[patch 40/45] sched: let +nice tasks have smaller impact

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
From: Greg KH
Date: Thursday, February 7, 2008 - 1:48 pm

2.6.24-stable review patch.  If anyone has any objections, please let us know.

------------------
From: Peter Zijlstra <a.p.zijlstra@chello.nl>

patch ef9884e6f29bbe1075204f962a00f7533bf7e8f3 in mainline.

Michel Dänzr has bisected an interactivity problem with
plus-reniced tasks back to this commit:

 810e95ccd58d91369191aa4ecc9e6d4a10d8d0c8 is first bad commit
 commit 810e95ccd58d91369191aa4ecc9e6d4a10d8d0c8
 Author: Peter Zijlstra <a.p.zijlstra@chello.nl>
 Date:   Mon Oct 15 17:00:14 2007 +0200

 sched: another wakeup_granularity fix

      unit mis-match: wakeup_gran was used against a vruntime

fix this by assymetrically scaling the vtime of positive reniced
tasks.

Bisected-by: Michel Dänzer <michel@tungstengraphics.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 kernel/sched_fair.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -867,7 +867,11 @@ static void check_preempt_wakeup(struct 
 	}
 
 	gran = sysctl_sched_wakeup_granularity;
-	if (unlikely(se->load.weight != NICE_0_LOAD))
+	/*
+	 * More easily preempt - nice tasks, while not making
+	 * it harder for + nice tasks.
+	 */
+	if (unlikely(se->load.weight > NICE_0_LOAD))
 		gran = calc_delta_fair(gran, &se->load);
 
 	if (pse->vruntime + gran < se->vruntime)

-- 
--
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
[patch 00/45] 2.6.24-stable review, Greg KH, (Thu Feb 7, 1:45 pm)
[patch 04/45] USB: keyspan: Fix oops, Greg KH, (Thu Feb 7, 1:46 pm)
[patch 06/45] USB: CP2101 New Device IDs, Greg KH, (Thu Feb 7, 1:46 pm)
[patch 10/45] USB: sierra driver - add devices, Greg KH, (Thu Feb 7, 1:47 pm)
[patch 17/45] USB: use GFP_NOIO in reset path, Greg KH, (Thu Feb 7, 1:47 pm)
[patch 21/45] USB: handle idVendor of 0x0000, Greg KH, (Thu Feb 7, 1:47 pm)
[patch 23/45] forcedeth: mac address mcp77/79, Greg KH, (Thu Feb 7, 1:47 pm)
[patch 24/45] lockdep: annotate epoll, Greg KH, (Thu Feb 7, 1:47 pm)
[patch 26/45] PCI: Fix fakephp deadlock, Greg KH, (Thu Feb 7, 1:47 pm)
[patch 27/45] ACPI: update ACPI blacklist, Greg KH, (Thu Feb 7, 1:47 pm)
[patch 31/45] b43: Fix suspend/resume, Greg KH, (Thu Feb 7, 1:47 pm)
[patch 34/45] b43legacy: fix PIO crash, Greg KH, (Thu Feb 7, 1:47 pm)
[patch 35/45] b43legacy: fix suspend/resume, Greg KH, (Thu Feb 7, 1:48 pm)
[patch 39/45] b43: Reject new firmware early, Greg KH, (Thu Feb 7, 1:48 pm)
[patch 40/45] sched: let +nice tasks have smaller impact, Greg KH, (Thu Feb 7, 1:48 pm)
Re: [patch 00/45] 2.6.24-stable review, S.Çağlar, (Thu Feb 7, 2:41 pm)
Re: [patch 00/45] 2.6.24-stable review, Greg KH, (Thu Feb 7, 2:58 pm)
Re: [patch 00/45] 2.6.24-stable review, Chuck Ebbert, (Thu Feb 7, 4:50 pm)
Re: [patch 00/45] 2.6.24-stable review, David Chinner, (Thu Feb 7, 5:44 pm)
Re: [stable] [patch 00/45] 2.6.24-stable review, David Chinner, (Thu Feb 7, 6:24 pm)
Re: [stable] [patch 00/45] 2.6.24-stable review, Greg KH, (Thu Feb 7, 10:31 pm)
Re: [patch 43/45] x86: replace LOCK_PREFIX in futex.h, Stefan Lippers-Hollmann, (Fri Feb 8, 11:35 am)
Re: [patch 43/45] x86: replace LOCK_PREFIX in futex.h, Chuck Ebbert, (Fri Feb 8, 2:45 pm)