linux-kernel - Re: [PATCH] btrfs: correctly escape subvol in btrfs_show

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: <20250319235251.GS32661@twin.jikos.cz>
Date: Thu, 20 Mar 2025 00:52:51 +0100
From: David Sterba <dsterba@...e.cz>
To: Johannes Kimmel <kernel@...eminimum.eu>
Cc: linux-btrfs@...r.kernel.org, clm@...com, josef@...icpanda.com,
	dsterba@...e.com, linux-kernel@...r.kernel.org,
	Calvin Walton <calvin.walton@...stin.ca>
Subject: Re: [PATCH] btrfs: correctly escape subvol in btrfs_show_options

On Wed, Mar 19, 2025 at 10:49:00PM +0100, Johannes Kimmel wrote:
> Currently, displaying the btrfs subvol mount option doesn't escape `,`.
> This makes parsing /proc/self/mounts and /proc/self/mountinfo
> ambigious for subvolume names that contain commas. The text after the
> comma could be mistaken for another option (think "subvol=foo,ro", where
> ro is actually part of the subvolumes name).

The subvol= option was intentionally last so the path does not mix with
other options but yeah it still can be confused if it's parsed in a
generic way and not assuming anything about the ordering.

I've checked util-linux/libmount, seems that it understands the "\," as
part of the option value so it should not break anything.

> This patch replaces the manual escape characters list with a call to
> seq_show_option. Thanks to Calvin Walton for suggesting this approach.
> 
> Fixes: c8d3fe028f64 ("Btrfs: show subvol= and subvolid= in /proc/mounts")
> Suggested-by: Calvin Walton <calvin.walton@...stin.ca>
> Signed-off-by: Johannes Kimmel <kernel@...eminimum.eu>

Added to for-next, thanks.