All of lore.kernel.org
 help / color / mirror / Atom feed
From: jamal <hadi@cyberus.ca>
To: David Miller <davem@davemloft.net>
Cc: johnpol@2ka.mipt.ru, herbert@gondor.apana.org.au,
	kaber@trash.net, shemminger@linux-foundation.org,
	jagana@us.ibm.com, Robert.Olsson@data.slu.se, rick.jones2@hp.com,
	xma@us.ibm.com, gaagaan@gmail.com, netdev@vger.kernel.org,
	rdreier@cisco.com, peter.p.waskiewicz.jr@intel.com,
	mcarlson@broadcom.com, mchan@broadcom.com,
	general@lists.openfabrics.org, kumarkr@linux.ibm.com,
	tgraf@suug.ch, randy.dunlap@oracle.com, sri@us.ibm.com
Subject: Re: [PATCH 0/10 REV5] Implement skb batching and support in IPoIB/E1000
Date: Mon, 17 Sep 2007 08:51:40 -0400	[thread overview]
Message-ID: <1190033500.4230.102.camel@localhost> (raw)
In-Reply-To: <20070916.201318.71091570.davem@davemloft.net>

On Sun, 2007-16-09 at 20:13 -0700, David Miller wrote:

> What Herbert and I want to do is basically turn on TSO for
> devices that can't do it in hardware, and rely upon the GSO
> framework to do the segmenting in software right before we
> hit the device.

Sensible. 

> This only makes sense for devices which can 1) scatter-gather
> and 2) checksum on transmit.  

If you have knowledge there are enough descriptors in the driver to
cover all skbs you are passing, do you need to have #1? 
Note i dont touch fragments, i am assuming the driver is smart enough to
handle them otherwise it wont advertise it can handle scatter-gather

> Otherwise we make too many copies and/or passes over the data.

I didnt understand this last bit - you are still going to go over the
list regardless of whether you call ->hard_start_xmit() once or
multiple times over the same list, no? In the later case i am assuming
a trimmed down ->hard_start_xmit()

> UDP is too easy a test case in fact :-)

I learnt a lot about the behavior out of doing udp (and before that with
pktgen); theres a lot of driver habits that may need to be tuned before
batching becomes really effective - which is easier to see with udp than
with tcp.

cheers,
jamal


  reply	other threads:[~2007-09-17 12:51 UTC|newest]

