size;
-	uint64_t	addr_ptr;
+	__u32	handle;
+	__u32	pad;
+	__u64	offset;
+	__u64	size;
+	__u64	addr_ptr;
 };
 
 struct drm_radeon_gem_set_domain {
-	uint32_t	handle;
-	uint32_t	read_domains;
-	uint32_t	write_domain;
+	__u32	handle;
+	__u32	read_domains;
+	__u32	write_domain;
 };
 
 struct drm_radeon_gem_wait_idle {
-	uint32_t	handle;
-	uint32_t	pad;
+	__u32	handle;
+	__u32	pad;
 };
 
 struct drm_radeon_gem_busy {
-	uint32_t	handle;
-	uint32_t        domain;
+	__u32	handle;
+	__u32        domain;
 };
 
 struct drm_radeon_gem_pread {
 	/** Handle for the object being read. */
-	uint32_t handle;
-	uint32_t pad;
+	__u32 handle;
+	__u32 pad;
 	/** Offset into the object to read from */
-	uint64_t offset;
+	__u64 offset;
 	/** Length of data to read */
-	uint64_t size;
+	__u64 size;
 	/** Pointer to write the data into. */
 	/* void *, but pointers are not 32/64 compatible */
-	uint64_t data_ptr;
+	__u64 data_ptr;
 };
 
 struct drm_radeon_gem_pwrite {
 	/** Handle for the object being written to. */
-	uint32_t handle;
-	uint32_t pad;
+	__u32 handle;
+	__u32 pad;
 	/** Offset into the object to write to */
-	uint64_t offset;
+	__u64 offset;
 	/** Length of data to write */
-	uint64_t size;
+	__u64 size;
 	/** Pointer to read the data from. */
 	/* void *, but pointers are not 32/64 compatible */
-	uint64_t data_ptr;
+	__u64 data_ptr;
 };
 
 /* Sets or returns a value associated with a buffer. */
 struct drm_radeon_gem_op {
-	uint32_t	handle; /* buffer */
-	uint32_t	op;     /* RADEON_GEM_OP_* */
-	uint64_t	value;  /* input or return value */
+	__u32	handle; /* buffer */
+	__u32	op;     /* RADEON_GEM_OP_* */
+	__u64	value;  /* input or return value */
 };
 
 #define RADEON_GEM_OP_GET_INITIAL_DOMAIN	0
@@ -935,11 +935,11 @@ struct drm_radeon_gem_op {
 #define RADEON_VM_PAGE_SNOOPED		(1 << 4)
 
 struct drm_radeon_gem_va {
-	uint32_t		handle;
-	uint32_t		operation;
-	uint32_t		vm_id;
-	uint32_t		flags;
-	uint64_t		offset;
+	__u32		handle;
+	__u32		operation;
+	__u32		vm_id;
+	__u32		flags;
+	__u64		offset;
 };
 
 #define RADEON_CHUNK_ID_RELOCS	0x01
@@ -961,29 +961,29 @@ struct drm_radeon_gem_va {
 /* 0 = normal, + = higher priority, - = lower priority */
 
 struct drm_radeon_cs_chunk {
-	uint32_t		chunk_id;
-	uint32_t		length_dw;
-	uint64_t		chunk_data;
+	__u32		chunk_id;
+	__u32		length_dw;
+	__u64		chunk_data;
 };
 
 /* drm_radeon_cs_reloc.flags */
 #define RADEON_RELOC_PRIO_MASK		(0xf << 0)
 
 struct drm_radeon_cs_reloc {
-	uint32_t		handle;
-	uint32_t		read_domains;
-	uint32_t		write_domain;
-	uint32_t		flags;
+	__u32		handle;
+	__u32		read_domains;
+	__u32		write_domain;
+	__u32		flags;
 };
 
 struct drm_radeon_cs {
-	uint32_t		num_chunks;
-	uint32_t		cs_id;
-	/* this points to uint64_t * which point to cs chunks */
-	uint64_t		chunks;
+	__u32		num_chunks;
+	__u32		cs_id;
+	/* this points to __u64 * which point to cs chunks */
+	__u64		chunks;
 	/* updates to the limits after this CS ioctl */
-	uint64_t		gart_limit;
-	uint64_t		vram_limit;
+	__u64		gart_limit;
+	__u64		vram_limit;
 };
 
 #define RADEON_INFO_DEVICE_ID		0x00
@@ -1042,9 +1042,9 @@ struct drm_radeon_cs {
 #define RADEON_INFO_GPU_RESET_COUNTER	0x26
 
 struct drm_radeon_info {
-	uint32_t		request;
-	uint32_t		pad;
-	uint64_t		value;
+	__u32		request;
+	__u32		pad;
+	__u64		value;
 };
 
 /* Those correspond to the tile index to use, this is to explicitly state
diff --git a/include/uapi/drm/savage_drm.h b/include/uapi/drm/savage_drm.h
index 9dc9dc1a7753..574147489c60 100644
--- a/include/uapi/drm/savage_drm.h
+++ b/include/uapi/drm/savage_drm.h
@@ -26,7 +26,7 @@
 #ifndef __SAVAGE_DRM_H__
 #define __SAVAGE_DRM_H__
 
