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: <20190520163131.7c143ebe@xps13>
Date:   Mon, 20 May 2019 16:31:31 +0200
From:   Miquel Raynal <miquel.raynal@...tlin.com>
To:     Stefan Agner <stefan@...er.ch>
Cc:     bbrezillon@...nel.org, richard@....at, dwmw2@...radead.org,
        computersforpeace@...il.com, marek.vasut@...il.com,
        linux-mtd@...ts.infradead.org, linux-kernel@...r.kernel.org,
        Sascha Hauer <s.hauer@...gutronix.de>,
        Boris Brezillon <boris.brezillon@...labora.com>
Subject: Re: [PATCH] mtd: rawnand: use longest matching pattern

Hi Stefan,

Stefan Agner <stefan@...er.ch> wrote on Fri, 19 Apr 2019 09:47:17 +0200:

> Sometimes the exec_op parser does not choose the optimal pattern if
> multiple patterns with optional elements are available. Since the stack
> automatically splits operations in multiple exec_op calls, a non-optimal
> pattern gets broken up into multiple calls. E.g. an OOB read using the
> vf610 driver:
>   nand: executing subop:
>   nand:     ->CMD      [0x00]
>   nand:     ->ADDR     [5 cyc: 00 08 ea 94 02]
>   nand:     ->CMD      [0x30]
>   nand:     ->WAITRDY  [max 200000 ms]
>   nand:       DATA_IN  [64 B]
>   nand: executing subop:
>   nand:       CMD      [0x00]
>   nand:       ADDR     [5 cyc: 00 08 ea 94 02]
>   nand:       CMD      [0x30]
>   nand:       WAITRDY  [max 200000 ms]
>   nand:     ->DATA_IN  [64 B]
> 
> However, the vf610 driver has a pattern which can execute the complete
> command in a single go...
> 
> This patch makes sure that the longest matching pattern is chosen
> instead of the first (potentially only partial) match. With this
> change the vf610 reads the OOB in a single exec_op call:
>   nand: executing subop:
>   nand:     ->CMD      [0x00]
>   nand:     ->ADDR     [5 cyc: 00 08 c0 1d 00]
>   nand:     ->CMD      [0x30]
>   nand:     ->WAITRDY  [max 200000 ms]
>   nand:     ->DATA_IN  [64 B]
> 
> Reported-by: Sascha Hauer <s.hauer@...gutronix.de>
> Suggested-by: Boris Brezillon <boris.brezillon@...labora.com>
> Tested-by: Stefan Agner <stefan@...er.ch>
> Signed-off-by: Stefan Agner <stefan@...er.ch>
> ---

Applied, thanks.

Miquèl

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