From: James Harper <james.harper@bendigoit.com.au>
To: Wei Liu <liuw@liuw.name>
Cc: Wei Liu <wei.liu2@citrix.com>,
"ian.campbell@citrix.com" <ian.campbell@citrix.com>,
"konrad.wilk@oracle.com" <konrad.wilk@oracle.com>,
"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
"xen-devel@lists.xen.org" <xen-devel@lists.xen.org>,
"annie.li@oracle.com" <annie.li@oracle.com>
Subject: Re: [PATCH 4/4] xen-netback: coalesce slots before copying
Date: Thu, 21 Mar 2013 22:14:17 +0000 [thread overview]
Message-ID: <6035A0D088A63A46850C3988ED045A4B387FB199__30929.9831548289$1363904266$gmane$org@BITCOM1.int.sbss.com.au> (raw)
In-Reply-To: <CAOsiSVWSRmo1+Ubo7GyAKcvUr8XUenkMvnAqt607gPoJm-sXFQ@mail.gmail.com>
>
>> Actually it turns out GPLPV just stops counting at 20. If I keep
>> counting I can sometimes see over 1000 buffers per GSO packet under
>> Windows using "iperf -
>
> Do you think it is necessary to increase MAX_SKB_SLOTS_DEFAULT to 21?
>
Doesn't really matter. Under windows you have to coalesce anyway and the number of cases where the skb count is 20 or 21 is very small so there will be negligible gain and it will break guests that can't handle more than 19.
Has anyone done the benchmarks on if memcpy to coalesce is better or worse than consuming additional ring slots? Probably OT here but I'm talking about packets that might have 19 buffers but could fit on a page or two of coalesced.
James
next prev parent reply other threads:[~2013-03-21 22:14 UTC|newest]
Thread overview: 97+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-03-18 10:35 [PATCH 0/4] Bundle fixes for xen-netfront/back Wei Liu
2013-03-18 10:35 ` [PATCH 1/4] xen-netfront: remove unused variable `extra' Wei Liu
2013-03-18 10:35 ` Wei Liu
2013-03-18 11:42 ` Ian Campbell
2013-03-18 11:42 ` Ian Campbell
2013-03-18 12:04 ` Wei Liu
2013-03-18 12:14 ` Ian Campbell
2013-03-18 12:14 ` Ian Campbell
2013-03-19 2:39 ` annie li
2013-03-19 2:39 ` annie li
2013-03-19 3:02 ` [Xen-devel] " James Harper
2013-03-19 3:02 ` James Harper
2013-03-19 9:28 ` Paul Durrant
2013-03-19 9:28 ` [Xen-devel] " Paul Durrant
2013-03-19 9:53 ` annie li
2013-03-19 9:53 ` [Xen-devel] " annie li
2013-03-19 10:03 ` Paul Durrant
2013-03-19 10:03 ` Paul Durrant
2013-03-19 15:26 ` Wei Liu
2013-03-19 15:26 ` [Xen-devel] " Wei Liu
2013-04-09 14:28 ` Ian Campbell
2013-04-09 14:28 ` Ian Campbell
2013-03-18 12:04 ` Wei Liu
2013-03-18 10:35 ` [PATCH 2/4] xen-netfront: drop skb when skb->len > 65535 Wei Liu
2013-03-18 11:42 ` Ian Campbell
2013-03-18 14:40 ` Wei Liu
2013-03-18 14:40 ` Wei Liu
2013-03-18 14:54 ` Ian Campbell
2013-03-18 14:54 ` Ian Campbell
2013-03-18 15:04 ` Wei Liu
2013-03-18 15:07 ` Ian Campbell
