Skip to contentSkip to conversation

drm/i915: pass the vma to insert_entries

PATCH In review Matthew Auld 3 participants 3 messages Jun 22, 2017

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 drm/i915: pass the vma to insert_entries

The vma already contains most of the information we need for insertion. But also in preparation for supporting huge gtt pages, it would be useful to know the details of the vma, such that we can we can easily determine the page sizes we are allowed to use when inserting into the 48b PPGTT. This is especially true for 64K where we can't just arbitrarily use it, since we require aligning/padding the vm space to 2M, which sometimes we can't enforce in the upper levels.

---
 drivers/gpu/drm/i915/i915_gem_gtt.c           | 68 +++++++++++----------------
 drivers/gpu/drm/i915/i915_gem_gtt.h           |  3 +-
 drivers/gpu/drm/i915/selftests/i915_gem_gtt.c | 10 +++-
 drivers/gpu/drm/i915/selftests/mock_gtt.c     |  3 +-
 4 files changed, 38 insertions(+), 46 deletions(-)
drivers/gpu/drm/i915/i915_gem_gtt.c +28 -40
index 61fc7e90a7da..de67084d5fcf 100644
@@ -207,8 +207,7 @@ static int ppgtt_bind_vma(struct i915_vma *vma,
 	if (vma->obj->gt_ro)
 		pte_flags |= PTE_READ_ONLY;
 
-	vma->vm->insert_entries(vma->vm, vma->pages, vma->node.start,
-				cache_level, pte_flags);
+	vma->vm->insert_entries(vma->vm, vma, cache_level, pte_flags);
 
 	return 0;
 }
@@ -907,37 +906,35 @@ gen8_ppgtt_insert_pte_entries(struct i915_hw_ppgtt *ppgtt,
 }
 
 static void gen8_ppgtt_insert_3lvl(struct i915_address_space *vm,
-				   struct sg_table *pages,
-				   u64 start,
+				   struct i915_vma *vma,
 				   enum i915_cache_level cache_level,
 				   u32 unused)
 {
-	struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm);
+	struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vma->vm);
 	struct sgt_dma iter = {
-		.sg = pages->sgl,
+		.sg = vma->pages->sgl,
 		.dma = sg_dma_address(iter.sg),
 		.max = iter.dma + iter.sg->length,
 	};
-	struct gen8_insert_pte idx = gen8_insert_pte(start);
+	struct gen8_insert_pte idx = gen8_insert_pte(vma->node.start);
 
 	gen8_ppgtt_insert_pte_entries(ppgtt, &ppgtt->pdp, &iter, &idx,
 				      cache_level);
 }
 
 static void gen8_ppgtt_insert_4lvl(struct i915_address_space *vm,
-				   struct sg_table *pages,
-				   u64 start,
+				   struct i915_vma *vma,
 				   enum i915_cache_level cache_level,
 				   u32 unused)
 {
 	struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm);
 	struct sgt_dma iter = {
-		.sg = pages->sgl,
+		.sg = vma->pages->sgl,
 		.dma = sg_dma_address(iter.sg),
 		.max = iter.dma + iter.sg->length,
 	};
 	struct i915_page_directory_pointer **pdps = ppgtt->pml4.pdps;
-	struct gen8_insert_pte idx = gen8_insert_pte(start);
+	struct gen8_insert_pte idx = gen8_insert_pte(vma->node.start);
 
 	while (gen8_ppgtt_insert_pte_entries(ppgtt, pdps[idx.pml4e++], &iter,
 					     &idx, cache_level))
@@ -1621,13 +1618,12 @@ static void gen6_ppgtt_clear_range(struct i915_address_space *vm,
 }
 
 static void gen6_ppgtt_insert_entries(struct i915_address_space *vm,
-				      struct sg_table *pages,
-				      u64 start,
+				      struct i915_vma *vma,
 				      enum i915_cache_level cache_level,
 				      u32 flags)
 {
 	struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm);
-	unsigned first_entry = start >> PAGE_SHIFT;
+	unsigned first_entry = vma->node.start >> PAGE_SHIFT;
 	unsigned act_pt = first_entry / GEN6_PTES;
 	unsigned act_pte = first_entry % GEN6_PTES;
 	const u32 pte_encode = vm->pte_encode(0, cache_level, flags);
@@ -1635,7 +1631,7 @@ static void gen6_ppgtt_insert_entries(struct i915_address_space *vm,
 	gen6_pte_t *vaddr;
 
