On 11/09/2012 01:10 PM, Akinobu Mita wrote:
It is unnecessary to disable preemption explicitly while calling copy_highpage(). Because copy_highpage() will do it again through kmap_atomic/kunmap_atomic.
Signed-off-by: Akinobu Mita akinobu.mita@gmail.com Cc: dri-devel@lists.freedesktop.org Cc: David Airlie airlied@linux.ie
drivers/gpu/drm/ttm/ttm_tt.c | 4 ---- 1 file changed, 4 deletions(-)
diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c index bf82601..7d759a4 100644 --- a/drivers/gpu/drm/ttm/ttm_tt.c +++ b/drivers/gpu/drm/ttm/ttm_tt.c @@ -308,9 +308,7 @@ int ttm_tt_swapin(struct ttm_tt *ttm) if (unlikely(to_page == NULL)) goto out_err;
copy_highpage(to_page, from_page);preempt_disable();
page_cache_release(from_page); }preempt_enable();
@@ -358,9 +356,7 @@ int ttm_tt_swapout(struct ttm_tt *ttm, struct file *persistent_swap_storage) ret = PTR_ERR(to_page); goto out_err; }
copy_highpage(to_page, from_page);preempt_disable();
set_page_dirty(to_page); mark_page_accessed(to_page); page_cache_release(to_page);preempt_enable();
Reviewed-by: Thomas Hellstrom thellstrom@vmware.com