linux-kernel - Re: [PATCH] fat: Prevent the race of read/write the FAT16 and FAT32 entry

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: <tencent_341B732549BA50BB6733349E621B0D4B7A08@qq.com>
Date: Tue, 29 Jul 2025 13:32:26 +0800
From: Edward Adam Davis <eadavis@...com>
To: viro@...iv.linux.org.uk
Cc: eadavis@...com,
	hirofumi@...l.parknet.co.jp,
	linkinjeon@...nel.org,
	linux-fsdevel@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	sj1557.seo@...sung.com,
	syzbot+d3c29ed63db6ddf8406e@...kaller.appspotmail.com,
	syzkaller-bugs@...glegroups.com
Subject: Re: [PATCH] fat: Prevent the race of read/write the FAT16 and FAT32 entry

On Tue, 29 Jul 2025 05:53:23 +0100, Al Viro wrote:
>FAT12 problem is that FAT entries being accessed there are 12-bit, packed in
>pairs into an array of 3-byte values.  Have you actually read what the functions
I learned it. I didn't really understand this code, but after your hint,
I understood that the 12-bit entry FAT12 has 3 bytes, and cacheline will
not be an integer multiple of 3. Finally, the entry with fat12 may exceed
the judgment of cacheline.
>are doing?  There we *must* serialize the access to bytes that have 4 bits
>from one entry and 4 from another - there's no such thing as atomically
>update half a byte; it has to be read, modified and stored back.  If two
>threads try to do that to upper and lower halves of the same byte at the
>same time, the value will be corrupted.
I will change the fix to READ_ONCE/WRITE_ONCE later.

BR,
Edward