linux-hotplug.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Patrick Mansfield <patmans@us.ibm.com>
To: linux-hotplug@vger.kernel.org
Subject: Re: some timings with/without hotplug scripts and udev
Date: Wed, 17 Mar 2004 01:41:55 +0000	[thread overview]
Message-ID: <20040316174155.A3605@beaverton.ibm.com> (raw)
In-Reply-To: <20040316090613.A31508@beaverton.ibm.com>

On Tue, Mar 16, 2004 at 11:33:00PM +0100, Kay Sievers wrote:
> On Tue, Mar 16, 2004 at 09:06:13AM -0800, Patrick Mansfield wrote:
> > Times running without hotplug scripts, hotplug scripts without udev, and
> > with udev using 512 scsi_debug LUNs.
> > 
> > insmod was really this:
> > 
> > 	modprobe scsi_debug num_tgts2 max_luns\x16 delay=0
> > 
> > No hotplug scripts means using /bin/true for the kernel hotplug.
> > 
> > 			insmod		rmmod
> > no hotplug scripts	0:05.46		0:04.63
> > hotplug no udev		0:30.95		0:26.40
> > udev w/glibc current	0:40.84		0:36.20
> > 
> > So hotplug scripts take about 25 seconds, and udev adds about 10 seconds
> > on top of that.

> Are you running udevsend/udevd in the test?

Yes, I assume udevsend per the link installed by udev:

[elm3b79 misc-io]$ ls -l /etc/hotplug.d/default/udev.hotplug 
lrwxr-xr-x    1 root     root           14 Mar 16 16:02 /etc/hotplug.d/default/udev.hotplug -> /sbin/udevsend

> Any chance to run the same test with the applied patch, which lets
> udevsend wait until the real udev comes back?
> 
> I'm interested if this takes much longer. If we ensure that udevsend is
> the first program that runs by /sbin/hotplug, we can be sure to know any
> node name and attribute in any of the following scripts.

Not much change, so I guess that is a good for your idea, though I prefer
udev running scripts under /etc/udev.d. 

Ran on top of your previous mmap patch (that should not matter for glibc),
using glibc + scsi_debug LUNs, and ran with 1024 LUNs (did not measure
that before):

			insmod		rmmod
512 LUNs 		0:42.09		0:32.94
1024 LUNs		1:31.25		1:05.94

Figuring out what is run when with the current hotplug scripts is
confusing.

-- Patrick Mansfield


-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials
Free Linux tutorial presented by Daniel Robbins, President and CEO of
GenToo technologies. Learn everything from fundamentals to system
administration.http://ads.osdn.com/?ad_id\x1470&alloc_id638&op=click
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

      parent reply	other threads:[~2004-03-17  1:41 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-03-16 17:06 some timings with/without hotplug scripts and udev Patrick Mansfield
2004-03-16 22:33 ` Kay Sievers
2004-03-17  1:41 ` Patrick Mansfield [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=20040316174155.A3605@beaverton.ibm.com \
    --to=patmans@us.ibm.com \
    --cc=linux-hotplug@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 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).