2013-03-18 15:10 ` Wei Liu
2013-03-18 15:10 ` Wei Liu
2013-03-19 21:24 ` Ben Hutchings
2013-03-19 21:24 ` Ben Hutchings
2013-03-19 21:28 ` Ben Hutchings
2013-04-09 14:30 ` Ian Campbell
2013-04-09 14:30 ` Ian Campbell
2013-04-09 14:45 ` Ben Hutchings
2013-04-09 14:53 ` [Xen-devel] " Christoph Egger
2013-04-09 14:59 ` Ben Hutchings
2013-04-09 14:59 ` [Xen-devel] " Ben Hutchings
2013-04-09 14:53 ` Christoph Egger
2013-04-09 14:45 ` Ben Hutchings
2013-03-19 21:28 ` Ben Hutchings
2013-03-18 15:07 ` Ian Campbell
2013-03-18 15:04 ` Wei Liu
2013-03-18 11:42 ` Ian Campbell
2013-03-18 13:44 ` Konrad Rzeszutek Wilk
2013-03-18 13:44 ` Konrad Rzeszutek Wilk
2013-03-18 13:46 ` David Vrabel
2013-03-18 13:46 ` [Xen-devel] " David Vrabel
2013-03-18 13:48 ` Ian Campbell
2013-03-18 13:48 ` [Xen-devel] " Ian Campbell
2013-03-18 14:00 ` David Vrabel
2013-03-18 14:19 ` Wei Liu
2013-03-19 13:40 ` David Vrabel
2013-03-19 13:40 ` [Xen-devel] " David Vrabel
2013-03-19 15:23 ` Wei Liu
2013-03-19 15:23 ` [Xen-devel] " Wei Liu
2013-03-18 14:19 ` Wei Liu
2013-03-20 20:02 ` David Vrabel
2013-03-20 20:02 ` [Xen-devel] " David Vrabel
2013-03-21 13:40 ` Wei Liu
2013-03-21 14:11 ` David Vrabel
2013-03-21 14:11 ` [Xen-devel] " David Vrabel
2013-03-21 14:15 ` Wei Liu
2013-03-21 14:15 ` [Xen-devel] " Wei Liu
2013-03-21 14:20 ` Wei Liu
2013-03-21 14:20 ` [Xen-devel] " Wei Liu
2013-03-21 13:40 ` Wei Liu
2013-03-18 14:00 ` David Vrabel
2013-03-19 1:35 ` [Xen-devel] " annie li
2013-03-19 1:35 ` annie li
2013-03-19 20:13 ` Nick Pegg
2013-03-18 10:35 ` Wei Liu
2013-03-18 10:35 ` [PATCH 3/4] xen-netback: remove skb in xen_netbk_alloc_page Wei Liu
2013-03-18 11:37 ` Ian Campbell
2013-03-18 11:37 ` Ian Campbell
2013-03-18 10:35 ` Wei Liu
2013-03-18 10:35 ` [PATCH 4/4] xen-netback: coalesce slots before copying Wei Liu
2013-03-18 12:07 ` Ian Campbell
2013-03-18 12:07 ` Ian Campbell
2013-03-21 18:37 ` Wei Liu
2013-03-21 18:37 ` Wei Liu
2013-03-18 13:09 ` James Harper
2013-03-18 13:27 ` James Harper
2013-03-21 19:08 ` Wei Liu
2013-03-21 22:14 ` James Harper [this message]
2013-03-21 22:14 ` [Xen-devel] " James Harper
2013-03-22 11:06 ` Wei Liu
2013-03-22 11:19 ` James Harper
2013-03-22 11:19 ` [Xen-devel] " James Harper
2013-03-22 11:28 ` Wei Liu
2013-03-22 11:28 ` [Xen-devel] " Wei Liu
2013-03-22 11:06 ` Wei Liu
2013-03-18 10:35 ` Wei Liu
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='6035A0D088A63A46850C3988ED045A4B387FB199__30929.9831548289$1363904266$gmane$org@BITCOM1.int.sbss.com.au' \
--to=james.harper@bendigoit.com.au \
--cc=annie.li@oracle.com \
--cc=ian.campbell@citrix.com \
--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.