From: Sander Eikelenboom <linux@eikelenboom•it>
To: Rick Jones <rick.jones2@hp•com>
Cc: Eric Dumazet <erdnetdev@gmail•com>,
Ian Campbell <Ian.Campbell@citrix•com>,
"netdev@vger•kernel.org" <netdev@vger•kernel.org>,
Konrad Rzeszutek Wilk <konrad.wilk@oracle•com>,
annie li <annie.li@oracle•com>,
"xen-devel@lists•xensource.com" <xen-devel@lists•xensource.com>
Subject: Re: [PATCH] xen/netfront: improve truesize tracking
Date: Thu, 3 Jan 2013 21:40:30 +0100 [thread overview]
Message-ID: <561084196.20130103214030@eikelenboom.it> (raw)
In-Reply-To: <50D4AB87.8050601@hp.com>
Friday, December 21, 2012, 7:33:43 PM, you wrote:
> I'm guessing that trusize checks matter more on the "inbound" path than
> the outbound path? If that is indeed the case, then instead of, or in
> addition to using the -s option to set the local (netperf side) socket
> buffer size, you should use a -S option to set the remote (netserver
> side) socket buffer size.
> happy benchmarking,
> rick jones
OK, ran them with -S as well:
"current" is with netfront as is (skb->truesize += skb->data_len - RX_COPY_THRESHOLD;)
"patched" is with IanC's latest patch (skb->truesize += PAGE_SIZE * skb_shinfo(skb)->nr_frags;)
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 (192.168.1.1) port 0 AF_INET : +/-2.500% @ 95% conf. : demo
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. KBytes/sec
current 87380 16384 16384 60.00 963213.91
patched 87380 16384 16384 60.00 965904.01
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 (192.168.1.1) port 0 AF_INET : +/-2.500% @ 95% conf. : demo
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. KBytes/sec
current 2280 16384 16384 60.00 10029.44
patched 2280 16384 16384 60.00 10453.63
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 (192.168.1.1) port 0 AF_INET : +/-2.500% @ 95% conf. : demo
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. KBytes/sec
current 2280 16384 1432 60.00 10085.23
patched 2280 16384 1432 60.00 10089.98
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 (192.168.1.1) port 0 AF_INET : +/-2.500% @ 95% conf. : demo
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. KBytes/sec
current 18000 16384 16384 60.00 53644.35
patched 18000 16384 16384 60.00 51892.59
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 (192.168.1.1) port 0 AF_INET : +/-2.500% @ 95% conf. : demo
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. KBytes/sec
current 18000 16384 1432 60.00 37559.94
patched 18000 16384 1432 60.00 40630.66
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 (192.168.1.1) port 0 AF_INET : +/-2.500% @ 95% conf. : demo
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. KBytes/sec
current 28000 16384 16384 60.00 103766.68
patched 28000 16384 16384 60.00 93277.98
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 (192.168.1.1) port 0 AF_INET : +/-2.500% @ 95% conf. : demo
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. KBytes/sec
current 28000 16384 1432 60.00 73494.97
patched 28000 16384 1432 60.00 65312.94
UDP UNIDIRECTIONAL SEND TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 (192.168.1.1) port 0 AF_INET : +/-2.500% @ 95% conf. : demo
Socket Message Elapsed Messages
Size Size Time Okay Errors Throughput
bytes bytes secs # # KBytes/sec
current 212992 65507 60.00 252087 0 268773.83
current 212992 60.00 237435 253151.28
patched 212992 65507 60.00 254997 0 271876.61
patched 212992 60.00 226636 241637.51
UDP UNIDIRECTIONAL SEND TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 (192.168.1.1) port 0 AF_INET : +/-2.500% @ 95% conf. : demo
Socket Message Elapsed Messages
Size Size Time Okay Errors Throughput
bytes bytes secs # # KBytes/sec
current 212992 65507 60.00 252586 0 269305.73
current 2280 60.00 229371 244553.96
patched 212992 65507 60.00 256209 0 273168.32
patched 2280 60.00 201670 215019.54
UDP UNIDIRECTIONAL SEND TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 (192.168.1.1) port 0 AF_INET : +/-2.500% @ 95% conf. : demo
Socket Message Elapsed Messages
Size Size Time Okay Errors Throughput
bytes bytes secs # # KBytes/sec
current 212992 1432 60.00 5474009 0 127583.97
current 2280 60.00 4282184 99805.83
patched 212992 1432 60.00 5521903 0 128700.40
patched 2280 60.00 4522227 105400.70
UDP UNIDIRECTIONAL SEND TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 (192.168.1.1) port 0 AF_INET : +/-2.500% @ 95% conf. : demo
Socket Message Elapsed Messages
Size Size Time Okay Errors Throughput
bytes bytes secs # # KBytes/sec
current 212992 65507 60.00 251315 0 267950.39
current 18000 60.00 227856 242938.94
patched 212992 65507 60.00 256529 0 273509.57
patched 18000 60.00 224780 239658.47
UDP UNIDIRECTIONAL SEND TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 (192.168.1.1) port 0 AF_INET : +/-2.500% @ 95% conf. : demo
Socket Message Elapsed Messages
Size Size Time Okay Errors Throughput
bytes bytes secs # # KBytes/sec
current 212992 1432 60.00 5484882 0 127837.39
current 18000 60.00 5458678 127226.64
patched 212992 1432 60.00 5585209 0 130175.81
patched 18000 60.00 5560244 129593.94
UDP UNIDIRECTIONAL SEND TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 (192.168.1.1) port 0 AF_INET : +/-2.500% @ 95% conf. : demo
Socket Message Elapsed Messages
Size Size Time Okay Errors Throughput
bytes bytes secs # # KBytes/sec
current 212992 65507 60.00 250322 0 266891.35
current 28000 60.00 191711 204401.50
patched 212992 65507 60.00 257466 0 274508.96
patched 28000 60.00 241620 257613.85
UDP UNIDIRECTIONAL SEND TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.1.1 (192.168.1.1) port 0 AF_INET : +/-2.500% @ 95% conf. : demo
Socket Message Elapsed Messages
Size Size Time Okay Errors Throughput
bytes bytes secs # # KBytes/sec
current 212992 1432 60.00 5532780 0 128953.85
current 28000 60.00 5507094 128355.17
patched 212992 1432 60.00 5576824 0 129980.46
patched 28000 60.00 5555917 129493.16
next prev parent reply other threads:[~2013-01-03 20:40 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-12-18 13:51 [PATCH] xen/netfront: improve truesize tracking Ian Campbell
2012-12-18 14:10 ` Konrad Rzeszutek Wilk
2012-12-18 14:15 ` Ian Campbell
2012-12-18 15:12 ` Eric Dumazet
2012-12-18 15:26 ` Ian Campbell
2012-12-18 16:13 ` Eric Dumazet
2012-12-18 16:22 ` Ian Campbell
2012-12-18 16:35 ` Eric Dumazet
2012-12-19 11:34 ` Sander Eikelenboom
2012-12-19 16:17 ` Eric Dumazet
2012-12-20 12:51 ` Sander Eikelenboom
2012-12-20 14:23 ` [Xen-devel] " Sander Eikelenboom
2012-12-20 14:58 ` Sander Eikelenboom
2012-12-20 15:39 ` Eric Dumazet
2012-12-21 11:21 ` Sander Eikelenboom
2012-12-21 18:33 ` Rick Jones
2013-01-03 20:40 ` Sander Eikelenboom [this message]
2013-01-07 13:41 ` Ian Campbell
2013-01-07 14:11 ` Sander Eikelenboom
2013-01-07 14:22 ` Ian Campbell
2013-01-07 15:26 ` Eric Dumazet
2013-01-07 16:17 ` Eric Dumazet
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=561084196.20130103214030@eikelenboom.it \
--to=linux@eikelenboom$(echo .)it \
--cc=Ian.Campbell@citrix$(echo .)com \
--cc=annie.li@oracle$(echo .)com \
--cc=erdnetdev@gmail$(echo .)com \
--cc=konrad.wilk@oracle$(echo .)com \
--cc=netdev@vger$(echo .)kernel.org \
--cc=rick.jones2@hp$(echo .)com \
--cc=xen-devel@lists$(echo .)xensource.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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