| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <CAAkQn5JVPnN_dowQAjexom9O+2vThSOiNiY4woCgRPwGNNmt7w@mail.gmail.com>
Date: Thu, 31 Jul 2025 16:01:04 -0400
From: Marie Zhussupova <marievic@...gle.com>
To: Dan Carpenter <dan.carpenter@...aro.org>
Cc: oe-kbuild@...ts.linux.dev, rmoar@...gle.com, davidgow@...gle.com,
shuah@...nel.org, brendan.higgins@...ux.dev, lkp@...el.com,
oe-kbuild-all@...ts.linux.dev, elver@...gle.com, dvyukov@...gle.com,
lucas.demarchi@...el.com, thomas.hellstrom@...ux.intel.com,
rodrigo.vivi@...el.com, linux-kselftest@...r.kernel.org,
kunit-dev@...glegroups.com, kasan-dev@...glegroups.com,
intel-xe@...ts.freedesktop.org, dri-devel@...ts.freedesktop.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH 6/9] kunit: Enable direct registration of parameter arrays
to a KUnit test
On Thu, Jul 31, 2025 at 11:58 AM Dan Carpenter <dan.carpenter@...aro.org> wrote:
>
> Hi Marie,
>
> kernel test robot noticed the following build warnings:
>
> https://git-scm.com/docs/git-format-patch#_base_tree_information]
>
> url: https://github.com/intel-lab-lkp/linux/commits/Marie-Zhussupova/kunit-Add-parent-kunit-for-parameterized-test-context/20250730-033818
> base: https://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest.git kunit
> patch link: https://lore.kernel.org/r/20250729193647.3410634-7-marievic%40google.com
> patch subject: [PATCH 6/9] kunit: Enable direct registration of parameter arrays to a KUnit test
> config: nios2-randconfig-r072-20250731 (https://download.01.org/0day-ci/archive/20250731/202507310854.pZvIcswn-lkp@intel.com/config)
> compiler: nios2-linux-gcc (GCC) 8.5.0
>
> If you fix the issue in a separate patch/commit (i.e. not just a new version of
> the same patch/commit), kindly add following tags
> | Reported-by: kernel test robot <lkp@...el.com>
> | Reported-by: Dan Carpenter <dan.carpenter@...aro.org>
> | Closes: https://lore.kernel.org/r/202507310854.pZvIcswn-lkp@intel.com/
>
> New smatch warnings:
> lib/kunit/test.c:723 kunit_run_tests() error: we previously assumed 'test_case->generate_params' could be null (see line 714)
>
> vim +723 lib/kunit/test.c
>
> 914cc63eea6fbe Brendan Higgins 2019-09-23 681 int kunit_run_tests(struct kunit_suite *suite)
> 914cc63eea6fbe Brendan Higgins 2019-09-23 682 {
> fadb08e7c7501e Arpitha Raghunandan 2020-11-16 683 char param_desc[KUNIT_PARAM_DESC_SIZE];
> 914cc63eea6fbe Brendan Higgins 2019-09-23 684 struct kunit_case *test_case;
> acd8e8407b8fcc David Gow 2021-08-03 685 struct kunit_result_stats suite_stats = { 0 };
> acd8e8407b8fcc David Gow 2021-08-03 686 struct kunit_result_stats total_stats = { 0 };
> 8631cd2cf5fbf2 Marie Zhussupova 2025-07-29 687 const void *curr_param;
> 914cc63eea6fbe Brendan Higgins 2019-09-23 688
> c272612cb4a2f7 David Gow 2022-07-01 689 /* Taint the kernel so we know we've run tests. */
> c272612cb4a2f7 David Gow 2022-07-01 690 add_taint(TAINT_TEST, LOCKDEP_STILL_OK);
> c272612cb4a2f7 David Gow 2022-07-01 691
> 1cdba21db2ca31 Daniel Latypov 2022-04-29 692 if (suite->suite_init) {
> 1cdba21db2ca31 Daniel Latypov 2022-04-29 693 suite->suite_init_err = suite->suite_init(suite);
> 1cdba21db2ca31 Daniel Latypov 2022-04-29 694 if (suite->suite_init_err) {
> 1cdba21db2ca31 Daniel Latypov 2022-04-29 695 kunit_err(suite, KUNIT_SUBTEST_INDENT
> 1cdba21db2ca31 Daniel Latypov 2022-04-29 696 "# failed to initialize (%d)", suite->suite_init_err);
> 1cdba21db2ca31 Daniel Latypov 2022-04-29 697 goto suite_end;
> 1cdba21db2ca31 Daniel Latypov 2022-04-29 698 }
> 1cdba21db2ca31 Daniel Latypov 2022-04-29 699 }
> 1cdba21db2ca31 Daniel Latypov 2022-04-29 700
