From: Zwane Mwaikambo <zwane@arm.linux.org.uk>
To: "Luck, Tony" <tony.luck@intel.com>
Cc: hawkes@sgi.com, Tony Luck <tony.luck@gmail.com>,
Andrew Morton <akpm@osdl.org>,
linux-ia64@vger.kernel.org, linux-kernel@vger.kernel.org,
Jack Steiner <steiner@sgi.com>, Keith Owens <kaos@sgi.com>,
Dimitri Sivanich <sivanich@sgi.com>
Subject: Re: [PATCH] ia64: disable preemption in udelay()
Date: Thu, 15 Dec 2005 17:52:02 -0800 (PST) [thread overview]
Message-ID: <Pine.LNX.4.64.0512151750500.1678@montezuma.fsmlabs.com> (raw)
In-Reply-To: <20051215225040.GA9086@agluck-lia64.sc.intel.com>
On Thu, 15 Dec 2005, Luck, Tony wrote:
> On Wed, Dec 14, 2005 at 03:25:26PM -0800, hawkes@sgi.com wrote:
> > Sending this to a wider audience:
> >
> > The udelay() inline for ia64 uses the ITC. If CONFIG_PREEMPT is enabled
> > and the platform has unsynchronized ITCs and the calling task migrates
> > to another CPU while doing the udelay loop, then the effective delay may
> > be too short or very, very long.
> >
> > The most simple fix is to disable preemption around the udelay looping.
> > The downside is that this inhibits realtime preemption for cases of long
> > udelays. One datapoint: an SGI realtime engineer reports that if
> > CONFIG_PREEMPT is turned off, that no significant holdoffs are
> > are attributed to udelay().
> >
> > I am reluctant to propose a much more complicated patch (that disables
> > preemption only for "short" delays, and uses the global RTC as the time
> > base for longer, preemptible delays) unless this patch introduces
> > significant and unacceptable preemption delays.
>
> Stuck between a rock and the proverbial hard place.
>
> I think that the more complex patch is needed though. If some crazy
> driver has a pre-emptible udelay(10000), then you really don't want
> to spin for that long without allowing preemption.
If it's a preemptible sleep period it should just use msleep.
next prev parent reply other threads:[~2005-12-16 1:46 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-12-14 23:25 [PATCH] ia64: disable preemption in udelay() hawkes
2005-12-15 22:50 ` Luck, Tony
2005-12-16 1:04 ` Luck, Tony
2005-12-16 8:20 ` Christian Hildner
2005-12-16 14:14 ` Alan Cox
2005-12-16 1:52 ` Zwane Mwaikambo [this message]
2005-12-16 2:03 ` Lee Revell
2005-12-16 2:12 ` John Hawkes
2005-12-16 2:40 ` Zwane Mwaikambo
2005-12-16 3:19 ` Lee Revell
2005-12-22 21:45 ` Bill Davidsen
2005-12-23 0:14 ` Keith Owens
2005-12-23 6:03 ` Zwane Mwaikambo
2005-12-16 2:37 ` Zwane Mwaikambo
2005-12-16 2:42 hawkes
2005-12-16 12:28 ` Robin Holt
2005-12-16 17:33 ` Luck, Tony
2005-12-16 18:39 ` John Hawkes
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=Pine.LNX.4.64.0512151750500.1678@montezuma.fsmlabs.com \
--to=zwane@arm.linux.org.uk \
--cc=akpm@osdl.org \
--cc=hawkes@sgi.com \
--cc=kaos@sgi.com \
--cc=linux-ia64@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=sivanich@sgi.com \
--cc=steiner@sgi.com \
--cc=tony.luck@gmail.com \
--cc=tony.luck@intel.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).