linux-kernel - Re: [PATCH] vfs: avoid spurious dentry ref/unref cycle on open

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: <ZrKo23cfS2jtN9wF@dread.disaster.area>
Date: Wed, 7 Aug 2024 08:51:07 +1000
From: Dave Chinner <david@...morbit.com>
To: Mateusz Guzik <mjguzik@...il.com>
Cc: brauner@...nel.org, viro@...iv.linux.org.uk, jack@...e.cz,
	linux-kernel@...r.kernel.org, linux-fsdevel@...r.kernel.org
Subject: Re: [PATCH] vfs: avoid spurious dentry ref/unref cycle on open

On Tue, Aug 06, 2024 at 04:46:28PM +0200, Mateusz Guzik wrote:
>  	error = may_open(idmap, &nd->path, acc_mode, open_flag);
> -	if (!error && !(file->f_mode & FMODE_OPENED))
> -		error = vfs_open(&nd->path, file);
> +	if (!error && !(file->f_mode & FMODE_OPENED)) {
> +		BUG_ON(nd->state & ND_PATH_CONSUMED);

Please don't litter new code with random BUG_ON() checks. If this
every happens, it will panic a production kernel and the fix will
generate a CVE.

Given that these checks should never fire in a production kernel
unless something is corrupting memory (i.e. the end is already
near), these should be considered debug assertions and we should
treat them that way from the start.

i.e. we really should have a VFS_ASSERT() or VFS_BUG_ON() (following
the VM_BUG_ON() pattern) masked by a CONFIG_VFS_DEBUG option so they
are only included into debug builds where there is a developer
watching to debug the system when one of these things fires.

This is a common pattern for subsystem specific assertions.  We do
this in all the major filesystems, the MM subsystem does this
(VM_BUG_ON), etc.  Perhaps it is time to do this in the VFS code as
well....

-Dave.
-- 
Dave Chinner
david@...morbit.com