[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20240731193318.67312-1-kuniyu@amazon.com>
Date: Wed, 31 Jul 2024 12:33:18 -0700
From: Kuniyuki Iwashima <kuniyu@...zon.com>
To: <dmantipov@...dex.ru>
CC: <davem@...emloft.net>, <edumazet@...gle.com>, <kuba@...nel.org>,
<kuni1840@...il.com>, <kuniyu@...zon.com>, <linux-sctp@...r.kernel.org>,
<lucien.xin@...il.com>, <marcelo.leitner@...il.com>,
<netdev@...r.kernel.org>, <pabeni@...hat.com>,
<syzbot+e6979a5d2f10ecb700e4@...kaller.appspotmail.com>
Subject: Re: [PATCH v1 net] sctp: Fix null-ptr-deref in reuseport_add_sock().
From: Dmitry Antipov <dmantipov@...dex.ru>
Date: Wed, 31 Jul 2024 22:15:12 +0300
> On 7/31/24 10:06 PM, Kuniyuki Iwashima wrote:
>
> > reuseport_lock is to synchronise operations within the same
> > reuseport group, but which socket should belong to which group
> > is out of scope.
>
> Hm... then the lock should be a member of 'struct sock_reuseport'
> rather than global, isn't?
I thought that before and it would be doable with more complex
locking since we still need another synchronisation for add/detach
and alloc. Such operations are most unlikely done concurrently
nor frequently, and the global lock would be enough.
I'll check that again later, but anyway, it's a different topic
for net-next.
Powered by blists - more mailing lists