 	vaddr = kmap_atomic_px(ppgtt->pd.page_table[act_pt]);
-	iter.sg = pages->sgl;
+	iter.sg = vma->pages->sgl;
 	iter.dma = sg_dma_address(iter.sg);
 	iter.max = iter.dma + iter.sg->length;
 	do {
@@ -2090,8 +2086,7 @@ static void gen8_ggtt_insert_page(struct i915_address_space *vm,
 }
 
 static void gen8_ggtt_insert_entries(struct i915_address_space *vm,
-				     struct sg_table *st,
-				     u64 start,
+				     struct i915_vma *vma,
 				     enum i915_cache_level level,
 				     u32 unused)
 {
@@ -2102,8 +2097,8 @@ static void gen8_ggtt_insert_entries(struct i915_address_space *vm,
 	dma_addr_t addr;
 
 	gtt_entries = (gen8_pte_t __iomem *)ggtt->gsm;
-	gtt_entries += start >> PAGE_SHIFT;
-	for_each_sgt_dma(addr, sgt_iter, st)
+	gtt_entries += vma->node.start >> PAGE_SHIFT;
+	for_each_sgt_dma(addr, sgt_iter, vma->pages)
 		gen8_set_pte(gtt_entries++, pte_encode | addr);
 
 	wmb();
@@ -2137,17 +2132,16 @@ static void gen6_ggtt_insert_page(struct i915_address_space *vm,
  * mapped BAR (dev_priv->mm.gtt->gtt).
  */
 static void gen6_ggtt_insert_entries(struct i915_address_space *vm,
-				     struct sg_table *st,
-				     u64 start,
+				     struct i915_vma *vma,
 				     enum i915_cache_level level,
 				     u32 flags)
 {
 	struct i915_ggtt *ggtt = i915_vm_to_ggtt(vm);
 	gen6_pte_t __iomem *entries = (gen6_pte_t __iomem *)ggtt->gsm;
-	unsigned int i = start >> PAGE_SHIFT;
+	unsigned int i = vma->node.start >> PAGE_SHIFT;
 	struct sgt_iter iter;
 	dma_addr_t addr;
-	for_each_sgt_dma(addr, iter, st)
+	for_each_sgt_dma(addr, iter, vma->pages)
 		iowrite32(vm->pte_encode(addr, level, flags), &entries[i++]);
 	wmb();
 
@@ -2229,8 +2223,7 @@ static void bxt_vtd_ggtt_insert_page__BKL(struct i915_address_space *vm,
 
 struct insert_entries {
 	struct i915_address_space *vm;
-	struct sg_table *st;
-	u64 start;
+	struct i915_vma *vma;
 	enum i915_cache_level level;
 };
 
@@ -2238,19 +2231,18 @@ static int bxt_vtd_ggtt_insert_entries__cb(void *_arg)
 {
 	struct insert_entries *arg = _arg;
 
-	gen8_ggtt_insert_entries(arg->vm, arg->st, arg->start, arg->level, 0);
+	gen8_ggtt_insert_entries(arg->vm, arg->vma, arg->level, 0);
 	bxt_vtd_ggtt_wa(arg->vm);
 
 	return 0;
 }
 
 static void bxt_vtd_ggtt_insert_entries__BKL(struct i915_address_space *vm,
-					     struct sg_table *st,
-					     u64 start,
+					     struct i915_vma *vma,
 					     enum i915_cache_level level,
 					     u32 unused)
 {
-	struct insert_entries arg = { vm, st, start, level };
+	struct insert_entries arg = { vma->vm, vma, level };
 
 	stop_machine(bxt_vtd_ggtt_insert_entries__cb, &arg, NULL);
 }
@@ -2316,15 +2308,15 @@ static void i915_ggtt_insert_page(struct i915_address_space *vm,
 }
 
 static void i915_ggtt_insert_entries(struct i915_address_space *vm,
-				     struct sg_table *pages,
-				     u64 start,
+				     struct i915_vma *vma,
 				     enum i915_cache_level cache_level,
 				     u32 unused)
 {
 	unsigned int flags = (cache_level == I915_CACHE_NONE) ?
 		AGP_USER_MEMORY : AGP_USER_CACHED_MEMORY;
 
