linux-kernel - Re: 【Bug】use-after-free in NVMe request double completion detected by kasan

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: <ZQWSFgDrkLgw0tO2@debian.me>
Date:   Sat, 16 Sep 2023 18:31:34 +0700
From:   Bagas Sanjaya <bagasdotme@...il.com>
To:     xiaohui li <buhuiyuanyuan1984@...il.com>
Cc:     Linux NVMe <linux-nvme@...ts.infradead.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Will Deacon <will@...nel.org>,
        Ard Biesheuvel <ardb@...nel.org>,
        Kees Cook <keescook@...omium.org>,
        Hanjun Guo <guohanjun@...wei.com>,
        Christoph Hellwig <hch@....de>,
        Jesper Dangaard Brouer <brouer@...hat.com>,
        Tony Luck <tony.luck@...el.com>
Subject: Re: 【Bug】use-after-free in NVMe
 request double completion detected by kasan

On Sat, Sep 16, 2023 at 02:23:21PM +0800, xiaohui li wrote:
> This bug only appears during concurrent stress testing, It has nothing
> to do with what kinds of nvme disk to use.
> Especially when conducting DDR stress testing, the consumption of DDR
> bandwidth is high. Once the NVME driver is under pressure, the NVME IO
> request times out，
> and then this use-after-free bug appears with high probability.
> 
> and i find other pepole have also encountered this bug. such as below:
> https://access.redhat.com/solutions/6993136   (System crash due to
> use-after-free in NVMe request double completion)
> 
> i have no v6.6-rc1 kernel,
> shall some one can give a bug-fix patch on 5.10.x linux kernel ?   many thanks.
Please don't top-post; reply inline with appropriate context instead. Replying
inline allows readers to follow the conversation flow.

Again, as I repeat, please test mainline kernel first. Since you're about
to compile your own kernel, there is a kernel documentation on it [1]. Or
you can follow Rocky Linux's guide ([2]) (as it is based on RedHat).

Thanks.

[1]: https://docs.kernel.org/admin-guide/quickly-build-trimmed-linux.html
[2]: https://docs.rockylinux.org/guides/custom-linux-kernel/

-- 
An old man doll... just what I always wanted! - Clara

Download attachment "signature.asc" of type "application/pgp-signature" (229 bytes)