public inbox for netdev@vger.kernel.org 
 help / color / mirror / Atom feed
From: Andrew Gallatin <gallatin@myri•com>
To: Herbert Xu <herbert@gondor•apana.org.au>
Cc: David Miller <davem@davemloft•net>,
	brice@myri•com, sgruszka@redhat•com, netdev@vger•kernel.org
Subject: Re: [PATCH] myr10ge: again fix lro_gen_skb() alignment
Date: Tue, 28 Apr 2009 17:12:20 -0400	[thread overview]
Message-ID: <49F77134.9030907@myri.com> (raw)
In-Reply-To: <20090428152047.GB7549@gondor.apana.org.au>

For variety, I grabbed a different "slow" receiver.  This is another
2 CPU machine, but a dual-socket single-core opteron (Tyan S2895)

processor       : 0
vendor_id       : AuthenticAMD
cpu family      : 15
model           : 37
model name      : AMD Opteron(tm) Processor 252
stepping        : 1
cpu MHz         : 2611.738
cache size      : 1024 KB
fpu             : yes
fpu_exception   : yes
cpuid level     : 1
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge 
mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx mmxext fxsr_opt 
lm 3dnowext 3dnow rep_good pni lahf_lm
bogomips        : 5223.47
TLB size        : 1024 4K pages
clflush size    : 64
cache_alignment : 64
address sizes   : 40 bits physical, 48 bits virtual
power management: ts fid vid ttp

The sender was an identical machine running an ancient RHEL4 kernel
(2.6.9-42.ELsmp) and our downloadable (backported) driver.
(http://www.myri.com/ftp/pub/Myri10GE/myri10ge-linux.1.4.4.tgz)
I disabled LRO, on the sender.

Binding the IRQ to CPU0, and the netserver to CPU1 I see 8.1Gb/s with
LRO and 8.0Gb/s with GRO.

Binding the IRQ to CPU0, and the netserver to CPU0, I see 6.9Gb/s
with LRO and 5.5 Gb/s with GRO.  Monitoring the packet/byte counts
on the interface once per second, LRO looks like this:

        Ipkts       IBytes        Opkts       Obytes
       588992    891733888         9758       644028
       589610    892669540         9771       644886
       589079    891865606         9754       643764

And GRO looks like this:

       480309    727187826         7949       524634
       480032    726768448         7947       524502
       480000    726720000         7943       524238


Similarly, in this same scenario, binding the app/irq to the same
CPU and running mpstat -P 0 1 shows about 60%sys and 40% irq+softirq
while GRO shows about 45% sys and 55% irq+softirq.

I can't put my finger on it, but something about GRO is certainly
more expensive on these types of machines.  I wish there was some
way you could see it, since it happens on every older AMD I try
it on.  If you haven't been able to reproduce it, I'll see if I
can make it happen on a newer "slow" amd64 box I have tomorrow.


Drew

  parent reply	other threads:[~2009-04-28 21:13 UTC|newest]

Thread overview: 41+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-04-15  8:09 [PATCH] myr10ge: again fix lro_gen_skb() alignment Stanislaw Gruszka
2009-04-15  9:28 ` David Miller
2009-04-15  9:48   ` Brice Goglin
2009-04-15 10:02     ` David Miller
2009-04-15 13:01       ` Andrew Gallatin
2009-04-15 21:04         ` Andrew Gallatin
2009-04-15 23:42           ` David Miller
2009-04-16  8:50             ` Herbert Xu
2009-04-16  9:02               ` David Miller
2009-04-21 19:19               ` Andrew Gallatin
2009-04-22 10:48                 ` Herbert Xu
2009-04-22 15:37                   ` Andrew Gallatin
2009-04-24  5:45                     ` Herbert Xu
2009-04-24 12:45                       ` Andrew Gallatin
2009-04-24 12:51                         ` Herbert Xu
2009-04-24 17:13                         ` Rick Jones
2009-04-24 16:16                       ` Andrew Gallatin
2009-04-24 16:30                         ` Herbert Xu
2009-04-24 16:31                           ` Herbert Xu
2009-04-27  8:05                         ` Herbert Xu
2009-04-27  8:07                           ` Herbert Xu
2009-04-27  9:32                             ` David Miller
2009-04-27 11:01                               ` Herbert Xu
2009-04-27 12:45                             ` David Miller
2009-04-27 12:45                           ` David Miller
2009-04-28  6:12                           ` Herbert Xu
2009-04-28 15:00                             ` Andrew Gallatin
2009-04-28 15:02                               ` David Miller
2009-04-28 15:20                               ` Herbert Xu
2009-04-28 15:44                                 ` Andrew Gallatin
2009-04-28 21:12                                 ` Andrew Gallatin [this message]
2009-04-29 13:42                                   ` Andrew Gallatin
2009-04-29 13:53                                     ` Eric Dumazet
2009-04-29 14:18                                       ` Andrew Gallatin
2009-04-29 15:26                                         ` Eric Dumazet
2009-04-29 17:28                                           ` Andrew Gallatin
2009-04-30  8:10                                             ` Herbert Xu
2009-04-30  8:14                                               ` Herbert Xu
2009-04-30  8:17                                             ` Eric Dumazet
2009-04-30 19:14                                               ` Andrew Gallatin
2009-04-23  8:00                 ` Herbert Xu

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=49F77134.9030907@myri.com \
    --to=gallatin@myri$(echo .)com \
    --cc=brice@myri$(echo .)com \
    --cc=davem@davemloft$(echo .)net \
    --cc=herbert@gondor$(echo .)apana.org.au \
    --cc=netdev@vger$(echo .)kernel.org \
    --cc=sgruszka@redhat$(echo .)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