-	intel_gtt_insert_sg_entries(pages, start >> PAGE_SHIFT, flags);
+	intel_gtt_insert_sg_entries(vma->pages, vma->node.start >> PAGE_SHIFT,
+				    flags);
 }
 
 static void i915_ggtt_clear_range(struct i915_address_space *vm,
@@ -2353,8 +2345,7 @@ static int ggtt_bind_vma(struct i915_vma *vma,
 		pte_flags |= PTE_READ_ONLY;
 
 	intel_runtime_pm_get(i915);
-	vma->vm->insert_entries(vma->vm, vma->pages, vma->node.start,
-				cache_level, pte_flags);
+	vma->vm->insert_entries(vma->vm, vma, cache_level, pte_flags);
 	intel_runtime_pm_put(i915);
 
 	/*
@@ -2407,16 +2398,13 @@ static int aliasing_gtt_bind_vma(struct i915_vma *vma,
 				goto err_pages;
 		}
 
-		appgtt->base.insert_entries(&appgtt->base,
-					    vma->pages, vma->node.start,
-					    cache_level, pte_flags);
+		appgtt->base.insert_entries(&appgtt->base, vma, cache_level,
+					    pte_flags);
 	}
 
 	if (flags & I915_VMA_GLOBAL_BIND) {
 		intel_runtime_pm_get(i915);
-		vma->vm->insert_entries(vma->vm,
-					vma->pages, vma->node.start,
-					cache_level, pte_flags);
+		vma->vm->insert_entries(vma->vm, vma, cache_level, pte_flags);
 		intel_runtime_pm_put(i915);
 	}
 
drivers/gpu/drm/i915/i915_gem_gtt.h +1 -2
index 1b2a56c3e5d3..b4e3aa7c0ce1 100644
@@ -313,8 +313,7 @@ struct i915_address_space {
 			    enum i915_cache_level cache_level,
 			    u32 flags);
 	void (*insert_entries)(struct i915_address_space *vm,
-			       struct sg_table *st,
-			       u64 start,
+			       struct i915_vma *vma,
 			       enum i915_cache_level cache_level,
 			       u32 flags);
 	void (*cleanup)(struct i915_address_space *vm);
drivers/gpu/drm/i915/selftests/i915_gem_gtt.c +8 -2
index 50710e3f1caa..6b132caffa18 100644
@@ -197,6 +197,9 @@ static int lowlevel_hole(struct drm_i915_private *i915,
 {
 	I915_RND_STATE(seed_prng);
 	unsigned int size;
+	struct i915_vma mock_vma;
+
+	memset(&mock_vma, 0, sizeof(struct i915_vma));
 
 	/* Keep creating larger objects until one cannot fit into the hole */
 	for (size = 12; (hole_end - hole_start) >> size; size++) {
@@ -255,8 +258,11 @@ static int lowlevel_hole(struct drm_i915_private *i915,
 			    vm->allocate_va_range(vm, addr, BIT_ULL(size)))
 				break;
 
-			vm->insert_entries(vm, obj->mm.pages, addr,
-					   I915_CACHE_NONE, 0);
+			mock_vma.pages = obj->mm.pages;
+			mock_vma.node.size = BIT_ULL(size);
+			mock_vma.node.start = addr;
+
+			vm->insert_entries(vm, &mock_vma, I915_CACHE_NONE, 0);
 		}
 		count = n;
 
drivers/gpu/drm/i915/selftests/mock_gtt.c +1 -2
index a61309c7cb3e..f2118cf535a0 100644
@@ -33,8 +33,7 @@ static void mock_insert_page(struct i915_address_space *vm,
 }
 
 static void mock_insert_entries(struct i915_address_space *vm,
-				struct sg_table *st,
-				u64 start,
+				struct i915_vma *vma,
 				enum i915_cache_level level, u32 flags)
 {
 }
Signature · 6 lines
2.9.4

