netdev - Re: [PATCH] netlink: Fix the netlink socket malfunction due to concurrency

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] [thread-next>] [day] [month] [year] [list]

Message-Id: <20230819180448.2315-1-xqjcool@gmail.com>
Date: Sat, 19 Aug 2023 11:04:48 -0700
From: Qingjie Xing <xqjcool@...il.com>
To: edumazet@...gle.com
Cc: davem@...emloft.net,
	dhowells@...hat.com,
	fw@...len.de,
	johannes@...solutions.net,
	keescook@...omium.org,
	kuba@...nel.org,
	kuniyu@...zon.com,
	linux-kernel@...r.kernel.org,
	netdev@...r.kernel.org,
	pabeni@...hat.com,
	pctammela@...atatu.com,
	xqjcool@...il.com
Subject: Re: [PATCH] netlink: Fix the netlink socket malfunction due to concurrency

This piece of code has been present since the Linux code v2.6.12 was
incorporated into Git management.
I believe this modification could potentially address the concurrent issue
we've been discussing.
In netlink_rcv_wake(), as described in [2], the socket's receive queue is
empty, indicated by sk_rmem_alloc being 0. At this point, concurrent
netlink_attachskb() calls netlink_overrun(). In this critical state, the
sk_rmem_alloc of the socket will not instantly transition from 0 to a
value greater than sk_rcvbuf.