Thread overview: 107+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-09-14  9:00 [PATCH 0/10 REV5] Implement skb batching and support in IPoIB/E1000 Krishna Kumar
2007-09-14  9:01 ` [PATCH 1/10 REV5] [Doc] HOWTO Documentation for batching Krishna Kumar
2007-09-14 18:37   ` [ofa-general] " Randy Dunlap
2007-09-17  4:10     ` Krishna Kumar2
2007-09-17  4:13       ` [ofa-general] " Jeff Garzik
2007-09-14  9:01 ` [PATCH 2/10 REV5] [core] Add skb_blist & support " Krishna Kumar
2007-09-14 12:46   ` [ofa-general] " Evgeniy Polyakov
2007-09-17  3:51     ` Krishna Kumar2
2007-09-14  9:01 ` [PATCH 3/10 REV5] [sched] Modify qdisc_run to support batching Krishna Kumar
2007-09-14 12:15   ` [ofa-general] " Evgeniy Polyakov
2007-09-17  3:49     ` Krishna Kumar2
2007-09-14  9:02 ` [PATCH 4/10 REV5] [ethtool] Add ethtool support Krishna Kumar
2007-09-14  9:02 ` [PATCH 5/10 REV5] [IPoIB] Header file changes Krishna Kumar
2007-09-14  9:03 ` [PATCH 6/10 REV5] [IPoIB] CM & Multicast changes Krishna Kumar
2007-09-14  9:03 ` [PATCH 7/10 REV5] [IPoIB] Verbs changes Krishna Kumar
2007-09-14  9:03 ` [PATCH 8/10 REV5] [IPoIB] Post and work completion handler changes Krishna Kumar
2007-09-14  9:04 ` [PATCH 9/10 REV5] [IPoIB] Implement batching Krishna Kumar
2007-09-14  9:04 ` [PATCH 10/10 REV5] [E1000] " Krishna Kumar
2007-09-14 12:47   ` [ofa-general] " Evgeniy Polyakov
2007-09-17  3:56     ` Krishna Kumar2
2007-11-13 21:28   ` [ofa-general] " Kok, Auke
2007-11-14  8:30     ` Krishna Kumar2
2007-09-14 12:49 ` [ofa-general] Re: [PATCH 0/10 REV5] Implement skb batching and support in IPoIB/E1000 Evgeniy Polyakov
2007-09-16 23:17 ` David Miller
2007-09-17  0:29   ` jamal
2007-09-17  1:02     ` David Miller
2007-09-17  2:14       ` [ofa-general] " jamal
2007-09-17  2:25         ` David Miller
2007-09-17  3:01           ` jamal
2007-09-17  3:13             ` David Miller
2007-09-17 12:51               ` jamal [this message]
2007-09-17 16:37                 ` [ofa-general] " David Miller
2007-09-17  4:46           ` Krishna Kumar2
2007-09-23 17:53     ` [PATCHES] TX batching jamal
2007-09-23 17:56       ` [ofa-general] [PATCH 1/4] [NET_SCHED] explict hold dev tx lock jamal
2007-09-23 17:58         ` [ofa-general] [PATCH 2/4] [NET_BATCH] Introduce batching interface jamal
2007-09-23 18:00           ` [PATCH 3/4][NET_BATCH] net core use batching jamal
2007-09-23 18:02             ` [ofa-general] [PATCH 4/4][NET_SCHED] kill dev->gso_skb jamal
2007-09-30 18:53               ` [ofa-general] [PATCH 3/3][NET_SCHED] " jamal
2007-10-07 18:39               ` [ofa-general] [PATCH 3/3][NET_BATCH] " jamal
2007-09-30 18:52             ` [ofa-general] [PATCH 2/3][NET_BATCH] net core use batching jamal
2007-10-01  4:11               ` Bill Fink
2007-10-01 13:30                 ` jamal
2007-10-02  4:25                   ` [ofa-general] " Bill Fink
2007-10-02 13:20                     ` jamal
2007-10-03  5:29                       ` [ofa-general] " Bill Fink
2007-10-03 13:42                         ` jamal
2007-10-01 10:42               ` [ofa-general] " Patrick McHardy
2007-10-01 13:21                 ` jamal
2007-10-08  5:03                   ` Krishna Kumar2
2007-10-08 13:17                     ` jamal
2007-10-09  3:09                       ` [ofa-general] " Krishna Kumar2
2007-10-09 13:10                         ` jamal
2007-10-07 18:38             ` [ofa-general] " jamal
2007-09-30 18:51           ` [ofa-general] [PATCH 1/4] [NET_BATCH] Introduce batching interface jamal
2007-09-30 18:54             ` [ofa-general] Re: [PATCH 1/3] " jamal
2007-10-07 18:36           ` [ofa-general] " jamal
2007-10-08  9:59             ` Krishna Kumar2
2007-10-08 13:49               ` jamal
