linux-kernel - Re: [PATCH] driver/android/binderfs: fix race for binderfs' devices list

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: <aCjGSx2TvQdECBT9@google.com>
Date: Sat, 17 May 2025 17:24:27 +0000
From: Carlos Llamas <cmllamas@...gle.com>
To: Yeoreum Yun <yeoreum.yun@....com>
Cc: gregkh@...uxfoundation.org, arve@...roid.com, tkjos@...roid.com,
	maco@...roid.com, joel@...lfernandes.org, brauner@...nel.org,
	mllamas@...gle.com, surenb@...gle.com, linux-kernel@...r.kernel.org,
	Yeoreum Yun <yeoreun.yun@....com>, Yunseong Kim <ysk@...lloc.com>
Subject: Re: [PATCH] driver/android/binderfs: fix race for binderfs' devices
 list

On Tue, May 13, 2025 at 06:47:19PM +0100, Yeoreum Yun wrote:
> binderfs' devices list is global list and can be accesed by multi-thread
> while unmount binder device.
> Since there is no proper locking, it can meet datarace problem.
> for example it's one of case reported UAF while binderfs_evict_inode()
> removes binder device:
> 
> sudo ./stress-ng --binderfs 8 --binderfs-ops 10000 -t 15 \
>   --pathological --timestamp --tz --syslog --perf --no-rand-seed \
>   --times --metrics --klog-check --status 5 -x smi -v --interrupts --change-cpu

I just found this by chance (there is a typo on my email). Note this was
already fixed here:
https://lore.kernel.org/all/20250324132427.922495-1-dmantipov@yandex.ru/

Cheers,
Carlos Llamas