[RFC,libdrm] intel: 48b ppgtt support

Submitted by Michał Winiarski on Aug. 7, 2015, 11 a.m.

Details

Message ID 1438945229-19011-1-git-send-email-michal.winiarski@intel.com
State New
Headers show

Commit Message

Michał Winiarski Aug. 7, 2015, 11 a.m.
---
 include/drm/i915_drm.h    |  3 ++-
 intel/intel_bufmgr.c      | 11 +++++++++++
 intel/intel_bufmgr.h      |  1 +
 intel/intel_bufmgr_gem.c  | 43 +++++++++++++++++++++++++++++++++----------
 intel/intel_bufmgr_priv.h |  8 ++++++++
 5 files changed, 55 insertions(+), 11 deletions(-)

Patch hide | download patch | download mbox

diff --git a/include/drm/i915_drm.h b/include/drm/i915_drm.h
index ded43b1..426b25c 100644
--- a/include/drm/i915_drm.h
+++ b/include/drm/i915_drm.h
@@ -680,7 +680,8 @@  struct drm_i915_gem_exec_object2 {
 #define EXEC_OBJECT_NEEDS_FENCE (1<<0)
 #define EXEC_OBJECT_NEEDS_GTT	(1<<1)
 #define EXEC_OBJECT_WRITE	(1<<2)
-#define __EXEC_OBJECT_UNKNOWN_FLAGS -(EXEC_OBJECT_WRITE<<1)
+#define EXEC_OBJECT_SUPPORTS_48B_ADDRESS (1<<3)
+#define __EXEC_OBJECT_UNKNOWN_FLAGS -(EXEC_OBJECT_SUPPORTS_48B_ADDRESS<<1)
 	__u64 flags;
 
 	__u64 rsvd1;
diff --git a/intel/intel_bufmgr.c b/intel/intel_bufmgr.c
index 14ea9f9..97ea6ec 100644
--- a/intel/intel_bufmgr.c
+++ b/intel/intel_bufmgr.c
@@ -261,6 +261,17 @@  drm_intel_bo_get_tiling(drm_intel_bo *bo, uint32_t * tiling_mode,
 }
 
 int
+drm_intel_bo_use_full_range(drm_intel_bo *bo, uint32_t enable)
+{
+	if (bo->bufmgr->bo_use_full_range) {
+		bo->bufmgr->bo_use_full_range(bo, enable);
+		return 0;
+	}
+
+	return -ENODEV;
+}
+
+int
 drm_intel_bo_disable_reuse(drm_intel_bo *bo)
 {
 	if (bo->bufmgr->bo_disable_reuse)
diff --git a/intel/intel_bufmgr.h b/intel/intel_bufmgr.h
index 285919e..2635fa4 100644
--- a/intel/intel_bufmgr.h
+++ b/intel/intel_bufmgr.h
@@ -160,6 +160,7 @@  int drm_intel_bo_get_tiling(drm_intel_bo *bo, uint32_t * tiling_mode,
 int drm_intel_bo_flink(drm_intel_bo *bo, uint32_t * name);
 int drm_intel_bo_busy(drm_intel_bo *bo);
 int drm_intel_bo_madvise(drm_intel_bo *bo, int madv);
+int drm_intel_bo_use_full_range(drm_intel_bo *bo, uint32_t enable);
 
 int drm_intel_bo_disable_reuse(drm_intel_bo *bo);
 int drm_intel_bo_is_reusable(drm_intel_bo *bo);
diff --git a/intel/intel_bufmgr_gem.c b/intel/intel_bufmgr_gem.c
index 41de396..ef71686 100644
--- a/intel/intel_bufmgr_gem.c
+++ b/intel/intel_bufmgr_gem.c
@@ -237,6 +237,11 @@  struct _drm_intel_bo_gem {
 	bool is_userptr;
 
 	/**
+	 * Whether this buffer can be placed in full (2^48) ppgtt range on gen8+
+	 */
+	bool uses_full_range;
+
+	/**
 	 * Size in bytes of this buffer and its relocation descendents.
 	 *
 	 * Used to avoid costly tree walking in
@@ -395,8 +400,8 @@  drm_intel_gem_dump_validation_list(drm_intel_bufmgr_gem *bufmgr_gem)
 			drm_intel_bo_gem *target_gem =
 			    (drm_intel_bo_gem *) target_bo;
 
-			DBG("%2d: %d (%s)@0x%08llx -> "
-			    "%d (%s)@0x%08lx + 0x%08x\n",
+			DBG("%2d: %d (%s)@0x%016llx -> "
+			    "%d (%s)@0x%016lx + 0x%08x\n",
 			    i,
 			    bo_gem->gem_handle, bo_gem->name,
 			    (unsigned long long)bo_gem->relocs[j].offset,
@@ -468,11 +473,15 @@  drm_intel_add_validate_buffer2(drm_intel_bo *bo, int need_fence)
 	drm_intel_bufmgr_gem *bufmgr_gem = (drm_intel_bufmgr_gem *)bo->bufmgr;
 	drm_intel_bo_gem *bo_gem = (drm_intel_bo_gem *)bo;
 	int index;
+	int flags = 0;
+
+	if (need_fence)
+		flags |= EXEC_OBJECT_NEEDS_FENCE;
+	if (bo_gem->uses_full_range)
+		flags |= EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
 
 	if (bo_gem->validate_index != -1) {
-		if (need_fence)
-			bufmgr_gem->exec2_objects[bo_gem->validate_index].flags |=
-				EXEC_OBJECT_NEEDS_FENCE;
+		bufmgr_gem->exec2_objects[bo_gem->validate_index].flags |= flags;
 		return;
 	}
 
