Re: [PATCH v2 44/60] scsi: qla2xxx: Adjust feature gating in BSG paths for 29xx support

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] D-port diagnostics remain broken for QLA29xx due to a missing downstream feature check.
--

commit 8ffa642602cccd0180ae4aedb7562a144e25f71e
Author: Manish Rangankar <mrangankar@marvell.com>

scsi: qla2xxx: Adjust feature gating in BSG paths for 29xx support

This commit extends qla2xxx BSG command handling to recognize QLA29xx
adapters and align feature availability with hardware capabilities.
It updates several feature checks to allow QLA29xx in paths previously
restricted to QLA27xx/28xx.

> diff --git a/drivers/scsi/qla2xxx/qla_bsg.c b/drivers/scsi/qla2xxx/qla_bsg.c
> index 00e980f0cd782..7f4558beee2c3 100644
> --- a/drivers/scsi/qla2xxx/qla_bsg.c
> +++ b/drivers/scsi/qla2xxx/qla_bsg.c
> @@ -2741,7 +2741,7 @@ qla2x00_do_dport_diagnostics(struct bsg_job *bsg_job)
>  	struct qla_dport_diag *dd;
>  
>  	if (!IS_QLA83XX(vha->hw) && !IS_QLA27XX(vha->hw) &&
> -	    !IS_QLA28XX(vha->hw))
> +	    !IS_QLA28XX(vha->hw) && !IS_QLA29XX(vha->hw))

[Severity: Medium]
Does this change miss an update to the underlying mailbox execution function?

While qla2x00_do_dport_diagnostics() now allows QLA29xx adapters to proceed,
the lower-level function it calls, qla26xx_dport_diagnostics() in qla_mbx.c,
still appears to have the old hardware check:

drivers/scsi/qla2xxx/qla_mbx.c:qla26xx_dport_diagnostics() {
	...
	if (!IS_QLA83XX(vha->hw) && !IS_QLA27XX(vha->hw) &&
	    !IS_QLA28XX(vha->hw))
		return QLA_FUNCTION_FAILED;
	...
}

Will this cause the diagnostic command to unconditionally fail with
EXT_STATUS_MAILBOX for QLA29xx adapters despite the wrapper permitting it?

>  		return -EPERM;
>  
>  	dd = kmalloc_obj(*dd);

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260612095333.1666592-1-njavali@marvell.com?part=44