linux-kernel - Re: [PATCH] jffs2: check dstlen for jffs2_zlib

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

Message-ID: <3151900.SLYxvDXFDr@blindfold>
Date:   Mon, 21 Jan 2019 12:35:24 +0100
From:   Richard Weinberger <richard@....at>
To:     yang.yang29@....com.cn
Cc:     richard.weinberger@...il.com, dwmw2@...radead.org,
        wang.yi59@....com.cn, linux-mtd@...ts.infradead.org,
        linux-kernel@...r.kernel.org, xue.zhihong@....com.cn
Subject: Re: [PATCH] jffs2: check dstlen for jffs2_zlib_compress()

Am Montag, 21. Januar 2019, 09:59:10 CET schrieb yang.yang29@....com.cn:
> Thanks for reply.
> But I think if (*dstlen < 2) is not good enough.

Not enough in terms of what?
My patch does what the commit message says, it fixes the out of bound access.
What do I miss?

> Because if  the length of data_in is equals to 2, the length of output compressed data is also 2.
> So there's no meaning to do all the compress calculates.
> 
> And I had do an experiment:
> If dstlen equals to 3, and data_in has 3 same bytes, the length of output compressed data will be 2. 
> Which means compress is effective.

This is a different issue and a possible further optimization.

Thanks,
//richard