[PATCH 1/2] drm/amdgpu: use TTM_PL_FLAG_CONTIGUOUS v2

[PATCH 1/2] drm/amdgpu: use TTM_PL_FLAG_CONTIGUOUS v2

[Christian König]

From: Christian König <christian.koenig@amd.com>

Implement AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS using TTM_PL_FLAG_CONTIGUOUS
instead of a placement limit. That allows us to better handle CPU
accessible placements.

v2: prevent virtual BO start address from overflowing

Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Michel Dänzer <michel.daenzer@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c   | 11 +++++------
 drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c | 17 +++++++++++++----
 2 files changed, 18 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index d6b2de9..387d190 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -122,20 +122,19 @@ static void amdgpu_ttm_placement_init(struct amdgpu_device *adev,
 
 	if (domain & AMDGPU_GEM_DOMAIN_VRAM) {
 		unsigned visible_pfn = adev->mc.visible_vram_size >> PAGE_SHIFT;
-		unsigned lpfn = 0;
-
-		/* This forces a reallocation if the flag wasn't set before */
-		if (flags & AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS)
-			lpfn = adev->mc.real_vram_size >> PAGE_SHIFT;
 
 		places[c].fpfn = 0;
-		places[c].lpfn = lpfn;
+		places[c].lpfn = 0;
 		places[c].flags = TTM_PL_FLAG_WC | TTM_PL_FLAG_UNCACHED |
 			TTM_PL_FLAG_VRAM;
+
 		if (flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED)
 			places[c].lpfn = visible_pfn;
 		else
 			places[c].flags |= TTM_PL_FLAG_TOPDOWN;
+
+		if (flags & AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS)
+			places[c].flags |= TTM_PL_FLAG_CONTIGUOUS;
 		c++;
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
index d710226..977195f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
@@ -93,7 +93,6 @@ static int amdgpu_vram_mgr_new(struct ttm_mem_type_manager *man,
 			       const struct ttm_place *place,
 			       struct ttm_mem_reg *mem)
 {
-	struct amdgpu_bo *bo = container_of(tbo, struct amdgpu_bo, tbo);
 	struct amdgpu_vram_mgr *mgr = man->priv;
 	struct drm_mm *mm = &mgr->mm;
 	struct drm_mm_node *nodes;
@@ -107,8 +106,8 @@ static int amdgpu_vram_mgr_new(struct ttm_mem_type_manager *man,
 	if (!lpfn)
 		lpfn = man->size;
 
-	if (bo->flags & AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS ||
-	    place->lpfn || amdgpu_vram_page_split == -1) {
+	if (place->flags & TTM_PL_FLAG_CONTIGUOUS ||
+	    amdgpu_vram_page_split == -1) {
 		pages_per_node = ~0ul;
 		num_nodes = 1;
 	} else {
@@ -126,12 +125,14 @@ static int amdgpu_vram_mgr_new(struct ttm_mem_type_manager *man,
 		aflags = DRM_MM_CREATE_TOP;
 	}
 
+	mem->start = 0;
 	pages_left = mem->num_pages;
 
 	spin_lock(&mgr->lock);
 	for (i = 0; i < num_nodes; ++i) {
 		unsigned long pages = min(pages_left, pages_per_node);
 		uint32_t alignment = mem->page_alignment;
+		unsigned long start;
 
 		if (pages == pages_per_node)
 			alignment = pages_per_node;
@@ -145,11 +146,19 @@ static int amdgpu_vram_mgr_new(struct ttm_mem_type_manager *man,
 		if (unlikely(r))
 			goto error;
 
+		/* Calculate a virtual BO start address to easily check if
+		 * everything is CPU accessible.
+		 */
+		start = nodes[i].start + nodes[i].size;
+		if (start > mem->num_pages)
+			start -= mem->num_pages;
+		else
+			start = 0;
+		mem->start = max(mem->start, start);
 		pages_left -= pages;
 	}
 	spin_unlock(&mgr->lock);
 
-	mem->start = num_nodes == 1 ? nodes[0].start : AMDGPU_BO_INVALID_OFFSET;
 	mem->mm_node = nodes;
 
 	return 0;
-- 
2.5.0

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

[PATCH 2/2] drm/amdgpu: handle CPU access for split VRAM buffers

[Christian König]

From: Christian König <christian.koenig@amd.com>

This avoids merging them together on page fault.

Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Michel Dänzer <michel.daenzer@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c |  4 +---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c    | 16 ++++++++++++----
 2 files changed, 13 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 387d190..10237a8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -927,8 +927,7 @@ int amdgpu_bo_fault_reserve_notify(struct ttm_buffer_object *bo)
 	size = bo->mem.num_pages << PAGE_SHIFT;
 	offset = bo->mem.start << PAGE_SHIFT;
 	/* TODO: figure out how to map scattered VRAM to the CPU */
-	if ((offset + size) <= adev->mc.visible_vram_size &&
-	    (abo->flags & AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS))
+	if ((offset + size) <= adev->mc.visible_vram_size)
 		return 0;
 
 	/* Can't move a pinned BO to visible VRAM */
@@ -936,7 +935,6 @@ int amdgpu_bo_fault_reserve_notify(struct ttm_buffer_object *bo)
 		return -EINVAL;
 
