Re: [PATCH] selftests: sched: Add more core schedule prctl calls

[Shuah Khan]

Hi Ivan,

On 3/22/23 14:37, Ivan Orlov wrote:
> The core sched kselftest makes prctl calls only with correct
> parameters. This patch will extend this test with more core
> schedule prctl calls with wrong parameters to increase code
> coverage.
>
> Signed-off-by: Ivan Orlov <ivan.orlov0322@xxxxxxxxx>

Thank you for the patch.

> ---
>   tools/testing/selftests/sched/cs_prctl_test.c | 8 ++++++++
>   1 file changed, 8 insertions(+)
>
> diff --git a/tools/testing/selftests/sched/cs_prctl_test.c b/tools/testing/selftests/sched/cs_prctl_test.c
> index 25e0d95d3713..df8e4629b3b5 100644
> --- a/tools/testing/selftests/sched/cs_prctl_test.c
> +++ b/tools/testing/selftests/sched/cs_prctl_test.c
> @@ -334,6 +334,14 @@ int main(int argc, char *argv[])
>   	validate(get_cs_cookie(pid) != 0);
>   	validate(get_cs_cookie(pid) == get_cs_cookie(procs[pidx].thr_tids[0]));
>   
> +	printf("\n## Try to pass wrong subcommand PR_SCHED_CORE_MAX = %d to prctl (shouldn't succeed)\n", PR_SCHED_CORE_MAX);

This message is unnecessary. validate() prints an error message.

> +	validate(_prctl(PR_SCHED_CORE, PR_SCHED_CORE_MAX, 0, PIDTYPE_PGID, 0) < 0
> +		&& errno == EINVAL);
> +
> +	printf("\n## Try to pass not-null uaddr with forbidden command (shouldn't succeed)\n");

Same here.

> +	validate(_prctl(PR_SCHED_CORE, PR_SCHED_CORE_SHARE_TO, 0, PIDTYPE_PGID, 1) < 0
> +		&& errno == EINVAL);
> +
>   	if (errors) {
>   		printf("TESTS FAILED. errors: %d\n", errors);
>   		res = 10;

thanks,
-- Shuah