Re: 2.6.25.8-rt7

Re: 2.6.25.8-rt7

[Gregory Haskins]

(Sorry for toppost)

I believe this is my fault.  You must be building for uniprocessor?  Ingo has a fix for this issue in sched-devel that should be pulled into -rt.

Sorry for the inconvenience.  I will pull the patch myself and submit it when I get home (if someone doesn't beat me to it)

-greg 
-----Original Message-----
From: Juergen Beisert <jbe@pengutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: LKML <linux-kernel@vger.kernel.org>
To:  <linux-rt-users@vger.kernel.org>

Sent: 6/24/2008 8:15:49 AM
Subject: Re: 2.6.25.8-rt7

On Tuesday 24 June 2008 03:34, Steven Rostedt wrote:
> We are pleased to announce the 2.6.25.8-rt7 tree, which can be
> downloaded from the location:
>
>   http://rt.et.redhat.com/download/
>
> Information on the RT patch can be found at:
>
>   http://rt.wiki.kernel.org/index.php/Main_Page
>
> Changes since 2.6.25.4-rt6
>
>   - ported to 2.6.25.8
>
>   - rwlock pi_lock fixes on waking up multiple readers (Steven Rostedt)
>
>   - revert adaptive locks from RCU (Thomas Gleixner)
>      [ for some reason, using RCU causes some latency spikes, this
>        may be a sign of something going wrong in RCU itself ]
>
>   - rwlock mutex owner fix for waking up reader with readers held
>                                                   (Steven Rostedt)
>
>   - rwlock remove waiter from pi_list on above fix (Steven Rostedt and
>                                                     Thomas Gleixner)
>
>   - rwlock replace macro with inline (Thomas Gleixner)
>
>   - rwlock pending namespace updates (Thomas Gleixner)
>
>   - futex fix fault (Thomas Gleixner)
>
> to build a 2.6.25.8-rt7 tree, the following patches should be applied:
>
>   http://www.kernel.org/pub/linux/kernel/v2.6/linux-2.6.25.tar.bz2
>   http://kernel.org/pub/linux/kernel/v2.6/patch-2.6.25.8.bz2
>   http://rt.et.redhat.com/download/patch-2.6.25.8-rt7.bz2

This occurs when I try to build a PowerPC (MPC5200) kernel:

[...]
In file included from kernel/sched.c:1318:
kernel/sched.c: In function 'sched_init':
kernel/sched.c:7702: error: implicit declaration of function 'global_rt_runtime'
[...]

And when I'm using the patch from Uli Brueggemann to fix it, I get:

[...]
In file included from kernel/sched.c:1325:
kernel/sched_rt.c: In function 'inc_rt_tasks':
kernel/sched_rt.c:290: error: 'struct rq' has no member named 'online'
[...]

Any hints?

Juergen Beisert

-- 
Dipl.-Ing. Juergen Beisert | http://www.pengutronix.de
 Pengutronix - Linux Solutions for Science and Industry
    Handelsregister: Amtsgericht Hildesheim, HRA 2686
         Vertretung Sued/Muenchen, Germany
   Phone: +49-8766-939 228 |  Fax: +49-5121-206917-9
--
To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: 2.6.25.8-rt7

[Steven Rostedt]

On Tue, 24 Jun 2008, Gregory Haskins wrote:

>
> (Sorry for toppost)
>
> I believe this is my fault.  You must be building for uniprocessor?  Ingo has a fix for this issue in sched-devel that should be pulled into -rt.
>
> Sorry for the inconvenience.  I will pull the patch myself and submit it when I get home (if someone doesn't beat me to it)
>

I'm looking forward to it ;-)

>
> This occurs when I try to build a PowerPC (MPC5200) kernel:
>
> [...]
> In file included from kernel/sched.c:1318:
> kernel/sched.c: In function 'sched_init':
> kernel/sched.c:7702: error: implicit declaration of function 'global_rt_runtime'
> [...]

I'm waiting on a Signed-off-by to include the patch. If I don't get it
soon then I'll just sign it off my self and leave the "From" line intact.
But I would feel much more comfortable with an actual SOB.

-- Steve

Re: 2.6.25.8-rt7

[Juergen Beisert]

On Tuesday 24 June 2008 16:31, Gregory Haskins wrote:
> I believe this is my fault.  You must be building for uniprocessor?  Ingo
> has a fix for this issue in sched-devel that should be pulled into -rt.

I can avoid this failure, when I disable the "Group scheduling for 
SCHED_RR/FIFO" menu entry. Don't know the impact.

