Re: Bootup time regression from 2.6.27 to 2.6.28-rc3+

[Arjan van de Ven <arjan@infradead.org>]

On Sun, 9 Nov 2008 12:39:50 -0800 (PST)
Linus Torvalds <torvalds@linux-foundation.org> wrote:

> 
> 
> On Sun, 9 Nov 2008, Arjan van de Ven wrote:
> > 
> > for me, the plan is that we need to get Thomas' fixes tested by
> > someone who can reproduce this very reliably. If they fix it, great.
> > If they don't, time is running out and we need to remove the feature
> > for 2.6.28 (that is 1 line of change), realizing that this just
> > papers over stuff and isn't a real fix, and get back to it for
> > 2.6.29.
> 
> I'd like to see the one-liner fix, even if it turns out to just be a 
> workaround that hides the issue and makes us go back to 2.6.27
> behaviour. I'm about to release -rc4 today, we should be aggressive
> about regressions.

>From bf1003799b70f68f927d312841a05930856ad716 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Sun, 9 Nov 2008 12:45:10 -0800
Subject: [PATCH] regression: disable timer peek-ahead for 2.6.28

it's showing up as regressions; disabling it very likely just papers
over an underlying issue, but time is running out for 2.6.28,
lets get back to this for 2.6.29

Fixes: #11826 and #11893

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
---
 drivers/cpuidle/cpuidle.c |    4 +++-
 1 files changed, 3 insertions(+), 1 deletions(-)

diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c
index 5bed733..67ec4da 100644
--- a/drivers/cpuidle/cpuidle.c
+++ b/drivers/cpuidle/cpuidle.c
@@ -65,12 +65,14 @@ static void cpuidle_idle_call(void)
 		return;
 	}
 
+#if 0
+	/* shows regressions, re-enable for 2.6.29 */
 	/*
 	 * run any timers that can be run now, at this point
 	 * before calculating the idle duration etc.
 	 */
 	hrtimer_peek_ahead_timers();
-
+#endif
 	/* ask the governor for the next state */
 	next_state = cpuidle_curr_governor->select(dev);
 	if (need_resched())
-- 
1.6.0.3