All of lore.kernel.org
 help / color / mirror / Atom feed
From: "David Laight" <David.Laight@ACULAB.COM>
To: James Harper <james.harper@bendigoit.com.au>,
	Paul Durrant <Paul.Durrant@citrix.com>,
	Wei Liu <wei.liu2@citrix.com>,
	David Vrabel <david.vrabel@citrix.com>
Cc: Ian Campbell <Ian.Campbell@citrix.com>,
	konrad.wilk@oracle.com, netdev@vger.kernel.org,
	Wei Liu <liuw@liuw.name>,
	xen-devel@lists.xen.org, annie.li@oracle.com
Subject: Re: [PATCH 5/6] xen-netback: coalesce slots before copying
Date: Tue, 26 Mar 2013 11:27:48 -0000	[thread overview]
Message-ID: <AE90C24D6B3A694183C094C60CF0A2F6026B71B6__23155.7589759634$1364303766$gmane$org@saturn3.aculab.com> (raw)
In-Reply-To: <6035A0D088A63A46850C3988ED045A4B3880B4F4@BITCOM1.int.sbss.com.au>

> As stated previously, I've observed windows issuing staggering
> numbers of buffers to NDIS miniport drivers, so you will need
> to coalesce in a windows driver anyway. I'm not sure what the
> break even point is but I think it's safe to say that in the
> choice between using 1000 (worst case) ring slots (with the
> resulting mapping overheads) and coalescing in the frontend,
> coalescing is going to be the better option.

A long time ago we did some calculation on a sparc mbus/sbus
system (that has an iommu requiring setup for dma) and got
a breakeven point of (about) 1k.
(And I'm not sure we arrange to do aligned copies.)

Clearly that isn't directly relevant here...

It is even likely that the ethernet chips will underrun
if requested to do too many ring operations - especially
at their maximum speed.
I guess none of the modern ones require the first fragment
to be at least 100 bytes in order to guarantee retransmission
after a collision.

	David

  parent reply	other threads:[~2013-03-26 11:27 UTC|newest]

