linux-kernel - Re: ubiblock: null pointer dereference using scatterlist in work

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

Message-ID: <20230809141542.GA32614@lst.de>
Date:   Wed, 9 Aug 2023 16:15:42 +0200
From:   hch <hch@....de>
To:     Stephan Wurm <stephan.wurm@...berle.de>
Cc:     hch <hch@....de>, Richard Weinberger <richard@....at>,
        linux-block <linux-block@...r.kernel.org>,
        Miquel Raynal <miquel.raynal@...tlin.com>,
        Vignesh Raghavendra <vigneshr@...com>,
        linux-mtd <linux-mtd@...ts.infradead.org>,
        linux-kernel <linux-kernel@...r.kernel.org>,
        Johannes Eigner <johannes.eigner@...berle.de>
Subject: Re: ubiblock: null pointer dereference using scatterlist in
 work_queue

On Wed, Aug 09, 2023 at 04:10:59PM +0200, Stephan Wurm wrote:
> Am 09. Aug 15:53 hat hch geschrieben:
> > Well, a scatterlist culd contain a highmem page, in which case sg_virt
> > isn't going to cut it and you need to kmap.  Is this a 32-bit system
> > with highmem enabled?
> I can confirm it is a 32-bit system with highmem enabled.

Yes.  So the code needs to stop using sg_virt to work on highmem
setups and do the equivalent of scsi_kmap_atomic_sg.  Best way
forward is probably to move scsi_kmap_atomic_sg to lib/scatterlist.c
(and also convert it to kmap_local instead of kmap_atomic while you're
at it) and then use that in ubiblock