Merge drm/drm-next into drm-misc-next

Let's start the 5.19 development cycle.

Signed-off-by: Maxime Ripard <maxime@cerno.tech>

3 files changed, 79 insertions, 19 deletions

diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
index 0b94ec7b73e7..1d65c1fbc4ec 100644
--- a/include/uapi/drm/amdgpu_drm.h
+++ b/include/uapi/drm/amdgpu_drm.h
@@ -206,6 +206,8 @@ union drm_amdgpu_bo_list {
 #define AMDGPU_CTX_OP_FREE_CTX	2
 #define AMDGPU_CTX_OP_QUERY_STATE	3
 #define AMDGPU_CTX_OP_QUERY_STATE2	4
+#define AMDGPU_CTX_OP_GET_STABLE_PSTATE	5
+#define AMDGPU_CTX_OP_SET_STABLE_PSTATE	6
 
 /* GPU reset status */
 #define AMDGPU_CTX_NO_RESET		0
@@ -238,10 +240,18 @@ union drm_amdgpu_bo_list {
 #define AMDGPU_CTX_PRIORITY_HIGH        512
 #define AMDGPU_CTX_PRIORITY_VERY_HIGH   1023
 
+/* select a stable profiling pstate for perfmon tools */
+#define AMDGPU_CTX_STABLE_PSTATE_FLAGS_MASK  0xf
+#define AMDGPU_CTX_STABLE_PSTATE_NONE  0
+#define AMDGPU_CTX_STABLE_PSTATE_STANDARD  1
+#define AMDGPU_CTX_STABLE_PSTATE_MIN_SCLK  2
+#define AMDGPU_CTX_STABLE_PSTATE_MIN_MCLK  3
+#define AMDGPU_CTX_STABLE_PSTATE_PEAK  4
+
 struct drm_amdgpu_ctx_in {
 	/** AMDGPU_CTX_OP_* */
 	__u32	op;
-	/** For future use, no flags defined so far */
+	/** Flags */
 	__u32	flags;
 	__u32	ctx_id;
 	/** AMDGPU_CTX_PRIORITY_* */
@@ -262,6 +272,11 @@ union drm_amdgpu_ctx_out {
 			/** Reset status since the last call of the ioctl. */
 			__u32	reset_status;
 		} state;
+
+		struct {
+			__u32	flags;
+			__u32	_pad;
+		} pstate;
 };
 
 union drm_amdgpu_ctx {
@@ -728,6 +743,8 @@ struct drm_amdgpu_cs_chunk_data {
 	#define AMDGPU_INFO_FW_DMCUB		0x14
 	/* Subquery id: Query TOC firmware version */
 	#define AMDGPU_INFO_FW_TOC		0x15
+	/* Subquery id: Query CAP firmware version */
+	#define AMDGPU_INFO_FW_CAP		0x16
 
 /* number of bytes moved for TTM migration */
 #define AMDGPU_INFO_NUM_BYTES_MOVED		0x0f
@@ -1134,6 +1151,8 @@ struct drm_amdgpu_info_video_caps {
 #define AMDGPU_FAMILY_NV			143 /* Navi10 */
 #define AMDGPU_FAMILY_VGH			144 /* Van Gogh */
 #define AMDGPU_FAMILY_YC			146 /* Yellow Carp */
+#define AMDGPU_FAMILY_GC_10_3_6			149 /* GC 10.3.6 */
+#define AMDGPU_FAMILY_GC_10_3_7			151 /* GC 10.3.7 */
 
 #if defined(__cplusplus)
 }
diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h
index 914ebd9290e5..05c3642aaece 100644
--- a/include/uapi/drm/i915_drm.h
+++ b/include/uapi/drm/i915_drm.h
@@ -1118,10 +1118,16 @@ struct drm_i915_gem_exec_object2 {
 	/**
 	 * 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.
+	 *
+	 * See struct drm_i915_gem_create_ext for the rules when dealing with
+	 * alignment restrictions with I915_MEMORY_CLASS_DEVICE, on devices with
+	 * minimum page sizes, like DG2.
 	 */
 	__u64 offset;
 
@@ -3144,11 +3150,40 @@ struct drm_i915_gem_create_ext {
 	 *
 	 * The (page-aligned) allocated size for the object will be returned.
 	 *
-	 * Note that for some devices we have might have further minimum
-	 * page-size restrictions(larger than 4K), like for device local-memory.
-	 * However in general the final size here should always reflect any
-	 * rounding up, if for example using the I915_GEM_CREATE_EXT_MEMORY_REGIONS
-	 * extension to place the object in device local-memory.
+	 *
+	 * DG2 64K min page size implications:
+	 *
+	 * On discrete platforms, starting from DG2, we have to contend with GTT
+	 * page size restrictions when dealing with I915_MEMORY_CLASS_DEVICE
+	 * objects.  Specifically the hardware only supports 64K or larger GTT
+	 * page sizes for such memory. The kernel will already ensure that all
+	 * I915_MEMORY_CLASS_DEVICE memory is allocated using 64K or larger page
+	 * sizes underneath.
+	 *
+	 * Note that the returned size here will always reflect any required
+	 * rounding up done by the kernel, i.e 4K will now become 64K on devices
+	 * such as DG2.
+	 *
+	 * Special DG2 GTT address alignment requirement:
+	 *
+	 * The GTT alignment will also need to be at least 2M for such objects.
+	 *
+	 * Note that due to how the hardware implements 64K GTT page support, we
+	 * have some further complications:
+	 *
+	 *   1) The entire PDE (which covers a 2MB virtual address range), must
+	 *   contain only 64K PTEs, i.e mixing 4K and 64K PTEs in the same
+	 *   PDE is forbidden by the hardware.
+	 *
+	 *   2) We still need to support 4K PTEs for I915_MEMORY_CLASS_SYSTEM
+	 *   objects.
+	 *
+	 * To keep things simple for userland, we mandate that any GTT mappings
+	 * must be aligned to and rounded up to 2MB. The kernel will internally
+	 * pad them out to the next 2MB boundary. As this only wastes virtual
+	 * address space and avoids userland having to copy any needlessly
+	 * complicated PDE sharing scheme (coloring) and only affects DG2, this
+	 * is deemed to be a good compromise.
 	 */
 	__u64 size;
 	/**
diff --git a/include/uapi/drm/msm_drm.h b/include/uapi/drm/msm_drm.h
index 6b8fffc28a50..07efc8033492 100644
--- a/include/uapi/drm/msm_drm.h
+++ b/include/uapi/drm/msm_drm.h
@@ -67,16 +67,21 @@ struct drm_msm_timespec {
 	__s64 tv_nsec;         /* nanoseconds */
 };
 