Thread overview: 106+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-03-25 11:08 [PATCH 0/6 V2] Bundle fixes for Xen netfront / netback Wei Liu
2013-03-25 11:08 ` [PATCH 1/6] xen-netfront: remove unused variable `extra' Wei Liu
2013-03-25 11:08 ` Wei Liu
2013-03-25 14:21   ` [Xen-devel] " David Vrabel
2013-03-25 14:21   ` David Vrabel
2013-03-25 16:20   ` David Miller
2013-03-25 16:20   ` David Miller
2013-03-25 11:08 ` [PATCH 2/6] xen-netfront: reduce gso_max_size to account for ethernet header Wei Liu
2013-03-25 11:08 ` Wei Liu
2013-03-25 13:54   ` Malcolm Crossley
2013-03-25 14:23   ` [Xen-devel] " David Vrabel
2013-03-25 14:39     ` Jan Beulich
2013-03-25 14:39     ` Jan Beulich
2013-03-25 14:23   ` David Vrabel
2013-03-25 15:50   ` Sergei Shtylyov
2013-03-25 15:50   ` Sergei Shtylyov
2013-03-25 16:18   ` David Miller
2013-03-25 16:54     ` Eric Dumazet
2013-03-25 16:54     ` Eric Dumazet
2013-03-25 16:59       ` David Miller
2013-03-25 17:24         ` Eric Dumazet
2013-03-25 17:24         ` Eric Dumazet
2013-03-25 20:49         ` Ben Hutchings
2013-03-25 20:49         ` Ben Hutchings
2013-03-25 16:59       ` David Miller
2013-03-25 18:32     ` Wei Liu
2013-03-25 18:32     ` Wei Liu
2013-03-25 18:39       ` David Miller
2013-03-25 18:39       ` David Miller
2013-03-25 19:09         ` Malcolm Crossley
2013-03-25 16:18   ` David Miller
2013-03-25 16:56   ` Konrad Rzeszutek Wilk
2013-03-25 16:56   ` Konrad Rzeszutek Wilk
2013-03-25 11:08 ` [PATCH 3/6] xen-netfront: frags -> slots in xennet_get_responses Wei Liu
2013-03-25 11:08 ` Wei Liu
2013-03-25 14:25   ` [Xen-devel] " David Vrabel
2013-03-25 14:25   ` David Vrabel
2013-03-25 16:20   ` David Miller
2013-03-25 16:20   ` David Miller
2013-03-25 11:08 ` [PATCH 4/6] xen-netback: remove skb in xen_netbk_alloc_page Wei Liu
2013-03-25 16:20   ` David Miller
2013-03-25 16:20   ` David Miller
2013-03-25 11:08 ` Wei Liu
2013-03-25 11:08 ` [PATCH 5/6] xen-netback: coalesce slots before copying Wei Liu
2013-03-25 11:08 ` Wei Liu
2013-03-25 15:13   ` David Vrabel
2013-03-25 15:47     ` [Xen-devel] " Wei Liu
2013-03-25 16:34       ` David Vrabel
2013-03-25 16:58         ` Wei Liu
2013-03-25 17:06           ` [Xen-devel] " Wei Liu
2013-03-25 17:06           ` Wei Liu
2013-03-25 18:29           ` [Xen-devel] " David Vrabel
2013-03-25 19:09             ` Wei Liu
2013-03-26  9:16               ` Paul Durrant
2013-03-26  9:16               ` [Xen-devel] " Paul Durrant
2013-03-26  9:51                 ` Wei Liu
2013-03-26  9:51                 ` [Xen-devel] " Wei Liu
2013-03-26 11:00                 ` James Harper
2013-03-26 11:24                   ` Paul Durrant
2013-03-26 11:29                     ` James Harper
2013-03-26 11:38                       ` Paul Durrant
2013-03-26 11:38                       ` [Xen-devel] " Paul Durrant
2013-03-26 11:29                     ` James Harper
2013-03-26 11:24                   ` Paul Durrant
2013-03-26 11:27                   ` David Laight [this message]
2013-03-26 11:27                   ` [Xen-devel] " David Laight
2013-03-26 11:00                 ` James Harper
2013-03-26 10:52               ` James Harper
2013-03-26 10:52               ` [Xen-devel] " James Harper
2013-03-26 11:06                 ` David Vrabel
2013-03-26 11:15                   ` James Harper
2013-03-26 11:15                   ` [Xen-devel] " James Harper
2013-03-26 11:06                 ` David Vrabel
2013-03-26 11:13               ` David Vrabel
2013-03-26 11:13               ` [Xen-devel] " David Vrabel
2013-03-26 11:29                 ` Wei Liu
2013-03-26 11:29                 ` [Xen-devel] " Wei Liu
2013-03-25 19:09             ` Wei Liu
2013-03-25 18:29           ` David Vrabel
2013-03-25 16:34       ` David Vrabel
2013-03-25 15:47     ` Wei Liu
2013-03-25 15:13   ` David Vrabel
2013-03-25 16:20   ` David Miller
2013-03-25 16:20   ` David Miller
2013-03-25 16:57   ` [Xen-devel] " Konrad Rzeszutek Wilk
2013-04-09 15:10     ` Ian Campbell
2013-04-09 15:10     ` Ian Campbell
2013-03-25 16:57   ` Konrad Rzeszutek Wilk
2013-03-26 18:02   ` David Vrabel
2013-03-26 18:02   ` David Vrabel
2013-03-25 11:08 ` [PATCH 6/6] xen-netback: don't disconnect frontend when seeing oversize frame Wei Liu
2013-03-25 11:08 ` Wei Liu
2013-03-25 11:47   ` David Vrabel
2013-03-25 12:00     ` Wei Liu
2013-03-25 12:00     ` Wei Liu
2013-03-25 11:47   ` David Vrabel
2013-03-25 12:53   ` Jan Beulich
2013-03-25 12:53   ` [Xen-devel] " Jan Beulich
2013-03-25 13:52     ` Wei Liu
2013-03-25 13:52     ` [Xen-devel] " Wei Liu
2013-03-25 16:21   ` David Miller
2013-03-25 16:21   ` David Miller
2013-03-25 16:58   ` [Xen-devel] " Konrad Rzeszutek Wilk
2013-03-27 10:03     ` William Dauchy
2013-03-27 10:03     ` William Dauchy
2013-03-25 16:58   ` Konrad Rzeszutek Wilk

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='AE90C24D6B3A694183C094C60CF0A2F6026B71B6__23155.7589759634$1364303766$gmane$org@saturn3.aculab.com' \
    --to=david.laight@aculab.com \
    --cc=Ian.Campbell@citrix.com \
    --cc=Paul.Durrant@citrix.com \
    --cc=annie.li@oracle.com \
    --cc=david.vrabel@citrix.com \
    --cc=james.harper@bendigoit.com.au \
    --cc=konrad.wilk@oracle.com \
    --cc=liuw@liuw.name \
    --cc=netdev@vger.kernel.org \
    --cc=wei.liu2@citrix.com \
    --cc=xen-devel@lists.xen.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 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.