[PATCH] Allocate bo objects to GTT always.

[Pauli Nieminen]

Memmove/memcpy to vram is very slow operation. To avoid the slow copy
allocate all buffer objects to GTT.

This fixed tearing/slow Xv with KMS.
---
 src/radeon_legacy_memory.c |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)

diff --git a/src/radeon_legacy_memory.c b/src/radeon_legacy_memory.c
index 02b95ed..840115f 100644
--- a/src/radeon_legacy_memory.c
+++ b/src/radeon_legacy_memory.c
@@ -25,7 +25,8 @@ radeon_legacy_allocate_memory(ScrnInfoPtr pScrn,
     if (info->cs) {
 	struct radeon_bo *video_bo;
 
-	video_bo = radeon_bo_open(info->bufmgr, 0, size, 4096, 0, 0);
+	video_bo = radeon_bo_open(info->bufmgr, 0, size, 4096,
+			RADEON_GEM_DOMAIN_GTT, 0);
 
 	*mem_struct = video_bo;
 
-- 
1.6.3.3