All of lore.kernel.org
 help / color / mirror / Atom feed
From: David Ahern <dsa@cumulusnetworks.com>
To: alexander.h.duyck@redhat.com
Cc: "netdev@vger.kernel.org" <netdev@vger.kernel.org>
Subject: increase in time to delete an interface with 4.x kernels
Date: Mon, 27 Jul 2015 10:49:48 -0600	[thread overview]
Message-ID: <55B6612C.7050506@cumulusnetworks.com> (raw)

Hi Alex:

I believe you did the recent overhaul to the fib implementation. I am 
seeing dramatically higher times to delete an interface with an ipv4 
address in 4.2-rc3. perf-top points to update_suffix:

    PerfTop:   15834 irqs/sec  kernel:97.3%  exact:  0.0% [4000Hz 
cpu-clock],  (all, 4 CPUs)
-------------------------------------------------------------------------------------------

     74.69%  [kernel]       [k] update_suffix
      2.38%  [kernel]       [k] fib_table_flush
      2.20%  [kernel]       [k] fib6_walk_continue
      2.03%  [kernel]       [k] fib6_ifdown
      1.31%  [kernel]       [k] fib6_age


I have a simple script to create and assign an ipv4 address to 10k dummy 
interfaces:

l=0
for (( j = 1; j <= 40; j += 1))
do
	for (( k = 1 ; k <= 250  ; k += 1 ))
	do
		l=$((l + 1))
		ip link add dev dummy${l} type dummy
	  	ip addr add  72.$j.$k.1/24 dev dummy${l}
	  	ifconfig dummy${l} up
	done
done


and a counter script to delete them all:

k=$(ip link show | grep dummy | wc -l)
for (( j = 1; j <= k; j += 1))
do
	ip link del dev dummy${j}
done


Looking at v3.19:

# time ./tadd-dummy.sh

real    3m8.896s
user    0m7.104s
sys     0m22.020s


# time ./tdel-dummy.sh

real    7m18.207s
user    0m3.824s
sys     3m15.672s


And the time to delete 1 interface after all 10k have been created:
# time ip link del dev dummy6666

real    0m0.064s
user    0m0.000s
sys     0m0.020s


Contrast those times with 4.2.0-rc3+ running the exact same scripts

# time ./tadd-dummy.sh

real	2m51.044s
user	0m7.220s
sys	0m29.520s

#  time ip link del dev dummy6666

real	0m0.441s
user	0m0.000s
sys	0m0.416s

so here the time to delete 1 interface has gone up by more than 10x.


# time ./tdel-dummy.sh
^C

real	14m10.000s
user	0m0.528s
sys	13m14.728s

I killed the delete; after 14 minutes only ~2k+ interfaces had been deleted:

# ip link show | grep dummy | wc -l
7822

In 4.2.0-rc3 it seems to take about 60 seconds to delete 150 interfaces 
which is inline with the 1 interface time of 0.4 seconds.

David

             reply	other threads:[~2015-07-27 16:49 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-27 16:49 David Ahern [this message]
2015-07-27 17:36 ` increase in time to delete an interface with 4.x kernels Alexander Duyck
2015-07-27 20:08 ` [net PATCH] fib_trie: Drop unnecessary calls to leaf_pull_suffix Alexander Duyck
2015-07-27 21:02   ` David Ahern
2015-07-27 21:51   ` David Miller

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=55B6612C.7050506@cumulusnetworks.com \
    --to=dsa@cumulusnetworks.com \
    --cc=alexander.h.duyck@redhat.com \
    --cc=netdev@vger.kernel.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.