-#include <drm/drm.h>
+#include "drm.h"
 
 #ifndef __SAVAGE_SAREA_DEFINES__
 #define __SAVAGE_SAREA_DEFINES__
diff --git a/include/uapi/drm/tegra_drm.h b/include/uapi/drm/tegra_drm.h
index 5391780c2b05..27d0b054aed0 100644
--- a/include/uapi/drm/tegra_drm.h
+++ b/include/uapi/drm/tegra_drm.h
@@ -23,7 +23,7 @@
 #ifndef _UAPI_TEGRA_DRM_H_
 #define _UAPI_TEGRA_DRM_H_
 
-#include <drm/drm.h>
+#include "drm.h"
 
 #define DRM_TEGRA_GEM_CREATE_TILED     (1 << 0)
 #define DRM_TEGRA_GEM_CREATE_BOTTOM_UP (1 << 1)
diff --git a/include/uapi/drm/vc4_drm.h b/include/uapi/drm/vc4_drm.h
new file mode 100644
index 000000000000..eeb37e394f13
--- /dev/null
+++ b/include/uapi/drm/vc4_drm.h
@@ -0,0 +1,279 @@
+/*
+ * Copyright © 2014-2015 Broadcom
+ *
+ * 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 (including the next
+ * paragraph) 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 AUTHORS OR COPYRIGHT HOLDERS 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 _UAPI_VC4_DRM_H_
+#define _UAPI_VC4_DRM_H_
+
+#include "drm.h"
+
+#define DRM_VC4_SUBMIT_CL                         0x00
+#define DRM_VC4_WAIT_SEQNO                        0x01
+#define DRM_VC4_WAIT_BO                           0x02
+#define DRM_VC4_CREATE_BO                         0x03
+#define DRM_VC4_MMAP_BO                           0x04
+#define DRM_VC4_CREATE_SHADER_BO                  0x05
+#define DRM_VC4_GET_HANG_STATE                    0x06
+
+#define DRM_IOCTL_VC4_SUBMIT_CL           DRM_IOWR(DRM_COMMAND_BASE + DRM_VC4_SUBMIT_CL, struct drm_vc4_submit_cl)
+#define DRM_IOCTL_VC4_WAIT_SEQNO          DRM_IOWR(DRM_COMMAND_BASE + DRM_VC4_WAIT_SEQNO, struct drm_vc4_wait_seqno)
+#define DRM_IOCTL_VC4_WAIT_BO             DRM_IOWR(DRM_COMMAND_BASE + DRM_VC4_WAIT_BO, struct drm_vc4_wait_bo)
+#define DRM_IOCTL_VC4_CREATE_BO           DRM_IOWR(DRM_COMMAND_BASE + DRM_VC4_CREATE_BO, struct drm_vc4_create_bo)
+#define DRM_IOCTL_VC4_MMAP_BO             DRM_IOWR(DRM_COMMAND_BASE + DRM_VC4_MMAP_BO, struct drm_vc4_mmap_bo)
+#define DRM_IOCTL_VC4_CREATE_SHADER_BO    DRM_IOWR(DRM_COMMAND_BASE + DRM_VC4_CREATE_SHADER_BO, struct drm_vc4_create_shader_bo)
+#define DRM_IOCTL_VC4_GET_HANG_STATE      DRM_IOWR(DRM_COMMAND_BASE + DRM_VC4_GET_HANG_STATE, struct drm_vc4_get_hang_state)
+
+struct drm_vc4_submit_rcl_surface {
+	__u32 hindex; /* Handle index, or ~0 if not present. */
+	__u32 offset; /* Offset to start of buffer. */
+	/*
+	 * Bits for either render config (color_write) or load/store packet.
+	 * Bits should all be 0 for MSAA load/stores.
+	 */
+	__u16 bits;
+
+#define VC4_SUBMIT_RCL_SURFACE_READ_IS_FULL_RES		(1 << 0)
+	__u16 flags;
+};
+
+/**
+ * struct drm_vc4_submit_cl - ioctl argument for submitting commands to the 3D
+ * engine.
+ *
+ * Drivers typically use GPU BOs to store batchbuffers / command lists and
+ * their associated state.  However, because the VC4 lacks an MMU, we have to
+ * do validation of memory accesses by the GPU commands.  If we were to store
+ * our commands in BOs, we'd need to do uncached readback from them to do the
+ * validation process, which is too expensive.  Instead, userspace accumulates
+ * commands and associated state in plain memory, then the kernel copies the
+ * data to its own address space, and then validates and stores it in a GPU
+ * BO.
+ */
+struct drm_vc4_submit_cl {
+	/* Pointer to the binner command list.
+	 *
+	 * This is the first set of commands executed, which runs the
+	 * coordinate shader to determine where primitives land on the screen,
+	 * then writes out the state updates and draw calls necessary per tile
+	 * to the tile allocation BO.
+	 */
+	__u64 bin_cl;
+
+	/* Pointer to the shader records.
+	 *
+	 * Shader records are the structures read by the hardware that contain
+	 * pointers to uniforms, shaders, and vertex attributes.  The
+	 * reference to the shader record has enough information to determine
+	 * how many pointers are necessary (fixed number for shaders/uniforms,
+	 * and an attribute count), so those BO indices into bo_handles are
+	 * just stored as __u32s before each shader record passed in.
+	 */
+	__u64 shader_rec;
+
+	/* Pointer to uniform data and texture handles for the textures
+	 * referenced by the shader.
+	 *
+	 * For each shader state record, there is a set of uniform data in the
+	 * order referenced by the record (FS, VS, then CS).  Each set of
+	 * uniform data has a __u32 index into bo_handles per texture
+	 * sample operation, in the order the QPU_W_TMUn_S writes appear in
+	 * the program.  Following the texture BO handle indices is the actual
+	 * uniform data.
+	 *
+	 * The individual uniform state blocks don't have sizes passed in,
