Re: [PATCH v2 07/40] scsi: core: use 16-bits defined sense codes

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] Duplicate KUNIT_EXPECT_EQ execution in scsi_lib_test_multiple_sense
--

commit 7abd3d2760cfc6366060acbd670a7ba4cba45ea2
Author: Damien Le Moal <dlemoal@kernel.org>

scsi: core: use 16-bits defined sense codes
    
Refactor the SCSI core code to use the 16-bits sense_code field of
struct scsi_sense_hdr and struct scsi_failure. This replaces hard-coded
additional sense codes with enum values to improve code clarity and
testing.

> diff --git a/drivers/scsi/scsi_lib_test.c b/drivers/scsi/scsi_lib_test.c
> --- a/drivers/scsi/scsi_lib_test.c
> +++ b/drivers/scsi/scsi_lib_test.c
[ ... ]
> @@ -72,32 +73,37 @@ static void scsi_lib_test_multiple_sense(struct kunit *test)
[ ... ]
>  	/* Match using SCMD_FAILURE_ASC_ANY */
> -	scsi_build_sense(&sc, 0, HARDWARE_ERROR, 0x11, 0x22);
> +	scsi_set_sense(&sc, 0, HARDWARE_ERROR,
> +		       scsi_sense_code(ASC_UNRECOVERED_READ_ERROR, 0x22));
> +	KUNIT_EXPECT_EQ(test, -EAGAIN, scsi_check_passthrough(&sc, &failures));
>  	KUNIT_EXPECT_EQ(test, -EAGAIN, scsi_check_passthrough(&sc, &failures));

[Severity: Low]
Is this duplicated KUNIT_EXPECT_EQ call intentional?

Since scsi_check_passthrough() modifies internal state by incrementing the
retry counter in the failures array, calling it twice for the exact same
mock failure consumes two retries instead of one.

The test still passes because the retry limit isn't exceeded, but does
this unintentionally mask future issues by using up the retry budget?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260903034201.112211-1-dlemoal@kernel.org?part=7