Re: [PATCH v5 11/27] scsi-multipath: introduce basic SCSI device support

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] scsi_mpath_unique_lun_id incorrectly accepts a 0-length LUN ID as a valid unique identifier.
--

--- Patch [11]: [PATCH v5 11/27] scsi-multipath: introduce basic SCSI device support ---
commit b27d35f732d009c9f3684249b66e5403b7b120c1
Author: John Garry <john.g.garry@oracle.com>

scsi-multipath: introduce basic SCSI device support

This commit introduces the scsi_mpath_device structure to hold
multipath-specific details for SCSI devices. It also adds a module
parameter to control multipath modes (on, off, always).

> diff --git a/drivers/scsi/scsi_multipath.c b/drivers/scsi/scsi_multipath.c
> new file mode 100644
> index 0000000000000..947f3e2103e6b
> --- /dev/null
> +++ b/drivers/scsi/scsi_multipath.c

[ ... ]

> +static int scsi_mpath_unique_lun_id(struct scsi_device *sdev)
> +{
> +	struct scsi_mpath_device *scsi_mpath_dev = sdev->scsi_mpath_dev;
> +	int ret;
> +
> +	ret = scsi_vpd_lun_id(sdev, scsi_mpath_dev->device_id_str,
> +				SCSI_MPATH_DEVICE_ID_LEN);
> +	if (ret < 0)
> +		return ret;

[Severity: High]
If a SCSI device returns a 0-length designator in VPD page 83,
scsi_vpd_lun_id() will return 0. Does checking only for (ret < 0)
incorrectly accept a 0-length LUN ID as a valid unique identifier?

If multiple devices lack a unique ID and return 0, could they all be
assigned an empty string and incorrectly aggregated into the same
multipath group? This might risk routing I/Os intended for one device 
to a completely different device.

> +
> +	return 0;
> +}
> +

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260727144234.3802948-1-john.g.garry@oracle.com?part=11