netdev - Re: skb_can_coalesce() merges tcp frags with XFS-related slab objects

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: <fd3eabed-3686-83c7-0169-c25921390b5c@gmail.com>
Date:   Wed, 20 Feb 2019 08:35:05 -0800
From:   Eric Dumazet <eric.dumazet@...il.com>
To:     Vasily Averin <vvs@...tuozzo.com>,
        Eric Dumazet <eric.dumazet@...il.com>,
        "David S. Miller" <davem@...emloft.net>
Cc:     Linux Kernel Network Developers <netdev@...r.kernel.org>,
        Ilya Dryomov <idryomov@...il.com>
Subject: Re: skb_can_coalesce() merges tcp frags with XFS-related slab objects



On 02/20/2019 08:19 AM, Vasily Averin wrote:

> Thank you for explanation, 
> though this happen in real life and triggers BUG_ON only if receiving side is located on the same host.
> Is it probably makes sense to add WARN_ON into skb_can_coalesce to detect such cases?

Yes, but please do it only in the sendpage() path, or only in CONFIG_DEBUG_PAGEALLOC / CONFIG_DEBUG_VM cases.

tcp_sendmsg() uses a per task page (look at sk_page_frag()), and it seems
strange to recheck what we already know (it is a page not backed/used by SLAB)