linux-kernel - Re: [PATCH] kernfs: Fix UAF in PSI polling when open file is released

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 for Android: free password hash cracker in your pocket

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

Message-ID: <aKUo7BuX-teh4IzF@slm.duckdns.org>
Date: Tue, 19 Aug 2025 15:46:20 -1000
From: Tejun Heo <tj@...nel.org>
To: Chen Ridong <chenridong@...weicloud.com>
Cc: Baokun Li <libaokun1@...wei.com>, cgroups@...r.kernel.org,
	chenridong@...wei.com, gregkh@...uxfoundation.org,
	hannes@...xchg.org, linux-kernel@...r.kernel.org,
	lujialin4@...wei.com, mkoutny@...e.com, peterz@...radead.org,
	zhouchengming@...edance.com, Yang Erkun <yangerkun@...wei.com>
Subject: Re: [PATCH] kernfs: Fix UAF in PSI polling when open file is released

Hello,

On Mon, Aug 18, 2025 at 04:00:08PM +0800, Chen Ridong wrote:
> > A potential solution is to make the lifecycles of cgroup_file_ctx and
> > psi_trigger match the struct kernfs_open_file they're associated with.
> > Maybe we could just get rid of the kernfs_release_file call in
> > kernfs_drain_open_files?
> >
> 
> Hi, Tj, what do you think about this solution?

So, I think it's really fragile for a killed (drained) kernfs_open_file to
be reused after the corresponding @kn is resurrected. Once killed, that file
should stay dead. I think it'd be best if we can do this in a generic manner
rather than trying to fix it only for poll.

kernfs_get_active() is the thing which gates active operations on the file.
Maybe we can add a wrapper, say, kernfs_get_active_of(struct
kernfs_open_file *of) which returns NULL if @of has already been killed or
the underlying @kn can't be activated?

Thanks.

-- 
tejun