netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Robert Olsson <Robert.Olsson@data.slu.se>
To: Andrea Arcangeli <andrea@suse.de>
Cc: Robert Olsson <Robert.Olsson@data.slu.se>,
	Dipankar Sarma <dipankar@in.ibm.com>,
	"David S. Miller" <davem@redhat.com>,
	kuznet@ms2.inr.ac.ru, linux-kernel@vger.kernel.org,
	netdev@oss.sgi.com, paulmck@us.ibm.com, akpm@osdl.org
Subject: Re: route cache DoS testing and softirqs
Date: Thu, 1 Apr 2004 15:44:57 +0200	[thread overview]
Message-ID: <16492.7385.765579.771428@robur.slu.se> (raw)
In-Reply-To: <20040331225259.GT2143@dualathlon.random>


Andrea Arcangeli writes:

 > But I feel like we should change the softirq code so that the irqexit
 > runs only the softirq setup by the current (or nested) irq handler. This
 > way the timer irq will stop executing the softirqs posted by the network
 > stack and it may very well fix the problem completely. This definitely
 > will help fairness too.

 How would softirq's scheduled be softirq's be run? 

 For a definitive solution a think Alexey analysis make sense:

 "All of them happening outside of ksoftirqd are equally bad, unaccountable,
 uncontrollable and will show up in some situation."

FYI. I did some more experiments:

Simple sycall test: user app = ank-time-loop (does gettimeofday in a loop)
=========================================================================
Before

total    droppped tsquz    throttl  bh_enbl  ksoftird irqexit  other   
00000099 00000000 00000000 00000000 0000005d 00000005 000163a2 00000000
00000008 00000000 00000000 00000000 00000009 00000000 000162f6 00000000

ank-time-loop 

After 10 sec. 
000000c0 00000000 00000000 00000000 0000005d 00000005 0001c3d2 00000000
00000008 00000000 00000000 00000000 00000009 00000000 0001c2f3 00000000

So syscalls "in this context" seems to do_softirq() via irqexit.

Route DoS while BGP tables are loading. ank-time-loop is running.
================================================================
total    droppped tsquz    throttl  bh_enbl  ksoftird irqexit  other   
000000c2 00000000 00000000 00000000 00000042 00000005 0001aef8 00000000
00000008 00000000 00000000 00000000 00000017 00000001 0001adba 00000000

After run:
000b3a10 00000000 00000eda 00000000 000006f6 0000052e 00041c06 00000000
000b0500 00000000 00000ec1 00000000 0000075e 00000197 000419f2 00000000

Most softirq's are run from irqexit but we see local_bh_enable is running
softirq's too. And only a minority of softirq's is running under scheduler 
control. As discussed we before with 100Hz timer we would expect lower 
rates from irqexit.

Cheers.
						--ro

 

      parent reply	other threads:[~2004-04-01 13:44 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-03-29 18:45 route cache DoS testing and softirqs Dipankar Sarma
2004-03-29 22:29 ` Andrea Arcangeli
2004-03-30  5:06   ` Srivatsa Vaddagiri
2004-03-30  5:35     ` Srivatsa Vaddagiri
2004-03-30 15:11       ` Andrea Arcangeli
2004-03-31  2:36     ` Rusty Russell
2004-03-30 14:43   ` Dipankar Sarma
2004-03-30 19:53     ` Dipankar Sarma
2004-03-30 20:47       ` Andrea Arcangeli
2004-03-30 21:06         ` Dipankar Sarma
2004-03-30 21:27           ` Andrea Arcangeli
2004-03-30 21:29         ` Robert Olsson
2004-03-31  7:36           ` Dipankar Sarma
2004-03-30 20:05   ` kuznet
2004-03-30 20:28     ` Dipankar Sarma
2004-04-01  6:00       ` kuznet
2004-03-30 21:14     ` Andrea Arcangeli
2004-03-30 21:30       ` David S. Miller
2004-03-30 21:37         ` Andrea Arcangeli
2004-03-30 22:22           ` David S. Miller
2004-03-30 22:49             ` Andrea Arcangeli
2004-03-31 20:46               ` Dipankar Sarma
2004-03-31 21:31                 ` Andrea Arcangeli
2004-03-31 21:52                   ` Dipankar Sarma
2004-03-30 22:33           ` Robert Olsson
2004-03-31 17:10           ` Dipankar Sarma
2004-03-31 18:46             ` Robert Olsson
2004-03-31 20:37               ` Dipankar Sarma
2004-03-31 21:28                 ` Andrea Arcangeli
2004-03-31 21:43                   ` Dipankar Sarma
2004-04-05 17:11                     ` Robert Olsson
2004-04-05 21:22                       ` Dipankar Sarma
2004-04-06 12:55                         ` Robert Olsson
2004-04-06 19:52                           ` Dipankar Sarma
2004-04-07 15:23                             ` Robert Olsson
2004-04-07 19:48                               ` Dipankar Sarma
2004-04-08 13:29                           ` kuznet
2004-04-08 14:07                             ` Robert Olsson
2004-03-31 22:36                   ` Robert Olsson
2004-03-31 22:52                     ` Andrea Arcangeli
2004-04-01  6:43                       ` kuznet
2004-04-01 13:16                         ` Andrea Arcangeli
2004-04-08 13:38                           ` kuznet
2004-04-01 13:44                       ` Robert Olsson [this message]

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=16492.7385.765579.771428@robur.slu.se \
    --to=robert.olsson@data.slu.se \
    --cc=akpm@osdl.org \
    --cc=andrea@suse.de \
    --cc=davem@redhat.com \
    --cc=dipankar@in.ibm.com \
    --cc=kuznet@ms2.inr.ac.ru \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@oss.sgi.com \
    --cc=paulmck@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).