Re: [PATCH 04/37] scsi: rename sense field of struct scsi_failure

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] The unit test `scsi_lib_test_any_sense()` is broken because `.result` is incorrectly initialized to `SCMD_FAILURE_RESULT_ANY`, completely bypassing the sense key check.
--

commit be75fcae059f5a6c0fee14ecc0a6ae4e8a66d23d
Author: Damien Le Moal <dlemoal@kernel.org>

scsi: rename sense field of struct scsi_failure

This commit renames the sense field of struct scsi_failure to sense_key
to make its purpose clearer and align it with struct scsi_sense_hdr. It
also renames the associated SCMD_FAILURE_SENSE_ANY macro and updates
initializations in tests.

> diff --git a/drivers/scsi/scsi_lib_test.c b/drivers/scsi/scsi_lib_test.c
> index ae8af0e0047a8..4558dc853e26e 100644
> --- a/drivers/scsi/scsi_lib_test.c
> +++ b/drivers/scsi/scsi_lib_test.c
[ ... ]
> @@ -116,8 +116,9 @@ static void scsi_lib_test_any_sense(struct kunit *test)
>  {
>  	struct scsi_failure any_sense_failure_defs[] = {
>  		{
> -			.result = SCMD_FAILURE_SENSE_ANY,
> +			.sense_key = SCMD_FAILURE_SENSE_KEY_ANY,
>  			.allowed = SCSI_LIB_TEST_MAX_ALLOWED,
> +			.result = SCMD_FAILURE_RESULT_ANY,
>  		},
>  		{}
>  	};

[Severity: Medium]
Does setting .result to SCMD_FAILURE_RESULT_ANY here cause the test to
bypass the sense key checking logic completely?

Looking at scsi_check_passthrough() in drivers/scsi/scsi_lib.c, the check
short-circuits when .result is SCMD_FAILURE_RESULT_ANY:

drivers/scsi/scsi_lib.c:scsi_check_passthrough() {
	...
	for (failure = failures->failure_definitions; failure->result;
	     failure++) {
		if (failure->result == SCMD_FAILURE_RESULT_ANY)
			goto maybe_retry;
	...
}

Because of this short-circuit, the sense key check logic is never actually
evaluated. To properly test SCMD_FAILURE_SENSE_KEY_ANY, shouldn't .result be
initialized to SAM_STAT_CHECK_CONDITION instead?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260831020451.585944-1-dlemoal@kernel.org?part=4