A regression was introduced in the 3.3 rc series, commit
"drm/ttm: simplify memory accounting for ttm user v2",
causing the metadata of buffer objects created using the ttm_bo_create()
function to be accounted twice.
That causes massive leaks with the vmwgfx driver running for example
SpecViewperf Catia-03 test 2, eventually killing the app.
Furthermore, the same commit introduces a regression where
metadata accounting is leaked if a buffer object is
initialized with an illegal size. This is also fixed with this commit.
Signed-off-by: Thomas Hellstrom <thellst...@vmware.com>
Cc: Jerome Glisse <jgli...@redhat.com>
Cc: sta...@vger.kernel.org
---
drivers/gpu/drm/ttm/ttm_bo.c | 5 +----
1 files changed, 1 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index b67cfca..4baea2f 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -1204,6 +1204,7 @@ int ttm_bo_init(struct ttm_bo_device *bdev,
(*destroy)(bo);
else
kfree(bo);
+ ttm_mem_global_free(mem_glob, acc_size);
return -EINVAL;
}
bo->destroy = destroy;
@@ -1312,10 +1313,6 @@ int ttm_bo_create(struct ttm_bo_device *bdev,
int ret;
acc_size = ttm_bo_acc_size(bdev, size, sizeof(struct
ttm_buffer_object));
- ret = ttm_mem_global_alloc(mem_glob, acc_size, false, false);
- if (unlikely(ret != 0))
- return ret;
-
bo = kzalloc(sizeof(*bo), GFP_KERNEL);
if (unlikely(bo == NULL)) {
--
1.7.7.5
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel
