From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Glanzmann Subject: Re: [PATCH] This extends tx_data and and iscsit_do_tx_data with the additional parameter flags and avoids sending multiple TCP packets in iscsit_fe_sendpage_sg Date: Mon, 10 Feb 2014 21:56:34 +0100 Message-ID: <20140210205634.GC19621@glanzmann.de> References: <20140208150001.GI20512@glanzmann.de> <1391871986.10160.105.camel@edumazet-glaptop2.roam.corp.google.com> <20140208165732.GB22359@glanzmann.de> <1391879318.10160.108.camel@edumazet-glaptop2.roam.corp.google.com> <20140208171531.GA23798@glanzmann.de> <1391886759.10160.114.camel@edumazet-glaptop2.roam.corp.google.com> <20140209074027.GA8105@glanzmann.de> <20140209074227.GA8219@glanzmann.de> <1391949039.10160.129.camel@edumazet-glaptop2.roam.corp.google.com> <1392058718.17867.8.camel@haakon3.risingtidesystems.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Eric Dumazet , John Ogness , Eric Dumazet , "David S. Miller" , target-devel , Linux Network Development , LKML To: "Nicholas A. Bellinger" Return-path: Content-Disposition: inline In-Reply-To: <1392058718.17867.8.camel@haakon3.risingtidesystems.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Hello Nab, > This looks correct to me. Thomas, once your able to confirm please > include your 'Tested-by' and I'll include for the next -rc3 PULL > request. Eric is currently reviewing our latest iteration with MSG_MORE for kernel_sendmsg and MSG_MORE | MSG_SENDPAGE_NOTLAST for sendpage. However with the last iteration we had again a high RTT for some packets. But than Eric let me tune net.ipv4.tcp_min_tso_segs to 8 and the RTT went down to what it used before auto corking was enabled. At least almost. I'm having a steep learning curve but Eric hopefully knows how to get this back in check. Nevertheless the regression I saw are history because I saw that Eric has submitted the patch to David S. Miller which fixes the two bugs that killed the iSCSI performance when tcp auto corking was on. So currently we're just optimizing to get the last 20% or so out of it. Quite interesting. Especially how much bandwidth can be saved by coalescing packets. Cheers, Thomas