+	 * because the kernel has to determine the sizes anyway during shader
+	 * code validation.
+	 */
+	__u64 uniforms;
+	__u64 bo_handles;
+
+	/* Size in bytes of the binner command list. */
+	__u32 bin_cl_size;
+	/* Size in bytes of the set of shader records. */
+	__u32 shader_rec_size;
+	/* Number of shader records.
+	 *
+	 * This could just be computed from the contents of shader_records and
+	 * the address bits of references to them from the bin CL, but it
+	 * keeps the kernel from having to resize some allocations it makes.
+	 */
+	__u32 shader_rec_count;
+	/* Size in bytes of the uniform state. */
+	__u32 uniforms_size;
+
+	/* Number of BO handles passed in (size is that times 4). */
+	__u32 bo_handle_count;
+
+	/* RCL setup: */
+	__u16 width;
+	__u16 height;
+	__u8 min_x_tile;
+	__u8 min_y_tile;
+	__u8 max_x_tile;
+	__u8 max_y_tile;
+	struct drm_vc4_submit_rcl_surface color_read;
+	struct drm_vc4_submit_rcl_surface color_write;
+	struct drm_vc4_submit_rcl_surface zs_read;
+	struct drm_vc4_submit_rcl_surface zs_write;
+	struct drm_vc4_submit_rcl_surface msaa_color_write;
+	struct drm_vc4_submit_rcl_surface msaa_zs_write;
+	__u32 clear_color[2];
+	__u32 clear_z;
+	__u8 clear_s;
+
+	__u32 pad:24;
+
+#define VC4_SUBMIT_CL_USE_CLEAR_COLOR			(1 << 0)
+	__u32 flags;
+
+	/* Returned value of the seqno of this render job (for the
+	 * wait ioctl).
+	 */
+	__u64 seqno;
+};
+
+/**
+ * struct drm_vc4_wait_seqno - ioctl argument for waiting for
+ * DRM_VC4_SUBMIT_CL completion using its returned seqno.
+ *
+ * timeout_ns is the timeout in nanoseconds, where "0" means "don't
+ * block, just return the status."
+ */
+struct drm_vc4_wait_seqno {
+	__u64 seqno;
+	__u64 timeout_ns;
+};
+
+/**
+ * struct drm_vc4_wait_bo - ioctl argument for waiting for
+ * completion of the last DRM_VC4_SUBMIT_CL on a BO.
+ *
+ * This is useful for cases where multiple processes might be
+ * rendering to a BO and you want to wait for all rendering to be
+ * completed.
+ */
+struct drm_vc4_wait_bo {
+	__u32 handle;
+	__u32 pad;
+	__u64 timeout_ns;
+};
+
+/**
+ * struct drm_vc4_create_bo - ioctl argument for creating VC4 BOs.
+ *
+ * There are currently no values for the flags argument, but it may be
+ * used in a future extension.
+ */
+struct drm_vc4_create_bo {
+	__u32 size;
+	__u32 flags;
+	/** Returned GEM handle for the BO. */
+	__u32 handle;
+	__u32 pad;
+};
+
+/**
+ * struct drm_vc4_mmap_bo - ioctl argument for mapping VC4 BOs.
+ *
+ * This doesn't actually perform an mmap.  Instead, it returns the
+ * offset you need to use in an mmap on the DRM device node.  This
+ * means that tools like valgrind end up knowing about the mapped
+ * memory.
+ *
+ * There are currently no values for the flags argument, but it may be
+ * used in a future extension.
+ */
+struct drm_vc4_mmap_bo {
+	/** Handle for the object being mapped. */
+	__u32 handle;
+	__u32 flags;
+	/** offset into the drm node to use for subsequent mmap call. */
+	__u64 offset;
+};
+
+/**
+ * struct drm_vc4_create_shader_bo - ioctl argument for creating VC4
+ * shader BOs.
+ *
+ * Since allowing a shader to be overwritten while it's also being
+ * executed from would allow privlege escalation, shaders must be
+ * created using this ioctl, and they can't be mmapped later.
+ */
+struct drm_vc4_create_shader_bo {
+	/* Size of the data argument. */
+	__u32 size;
+	/* Flags, currently must be 0. */
+	__u32 flags;
+
+	/* Pointer to the data. */
+	__u64 data;
+
+	/** Returned GEM handle for the BO. */
+	__u32 handle;
+	/* Pad, must be 0. */
+	__u32 pad;
+};
+
+struct drm_vc4_get_hang_state_bo {
+	__u32 handle;
+	__u32 paddr;
+	__u32 size;
+	__u32 pad;
+};
+
+/**
+ * struct drm_vc4_hang_state - ioctl argument for collecting state
+ * from a GPU hang for analysis.
+*/
+struct drm_vc4_get_hang_state {
+	/** Pointer to array of struct drm_vc4_get_hang_state_bo. */
+	__u64 bo;
+	/**
+	 * On input, the size of the bo array.  Output is the number
+	 * of bos to be returned.
+	 */
+	__u32 bo_count;
+
+	__u32 start_bin, start_render;
+
+	__u32 ct0ca, ct0ea;
+	__u32 ct1ca, ct1ea;
+	__u32 ct0cs, ct1cs;
+	__u32 ct0ra0, ct1ra0;
+
+	__u32 bpca, bpcs;
+	__u32 bpoa, bpos;
+
+	__u32 vpmbase;
+
+	__u32 dbge;
+	__u32 fdbgo;
+	__u32 fdbgb;
+	__u32 fdbgr;
+	__u32 fdbgs;
+	__u32 errstat;
+
+	/* Pad that we may save more registers into in the future. */
+	__u32 pad[16];
+};
+
+#endif /* _UAPI_VC4_DRM_H_ */
diff --git a/include/uapi/drm/via_drm.h b/include/uapi/drm/via_drm.h
index 45bc80c3714b..fa21ed185520 100644
--- a/include/uapi/drm/via_drm.h
+++ b/include/uapi/drm/via_drm.h
@@ -24,7 +24,7 @@
 #ifndef _VIA_DRM_H_
 #define _VIA_DRM_H_
 