 	/* hurrah the memory is not visible ! */
-	abo->flags |= AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS;
 	amdgpu_ttm_placement_from_domain(abo, AMDGPU_GEM_DOMAIN_VRAM);
 	lpfn =	adev->mc.visible_vram_size >> PAGE_SHIFT;
 	for (i = 0; i < abo->placement.num_placement; i++) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 524abca..10b793a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -529,9 +529,6 @@ static int amdgpu_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_
 	case TTM_PL_TT:
 		break;
 	case TTM_PL_VRAM:
-		if (mem->start == AMDGPU_BO_INVALID_OFFSET)
-			return -EINVAL;
-
 		mem->bus.offset = mem->start << PAGE_SHIFT;
 		/* check if it's visible */
 		if ((mem->bus.offset + mem->bus.size) > adev->mc.visible_vram_size)
@@ -549,6 +546,17 @@ static void amdgpu_ttm_io_mem_free(struct ttm_bo_device *bdev, struct ttm_mem_re
 {
 }
 
+static unsigned long amdgpu_ttm_io_mem_pfn(struct ttm_buffer_object *bo,
+					   unsigned long page_offset)
+{
+	struct drm_mm_node *mm = bo->mem.mm_node;
+	uint64_t size = mm->size;
+
+	mm += page_offset / size;
+	page_offset %= size;
+	return (bo->mem.bus.base >> PAGE_SHIFT) + mm->start + page_offset;
+}
+
 /*
  * TTM backend functions.
  */
@@ -1064,7 +1072,7 @@ static struct ttm_bo_driver amdgpu_bo_driver = {
 	.fault_reserve_notify = &amdgpu_bo_fault_reserve_notify,
 	.io_mem_reserve = &amdgpu_ttm_io_mem_reserve,
 	.io_mem_free = &amdgpu_ttm_io_mem_free,
-	.io_mem_pfn = ttm_bo_default_io_mem_pfn,
+	.io_mem_pfn = amdgpu_ttm_io_mem_pfn,
 };
 
 int amdgpu_ttm_init(struct amdgpu_device *adev)
-- 
2.5.0

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

Re: [PATCH 2/2] drm/amdgpu: handle CPU access for split VRAM buffers

[Nicolai Hähnle]

On 04.04.2017 14:48, Christian König wrote:
> From: Christian König <christian.koenig@amd.com>
>
> This avoids merging them together on page fault.
>
> Signed-off-by: Christian König <christian.koenig@amd.com>
> Acked-by: Michel Dänzer <michel.daenzer@amd.com>

Thanks for the explanations in the earlier thread. Both patches are

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>


> ---
>  drivers/gpu/drm/amd/amdgpu/amdgpu_object.c |  4 +---
>  drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c    | 16 ++++++++++++----
>  2 files changed, 13 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
> index 387d190..10237a8 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
> @@ -927,8 +927,7 @@ int amdgpu_bo_fault_reserve_notify(struct ttm_buffer_object *bo)
>  	size = bo->mem.num_pages << PAGE_SHIFT;
>  	offset = bo->mem.start << PAGE_SHIFT;
>  	/* TODO: figure out how to map scattered VRAM to the CPU */
> -	if ((offset + size) <= adev->mc.visible_vram_size &&
> -	    (abo->flags & AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS))
> +	if ((offset + size) <= adev->mc.visible_vram_size)
>  		return 0;
>
>  	/* Can't move a pinned BO to visible VRAM */
> @@ -936,7 +935,6 @@ int amdgpu_bo_fault_reserve_notify(struct ttm_buffer_object *bo)
>  		return -EINVAL;
>
>  	/* hurrah the memory is not visible ! */
> -	abo->flags |= AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS;
>  	amdgpu_ttm_placement_from_domain(abo, AMDGPU_GEM_DOMAIN_VRAM);
>  	lpfn =	adev->mc.visible_vram_size >> PAGE_SHIFT;
>  	for (i = 0; i < abo->placement.num_placement; i++) {
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> index 524abca..10b793a 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> @@ -529,9 +529,6 @@ static int amdgpu_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_
>  	case TTM_PL_TT:
>  		break;
>  	case TTM_PL_VRAM:
> -		if (mem->start == AMDGPU_BO_INVALID_OFFSET)
> -			return -EINVAL;
> -
>  		mem->bus.offset = mem->start << PAGE_SHIFT;
>  		/* check if it's visible */
>  		if ((mem->bus.offset + mem->bus.size) > adev->mc.visible_vram_size)
> @@ -549,6 +546,17 @@ static void amdgpu_ttm_io_mem_free(struct ttm_bo_device *bdev, struct ttm_mem_re
>  {
>  }
>
> +static unsigned long amdgpu_ttm_io_mem_pfn(struct ttm_buffer_object *bo,
> +					   unsigned long page_offset)
> +{
> +	struct drm_mm_node *mm = bo->mem.mm_node;
> +	uint64_t size = mm->size;
> +
> +	mm += page_offset / size;
> +	page_offset %= size;
> +	return (bo->mem.bus.base >> PAGE_SHIFT) + mm->start + page_offset;
> +}
> +
>  /*
>   * TTM backend functions.
>   */
> @@ -1064,7 +1072,7 @@ static struct ttm_bo_driver amdgpu_bo_driver = {
>  	.fault_reserve_notify = &amdgpu_bo_fault_reserve_notify,
>  	.io_mem_reserve = &amdgpu_ttm_io_mem_reserve,
>  	.io_mem_free = &amdgpu_ttm_io_mem_free,
> -	.io_mem_pfn = ttm_bo_default_io_mem_pfn,
> +	.io_mem_pfn = amdgpu_ttm_io_mem_pfn,
>  };
>
>  int amdgpu_ttm_init(struct amdgpu_device *adev)
>


-- 
Lerne, wie die Welt wirklich ist,
Aber vergiss niemals, wie sie sein sollte.
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx