[Bug 268390] Kyua KTLS tests (*bad_minor, *bad_major) fail randomly
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 268390] Kyua KTLS tests (*bad_minor, *bad_major) fail randomly"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 268390] Kyua KTLS tests (*bad_minor, *bad_major) fail randomly"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 268390] Kyua KTLS tests (*bad_minor, *bad_major) fail randomly"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 268390] Kyua KTLS tests (*bad_minor, *bad_major) fail randomly"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 268390] Kyua KTLS tests (*bad_minor, *bad_major) fail randomly"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 268390] Kyua KTLS tests (*bad_minor, *bad_major) fail randomly"
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Thu, 15 Dec 2022 10:47:25 UTC
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=268390 Bug ID: 268390 Summary: Kyua KTLS tests (*bad_minor, *bad_major) fail randomly Product: Base System Version: CURRENT Hardware: i386 OS: Any Status: New Severity: Affects Some People Priority: --- Component: tests Assignee: testing@FreeBSD.org Reporter: michalx.gulbicki@intel.com Following KTLS test can fail randomly - no matter if any HW OCF backend is registered or not: ktls_test:ktls_receive_aes128_gcm_1_3_bad_major -> failed: /usr/src/tests/sys/kern/ktls_test.c:258: close(sv[1]) == 0 not met [0.022s] ktls_test:ktls_receive_aes128_gcm_1_3_bad_minor -> failed: /usr/src/tests/sys/kern/ktls_test.c:258: close(sv[1]) == 0 not met [0.018s] ktls_test:ktls_receive_aes128_gcm_1_3_bad_type -> failed: /usr/src/tests/sys/kern/ktls_test.c:258: close(sv[1]) == 0 not met [0.020s] ktls_test:ktls_receive_aes256_cbc_1_2_sha256_bad_minor -> failed: /usr/src/tests/sys/kern/ktls_test.c:258: close(sv[1]) == 0 not met [0.017s] ktls_test:ktls_receive_chacha20_poly1305_1_3_bad_major -> failed: /usr/src/tests/sys/kern/ktls_test.c:258: close(sv[1]) == 0 not met [0.017s] Adding a few ms delay between sending request to ktls socket and calling rcvmsg makes code much more robust (10+ consecutive runs without any radom fail). usleep has been added here: (ktls_test.c) diff --git a/tests/sys/kern/ktls_test.c b/tests/sys/kern/ktls_test.c index 09fb96ed11e..8a7652a6645 100644 --- a/tests/sys/kern/ktls_test.c +++ b/tests/sys/kern/ktls_test.c @@ -1506,6 +1506,7 @@ ktls_receive_tls_error(int fd, int expected_error) msg.msg_iov = &iov; msg.msg_iovlen = 1; + usleep(5 * 1000); // 5ms ATF_REQUIRE(recvmsg(fd, &msg, 0) == -1); if (expected_error != 0) ATF_REQUIRE(errno == expected_error); -- You are receiving this mail because: You are the assignee for the bug.