drm/ttm: Optimize reservation slightly
authorThomas Hellstrom <thellstrom@vmware.com>
Tue, 6 Nov 2012 11:31:51 +0000 (11:31 +0000)
committerDave Airlie <airlied@redhat.com>
Tue, 20 Nov 2012 06:15:06 +0000 (16:15 +1000)
commit6c1e963cc5771c93d4ed7aa8bdd4322a7c918e9b
tree435fec38d156d82a0f7497431712f39f59141919
parentcdad05216c2b2edfe92a9f87d6ae51aab277f3b2
drm/ttm: Optimize reservation slightly

Reservation locking currently always takes place under the LRU spinlock.
Hence, strictly there is no need for an atomic_cmpxchg call; we can use
atomic_read followed by atomic_write since nobody else will ever reserve
without the lru spinlock held.
At least on Intel this should remove a locked bus cycle on successful
reserve.

Note that thit commit may be obsoleted by the cross-device reservation work.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
drivers/gpu/drm/ttm/ttm_bo.c