-#include <drm/drm.h>
+#include "drm.h"
 
 /* WARNING: These defines must be the same as what the Xserver uses.
  * if you change them, you must change the defines in the Xserver.
@@ -33,9 +33,6 @@
 #ifndef _VIA_DEFINES_
 #define _VIA_DEFINES_
 
-#ifndef __KERNEL__
-#include "via_drmclient.h"
-#endif
 
 #define VIA_NR_SAREA_CLIPRECTS		8
 #define VIA_NR_XVMC_PORTS               10
diff --git a/include/uapi/drm/virtgpu_drm.h b/include/uapi/drm/virtgpu_drm.h
index fc9e2d6e5e2f..c74f1f90cb37 100644
--- a/include/uapi/drm/virtgpu_drm.h
+++ b/include/uapi/drm/virtgpu_drm.h
@@ -24,13 +24,12 @@
 #ifndef VIRTGPU_DRM_H
 #define VIRTGPU_DRM_H
 
-#include <stddef.h>
-#include "drm/drm.h"
+#include "drm.h"
 
 /* Please note that modifications to all structs defined here are
  * subject to backwards-compatibility constraints.
  *
- * Do not use pointers, use uint64_t instead for 32 bit / 64 bit user/kernel
+ * Do not use pointers, use __u64 instead for 32 bit / 64 bit user/kernel
  * compatibility Keep fields aligned to their size
  */
 
@@ -45,88 +44,88 @@
 #define DRM_VIRTGPU_GET_CAPS  0x09
 
 struct drm_virtgpu_map {
-	uint64_t offset; /* use for mmap system call */
-	uint32_t handle;
-	uint32_t pad;
+	__u64 offset; /* use for mmap system call */
+	__u32 handle;
+	__u32 pad;
 };
 
 struct drm_virtgpu_execbuffer {
-	uint32_t		flags;		/* for future use */
-	uint32_t size;
-	uint64_t command; /* void* */
-	uint64_t bo_handles;
-	uint32_t num_bo_handles;
-	uint32_t pad;
+	__u32		flags;		/* for future use */
+	__u32 size;
+	__u64 command; /* void* */
+	__u64 bo_handles;
+	__u32 num_bo_handles;
+	__u32 pad;
 };
 
 #define VIRTGPU_PARAM_3D_FEATURES 1 /* do we have 3D features in the hw */
 
 struct drm_virtgpu_getparam {
-	uint64_t param;
-	uint64_t value;
+	__u64 param;
+	__u64 value;
 };
 
