drm/radeon: fix ring commit padding
authorChristian König <deathsimple@vodafone.de>
Sat, 7 Jul 2012 10:11:32 +0000 (12:11 +0200)
committerChristian König <deathsimple@vodafone.de>
Tue, 17 Jul 2012 08:32:08 +0000 (10:32 +0200)
We don't need to pad anything if the number of dwords
written to the ring already matches the requirements.

Fixes some "writting more dword to ring than expected"
warnings.

Signed-off-by: Christian König <deathsimple@vodafone.de>
Reviewed-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
drivers/gpu/drm/radeon/radeon_ring.c

index 0826e77..674aaba 100644 (file)
@@ -272,13 +272,8 @@ int radeon_ring_lock(struct radeon_device *rdev, struct radeon_ring *ring, unsig
 
 void radeon_ring_commit(struct radeon_device *rdev, struct radeon_ring *ring)
 {
-       unsigned count_dw_pad;
-       unsigned i;
-
        /* We pad to match fetch size */
-       count_dw_pad = (ring->align_mask + 1) -
-                      (ring->wptr & ring->align_mask);
-       for (i = 0; i < count_dw_pad; i++) {
+       while (ring->wptr & ring->align_mask) {
                radeon_ring_write(ring, ring->nop);
        }
        DRM_MEMORYBARRIER();