Use kselftest wrapper to mark tests pass/fail instead of manually counting. This is needed to return correct exit status. This also improves readability and mainability.
Signed-off-by: Muhammad Usama Anjum usama.anjum@collabora.com --- tools/testing/selftests/x86/vdso_restorer.c | 20 +++++++------------- 1 file changed, 7 insertions(+), 13 deletions(-)
diff --git a/tools/testing/selftests/x86/vdso_restorer.c b/tools/testing/selftests/x86/vdso_restorer.c index fe99f24341554..8e173d71291f6 100644 --- a/tools/testing/selftests/x86/vdso_restorer.c +++ b/tools/testing/selftests/x86/vdso_restorer.c @@ -21,6 +21,7 @@ #include <unistd.h> #include <syscall.h> #include <sys/syscall.h> +#include "../kselftest.h"
/* Open-code this -- the headers are too messy to easily use them. */ struct real_sigaction { @@ -44,9 +45,10 @@ static void handler_without_siginfo(int sig)
int main() { - int nerrs = 0; struct real_sigaction sa;
+ ksft_print_header(); + void *vdso = dlopen("linux-vdso.so.1", RTLD_LAZY | RTLD_LOCAL | RTLD_NOLOAD); if (!vdso) @@ -57,6 +59,8 @@ int main() return 0; }
+ ksft_set_plan(2); + memset(&sa, 0, sizeof(sa)); sa.handler = handler_with_siginfo; sa.flags = SA_SIGINFO; @@ -69,12 +73,7 @@ int main()
raise(SIGUSR1);
- if (handler_called) { - printf("[OK]\tSA_SIGINFO handler returned successfully\n"); - } else { - printf("[FAIL]\tSA_SIGINFO handler was not called\n"); - nerrs++; - } + ksft_test_result(handler_called, "SA_SIGINFO handler returned\n");
printf("[RUN]\tRaise a signal, !SA_SIGINFO, sa.restorer == NULL\n");
@@ -86,10 +85,5 @@ int main()
raise(SIGUSR1);
- if (handler_called) { - printf("[OK]\t!SA_SIGINFO handler returned successfully\n"); - } else { - printf("[FAIL]\t!SA_SIGINFO handler was not called\n"); - nerrs++; - } + ksft_test_result(handler_called, "SA_SIGINFO handler returned\n"); }
Return correct exit status, KSFT_SKIP if the pre-conditions aren't met. Return KSFT_FAIL if error occurs. Use ksft_finished() which will compmare the total planned tests with passed tests to return the exit value.
Signed-off-by: Muhammad Usama Anjum usama.anjum@collabora.com --- tools/testing/selftests/x86/vdso_restorer.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/tools/testing/selftests/x86/vdso_restorer.c b/tools/testing/selftests/x86/vdso_restorer.c index 8e173d71291f6..54f33e8cda5cc 100644 --- a/tools/testing/selftests/x86/vdso_restorer.c +++ b/tools/testing/selftests/x86/vdso_restorer.c @@ -56,7 +56,7 @@ int main() RTLD_LAZY | RTLD_LOCAL | RTLD_NOLOAD); if (!vdso) { printf("[SKIP]\tFailed to find vDSO. Tests are not expected to work.\n"); - return 0; + return KSFT_SKIP; }
ksft_set_plan(2); @@ -69,7 +69,7 @@ int main() printf("[RUN]\tRaise a signal, SA_SIGINFO, sa.restorer == NULL\n");
if (syscall(SYS_rt_sigaction, SIGUSR1, &sa, NULL, 8) != 0) - err(1, "raw rt_sigaction syscall"); + err(KSFT_FAIL, "raw rt_sigaction syscall");
raise(SIGUSR1);
@@ -80,10 +80,12 @@ int main() sa.flags = 0; sa.handler = handler_without_siginfo; if (syscall(SYS_sigaction, SIGUSR1, &sa, 0) != 0) - err(1, "raw sigaction syscall"); + err(KSFT_FAIL, "raw sigaction syscall"); handler_called = 0;
raise(SIGUSR1);
ksft_test_result(handler_called, "SA_SIGINFO handler returned\n"); + + ksft_finished(); }
On 7/17/24 3:01 AM, Shuah Khan wrote:
Following results have been generated in the case when both tests fail:
# selftests: x86: vdso_restorer_32 # ERROR: ld.so: object '/usr/libexec/coreutils/libstdbuf.so' from LD_PRELOAD cannot be preloaded (wrong ELF class: ELFCLASS64): ignored. # [RUN] Raise a signal, SA_SIGINFO, sa.restorer == NULL # [FAIL] SA_SIGINFO handler was not called # [RUN] Raise a signal, !SA_SIGINFO, sa.restorer == NULL # [FAIL] !SA_SIGINFO handler was not called not ok 21 selftests: x86: vdso_restorer_32 # exit=2
# selftests: x86: vdso_restorer_32 # ERROR: ld.so: object '/usr/libexec/coreutils/libstdbuf.so' from LD_PRELOAD cannot be preloaded (wrong ELF class: ELFCLASS64): ignored. # TAP version 13 # 1..2 # [RUN] Raise a signal, SA_SIGINFO, sa.restorer == NULL # not ok 1 SA_SIGINFO handler returned # [RUN] Raise a signal, !SA_SIGINFO, sa.restorer == NULL # not ok 2 SA_SIGINFO handler returned # # Totals: pass:0 fail:2 xfail:0 xpass:0 skip:0 error:0 not ok 21 selftests: x86: vdso_restorer_32 # exit=1
Please let me know what you think?