> cae56e1740f559 Daniel Latypov 2022-04-29 701 kunit_print_suite_start(suite);
> 914cc63eea6fbe Brendan Higgins 2019-09-23 702
> fadb08e7c7501e Arpitha Raghunandan 2020-11-16 703 kunit_suite_for_each_test_case(suite, test_case) {
> fadb08e7c7501e Arpitha Raghunandan 2020-11-16 704 struct kunit test = { .param_value = NULL, .param_index = 0 };
> acd8e8407b8fcc David Gow 2021-08-03 705 struct kunit_result_stats param_stats = { 0 };
> fadb08e7c7501e Arpitha Raghunandan 2020-11-16 706
> 887d85a0736ff3 Rae Moar 2023-03-08 707 kunit_init_test(&test, test_case->name, test_case->log);
> 03806177fa4cbb Marie Zhussupova 2025-07-29 708 __kunit_init_parent_test(test_case, &test);
> 03806177fa4cbb Marie Zhussupova 2025-07-29 709
> 529534e8cba3e6 Rae Moar 2023-07-25 710 if (test_case->status == KUNIT_SKIPPED) {
> 529534e8cba3e6 Rae Moar 2023-07-25 711 /* Test marked as skip */
> 529534e8cba3e6 Rae Moar 2023-07-25 712 test.status = KUNIT_SKIPPED;
> 529534e8cba3e6 Rae Moar 2023-07-25 713 kunit_update_stats(¶m_stats, test.status);
> 44c50ed8e59936 Marie Zhussupova 2025-07-29 @714 } else if (!test_case->generate_params && !test.params_data.params) {
> ^^^^^^^^^^^^^^^^^^^^^^^^^^
> Imagine ->generate_parms is NULL but test.params_data.params is
> non-NULL.
>
> 37dbb4c7c7442d David Gow 2021-11-02 715 /* Non-parameterised test. */
> 529534e8cba3e6 Rae Moar 2023-07-25 716 test_case->status = KUNIT_SKIPPED;
> 37dbb4c7c7442d David Gow 2021-11-02 717 kunit_run_case_catch_errors(suite, test_case, &test);
> 37dbb4c7c7442d David Gow 2021-11-02 718 kunit_update_stats(¶m_stats, test.status);
> 03806177fa4cbb Marie Zhussupova 2025-07-29 719 } else if (test_case->status != KUNIT_FAILURE) {
> fadb08e7c7501e Arpitha Raghunandan 2020-11-16 720 /* Get initial param. */
> fadb08e7c7501e Arpitha Raghunandan 2020-11-16 721 param_desc[0] = '\0';
> 8631cd2cf5fbf2 Marie Zhussupova 2025-07-29 722 /* TODO: Make generate_params try-catch */
> 13ee0c64bd88a3 Marie Zhussupova 2025-07-29 @723 curr_param = test_case->generate_params(&test, NULL, param_desc);
> ^^^^^^^^^^^^^^^^^^^^^^^^^^
> Then this could crash.
>
> I suspect that this is fine, but I bet that in the previous
> condition, just testing one would probably have been sufficient
> or maybe we could have change && to ||.
Hello Dan,
My apologies for the HTML version of this email earlier. Here is the
plain text version.
As of now, test.params_data.params can only be populated in a param_init
function, which can only be used if we register the test case with a
KUNIT_CASE_PARAM_WITH_INIT macro. That macro auto populates
test_case->generate_params with a function called
kunit_get_next_param_and_desc()
(which iterates over the parameter array) if the test user didn't provide their
own generator function. So, there shouldn't be a case where
test_case->generate_params is NULL but test.params_data.params is NON-NULL.
However, to be robust, we could add a NULL check before calling
test_case->generate_params on line 723.
Thank you!
-Marie
>
> 529534e8cba3e6 Rae Moar 2023-07-25 724 test_case->status = KUNIT_SKIPPED;
> 6c738b52316c58 Rae Moar 2022-11-23 725 kunit_log(KERN_INFO, &test, KUNIT_SUBTEST_INDENT KUNIT_SUBTEST_INDENT
> 6c738b52316c58 Rae Moar 2022-11-23 726 "KTAP version 1\n");
> 44b7da5fcd4c99 David Gow 2021-11-02 727 kunit_log(KERN_INFO, &test, KUNIT_SUBTEST_INDENT KUNIT_SUBTEST_INDENT
> 44b7da5fcd4c99 David Gow 2021-11-02 728 "# Subtest: %s", test_case->name);
> fadb08e7c7501e Arpitha Raghunandan 2020-11-16 729
> 8631cd2cf5fbf2 Marie Zhussupova 2025-07-29 730 while (curr_param) {
>
> --
> 0-DAY CI Kernel Test Service
> https://github.com/intel/lkp-tests/wiki
>
Powered by blists - more mailing lists