[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <AANLkTim0+7mTYrQF3mttUW+JqOmvTyMoDXOdeCX_SSdp@mail.gmail.com>
Date: Tue, 1 Mar 2011 23:12:29 -0800
From: Tom Herbert <therbert@...gle.com>
To: Herbert Xu <herbert@...dor.hengli.com.au>
Cc: Eric Dumazet <eric.dumazet@...il.com>,
Thomas Graf <tgraf@...radead.org>,
David Miller <davem@...emloft.net>, rick.jones2@...com,
wsommerfeld@...gle.com, daniel.baluta@...il.com,
netdev@...r.kernel.org
Subject: Re: SO_REUSEPORT - can it be done in kernel?
On Tue, Mar 1, 2011 at 6:39 PM, Herbert Xu <herbert@...dor.apana.org.au> wrote:
> On Wed, Mar 02, 2011 at 03:00:03AM +0100, Eric Dumazet wrote:
>> >
>> > Think about it, a TCP socket cannot be used by a multi-threaded app
>> > in a scalable way.
>>
>> Well...
>>
>> If you think about it, SO_REUSEPORT patch has exactly the same goal :
>
In a sense. SO_RESUSEPORT for TCP is intended to provide a scalable
listener solution. Sharing an established socket is not very
efficient, something like a multiplexing socket layer on top of TCP
might be good.
> UDP is a datagram protocol, TCP is not.
>
> Anyway, here is an alternate proposal. When a TCP socket transmits
> for the first time (SYN or SYN-ACK), we pick a queue based on CPU and
> store it in the socket. From then on we stick to that selection.
>
> We would only allow changes if we can ensure that all transmitted
> packets have left the queue. Or we just never change it like we
> do now.
>
XPS does all this already.
> For datagram protocols we simply use the current CPU.
>
Probably need to set skb->ooo_okay (for UDP etc.) also so that XPS
will change queues.
Tom
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists