I have a cunning plan to make the vma open/close lazy to cache frequent reallocations (as buffers are passed between applications, e.g. DRI). However, this will mean that we will not be immediately closing vma and so need to tell the kernel to process the idle handlers before checking for leaks. Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> Cc: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx> --- tests/gem_ppgtt.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/gem_ppgtt.c b/tests/gem_ppgtt.c index bed95db83..575b0e9d3 100644 --- a/tests/gem_ppgtt.c +++ b/tests/gem_ppgtt.c @@ -236,7 +236,7 @@ static void flink_and_close(void) gem_sync(fd2, flinked_bo); gem_close(fd2, flinked_bo); - igt_drop_caches_set(fd, DROP_RETIRE); + igt_drop_caches_set(fd, DROP_RETIRE | DROP_IDLE); /* the flinked bo VMA should have been cleared now, so a new bo of the * same size should get the same offset @@ -286,7 +286,7 @@ static void flink_and_exit(void) exec_and_get_offset(fd3, gem_create(fd3, 4096)); close(fd3); - igt_drop_caches_set(fd, DROP_ACTIVE | DROP_RETIRE); + igt_drop_caches_set(fd, DROP_ACTIVE | DROP_RETIRE | DROP_IDLE); igt_assert(!igt_debugfs_search(fd, "i915_gem_gtt", match)); close(fd); -- 2.17.0 _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx
