linux-kernel - Re: [PATCH v2] wifi: nl80211: fix bounds checker error in nl80211_parse_sched

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: <87ses6x8j6.fsf@kernel.org>
Date: Tue, 05 Nov 2024 12:46:37 +0200
From: Kalle Valo <kvalo@...nel.org>
To: Jeff Johnson <quic_jjohnson@...cinc.com>
Cc: Aleksei Vetrov <vvvvvv@...gle.com>,  Johannes Berg
 <johannes@...solutions.net>,  Kees Cook <kees@...nel.org>,  "Gustavo A. R.
 Silva" <gustavoars@...nel.org>,  Dmitry Antipov <dmantipov@...dex.ru>,
  <linux-wireless@...r.kernel.org>,  <linux-kernel@...r.kernel.org>,
  <linux-hardening@...r.kernel.org>,  <stable@...r.kernel.org>
Subject: Re: [PATCH v2] wifi: nl80211: fix bounds checker error in
 nl80211_parse_sched_scan

Jeff Johnson <quic_jjohnson@...cinc.com> writes:

> On 10/29/2024 6:22 AM, Aleksei Vetrov wrote:
>> The channels array in the cfg80211_scan_request has a __counted_by
>> attribute attached to it, which points to the n_channels variable. This
>> attribute is used in bounds checking, and if it is not set before the
>> array is filled, then the bounds sanitizer will issue a warning or a
>> kernel panic if CONFIG_UBSAN_TRAP is set.
>> 
>> This patch sets the size of allocated memory as the initial value for
>> n_channels. It is updated with the actual number of added elements after
>> the array is filled.
>> 
>> Fixes: aa4ec06c455d ("wifi: cfg80211: use __counted_by where appropriate")
>> Cc: stable@...r.kernel.org
>> Signed-off-by: Aleksei Vetrov <vvvvvv@...gle.com>
> Reviewed-by: Jeff Johnson <quic_jjohnson@...cinc.com>
>
> And it is exactly this kind of issue why I'm not accepting any __counted_by()
> changes in ath.git without actually testing the code that is modified.

That's a good rule. If we ever manage to write that "wireless cleanup
policy" document this is something we should add there.

-- 
https://patchwork.kernel.org/project/linux-wireless/list/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches