netdev - Re: [PATCH net-next 01/10] net: ipa: Don't error out in .remove()

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <11f2d507-232c-493f-8efd-882f0330f7bd@ieee.org>
Date: Fri, 17 Nov 2023 13:43:21 -0600
From: Alex Elder <elder@...e.org>
To: Uwe Kleine-König <u.kleine-koenig@...gutronix.de>
Cc: Alex Elder <elder@...nel.org>, "David S. Miller" <davem@...emloft.net>,
 Eric Dumazet <edumazet@...gle.com>, Jakub Kicinski <kuba@...nel.org>,
 Paolo Abeni <pabeni@...hat.com>, netdev@...r.kernel.org,
 kernel@...gutronix.de
Subject: Re: [PATCH net-next 01/10] net: ipa: Don't error out in .remove()

On 11/17/23 8:45 AM, Uwe Kleine-König wrote:
>>> Fixes: cdf2e9419dd9 ("soc: qcom: ipa: main code")
>> Is this really a bug fix?  This code was doing the right
>> thing even if the caller was not.
> Yes, since cdf2e9419dd9 the driver is leaking resources if
> ipa_modem_stop() fails. I call that a bug.

I understand that.  But the alternative is that we free
those resources and allow the hardware to (eventually)
complete an in-flight operation that touches one of those
resources, which is a use-after-free (rather than a leak),
and I call that a bug too.

The function was returning an error to the caller to
indicate the request failed.  I'm comfortable with
accepting that and just issuing a warning and returning
no error.

The reason I wanted more time to review was that I want to
walk through that code path again and decide which of the
bugs I'd rather keep--and I think it would be the resource
leak.

It's also possible the cleanup function can preclude any
later completion from causing a problem, which would be
the best.  I just don't remember without looking at it
again closely.

					-Alex