[Intel-gfx] [PATCH igt 1/2] igt/gem_busy: Remove repeated use of igt_spin_batch_new

Chris Wilson chris at chris-wilson.co.uk
Sat Dec 23 16:14:20 UTC 2017


igt_spin_batch_new() includes a throttling check that GEM works, which
breaks trying to create multiple spin batches, use
__igt_spin_batch_new() instead, after verifying GEM works.

Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
---
 tests/gem_busy.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/tests/gem_busy.c b/tests/gem_busy.c
index 4ba232411..c349c2919 100644
--- a/tests/gem_busy.c
+++ b/tests/gem_busy.c
@@ -459,17 +459,17 @@ static void close_race(int fd)
 		igt_assert(sched_setscheduler(getpid(), SCHED_RR, &rt) == 0);
 
 		for (i = 0; i < nhandles; i++) {
-			spin[i] = igt_spin_batch_new(fd, 0,
-						     engines[rand() % nengine], 0);
+			spin[i] = __igt_spin_batch_new(fd, 0,
+						       engines[rand() % nengine], 0);
 			handles[i] = spin[i]->handle;
 		}
 
 		igt_until_timeout(20) {
 			for (i = 0; i < nhandles; i++) {
 				igt_spin_batch_free(fd, spin[i]);
-				spin[i] = igt_spin_batch_new(fd, 0,
-							     engines[rand() % nengine],
-							     0);
+				spin[i] = __igt_spin_batch_new(fd, 0,
+							       engines[rand() % nengine],
+							       0);
 				handles[i] = spin[i]->handle;
 				__sync_synchronize();
 			}
-- 
2.15.1



More information about the Intel-gfx mailing list