linux-kernel - Re: [PATCH] wifi: cfg80211: Prevent comparison with invalid registered dev scan req

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: <20250623082626.3104528-1-lizhi.xu@windriver.com>
Date: Mon, 23 Jun 2025 16:26:25 +0800
From: Lizhi Xu <lizhi.xu@...driver.com>
To: <johannes@...solutions.net>
CC: <davem@...emloft.net>, <edumazet@...gle.com>, <kuba@...nel.org>,
        <linux-kernel@...r.kernel.org>, <linux-wireless@...r.kernel.org>,
        <lizhi.xu@...driver.com>, <netdev@...r.kernel.org>,
        <pabeni@...hat.com>,
        <syzbot+189dcafc06865d38178d@...kaller.appspotmail.com>,
        <syzkaller-bugs@...glegroups.com>
Subject: Re: [PATCH] wifi: cfg80211: Prevent comparison with invalid registered dev scan req

On Fri, 20 Jun 2025 13:01:51 +0200, Johannes Berg wrote:
> > The scan req of a registered device may have been released, so it should
> > be checked to be valid before comparing it with the current req.
> >
> 
> I don't understand the subject/commit log at all. You're now accepting
> scan_done() with a NULL scan request, why does that make sense?
It is meaningless to compare the registered device with NULL scan_req with
the current scan request.

Because there is a check for scan_req being NULL in ___cfg80211_scan_done(),
cfg80211_scan_done() is not directly exited when the scan_req of the registered
device is NULL.

Lizhi