From: Thomas Glanzmann <thomas@glanzmann.de>
To: "Nicholas A. Bellinger" <nab@linux-iscsi.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>,
John Ogness <john.ogness@linutronix.de>,
Eric Dumazet <edumazet@google.com>,
"David S. Miller" <davem@davemloft.net>,
target-devel <target-devel@vger.kernel.org>,
Linux Network Development <netdev@vger.kernel.org>,
LKML <linux-kernel@vger.kernel.org>
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 [thread overview]
Message-ID: <20140210205634.GC19621@glanzmann.de> (raw)
In-Reply-To: <1392058718.17867.8.camel@haakon3.risingtidesystems.com>
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
next prev parent reply other threads:[~2014-02-10 20:56 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20140206153640.GB4103@glanzmann.de>
[not found] ` <1391727771.14985.41.camel@haakon3.risingtidesystems.com>
[not found] ` <20140207051500.GB10916@glanzmann.de>
[not found] ` <20140207075536.GB17815@glanzmann.de>
[not found] ` <1391801597.1155.28.camel@haakon3.risingtidesystems.com>
[not found] ` <20140207205142.GA8609@glanzmann.de>
2014-02-08 9:18 ` REGRESSION f54b311142a92ea2e42598e347b84e1655caf8e3 tcp auto corking slows down iSCSI file system creation by factor of 70 [WAS: 4 TB VMFS creation takes 15 minutes vs 26 seconds] Thomas Glanzmann
2014-02-08 9:19 ` [PATCH] tcp: disable auto corking by default Thomas Glanzmann
2014-02-08 15:04 ` Eric Dumazet
2014-02-08 16:55 ` Thomas Glanzmann
2014-02-08 17:12 ` Eric Dumazet
2014-02-08 17:20 ` Thomas Glanzmann
2014-02-08 9:23 ` REGRESSION f54b311142a92ea2e42598e347b84e1655caf8e3 tcp auto corking slows down iSCSI file system creation by factor of 70 [WAS: 4 TB VMFS creation takes 15 minutes vs 26 seconds] Thomas Glanzmann
2014-02-08 9:38 ` Thomas Glanzmann
2014-02-08 13:14 ` Eric Dumazet
2014-02-08 13:33 ` Eric Dumazet
2014-02-08 13:38 ` Thomas Glanzmann
2014-02-08 13:50 ` Eric Dumazet
2014-02-08 14:13 ` Eric Dumazet
2014-02-08 14:19 ` Thomas Glanzmann
2014-02-08 14:30 ` Eric Dumazet
2014-02-08 15:00 ` Thomas Glanzmann
2014-02-08 15:06 ` Eric Dumazet
2014-02-08 16:57 ` Thomas Glanzmann
2014-02-08 17:08 ` Eric Dumazet
2014-02-08 17:15 ` Thomas Glanzmann
2014-02-08 19:12 ` Eric Dumazet
2014-02-08 21:36 ` Thomas Glanzmann
2014-02-09 0:15 ` Eric Dumazet
2014-02-09 7:45 ` Thomas Glanzmann
2014-02-09 7:40 ` RFC: Set MSG_MORE in iscsit_fe_sendpage_sg to avoid sending multiple TCP packets instead of one Thomas Glanzmann
2014-02-09 7:42 ` [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 Eric Dumazet
2014-02-09 12:30 ` Eric Dumazet
2014-02-09 15:07 ` Thomas Glanzmann
2014-02-10 18:58 ` Nicholas A. Bellinger
2014-02-10 20:56 ` Thomas Glanzmann [this message]
2014-02-10 21:01 ` Eric Dumazet
2014-02-10 21:14 ` Thomas Glanzmann
2014-02-08 13:37 ` REGRESSION f54b311142a92ea2e42598e347b84e1655caf8e3 tcp auto corking slows down iSCSI file system creation by factor of 70 [WAS: 4 TB VMFS creation takes 15 minutes vs 26 seconds] Thomas Glanzmann
2014-02-08 13:53 ` Eric Dumazet
2014-02-08 13:58 ` Thomas Glanzmann
2014-02-08 14:09 ` Eric Dumazet
2014-02-08 14:12 ` Thomas Glanzmann
2014-02-17 14:08 ` Thomas Glanzmann
2014-02-17 15:26 ` Eric Dumazet
2014-02-17 15:32 ` Thomas Glanzmann
2014-02-17 15:46 ` Eric Dumazet
2014-02-17 15:46 ` Thomas Glanzmann
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20140210205634.GC19621@glanzmann.de \
--to=thomas@glanzmann.de \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=eric.dumazet@gmail.com \
--cc=john.ogness@linutronix.de \
--cc=linux-kernel@vger.kernel.org \
--cc=nab@linux-iscsi.org \
--cc=netdev@vger.kernel.org \
--cc=target-devel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).