From mboxrd@z Thu Jan 1 00:00:00 1970 From: Daniel Borkmann Date: Wed, 25 Jun 2014 10:28:28 +0000 Subject: Re: [PATCH net-next] sctp: Add partially support for MSG_MORE to SCTP. Message-Id: <53AAA44C.5010206@redhat.com> List-Id: References: <063D6719AE5E284EB5DD2968C1650D6D1725FB90@AcuExch.aculab.com> <53A4B165.6050504@gmail.com> <063D6719AE5E284EB5DD2968C1650D6D1726152D@AcuExch.aculab.com> In-Reply-To: <063D6719AE5E284EB5DD2968C1650D6D1726152D@AcuExch.aculab.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: David Laight Cc: 'Vlad Yasevich' , "netdev@vger.kernel.org" , "linux-sctp@vger.kernel.org" On 06/23/2014 04:27 PM, David Laight wrote: > From: Vlad Yasevich >> On 06/20/2014 12:24 PM, David Laight wrote: >>> If MSG_MORE is set then buffer sends as if Nagle were enabled. >>> The first data chunk is still sent on its own, but subsequent chunks >>> will be bundled and full packets sent. >>> Full MSG_MORE support would require a timeout (preferably configurable >>> per-socket) to send the last chunk(s), instead of sending them >>> when there is nothing outstanding. >> >> Instead of using 1 and 2, can you define them as flags please > > Will do.... > > It is worth inverting nagle/nodelay bit; so that: > 0 => SCTP_NODELAY > 1 => Nagle (default) > 2 => MSG_MORE > 4 => reserved for corked > > That would require working out where the structure is initialised > (in order to default to Nagle). I think that should be fine, too.