 /* NO_BO flags? NO resource flag? */
 /* resource flag for y_0_top */
 struct drm_virtgpu_resource_create {
-	uint32_t target;
-	uint32_t format;
-	uint32_t bind;
-	uint32_t width;
-	uint32_t height;
-	uint32_t depth;
-	uint32_t array_size;
-	uint32_t last_level;
-	uint32_t nr_samples;
-	uint32_t flags;
-	uint32_t bo_handle; /* if this is set - recreate a new resource attached to this bo ? */
-	uint32_t res_handle;  /* returned by kernel */
-	uint32_t size;        /* validate transfer in the host */
-	uint32_t stride;      /* validate transfer in the host */
+	__u32 target;
+	__u32 format;
+	__u32 bind;
+	__u32 width;
+	__u32 height;
+	__u32 depth;
+	__u32 array_size;
+	__u32 last_level;
+	__u32 nr_samples;
+	__u32 flags;
+	__u32 bo_handle; /* if this is set - recreate a new resource attached to this bo ? */
+	__u32 res_handle;  /* returned by kernel */
+	__u32 size;        /* validate transfer in the host */
+	__u32 stride;      /* validate transfer in the host */
 };
 
 struct drm_virtgpu_resource_info {
-	uint32_t bo_handle;
-	uint32_t res_handle;
-	uint32_t size;
-	uint32_t stride;
+	__u32 bo_handle;
+	__u32 res_handle;
+	__u32 size;
+	__u32 stride;
 };
 
 struct drm_virtgpu_3d_box {
-	uint32_t x;
-	uint32_t y;
-	uint32_t z;
-	uint32_t w;
-	uint32_t h;
-	uint32_t d;
+	__u32 x;
+	__u32 y;
+	__u32 z;
+	__u32 w;
+	__u32 h;
+	__u32 d;
 };
 
 struct drm_virtgpu_3d_transfer_to_host {
-	uint32_t bo_handle;
+	__u32 bo_handle;
 	struct drm_virtgpu_3d_box box;
-	uint32_t level;
-	uint32_t offset;
+	__u32 level;
+	__u32 offset;
 };
 
 struct drm_virtgpu_3d_transfer_from_host {
-	uint32_t bo_handle;
+	__u32 bo_handle;
 	struct drm_virtgpu_3d_box box;
-	uint32_t level;
-	uint32_t offset;
+	__u32 level;
+	__u32 offset;
 };
 
 #define VIRTGPU_WAIT_NOWAIT 1 /* like it */
 struct drm_virtgpu_3d_wait {
-	uint32_t handle; /* 0 is an invalid handle */
-	uint32_t flags;
+	__u32 handle; /* 0 is an invalid handle */
+	__u32 flags;
 };
 
 struct drm_virtgpu_get_caps {
-	uint32_t cap_set_id;
-	uint32_t cap_set_ver;
-	uint64_t addr;
-	uint32_t size;
-	uint32_t pad;
+	__u32 cap_set_id;
+	__u32 cap_set_ver;
+	__u64 addr;
+	__u32 size;
+	__u32 pad;
 };
 
 #define DRM_IOCTL_VIRTGPU_MAP \
diff --git a/include/uapi/drm/vmwgfx_drm.h b/include/uapi/drm/vmwgfx_drm.h
index 05b204954d16..5b68b4d10884 100644
--- a/include/uapi/drm/vmwgfx_drm.h
+++ b/include/uapi/drm/vmwgfx_drm.h
@@ -28,9 +28,7 @@
 #ifndef __VMWGFX_DRM_H__
 #define __VMWGFX_DRM_H__
 
-#ifndef __KERNEL__
-#include <drm/drm.h>
-#endif
+#include "drm.h"
 
 #define DRM_VMW_MAX_SURFACE_FACES 6
 #define DRM_VMW_MAX_MIP_LEVELS 24
@@ -111,9 +109,9 @@ enum drm_vmw_handle_type {
  */
 
 struct drm_vmw_getparam_arg {
-	uint64_t value;
-	uint32_t param;
-	uint32_t pad64;
+	__u64 value;
+	__u32 param;
+	__u32 pad64;
 };
 
 /*************************************************************************/
@@ -134,8 +132,8 @@ struct drm_vmw_getparam_arg {
  */
 
 struct drm_vmw_context_arg {
-	int32_t cid;
-	uint32_t pad64;
+	__s32 cid;
+	__u32 pad64;
 };
 
 /*************************************************************************/
@@ -165,7 +163,7 @@ struct drm_vmw_context_arg {
  * @mip_levels: Number of mip levels for each face.
  * An unused face should have 0 encoded.
  * @size_addr: Address of a user-space array of sruct drm_vmw_size
- * cast to an uint64_t for 32-64 bit compatibility.
+ * cast to an __u64 for 32-64 bit compatibility.
  * The size of the array should equal the total number of mipmap levels.
  * @shareable: Boolean whether other clients (as identified by file descriptors)
  * may reference this surface.
@@ -177,12 +175,12 @@ struct drm_vmw_context_arg {
  */
 
 struct drm_vmw_surface_create_req {
-	uint32_t flags;
-	uint32_t format;
-	uint32_t mip_levels[DRM_VMW_MAX_SURFACE_FACES];
-	uint64_t size_addr;
-	int32_t shareable;
-	int32_t scanout;
+	__u32 flags;
+	__u32 format;
+	__u32 mip_levels[DRM_VMW_MAX_SURFACE_FACES];
+	__u64 size_addr;
+	__s32 shareable;
+	__s32 scanout;
 };
 
