iommufd/selftest: Update the fail_nth limit

There are more failure conditions now so 400 iterations is not enough pass
them all, up it to 1000. The limit exists so it doesn't infinite loop.

Link: https://patch.msgid.link/r/3-v1-02cd136829df+31-iommufd_syz_fput_jgg@nvidia.com
Reviewed-by: Nicolin Chen <nicolinc@nvidia.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

diff --git a/tools/testing/selftests/iommu/iommufd_fail_nth.c b/tools/testing/selftests/iommu/iommufd_fail_nth.c
index 651fc9f13c0810ba261ae6cc4681cb8fa89d4833..45c14323a6183c7887b63e03ed0e59240bc39db9 100644 (file)
--- a/tools/testing/selftests/iommu/iommufd_fail_nth.c
+++ b/tools/testing/selftests/iommu/iommufd_fail_nth.c
@@ -113,7 +113,7 @@ static bool fail_nth_next(struct __test_metadata *_metadata,
         * necessarily mean a test failure, just that the limit has to be made
         * bigger.
         */
-       ASSERT_GT(400, nth_state->iteration);
+       ASSERT_GT(1000, nth_state->iteration);
        if (nth_state->iteration != 0) {
                ssize_t res;
                ssize_t res2;