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#1 lore raw

✓ Fi.CI.BAT: success for drm/i915: pass the vma to insert_entries

== Series Details ==

== Series Details ==

Series: drm/i915: pass the vma to insert_entries
URL : https://patchwork.freedesktop.org/series/26208/
State : success

== Summary ==

Series 26208v1 drm/i915: pass the vma to insert_entries
https://patchwork.freedesktop.org/api/1.0/series/26208/revisions/1/mbox/

Test gem_exec_suspend:

        Subgroup basic-s4-devices:
                dmesg-warn -> PASS       (fi-kbl-7560u) fdo#100125
Test kms_cursor_legacy:
        Subgroup basic-busy-flip-before-cursor-atomic:
                fail       -> PASS       (fi-snb-2600) fdo#100215

fdo#100125 https://bugs.freedesktop.org/show_bug.cgi?id=100125
fdo#100215 https://bugs.freedesktop.org/show_bug.cgi?id=100215

fi-bdw-5557u     total:278  pass:267  dwarn:0   dfail:0   fail:0   skip:11  time:439s
fi-bdw-gvtdvm    total:278  pass:256  dwarn:8   dfail:0   fail:0   skip:14  time:430s
fi-bsw-n3050     total:278  pass:241  dwarn:1   dfail:0   fail:0   skip:36  time:532s
fi-bxt-j4205     total:278  pass:259  dwarn:0   dfail:0   fail:0   skip:19  time:506s
fi-byt-j1900     total:278  pass:252  dwarn:2   dfail:0   fail:0   skip:24  time:491s
fi-byt-n2820     total:278  pass:248  dwarn:2   dfail:0   fail:0   skip:28  time:481s
fi-glk-2a        total:278  pass:259  dwarn:0   dfail:0   fail:0   skip:19  time:594s
fi-hsw-4770      total:278  pass:262  dwarn:0   dfail:0   fail:0   skip:16  time:433s
fi-hsw-4770r     total:278  pass:262  dwarn:0   dfail:0   fail:0   skip:16  time:415s
fi-ilk-650       total:278  pass:228  dwarn:0   dfail:0   fail:0   skip:50  time:421s
fi-ivb-3520m     total:278  pass:260  dwarn:0   dfail:0   fail:0   skip:18  time:501s
fi-ivb-3770      total:278  pass:260  dwarn:0   dfail:0   fail:0   skip:18  time:477s
fi-kbl-7500u     total:278  pass:260  dwarn:0   dfail:0   fail:0   skip:18  time:469s
fi-kbl-7560u     total:278  pass:268  dwarn:0   dfail:0   fail:0   skip:10  time:573s
fi-kbl-r         total:278  pass:259  dwarn:1   dfail:0   fail:0   skip:18  time:579s
fi-skl-6260u     total:278  pass:268  dwarn:0   dfail:0   fail:0   skip:10  time:455s
fi-skl-6700hq    total:278  pass:220  dwarn:3   dfail:0   fail:30  skip:24  time:345s
fi-skl-6700k     total:278  pass:256  dwarn:4   dfail:0   fail:0   skip:18  time:468s
fi-skl-6770hq    total:278  pass:268  dwarn:0   dfail:0   fail:0   skip:10  time:482s
fi-skl-gvtdvm    total:278  pass:265  dwarn:0   dfail:0   fail:0   skip:13  time:435s
fi-snb-2520m     total:278  pass:250  dwarn:0   dfail:0   fail:0   skip:28  time:541s
fi-snb-2600      total:278  pass:249  dwarn:0   dfail:0   fail:0   skip:29  time:400s

7f2c92652872fbbe37c8bca92eccb7218100c21d drm-tip: 2017y-06m-22d-08h-47m-32s UTC integration manifest 4d2b804 drm/i915: pass the vma to insert_entries

== Logs ==

For more details see: https://intel-gfx-ci.01.org/CI/Patchwork_5021/
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#2 lore raw

Re: ✓ Fi.CI.BAT: success for drm/i915: pass the vma to insert_entries

Quoting Patchwork (2017-06-22 11:13:33)

Quoted 44 lines

== Series Details ==

Series: drm/i915: pass the vma to insert_entries
URL : https://patchwork.freedesktop.org/series/26208/
State : success

== Summary ==

Series 26208v1 drm/i915: pass the vma to insert_entries
https://patchwork.freedesktop.org/api/1.0/series/26208/revisions/1/mbox/

Test gem_exec_suspend:

        Subgroup basic-s4-devices:
                dmesg-warn -> PASS       (fi-kbl-7560u) fdo#100125
Test kms_cursor_legacy:
        Subgroup basic-busy-flip-before-cursor-atomic:
                fail       -> PASS       (fi-snb-2600) fdo#100215

fdo#100125 https://bugs.freedesktop.org/show_bug.cgi?id=100125
fdo#100215 https://bugs.freedesktop.org/show_bug.cgi?id=100215

fi-bdw-5557u     total:278  pass:267  dwarn:0   dfail:0   fail:0   skip:11  time:439s
fi-bdw-gvtdvm    total:278  pass:256  dwarn:8   dfail:0   fail:0   skip:14  time:430s
fi-bsw-n3050     total:278  pass:241  dwarn:1   dfail:0   fail:0   skip:36  time:532s
fi-bxt-j4205     total:278  pass:259  dwarn:0   dfail:0   fail:0   skip:19  time:506s
fi-byt-j1900     total:278  pass:252  dwarn:2   dfail:0   fail:0   skip:24  time:491s
fi-byt-n2820     total:278  pass:248  dwarn:2   dfail:0   fail:0   skip:28  time:481s
fi-glk-2a        total:278  pass:259  dwarn:0   dfail:0   fail:0   skip:19  time:594s
fi-hsw-4770      total:278  pass:262  dwarn:0   dfail:0   fail:0   skip:16  time:433s
fi-hsw-4770r     total:278  pass:262  dwarn:0   dfail:0   fail:0   skip:16  time:415s
fi-ilk-650       total:278  pass:228  dwarn:0   dfail:0   fail:0   skip:50  time:421s
fi-ivb-3520m     total:278  pass:260  dwarn:0   dfail:0   fail:0   skip:18  time:501s
fi-ivb-3770      total:278  pass:260  dwarn:0   dfail:0   fail:0   skip:18  time:477s
fi-kbl-7500u     total:278  pass:260  dwarn:0   dfail:0   fail:0   skip:18  time:469s
fi-kbl-7560u     total:278  pass:268  dwarn:0   dfail:0   fail:0   skip:10  time:573s
fi-kbl-r         total:278  pass:259  dwarn:1   dfail:0   fail:0   skip:18  time:579s
fi-skl-6260u     total:278  pass:268  dwarn:0   dfail:0   fail:0   skip:10  time:455s
fi-skl-6700hq    total:278  pass:220  dwarn:3   dfail:0   fail:30  skip:24  time:345s
fi-skl-6700k     total:278  pass:256  dwarn:4   dfail:0   fail:0   skip:18  time:468s
fi-skl-6770hq    total:278  pass:268  dwarn:0   dfail:0   fail:0   skip:10  time:482s
fi-skl-gvtdvm    total:278  pass:265  dwarn:0   dfail:0   fail:0   skip:13  time:435s
fi-snb-2520m     total:278  pass:250  dwarn:0   dfail:0   fail:0   skip:28  time:541s
fi-snb-2600      total:278  pass:249  dwarn:0   dfail:0   fail:0   skip:29  time:400s
Quoted 2 lines

7f2c92652872fbbe37c8bca92eccb7218100c21d drm-tip: 2017y-06m-22d-08h-47m-32s UTC integration manifest 4d2b804 drm/i915: pass the vma to insert_entries

And pushed because I love the smell of conflicts in the morning. -Chris
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx