Linux SCSI subsystem development
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Weibin Liu" <liuwb@xiaopeng.com>, linux-scsi@vger.kernel.org
Subject: Re: [PATCH] scsi: mpt3sas: validate data_out size before SMP link-reset check
Date: Thu, 15 Oct 2026 02:22:32 +0000	[thread overview]
Message-ID: <sashiko-outbox-172142@kernel.org> (raw)
In-Reply-To: <20261015-mpt3sas-dataout-sz-v1-1-0781d24bb7ad@xiaopeng.com>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] scsi: mpt3sas: UAPI regression breaking short SMP requests
--

commit c6e5c105f2c358ce97748e66aceca72796ee77b8
Author: Weibin Liu <liuwb@xiaopeng.com>

scsi: mpt3sas: validate data_out size before SMP link-reset check

This commit prevents an out-of-bounds read in _ctl_do_mpt_command() by
validating the size of the data_out buffer before checking for SMP link
reset requests on non-IMMEDIATE SMP PASSTHROUGH commands.

> diff --git a/drivers/scsi/mpt3sas/mpt3sas_ctl.c b/drivers/scsi/mpt3sas/mpt3sas_ctl.c
> index 8bb947004885b..ae9b2be51ce0f 100644
> --- a/drivers/scsi/mpt3sas/mpt3sas_ctl.c
> +++ b/drivers/scsi/mpt3sas/mpt3sas_ctl.c
> @@ -968,7 +968,7 @@ _ctl_do_mpt_command(struct MPT3SAS_ADAPTER *ioc, struct mpt3_ioctl_command karg,
>  		    MPI2_SMP_PT_REQ_PT_FLAGS_IMMEDIATE)
>  			data = (u8 *)&smp_request->SGL;
>  		else {
> -			if (unlikely(data_out == NULL)) {
> +			if (unlikely(!data_out || data_out_sz < 11)) {

[Severity: High]
Does this cause a regression for valid short SMP requests?

By enforcing data_out_sz >= 11 and returning -EINVAL on shorter payloads,
this change appears to prevent user-space applications (such as smp_utils
or lsiutil) from sending standard short SMP commands (like
SMP_REPORT_GENERAL, which has an 8-byte payload) via the MPT3COMMAND ioctl
using a non-immediate payload buffer.

Before this patch, short commands were successfully sent to the hardware,
even though the driver incorrectly read data[10] out of bounds.

Could this be resolved by instead checking the payload size only when
evaluating the link reset detection logic later in the function?

For example:
	if (data_out_sz >= 11 && data[1] == 0x91 && (data[10] == 1 || ...))

>  				pr_err("failure at %s:%d/%s()!\n",
>  				    __FILE__, __LINE__, __func__);
>  				mpt3sas_base_free_smid(ioc, smid);

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261015-mpt3sas-dataout-sz-v1-1-0781d24bb7ad@xiaopeng.com?part=1

      reply	other threads:[~2026-10-15  2:22 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-15  2:07 [PATCH] scsi: mpt3sas: validate data_out size before SMP link-reset check Weibin Liu
2026-10-15  2:22 ` sashiko-bot [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=sashiko-outbox-172142@kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=linux-scsi@vger.kernel.org \
    --cc=liuwb@xiaopeng.com \
    --cc=sashiko-reviews@lists.linux.dev \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox