Subject: Linux-RT 2.6.25-rt
From: http://www.kernel.org/pub/linux/kernel/projects/rt/
Acked-by: Sven-Thorsten Dietrich <sdietrich@suse.de>
From lclaudio@uudg.org Fri Aug 17 21:40:37 2007
Return-Path: <lclaudio@uudg.org>
X-Spam-Checker-Version: SpamAssassin 3.1.7-deb (2006-10-05) on debian
X-Spam-Level:
X-Spam-Status: No, score=0.0 required=5.0 tests=none autolearn=ham
version=3.1.7-deb
Received: from sr02-01.mta.terra.com.br (sr02-01.mta.terra.com.br
[200.154.152.30]) by mail.tglx.de (Postfix) with ESMTP id 2E9BC65C3D9 for
<tglx@linutronix.de>; Fri, 17 Aug 2007 21:40:37 +0200 (CEST)
Received: from tiaro.hst.terra.com.br (tiaro.hst.terra.com.br
[200.176.10.7]) by bundure.hst.terra.com.br (Postfix) with ESMTP id
459344D7005C; Fri, 17 Aug 2007 16:40:34 -0300 (BRT)
X-Terra-Karma: -2%
X-Terra-Hash: 9bbc9fa12a67f4c16ad599245ee6a8fb
Received-SPF: none (tiaro.hst.terra.com.br: 200.176.10.7 is neither
permitted nor denied by domain of uudg.org) client-ip=200.176.10.7;
envelope-from=lclaudio@uudg.org; helo=lclaudio.dyndns.org;
Received: from lclaudio.dyndns.org (unknown [189.4.11.102]) (authenticated
user lc_poa) by tiaro.hst.terra.com.br (Postfix) with ESMTP id 97492214136;
Fri, 17 Aug 2007 16:40:32 -0300 (BRT)
Received: by lclaudio.dyndns.org (Postfix, from userid 500) id 7530B117DC8;
Fri, 17 Aug 2007 16:37:07 -0300 (BRT)
Date: Fri, 17 Aug 2007 16:37:06 -0300
From: "Luis Claudio R. Goncalves" <lclaudio@uudg.org>
To: Thomas Gleixner <tglx@linutronix.de>, Ingo Molnar <mingo@elte.hu>
Subject: [PATCH] Fixes spurious system load spikes in /proc/loadavgrt
Message-ID: <20070817193706.GB18693@unix.sh>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.5.14 (2007-02-12)
X-Evolution-Source: imap://tglx%40linutronix.de@localhost:8993/
Content-Transfer-Encoding: 8bit
Hi!
The patch I sent to the list had a minor glitch in the path for the second
half of the diff. This is the fixed version.
Sorry for any disturbance!
Best regards,
Luis
Hello,
The values in /proc/loadavgrt are sometimes the real load and sometimes
garbage. As you can see in th tests below, it occurs from in 2.6.21.5-rt20
to 2.6.23-rc2-rt2. The code for calc_load(), in kernel/timer.c has not
changed much in -rt patches.
[lclaudio@lab sandbox]$ ls /proc/loadavg*
/proc/loadavg /proc/loadavgrt
[lclaudio@lab sandbox]$ uname -a
Linux lab.casa 2.6.21-34.el5rt #1 SMP PREEMPT RT Thu Jul 12 15:26:48 EDT 2007 x86_64 x86_64 x86_64 GNU/Linux
[lclaudio@lab sandbox]$ cat /proc/loadavg*
4.57 4.90 4.16 3/146 23499
0.44 0.98 1.78 0/146 23499
...
[lclaudio@lab sandbox]$ cat /proc/loadavg*
4.65 4.80 4.75 5/144 20720
23896.04 -898421.23 383170.94 2/144 20720
[root@neverland ~]# uname -a
Linux neverland.casa 2.6.21.5-rt20 #2 SMP PREEMPT RT Fri Jul 1318:31:38 BRT 2007 i686 athlon i386 GNU/Linux
[root@neverland ~]# cat /proc/loadavg*
0.16 0.16 0.15 1/184 11240
344.65 0.38 311.71 0/184 11240
[williams@torg ~]$ uname -a
Linux torg 2.6.23-rc2-rt2 #14 SMP PREEMPT RT Tue Aug 7 20:07:31 CDT 2007 x86_64 x86_64 x86_64 GNU/Linux
[williams@torg ~]$ cat /proc/loadavg*
0.88 0.76 0.57 1/257 7267
122947.70 103790.53 -564712.87 0/257 7267
---------->
Fixes spurious system load spikes observed in /proc/loadavgrt, as described in:
Bug 253103: /proc/loadavgrt issues weird results
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=253103
Signed-off-by: Luis Claudio R. Goncalves <lclaudio@uudg.org>
---
---
kernel/sched_rt.c | 7 +++++++
kernel/timer.c | 14 ++++++++------
2 files changed, 15 insertions(+), 6 deletions(-)
Index: linux-2.6.25.2-rt1/kernel/timer.c
===================================================================
--- linux-2.6.25.2-rt1.orig/kernel/timer.c 2008-05-08 16:19:11.000000000 -0400
+++ linux-2.6.25.2-rt1/kernel/timer.c 2008-05-08 16:19:11.000000000 -0400
@@ -973,23 +973,25 @@ unsigned long avenrun_rt[3];
static inline void calc_load(unsigned long ticks)
{
unsigned long active_tasks; /* fixed-point */
+ unsigned long active_rt_tasks; /* fixed-point */
static int count = LOAD_FREQ;
count -= ticks;
if (unlikely(count < 0)) {
active_tasks = count_active_tasks();
+ active_rt_tasks = count_active_rt_tasks();
do {
CALC_LOAD(avenrun[0], EXP_1, active_tasks);
CALC_LOAD(avenrun[1], EXP_5, active_tasks);
CALC_LOAD(avenrun[2], EXP_15, active_tasks);
- count += LOAD_FREQ;
- } while (count < 0);
#ifdef CONFIG_PREEMPT_RT
- active_tasks = count_active_rt_tasks();
- CALC_LOAD(avenrun_rt[0], EXP_1, active_tasks);
- CALC_LOAD(avenrun_rt[1], EXP_5, active_tasks);
- CALC_LOAD(avenrun_rt[2], EXP_15, active_tasks);
+ CALC_LOAD(avenrun_rt[0], EXP_1, active_tasks);
+ CALC_LOAD(avenrun_rt[1], EXP_5, active_tasks);
+ CALC_LOAD(avenrun_rt[2], EXP_15, active_tasks);
#endif
+ count += LOAD_FREQ;
+
+ } while (count < 0);
}
}
Index: linux-2.6.25.2-rt1/kernel/sched_rt.c
===================================================================
--- linux-2.6.25.2-rt1.orig/kernel/sched_rt.c 2008-05-08 16:19:06.000000000 -0400
+++ linux-2.6.25.2-rt1/kernel/sched_rt.c 2008-05-08 16:19:11.000000000 -0400
@@ -422,6 +422,13 @@ unsigned long rt_nr_uninterruptible(void
for_each_online_cpu(i)
sum += cpu_rq(i)->rt.rt_nr_uninterruptible;
+ /*
+ * Since we read the counters lockless, it might be slightly
+ * inaccurate. Do not allow it to go below zero though:
+ */
+ if (unlikely((long)sum < 0))
+ sum = 0;
+
return sum;
}