Re: [PATCH] selftests: runner.sh: avoid overriding kselftest_cmd_args with empty variable

[Hangbin Liu <hangbin.liu@linux.dev>]

On Tue, Sep 01, 2026 at 08:07:22PM +0800, Qianheng Peng wrote:
> Hi Hangbin,
> On Tue, 1 Sep 2026 17:46:14, Hangbin Liu wrote:
> >Hi Qianheng
> >On Tue, Sep 01, 2026 at 10:44:52AM +0800, Qianheng Peng wrote:
> >> On Tue, 14 Jul 2026 14:34:43, Qianheng Peng wrote:
> >> >Normally the value of kselftest_cmd_args can be read from settings file,
> >> >but it will be overridden by KSELFTEST_TEST_KMOD_SH_ARGS which can be
> >> >empty if user did not give it a real value.So it should be better to
> >> >check if KSELFTEST_TEST_KMOD_SH_ARGS is emty before referring its
> >> >value to kselftest_cmd_args.
> >
> >Where does KSELFTEST_TEST_KMOD_SH_ARGS from? kmod testing? If yes,
> >can we fix it there?
> 
> Sorry for my confusing commit message and poor English.
> In fact the problem is not KSELFTEST_TEST_KMOD_SH_ARGS but empty
> kselftest_cmd_args_ref.
> If user does not export "KSELFTEST_${BASENAME_SANITIZED}_ARGS" (the
> ${BASENAME_SANITIZED} here can be anything like KMOD or RTCTEST) explicitly,
> kselftest_cmd_args_ref will be empty by default and override kselftest_cmd_args.
> So this patch will check whether kselftest_cmd_args_ref is empty before
> starting one test.

Oh, I see what you mean now.

	kselftest_cmd_args_ref="KSELFTEST_${BASENAME_SANITIZED}_ARGS"

This is used to retrieve parameters for a *specific* test. When a test does not
require extra parameters, `KSELFTEST_${BASENAME_SANITIZED}_ARGS` is not set.
This has the same effect as when a user exports it but does not assign a value.

Therefore, I do not think we need to add a check for this case. Users should
understand the semantics when exporting the `KSELFTEST_${BASENAME_SANITIZED}_ARGS`
variable.

Thanks
Hangbin