On 02/09/2017 01:58 AM, Michel Dänzer wrote:
On 09/02/17 08:35 AM, Samuel Pitoiset wrote:
Like ttm_bo_validate(), ttm_bo_init() might need to move BO and
the number of bytes moved by TTM should be reported. This can help
the throttle buffer migration mechanism to make a better decision.
Signed-off-by: Samuel Pitoiset <[email protected]>
[...]
@@ -399,10 +400,15 @@ int amdgpu_bo_create_restricted(struct amdgpu_device
*adev,
locked = ww_mutex_trylock(&bo->tbo.ttm_resv.lock);
WARN_ON(!locked);
}
+
+ initial_bytes_moved = atomic64_read(&adev->num_bytes_moved);
r = ttm_bo_init(&adev->mman.bdev, &bo->tbo, size, type,
&bo->placement, page_align, !kernel, NULL,
acc_size, sg, resv ? resv : &bo->tbo.ttm_resv,
&amdgpu_ttm_bo_destroy);
+ amdgpu_cs_report_moved_bytes(adev,
+ initial_bytes_moved - atomic64_read(&adev->num_bytes_moved));
This looks backwards, should be
atomic64_read(&adev->num_bytes_moved) - initial_bytes_moved);
?
My mistake, thanks for noticing.
_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