@@ -501,13 +510,9 @@  drm_intel_add_validate_buffer2(drm_intel_bo *bo, int need_fence)
 	bufmgr_gem->exec2_objects[index].alignment = bo->align;
 	bufmgr_gem->exec2_objects[index].offset = 0;
 	bufmgr_gem->exec_bos[index] = bo;
-	bufmgr_gem->exec2_objects[index].flags = 0;
+	bufmgr_gem->exec2_objects[index].flags = flags;
 	bufmgr_gem->exec2_objects[index].rsvd1 = 0;
 	bufmgr_gem->exec2_objects[index].rsvd2 = 0;
-	if (need_fence) {
-		bufmgr_gem->exec2_objects[index].flags |=
-			EXEC_OBJECT_NEEDS_FENCE;
-	}
 	bufmgr_gem->exec_count++;
 }
 
@@ -780,6 +785,7 @@  retry:
 	bo_gem->used_as_reloc_target = false;
 	bo_gem->has_error = false;
 	bo_gem->reusable = true;
+	bo_gem->uses_full_range = false;
 
 	drm_intel_bo_gem_set_in_aperture_size(bufmgr_gem, bo_gem, alignment);
 
@@ -926,6 +932,7 @@  drm_intel_gem_bo_alloc_userptr(drm_intel_bufmgr *bufmgr,
 	bo_gem->used_as_reloc_target = false;
 	bo_gem->has_error = false;
 	bo_gem->reusable = false;
+	bo_gem->uses_full_range = false;
 
 	drm_intel_bo_gem_set_in_aperture_size(bufmgr_gem, bo_gem, 0);
 
@@ -1081,6 +1088,7 @@  drm_intel_bo_gem_create_from_name(drm_intel_bufmgr *bufmgr,
 	bo_gem->bo.handle = open_arg.handle;
 	bo_gem->global_name = handle;
 	bo_gem->reusable = false;
+	bo_gem->uses_full_range = false;
 
 	memclear(get_tiling);
 	get_tiling.handle = bo_gem->gem_handle;
@@ -2418,6 +2426,13 @@  drm_intel_gem_bo_get_tiling(drm_intel_bo *bo, uint32_t * tiling_mode,
 	return 0;
 }
 
+static void
+drm_intel_gem_bo_use_full_range(drm_intel_bo *bo, uint32_t enable)
+{
+	drm_intel_bo_gem *bo_gem = (drm_intel_bo_gem *) bo;
+	bo_gem->uses_full_range = enable;
+}
+
 drm_intel_bo *
 drm_intel_bo_gem_create_from_prime(drm_intel_bufmgr *bufmgr, int prime_fd, int size)
 {
@@ -2482,6 +2497,7 @@  drm_intel_bo_gem_create_from_prime(drm_intel_bufmgr *bufmgr, int prime_fd, int s
 	bo_gem->used_as_reloc_target = false;
 	bo_gem->has_error = false;
 	bo_gem->reusable = false;
+	bo_gem->uses_full_range = false;
 
 	DRMINITLISTHEAD(&bo_gem->vma_list);
 	DRMLISTADDTAIL(&bo_gem->name_list, &bufmgr_gem->named);
@@ -3278,6 +3294,13 @@  drm_intel_bufmgr_gem_init(int fd, int batch_size)
 		}
 	}
 
+	if (bufmgr_gem->gen >= 8) {
+		gp.param = I915_PARAM_HAS_ALIASING_PPGTT;
+		ret = drmIoctl(bufmgr_gem->fd, DRM_IOCTL_I915_GETPARAM, &gp);
+		if (ret == 0 && *gp.value == 3)
+			bufmgr_gem->bufmgr.bo_use_full_range = drm_intel_gem_bo_use_full_range;
+	}
+
 	/* Let's go with one relocation per every 2 dwords (but round down a bit
 	 * since a power of two will mean an extra page allocation for the reloc
 	 * buffer).
diff --git a/intel/intel_bufmgr_priv.h b/intel/intel_bufmgr_priv.h
index 59ebd18..3461d31 100644
--- a/intel/intel_bufmgr_priv.h
+++ b/intel/intel_bufmgr_priv.h
@@ -227,6 +227,14 @@  struct _drm_intel_bufmgr {
 			      uint32_t * swizzle_mode);
 
 	/**
+	 * Indicate that the buffer can be placed in full ppgtt range (2^48)
+	 * without restrictions.
+	 * \param bo Buffer to set the full_range flag for
+	 * \param enable Flag value
+	 */
+	void (*bo_use_full_range) (drm_intel_bo *bo, uint32_t enable);
+
+	/**
 	 * Create a visible name for a buffer which can be used by other apps
 	 *
 	 * \param buf Buffer to create a name for