[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CACSApvY697bqvPetoLUzAs7Ympq+Fgigyz-Z=JqdJ8fX-jfbhQ@mail.gmail.com>
Date: Thu, 2 Nov 2017 22:33:26 -0400
From: Soheil Hassas Yeganeh <soheil.kdev@...il.com>
To: Neal Cardwell <ncardwell@...gle.com>
Cc: Eric Dumazet <eric.dumazet@...il.com>,
David Miller <davem@...emloft.net>,
netdev <netdev@...r.kernel.org>,
Yuchung Cheng <ycheng@...gle.com>,
Alexei Starovoitov <ast@...nel.org>,
Priyaranjan Jha <priyarjha@...gle.com>
Subject: Re: [PATCH net-next] tcp: tcp_fragment() should not assume rtx skbs
On Thu, Nov 2, 2017 at 9:16 PM, Neal Cardwell <ncardwell@...gle.com> wrote:
> On Thu, Nov 2, 2017 at 9:10 PM, Eric Dumazet <eric.dumazet@...il.com> wrote:
>> From: Eric Dumazet <edumazet@...gle.com>
>>
>> While stress testing MTU probing, we had crashes in list_del() that we root-caused
>> to the fact that tcp_fragment() is unconditionally inserting the freshly allocated
>> skb into tsorted_sent_queue list.
>>
>> But this list is supposed to contain skbs that were sent.
>> This was mostly harmless until MTU probing was enabled.
>>
>> Fortunately we can use the tcp_queue enum added later (but in same linux version)
>> for rtx-rb-tree to fix the bug.
>>
>> Fixes: e2080072ed2d ("tcp: new list for sent but unacked skbs for RACK recovery")
>> Signed-off-by: Eric Dumazet <edumazet@...gle.com>
>
> Acked-by: Neal Cardwell <ncardwell@...gle.com>
Acked-by: Soheil Hassas Yeganeh <soheil@...gle.com>
Nice! Thank you, Eric!
Powered by blists - more mailing lists