2007-09-24 19:12         ` [ofa-general] RE: [PATCH 1/4] [NET_SCHED] explict hold dev tx lock Waskiewicz Jr, Peter P
2007-09-24 22:51           ` jamal
2007-09-24 22:57             ` Waskiewicz Jr, Peter P
2007-09-24 23:38               ` [ofa-general] " jamal
2007-09-24 23:47                 ` Waskiewicz Jr, Peter P
2007-09-25  0:14                   ` [ofa-general] " Stephen Hemminger
2007-09-25  0:31                     ` [ofa-general] " Waskiewicz Jr, Peter P
2007-09-25 13:15                     ` [ofa-general] " jamal
2007-09-25 15:24                       ` Stephen Hemminger
2007-09-25 22:14                         ` jamal
2007-09-25 22:43                           ` jamal
2007-09-25 13:08                   ` [ofa-general] " jamal
2007-10-08  4:51                 ` [ofa-general] " David Miller
2007-10-08 13:34                   ` jamal
2007-10-08 14:22                     ` parallel networking (was Re: [PATCH 1/4] [NET_SCHED] explict hold dev tx lock) Jeff Garzik
2007-10-08 15:18                       ` jamal
2007-10-08 15:18                         ` [ofa-general] " jamal
2007-10-08 21:11                       ` parallel networking David Miller
2007-10-08 21:11                         ` [ofa-general] " David Miller
2007-10-08 22:30                         ` jamal
2007-10-08 22:33                           ` David Miller
2007-10-08 22:35                             ` Waskiewicz Jr, Peter P
2007-10-08 22:35                               ` [ofa-general] " Waskiewicz Jr, Peter P
2007-10-08 23:42                             ` jamal
2007-10-08 23:42                               ` [ofa-general] " jamal
2007-10-09  1:53                         ` Jeff Garzik
2007-10-09  1:53                           ` [ofa-general] " Jeff Garzik
2007-10-09 14:59                           ` Michael Krause
2007-10-08 21:05                     ` [PATCH 1/4] [NET_SCHED] explict hold dev tx lock David Miller
2007-09-23 18:19       ` [PATCHES] TX batching Jeff Garzik
2007-09-23 19:11         ` [ofa-general] " jamal
2007-09-23 19:36           ` Kok, Auke
2007-09-23 21:20             ` jamal
2007-09-24  7:00               ` Kok, Auke
2007-09-24 22:38                 ` jamal
2007-09-24 22:52                   ` [ofa-general] " Kok, Auke
2007-09-24 22:54           ` [DOC] Net batching driver howto jamal
2007-09-25 20:16             ` [ofa-general] " Randy Dunlap
2007-09-25 22:28               ` jamal
2007-09-25  0:15           ` [PATCHES] TX batching Jeff Garzik
2007-09-30 18:50       ` [ofa-general] " jamal
2007-09-30 19:19         ` [ofa-general] " jamal
2007-10-07 18:34       ` [ofa-general] " jamal
2007-10-08 12:51         ` [ofa-general] " Evgeniy Polyakov
2007-10-08 14:05           ` jamal
2007-10-09  8:14             ` Krishna Kumar2
2007-10-09 13:25               ` jamal
2007-09-17  4:08   ` [PATCH 0/10 REV5] Implement skb batching and support in IPoIB/E1000 Krishna Kumar2

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=1190033500.4230.102.camel@localhost \
    --to=hadi@cyberus.ca \
    --cc=Robert.Olsson@data.slu.se \
    --cc=davem@davemloft.net \
    --cc=gaagaan@gmail.com \
    --cc=general@lists.openfabrics.org \
    --cc=herbert@gondor.apana.org.au \
    --cc=jagana@us.ibm.com \
    --cc=johnpol@2ka.mipt.ru \
    --cc=kaber@trash.net \
    --cc=kumarkr@linux.ibm.com \
    --cc=mcarlson@broadcom.com \
    --cc=mchan@broadcom.com \
    --cc=netdev@vger.kernel.org \
    --cc=peter.p.waskiewicz.jr@intel.com \
    --cc=randy.dunlap@oracle.com \
    --cc=rdreier@cisco.com \
    --cc=rick.jones2@hp.com \
    --cc=shemminger@linux-foundation.org \
    --cc=sri@us.ibm.com \
    --cc=tgraf@suug.ch \
    --cc=xma@us.ibm.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.