 /**
@@ -197,7 +195,7 @@ struct drm_vmw_surface_create_req {
  */
 
 struct drm_vmw_surface_arg {
-	int32_t sid;
+	__s32 sid;
 	enum drm_vmw_handle_type handle_type;
 };
 
@@ -213,10 +211,10 @@ struct drm_vmw_surface_arg {
  */
 
 struct drm_vmw_size {
-	uint32_t width;
-	uint32_t height;
-	uint32_t depth;
-	uint32_t pad64;
+	__u32 width;
+	__u32 height;
+	__u32 depth;
+	__u32 pad64;
 };
 
 /**
@@ -284,13 +282,13 @@ union drm_vmw_surface_reference_arg {
 /**
  * struct drm_vmw_execbuf_arg
  *
- * @commands: User-space address of a command buffer cast to an uint64_t.
+ * @commands: User-space address of a command buffer cast to an __u64.
  * @command-size: Size in bytes of the command buffer.
  * @throttle-us: Sleep until software is less than @throttle_us
  * microseconds ahead of hardware. The driver may round this value
  * to the nearest kernel tick.
  * @fence_rep: User-space address of a struct drm_vmw_fence_rep cast to an
- * uint64_t.
+ * __u64.
  * @version: Allows expanding the execbuf ioctl parameters without breaking
  * backwards compatibility, since user-space will always tell the kernel
  * which version it uses.
@@ -302,14 +300,14 @@ union drm_vmw_surface_reference_arg {
 #define DRM_VMW_EXECBUF_VERSION 2
 
 struct drm_vmw_execbuf_arg {
-	uint64_t commands;
-	uint32_t command_size;
-	uint32_t throttle_us;
-	uint64_t fence_rep;
-	uint32_t version;
-	uint32_t flags;
-	uint32_t context_handle;
-	uint32_t pad64;
+	__u64 commands;
+	__u32 command_size;
+	__u32 throttle_us;
+	__u64 fence_rep;
+	__u32 version;
+	__u32 flags;
+	__u32 context_handle;
+	__u32 pad64;
 };
 
 /**
@@ -338,12 +336,12 @@ struct drm_vmw_execbuf_arg {
  */
 
 struct drm_vmw_fence_rep {
-	uint32_t handle;
-	uint32_t mask;
-	uint32_t seqno;
-	uint32_t passed_seqno;
-	uint32_t pad64;
-	int32_t error;
+	__u32 handle;
+	__u32 mask;
+	__u32 seqno;
+	__u32 passed_seqno;
+	__u32 pad64;
+	__s32 error;
 };
 
 /*************************************************************************/
@@ -373,8 +371,8 @@ struct drm_vmw_fence_rep {
  */
 
 struct drm_vmw_alloc_dmabuf_req {
-	uint32_t size;
-	uint32_t pad64;
+	__u32 size;
+	__u32 pad64;
 };
 
 /**
@@ -391,11 +389,11 @@ struct drm_vmw_alloc_dmabuf_req {
  */
 
 struct drm_vmw_dmabuf_rep {
-	uint64_t map_handle;
-	uint32_t handle;
-	uint32_t cur_gmr_id;
-	uint32_t cur_gmr_offset;
-	uint32_t pad64;
+	__u64 map_handle;
+	__u32 handle;
+	__u32 cur_gmr_id;
+	__u32 cur_gmr_offset;
+	__u32 pad64;
 };
 
 /**
@@ -428,8 +426,8 @@ union drm_vmw_alloc_dmabuf_arg {
  */
 
 struct drm_vmw_unref_dmabuf_arg {
-	uint32_t handle;
-	uint32_t pad64;
+	__u32 handle;
+	__u32 pad64;
 };
 
 /*************************************************************************/
@@ -452,10 +450,10 @@ struct drm_vmw_unref_dmabuf_arg {
  */
 
 struct drm_vmw_rect {
-	int32_t x;
-	int32_t y;
-	uint32_t w;
-	uint32_t h;
+	__s32 x;
+	__s32 y;
+	__u32 w;
+	__u32 h;
 };
 
 /**
@@ -477,21 +475,21 @@ struct drm_vmw_rect {
  */
 
 struct drm_vmw_control_stream_arg {
-	uint32_t stream_id;
-	uint32_t enabled;
+	__u32 stream_id;
+	__u32 enabled;
 
-	uint32_t flags;
-	uint32_t color_key;
+	__u32 flags;
+	__u32 color_key;
 
-	uint32_t handle;
-	uint32_t offset;
-	int32_t format;
-	uint32_t size;
-	uint32_t width;
-	uint32_t height;
-	uint32_t pitch[3];
+	__u32 handle;
+	__u32 offset;
+	__s32 format;
+	__u32 size;
+	__u32 width;
+	__u32 height;
+	__u32 pitch[3];
 