-#define MSM_PARAM_GPU_ID     0x01
-#define MSM_PARAM_GMEM_SIZE  0x02
-#define MSM_PARAM_CHIP_ID    0x03
-#define MSM_PARAM_MAX_FREQ   0x04
-#define MSM_PARAM_TIMESTAMP  0x05
-#define MSM_PARAM_GMEM_BASE  0x06
-#define MSM_PARAM_PRIORITIES 0x07  /* The # of priority levels */
-#define MSM_PARAM_PP_PGTABLE 0x08  /* => 1 for per-process pagetables, else 0 */
-#define MSM_PARAM_FAULTS     0x09
-#define MSM_PARAM_SUSPENDS   0x0a
+/* Below "RO" indicates a read-only param, "WO" indicates write-only, and
+ * "RW" indicates a param that can be both read (GET_PARAM) and written
+ * (SET_PARAM)
+ */
+#define MSM_PARAM_GPU_ID     0x01  /* RO */
+#define MSM_PARAM_GMEM_SIZE  0x02  /* RO */
+#define MSM_PARAM_CHIP_ID    0x03  /* RO */
+#define MSM_PARAM_MAX_FREQ   0x04  /* RO */
+#define MSM_PARAM_TIMESTAMP  0x05  /* RO */
+#define MSM_PARAM_GMEM_BASE  0x06  /* RO */
+#define MSM_PARAM_PRIORITIES 0x07  /* RO: The # of priority levels */
+#define MSM_PARAM_PP_PGTABLE 0x08  /* RO: Deprecated, always returns zero */
+#define MSM_PARAM_FAULTS     0x09  /* RO */
+#define MSM_PARAM_SUSPENDS   0x0a  /* RO */
+#define MSM_PARAM_SYSPROF    0x0b  /* WO: 1 preserves perfcntrs, 2 also disables suspend */
 
 /* For backwards compat.  The original support for preemption was based on
  * a single ring per priority level so # of priority levels equals the #
@@ -227,6 +232,7 @@ struct drm_msm_gem_submit_bo {
 #define MSM_SUBMIT_SUDO          0x10000000 /* run submitted cmds from RB */
 #define MSM_SUBMIT_SYNCOBJ_IN    0x08000000 /* enable input syncobj */
 #define MSM_SUBMIT_SYNCOBJ_OUT   0x04000000 /* enable output syncobj */
+#define MSM_SUBMIT_FENCE_SN_IN   0x02000000 /* userspace passes in seqno fence */
 #define MSM_SUBMIT_FLAGS                ( \
 		MSM_SUBMIT_NO_IMPLICIT   | \
 		MSM_SUBMIT_FENCE_FD_IN   | \
@@ -234,6 +240,7 @@ struct drm_msm_gem_submit_bo {
 		MSM_SUBMIT_SUDO          | \
 		MSM_SUBMIT_SYNCOBJ_IN    | \
 		MSM_SUBMIT_SYNCOBJ_OUT   | \
+		MSM_SUBMIT_FENCE_SN_IN   | \
 		0)
 
 #define MSM_SUBMIT_SYNCOBJ_RESET 0x00000001 /* Reset syncobj after wait. */
@@ -253,7 +260,7 @@ struct drm_msm_gem_submit_syncobj {
  */
 struct drm_msm_gem_submit {
 	__u32 flags;          /* MSM_PIPE_x | MSM_SUBMIT_x */
-	__u32 fence;          /* out */
+	__u32 fence;          /* out (or in with MSM_SUBMIT_FENCE_SN_IN flag) */
 	__u32 nr_bos;         /* in, number of submit_bo's */
 	__u32 nr_cmds;        /* in, number of submit_cmd's */
 	__u64 bos;            /* in, ptr to array of submit_bo's */
@@ -333,9 +340,7 @@ struct drm_msm_submitqueue_query {
 };
 
 #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
@@ -351,6 +356,7 @@ struct drm_msm_submitqueue_query {
 #define DRM_MSM_SUBMITQUEUE_QUERY      0x0C
 
 #define DRM_IOCTL_MSM_GET_PARAM        DRM_IOWR(DRM_COMMAND_BASE + DRM_MSM_GET_PARAM, struct drm_msm_param)
+#define DRM_IOCTL_MSM_SET_PARAM        DRM_IOW (DRM_COMMAND_BASE + DRM_MSM_SET_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)