We multiply the memfd 64k to create a 2G arena which we then attempt to write into after marking read-only. Howver, when it comes to unlock the arena after the test, performance tanks as the kernel tries to resolve the 64k repeated mappings onto the same set of pages. (Must not be a very common operation!) We can get away with just mlocking the backing store to prevent its eviction, which should prevent the arena mapping from being freed as well. References: https://bugs.freedesktop.org/show_bug.cgi?id=108887 Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> --- tests/i915/gem_userptr_blits.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tests/i915/gem_userptr_blits.c b/tests/i915/gem_userptr_blits.c index 909dd19df..c4e60ba92 100644 --- a/tests/i915/gem_userptr_blits.c +++ b/tests/i915/gem_userptr_blits.c @@ -1124,7 +1124,7 @@ static void test_readonly(int i915) *(uint32_t *)(space + offset) = offset; } igt_assert_eq_u32(*(uint32_t *)pages, (uint32_t)(total - sz)); - igt_assert(mlock(space, total) == 0); + igt_assert(mlock(pages, sz) == 0); close(memfd); /* Check we can create a normal userptr bo wrapping the wrapper */ @@ -1176,6 +1176,7 @@ static void test_readonly(int i915) } igt_waitchildren(); + munlock(pages, sz); munmap(space, total); munmap(pages, sz); } -- 2.20.1 _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx
