[PATCH 1/2] drm/amdgpu: linear validate first then bind to GART

[Christian König]

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

For VM emulation for old UVD/VCE we need to validate the BO with linear
VRAM flag set first and then eventually bind it to GART.

Validating with linear VRAM flag set can move the BO to GART making
UVD/VCE read/write from an unbound GART BO.

Signed-off-by: Christian König <christian.koenig at amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 52dd78ee8fd0..0c52295e74d3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -1583,14 +1583,14 @@ int amdgpu_cs_find_mapping(struct amdgpu_cs_parser *parser,
 	if (READ_ONCE((*bo)->tbo.resv->lock.ctx) != &parser->ticket)
 		return -EINVAL;
 
-	r = amdgpu_ttm_bind(&(*bo)->tbo, &(*bo)->tbo.mem);
-	if (unlikely(r))
-		return r;
-
-	if ((*bo)->flags & AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS)
-		return 0;
+	if (!((*bo)->flags & AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS)) {
+		(*bo)->flags |= AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS;
+		amdgpu_ttm_placement_from_domain(*bo, (*bo)->allowed_domains);
+		r = ttm_bo_validate(&(*bo)->tbo, &(*bo)->placement, false,
+				    false);
+		if (r)
+			return r;
+	}
 
-	(*bo)->flags |= AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS;
-	amdgpu_ttm_placement_from_domain(*bo, (*bo)->allowed_domains);
-	return ttm_bo_validate(&(*bo)->tbo, &(*bo)->placement, false, false);
+	return amdgpu_ttm_bind(&(*bo)->tbo, &(*bo)->tbo.mem);
 }
-- 
2.11.0