-	uint32_t pad64;
+	__u32 pad64;
 	struct drm_vmw_rect src;
 	struct drm_vmw_rect dst;
 };
@@ -519,12 +517,12 @@ struct drm_vmw_control_stream_arg {
  */
 
 struct drm_vmw_cursor_bypass_arg {
-	uint32_t flags;
-	uint32_t crtc_id;
-	int32_t xpos;
-	int32_t ypos;
-	int32_t xhot;
-	int32_t yhot;
+	__u32 flags;
+	__u32 crtc_id;
+	__s32 xpos;
+	__s32 ypos;
+	__s32 xhot;
+	__s32 yhot;
 };
 
 /*************************************************************************/
@@ -542,8 +540,8 @@ struct drm_vmw_cursor_bypass_arg {
  */
 
 struct drm_vmw_stream_arg {
-	uint32_t stream_id;
-	uint32_t pad64;
+	__u32 stream_id;
+	__u32 pad64;
 };
 
 /*************************************************************************/
@@ -565,7 +563,7 @@ struct drm_vmw_stream_arg {
 /**
  * struct drm_vmw_get_3d_cap_arg
  *
- * @buffer: Pointer to a buffer for capability data, cast to an uint64_t
+ * @buffer: Pointer to a buffer for capability data, cast to an __u64
  * @size: Max size to copy
  *
  * Input argument to the DRM_VMW_GET_3D_CAP_IOCTL
@@ -573,9 +571,9 @@ struct drm_vmw_stream_arg {
  */
 
 struct drm_vmw_get_3d_cap_arg {
-	uint64_t buffer;
-	uint32_t max_size;
-	uint32_t pad64;
+	__u64 buffer;
+	__u32 max_size;
+	__u32 pad64;
 };
 
 /*************************************************************************/
@@ -624,14 +622,14 @@ struct drm_vmw_get_3d_cap_arg {
  */
 
 struct drm_vmw_fence_wait_arg {
-	uint32_t handle;
-	int32_t  cookie_valid;
-	uint64_t kernel_cookie;
-	uint64_t timeout_us;
-	int32_t lazy;
-	int32_t flags;
-	int32_t wait_options;
-	int32_t pad64;
+	__u32 handle;
+	__s32  cookie_valid;
+	__u64 kernel_cookie;
+	__u64 timeout_us;
+	__s32 lazy;
+	__s32 flags;
+	__s32 wait_options;
+	__s32 pad64;
 };
 
 /*************************************************************************/
@@ -655,12 +653,12 @@ struct drm_vmw_fence_wait_arg {
  */
 
 struct drm_vmw_fence_signaled_arg {
-	 uint32_t handle;
-	 uint32_t flags;
-	 int32_t signaled;
-	 uint32_t passed_seqno;
-	 uint32_t signaled_flags;
-	 uint32_t pad64;
+	 __u32 handle;
+	 __u32 flags;
+	 __s32 signaled;
+	 __u32 passed_seqno;
+	 __u32 signaled_flags;
+	 __u32 pad64;
 };
 
 /*************************************************************************/
@@ -681,8 +679,8 @@ struct drm_vmw_fence_signaled_arg {
  */
 
 struct drm_vmw_fence_arg {
-	 uint32_t handle;
-	 uint32_t pad64;
+	 __u32 handle;
+	 __u32 pad64;
 };
 
 
@@ -703,9 +701,9 @@ struct drm_vmw_fence_arg {
 
 struct drm_vmw_event_fence {
 	struct drm_event base;
-	uint64_t user_data;
-	uint32_t tv_sec;
-	uint32_t tv_usec;
+	__u64 user_data;
+	__u32 tv_sec;
+	__u32 tv_usec;
 };
 
 /*
@@ -717,17 +715,17 @@ struct drm_vmw_event_fence {
 /**
  * struct drm_vmw_fence_event_arg
  *
- * @fence_rep: Pointer to fence_rep structure cast to uint64_t or 0 if
+ * @fence_rep: Pointer to fence_rep structure cast to __u64 or 0 if
  * the fence is not supposed to be referenced by user-space.
  * @user_info: Info to be delivered with the event.
  * @handle: Attach the event to this fence only.
  * @flags: A set of flags as defined above.
  */
 struct drm_vmw_fence_event_arg {
-	uint64_t fence_rep;
-	uint64_t user_data;
-	uint32_t handle;
-	uint32_t flags;
+	__u64 fence_rep;
+	__u64 user_data;
+	__u32 handle;
+	__u32 flags;
 };
 
 
@@ -747,7 +745,7 @@ struct drm_vmw_fence_event_arg {
  * @sid: Surface id to present from.
  * @dest_x: X placement coordinate for surface.
  * @dest_y: Y placement coordinate for surface.
- * @clips_ptr: Pointer to an array of clip rects cast to an uint64_t.
+ * @clips_ptr: Pointer to an array of clip rects cast to an __u64.
  * @num_clips: Number of cliprects given relative to the framebuffer origin,
  * in the same coordinate space as the frame buffer.
  * @pad64: Unused 64-bit padding.
@@ -756,13 +754,13 @@ struct drm_vmw_fence_event_arg {
  */
 
 struct drm_vmw_present_arg {
-	uint32_t fb_id;
-	uint32_t sid;
-	int32_t dest_x;
-	int32_t dest_y;
-	uint64_t clips_ptr;
-	uint32_t num_clips;
-	uint32_t pad64;
+	__u32 fb_id;
+	__u32 sid;
+	__s32 dest_x;
+	__s32 dest_y;
+	__u64 clips_ptr;
+	__u32 num_clips;
+	__u32 pad64;
 };
 
 
@@ -780,16 +778,16 @@ struct drm_vmw_present_arg {
  * struct drm_vmw_present_arg
  * @fb_id: fb_id to present / read back from.
  * @num_clips: Number of cliprects.
- * @clips_ptr: Pointer to an array of clip rects cast to an uint64_t.
- * @fence_rep: Pointer to a struct drm_vmw_fence_rep, cast to an uint64_t.
+ * @clips_ptr: Pointer to an array of clip rects cast to an __u64.
+ * @fence_rep: Pointer to a struct drm_vmw_fence_rep, cast to an __u64.
  * If this member is NULL, then the ioctl should not return a fence.
  */
 
 struct drm_vmw_present_readback_arg {
-	 uint32_t fb_id;
-	 uint32_t num_clips;
-	 uint64_t clips_ptr;
-	 uint64_t fence_rep;
+	 __u32 fb_id;
+	 __u32 num_clips;
+	 __u64 clips_ptr;
+	 __u64 fence_rep;
 };
 
 /*************************************************************************/
@@ -805,14 +803,14 @@ struct drm_vmw_present_readback_arg {
  * struct drm_vmw_update_layout_arg
  *
  * @num_outputs: number of active connectors
- * @rects: pointer to array of drm_vmw_rect cast to an uint64_t
+ * @rects: pointer to array of drm_vmw_rect cast to an __u64
  *
  * Input argument to the DRM_VMW_UPDATE_LAYOUT Ioctl.
  */
 struct drm_vmw_update_layout_arg {
-	uint32_t num_outputs;
-	uint32_t pad64;
-	uint64_t rects;
+	__u32 num_outputs;
+	__u32 pad64;
+	__u64 rects;
 };
 
 
@@ -849,10 +847,10 @@ enum drm_vmw_shader_type {
  */
 struct drm_vmw_shader_create_arg {
 	enum drm_vmw_shader_type shader_type;
-	uint32_t size;
-	uint32_t buffer_handle;
-	uint32_t shader_handle;
-	uint64_t offset;
+	__u32 size;
+	__u32 buffer_handle;
+	__u32 shader_handle;
+	__u64 offset;
 };
 
 /*************************************************************************/
@@ -871,8 +869,8 @@ struct drm_vmw_shader_create_arg {
  * Input argument to the DRM_VMW_UNREF_SHADER ioctl.
  */
 struct drm_vmw_shader_arg {
-	uint32_t handle;
-	uint32_t pad64;
+	__u32 handle;
+	__u32 pad64;
 };
 
 /*************************************************************************/
@@ -918,14 +916,14 @@ enum drm_vmw_surface_flags {
  * Part of output argument for the DRM_VMW_GB_SURFACE_REF Ioctl.
  */
 struct drm_vmw_gb_surface_create_req {
-	uint32_t svga3d_flags;
-	uint32_t format;
-	uint32_t mip_levels;
+	__u32 svga3d_flags;
+	__u32 format;
+	__u32 mip_levels;
 	enum drm_vmw_surface_flags drm_surface_flags;
-	uint32_t multisample_count;
-	uint32_t autogen_filter;
-	uint32_t buffer_handle;
-	uint32_t array_size;
+	__u32 multisample_count;
+	__u32 autogen_filter;
+	__u32 buffer_handle;
+	__u32 array_size;
 	struct drm_vmw_size base_size;
 };
 
@@ -944,11 +942,11 @@ struct drm_vmw_gb_surface_create_req {
  * Output argument for the DRM_VMW_GB_SURFACE_CREATE ioctl.
  */
 struct drm_vmw_gb_surface_create_rep {
-	uint32_t handle;
-	uint32_t backup_size;
-	uint32_t buffer_handle;
-	uint32_t buffer_size;
-	uint64_t buffer_map_handle;
+	__u32 handle;
+	__u32 backup_size;
+	__u32 buffer_handle;
+	__u32 buffer_size;
+	__u64 buffer_map_handle;
 };
 
 /**
@@ -1061,8 +1059,8 @@ enum drm_vmw_synccpu_op {
 struct drm_vmw_synccpu_arg {
 	enum drm_vmw_synccpu_op op;
 	enum drm_vmw_synccpu_flags flags;
-	uint32_t handle;
-	uint32_t pad64;
+	__u32 handle;
+	__u32 pad64;
 };
 
 /*************************************************************************/