RE: [PATCH v3 2/5] scsi: fnic: Fix crash in fnic_wq_cmpl_handler when FDMI times out

[Karan Tilak Kumar (kartilak)]

On Thursday, June 12, 2025 1:39 AM, Dan Carpenter <dan.carpenter@xxxxxxxxxx> wrote:
>
> On Thu, Jun 12, 2025 at 10:42:30AM +0300, Dan Carpenter wrote:
> > On Wed, Jun 11, 2025 at 05:44:23PM -0700, Karan Tilak Kumar wrote:
> > > When both the RHBA and RPA FDMI requests time out, fnic reuses a frame
> > > to send ABTS for each of them. On send completion, this causes an
> > > attempt to free the same frame twice that leads to a crash.
> > >
> > > Fix crash by allocating separate frames for RHBA and RPA,
> > > and modify ABTS logic accordingly.
> > >
> > > Tested by checking MDS for FDMI information.
> > > Tested by using instrumented driver to:
> > > Drop PLOGI response
> > > Drop RHBA response
> > > Drop RPA response
> > > Drop RHBA and RPA response
> > > Drop PLOGI response + ABTS response
> > > Drop RHBA response + ABTS response
> > > Drop RPA response + ABTS response
> > > Drop RHBA and RPA response + ABTS response for both of them
> > >
> > > Fixes: 09c1e6ab4ab2 ("scsi: fnic: Add and integrate support for FDMI")
> > > Reviewed-by: Sesidhar Baddela <sebaddel@xxxxxxxxx>
> > > Reviewed-by: Arulprabhu Ponnusamy <arulponn@xxxxxxxxx>
> > > Reviewed-by: Gian Carlo Boffa <gcboffa@xxxxxxxxx>
> > > Tested-by: Arun Easi <aeasi@xxxxxxxxx>
> > > Co-developed-by: Arun Easi <aeasi@xxxxxxxxx>
> > > Signed-off-by: Arun Easi <aeasi@xxxxxxxxx>
> > > Tested-by: Karan Tilak Kumar <kartilak@xxxxxxxxx>
> > > Cc: <stable@xxxxxxxxxxxxxxx> # 6.14.x Please see patch description
> >
> > I'm a bit confused.  Why do we need to specify 6.14.x?  I would have
> > assumed that the Fixes tag was enough information.  What are we supposed
> > to see in the patch description?
> >
> > I suspect you're making this too complicated...  Just put
> > Cc: <stable@xxxxxxxxxxxxxxx> and a Fixes tag and let the scripts figure
> > it out.  Or put in the commit description, "The Fixes tag points to
> > an older kernel because XXX but really this should only be backported
> > to 6.14.x because YYY."
>
> But here even with the comment in the commit description, you would still
> just say:
>
> Cc: <stable@xxxxxxxxxxxxxxx> # 6.14.x
>
> The stable maintainers trust you to list the correct kernel and don't
> need to know the reasoning.

Thanks for clarifying this Dan. 

> I much prefer to keep it simple whenever possible.  We had bad CVE where
> someone left off the Fixes tag and instead specified
> "Cc: <stable@xxxxxxxxxxxxxxx> # 4.1" where 4.1 was the oldest supported
> kernel on kernel.org.  The patch should have been applied to the older
> vendor kernels but it wasn't because the the tag was wrong.
>

I understand and agree with you. I prefer to keep it simple as well.
In V4, as you suggested, I'll keep the fixes tag and the Cc: tag and
remove the comments section.

> regards,
> dan carpenter
>
>

Thanks for your review comments.

Regards,
Karan