Due to a type mismatch that causes an implicit type conversion, the upper 32 bits of the GPU address have been zeroed out when adding to the command buffer.
Picked up by Coverity - CID 1198624.
Signed-off-by: Christoph Jaeger christophjaeger@linux.com --- drivers/gpu/drm/radeon/radeon_vce.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/radeon/radeon_vce.c b/drivers/gpu/drm/radeon/radeon_vce.c index 76e9904..ced53dd 100644 --- a/drivers/gpu/drm/radeon/radeon_vce.c +++ b/drivers/gpu/drm/radeon/radeon_vce.c @@ -613,7 +613,7 @@ void radeon_vce_fence_emit(struct radeon_device *rdev, struct radeon_fence *fence) { struct radeon_ring *ring = &rdev->ring[fence->ring]; - uint32_t addr = rdev->fence_drv[fence->ring].gpu_addr; + uint64_t addr = rdev->fence_drv[fence->ring].gpu_addr;
radeon_ring_write(ring, VCE_CMD_FENCE); radeon_ring_write(ring, addr);
-----Original Message----- From: Christoph Jaeger [mailto:christophjaeger@linux.com] Sent: Monday, April 14, 2014 6:10 PM To: Deucher, Alexander; Koenig, Christian; airlied@linux.ie Cc: dri-devel@lists.freedesktop.org; linux-kernel@vger.kernel.org; Christoph Jaeger Subject: [PATCH] drm/radeon: fix VCE fence command
Due to a type mismatch that causes an implicit type conversion, the upper 32 bits of the GPU address have been zeroed out when adding to the command buffer.
Picked up by Coverity - CID 1198624.
Signed-off-by: Christoph Jaeger christophjaeger@linux.com
Good catch!
Reviewed-by: Alex Deucher alexander.deucher@amd.com
drivers/gpu/drm/radeon/radeon_vce.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/radeon/radeon_vce.c b/drivers/gpu/drm/radeon/radeon_vce.c index 76e9904..ced53dd 100644 --- a/drivers/gpu/drm/radeon/radeon_vce.c +++ b/drivers/gpu/drm/radeon/radeon_vce.c @@ -613,7 +613,7 @@ void radeon_vce_fence_emit(struct radeon_device *rdev, struct radeon_fence *fence) { struct radeon_ring *ring = &rdev->ring[fence->ring];
- uint32_t addr = rdev->fence_drv[fence->ring].gpu_addr;
uint64_t addr = rdev->fence_drv[fence->ring].gpu_addr;
radeon_ring_write(ring, VCE_CMD_FENCE); radeon_ring_write(ring, addr);
-- 1.9.0
Am 15.04.2014 00:13, schrieb Deucher, Alexander:
-----Original Message----- From: Christoph Jaeger [mailto:christophjaeger@linux.com] Sent: Monday, April 14, 2014 6:10 PM To: Deucher, Alexander; Koenig, Christian; airlied@linux.ie Cc: dri-devel@lists.freedesktop.org; linux-kernel@vger.kernel.org; Christoph Jaeger Subject: [PATCH] drm/radeon: fix VCE fence command
Due to a type mismatch that causes an implicit type conversion, the upper 32 bits of the GPU address have been zeroed out when adding to the command buffer.
Picked up by Coverity - CID 1198624.
Signed-off-by: Christoph Jaeger christophjaeger@linux.com
Good catch!
Indeed.
Reviewed-by: Alex Deucher alexander.deucher@amd.com
Added to my 3.15 fixes queue.
Thanks, Christian.
drivers/gpu/drm/radeon/radeon_vce.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/radeon/radeon_vce.c b/drivers/gpu/drm/radeon/radeon_vce.c index 76e9904..ced53dd 100644 --- a/drivers/gpu/drm/radeon/radeon_vce.c +++ b/drivers/gpu/drm/radeon/radeon_vce.c @@ -613,7 +613,7 @@ void radeon_vce_fence_emit(struct radeon_device *rdev, struct radeon_fence *fence) { struct radeon_ring *ring = &rdev->ring[fence->ring];
- uint32_t addr = rdev->fence_drv[fence->ring].gpu_addr;
uint64_t addr = rdev->fence_drv[fence->ring].gpu_addr;
radeon_ring_write(ring, VCE_CMD_FENCE); radeon_ring_write(ring, addr);
-- 1.9.0
dri-devel@lists.freedesktop.org
-
Christian König -
Christoph Jaeger -
Deucher, Alexander