thanks, -- Shuah
On 7/12/24 01:30, Muhammad Usama Anjum wrote:
Use kselftest wrapper to mark tests pass/fail instead of manually counting.
You care combining two changes in the patch.
This is needed to return correct exit status. This also
improves readability and mainability.
Spelling - "mainability" - checkpatch would have helped you catch this.
The change to return the correct error fine and but not the change thaT ADDS DUPLICATE tap header.
The problem with adding this header here is when make kselftest TARGETS=vDSO is run there will be duplicate TAP 13 headers.
thanks, -- Shuah
On 7/19/24 9:40 PM, Shuah Khan wrote:
Sorry I'll fix it after following discussion. I use checkpatch with spelling checker. I may have missed it for this patch.
Usually all TAP compliant tests print TAP 13 header at the start. These tests when run from make run_tests have duplicate TAP 13 headers. I don't think that this is the issue. Why do you think it is wrong?
For example, I've attached the logs of vDSO test suite. TAP header is printed at the start. Then it is printed again at the start of the test if it is TAP compliant e.g., vdso_test_abi and vdso_test_getrandom. These tests are already TAP compliant. Other tests in this suite aren't TAP compliant.
On 7/21/24 9:24 PM, Muhammad Usama Anjum wrote:
On CIs (make runtests or make kselftest) is used to run the tests. I'm not aware of the ancient history. AFAIU following is the format of messages (make kselftests). The TAP header mention that a new test has started. One test may have multiple sub-tests. For example:
TAP version 13 1..4 # selftests: vDSO: test1 # TAP version 13 1..5 ok 1 ok 2 ok 3 ok 4 ok 5 # # Totals: pass:5 fail:0 xfail:0 xpass:0 skip:0 error:0 ok 1 selftests: vDSO: test1 # selftests: vDSO: test2 # TAP version 13 1..5 ok 1 ok 2 not ok 3 # # Totals: pass:2 fail:1 xfail:0 xpass:0 skip:0 error:0 not ok 2 selftests: vDSO: test2 # exit=1 # selftests: vDSO: test3 ok 1 ok 2 ok 3 not ok 3 selftests: vDSO: test3 # selftests: vDSO: test4 ok 1 not ok 3 selftests: vDSO: test4
The test1 and test2 are TAP compliant and print header and footer of the tests mentioning total number of tests. The test3 and test4 don't print TAP header and footer. The boundary between test3 and test4 isn't that clear, but seems fine. Overall I would say TAP compliant tests have better boundry when they print header/footer and total number of tests.
Do you agree with above layout's current state because we have both TAP compliant and non-compliant tests.
On 7/21/24 10:37, Muhammad Usama Anjum wrote:
Usama,
There is a reason for not using nested TAP headers - this is what the outcome would be with the changes you are making. The reason went with the direction of adding TAP headers once for each test suite is to avoid nested TAP headers.
Unless that changes and we can stay backwards compatible with TAP 13, I am not going accept TAP headers for individual patches.
You can take it as a nack from me for all such patches.
thanks, -- Shuah
linux-kselftest-mirror@lists.linaro.org
-
Muhammad Usama Anjum -
Shuah Khan