linux-kernel - Re: [PATCH 5/5] ubi: Allow to use read-only UBI volume with not enough PEBs

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] [day] [month] [year] [list]

Message-Id: <201708061230.02997@pali>
Date:   Sun, 6 Aug 2017 12:30:02 +0200
From:   Pali Rohár <pali.rohar@...il.com>
To:     Richard Weinberger <richard@....at>
Cc:     David Woodhouse <dwmw2@...radead.org>,
        Brian Norris <computersforpeace@...il.com>,
        Boris Brezillon <boris.brezillon@...e-electrons.com>,
        Marek Vasut <marek.vasut@...il.com>,
        Artem Bityutskiy <dedekind1@...il.com>,
        "linux-mtd@...ts.infradead.org" <linux-mtd@...ts.infradead.org>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 5/5] ubi: Allow to use read-only UBI volume with not enough PEBs

On Sunday 06 August 2017 11:43:25 Richard Weinberger wrote:
> Pali,
> 
> Am 25.07.2017 um 16:27 schrieb Pali Rohár:
> >> I fear this is not correct, it will disable a legit self-check of
> >> UBI volumes. If the read-only volume is corrupted/truncated and
> >> you miss PEBs, this check will no longer
> >> trigger.
> >> 
> >> Especially when dealing with nanddumps, truncation is a common
> >> problem.
> > 
> > Any idea how to fix it? Or how to handle read-only images which are
> > marked for auto-resize?
> 
> I'd vote for rejecting images that have auto-resize set when the MTD
> is read-only. In fact, using UBI on top of a read-only MTD is very
> uncommon and not recommended (for NAND). The auto-resize flag should
> be also only set when you just have created it using mkfs.ubifs. Why
> would you inspect such an image with the kernel UBIFS unless you're
> hunting down a bug in mkfs.ubifs?
> 
> Thanks,
> //richard

E.g. because when I get UBIFS image and I want to unpack it.

IMO UBIFS image which have auto-resize set is also valid UBIFS image and 
kernel should be able to read it too, even in R/O mode.

-- 
Pali Rohár
pali.rohar@...il.com

Download attachment "signature.asc " of type "application/pgp-signature" (199 bytes)