linux-kernel - Re: [PATCH] media:cec:fix double free and uaf issue when cancel data during noblocking

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] [day] [month] [year] [list]

Message-ID: <ccedda97-ed68-9453-30d3-24506d771b44@xs4all.nl>
Date:   Thu, 19 Jan 2023 08:46:39 +0100
From:   Hans Verkuil <hverkuil-cisco@...all.nl>
To:     Xinghui Li <korantwork@...il.com>, mchehab@...nel.org
Cc:     linux-kernel@...r.kernel.org, linux-media@...r.kernel.org,
        Xinghui Li <korantli@...cent.com>, loydlv <loydlv@...cent.com>
Subject: Re: [PATCH] media:cec:fix double free and uaf issue when cancel data
 during noblocking

On 19/01/2023 05:49, Xinghui Li wrote:
> 在 2023/1/18 18:18，“Hans Verkuil”<hverkuil-cisco@...all.nl <mailto:hverkuil-cisco@...all.nl>> 写入:
> 
>> ...while this free is called if data->blocking is true. (see the 'if (!block) return 0;'
>> further up).
> Do you mean this code?
> 
> 	/* All done if we don't need to block waiting for completion */
> 	if (!block)
> 		return 0;

Yes.

> I notice this part code. But I'm not sure if 'block' will be modified in other sync operations. 
> So I sent this patch for community to review.

It's not modified anywhere else.

> 
>> So I have my doubts if this patch actually addresses the correct issue.
>> Do you have an actual debug trace of the UAF? Or even better, code to reproduce
>> this issue.
> 
> And we found this issue by the code scanning tool developed by loydlv and filtered from 200 issue by human.
> So it could be the none-issue. If so, I hope I didn't waste too much of your time.

I'll reject this patch since I believe this to be a false report. For future reference:
if a patch is based on code scanning tools then it's good to mention that in the commit
log. I wasn't aware that 'loydlv' is such a tool.

Regards,

	Hans