Juergen

-- 
Dipl.-Ing. Juergen Beisert | http://www.pengutronix.de
 Pengutronix - Linux Solutions for Science and Industry
    Handelsregister: Amtsgericht Hildesheim, HRA 2686
         Vertretung Sued/Muenchen, Germany
   Phone: +49-8766-939 228 |  Fax: +49-5121-206917-9

Re: 2.6.25.8-rt7

[Gregory Haskins]

[-- Attachment #1: Type: text/plain, Size: 769 bytes --]

>>> On Tue, Jun 24, 2008 at 10:44 AM, in message
<Pine.LNX.4.58.0806241043070.32368@gandalf.stny.rr.com>, Steven Rostedt
<rostedt@goodmis.org> wrote: 

> On Tue, 24 Jun 2008, Gregory Haskins wrote:
> 
>>
>> (Sorry for toppost)
>>
>> I believe this is my fault.  You must be building for uniprocessor?  Ingo 
> has a fix for this issue in sched-devel that should be pulled into -rt.
>>
>> Sorry for the inconvenience.  I will pull the patch myself and submit it 
> when I get home (if someone doesn't beat me to it)
>>
> 
> I'm looking forward to it ;-)

Patch is attached (from linux-2.6-tip).  I meant to send this back around when the
offending patch went in, but it slipped through the cracks.  Sorry for the
inconvenience.

Regards,
-Greg


[-- Attachment #2: sched-fix-cpuprio-build-bug.patch --]
[-- Type: text/plain, Size: 1246 bytes --]

>From 1100ac91b6af02d8639d518fad5b434b1bf44ed6 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Thu, 5 Jun 2008 12:25:37 +0200
Subject: [PATCH] sched: fix cpuprio build bug

this patch was not built on !SMP:

 kernel/sched_rt.c: In function 'inc_rt_tasks':
 kernel/sched_rt.c:404: error: 'struct rq' has no member named 'online'

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched_rt.c |    5 ++++-
 1 files changed, 4 insertions(+), 1 deletions(-)

diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index e482159..eaa6060 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -399,16 +399,19 @@ void inc_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
 #if defined CONFIG_SMP || defined CONFIG_RT_GROUP_SCHED
 	if (rt_se_prio(rt_se) < rt_rq->highest_prio) {
 		struct rq *rq = rq_of_rt_rq(rt_rq);
-		rt_rq->highest_prio = rt_se_prio(rt_se);
 
+		rt_rq->highest_prio = rt_se_prio(rt_se);
+#ifdef CONFIG_SMP
 		if (rq->online)
 			cpupri_set(&rq->rd->cpupri, rq->cpu,
 				   rt_se_prio(rt_se));
+#endif
 	}
 #endif
 #ifdef CONFIG_SMP
 	if (rt_se->nr_cpus_allowed > 1) {
 		struct rq *rq = rq_of_rt_rq(rt_rq);
+
 		rq->rt.rt_nr_migratory++;
 	}
 
-- 
1.5.4.5

Re: 2.6.25.8-rt7

[Steven Rostedt]

On Tue, 24 Jun 2008, Juergen Beisert wrote:

>
> On Tuesday 24 June 2008 16:31, Gregory Haskins wrote:
> > I believe this is my fault.  You must be building for uniprocessor?  Ingo
> > has a fix for this issue in sched-devel that should be pulled into -rt.
>
> I can avoid this failure, when I disable the "Group scheduling for
> SCHED_RR/FIFO" menu entry. Don't know the impact.
>

Disabling that shouldn't effect you. [ CC'd Peter who can tell you better]

-- Steve

Re: 2.6.25.8-rt7

[Juergen Beisert]

On Tuesday 24 June 2008 03:34, Steven Rostedt wrote:
> We are pleased to announce the 2.6.25.8-rt7 tree, which can be
> downloaded from the location:
>
>   http://rt.et.redhat.com/download/
>
> Information on the RT patch can be found at:
>
>   http://rt.wiki.kernel.org/index.php/Main_Page
>
> Changes since 2.6.25.4-rt6
>
>   - ported to 2.6.25.8
>
>   - rwlock pi_lock fixes on waking up multiple readers (Steven Rostedt)
>
>   - revert adaptive locks from RCU (Thomas Gleixner)
>      [ for some reason, using RCU causes some latency spikes, this
>        may be a sign of something going wrong in RCU itself ]
>
>   - rwlock mutex owner fix for waking up reader with readers held
>                                                   (Steven Rostedt)
>
>   - rwlock remove waiter from pi_list on above fix (Steven Rostedt and
>                                                     Thomas Gleixner)
>
>   - rwlock replace macro with inline (Thomas Gleixner)
>
>   - rwlock pending namespace updates (Thomas Gleixner)
>
>   - futex fix fault (Thomas Gleixner)
>
> to build a 2.6.25.8-rt7 tree, the following patches should be applied:
>
>   http://www.kernel.org/pub/linux/kernel/v2.6/linux-2.6.25.tar.bz2
>   http://kernel.org/pub/linux/kernel/v2.6/patch-2.6.25.8.bz2
>   http://rt.et.redhat.com/download/patch-2.6.25.8-rt7.bz2

This occurs when I try to build a PowerPC (MPC5200) kernel:

[...]
In file included from kernel/sched.c:1318:
kernel/sched.c: In function 'sched_init':
kernel/sched.c:7702: error: implicit declaration of function 'global_rt_runtime'
[...]

And when I'm using the patch from Uli Brueggemann to fix it, I get:

[...]
In file included from kernel/sched.c:1325:
kernel/sched_rt.c: In function 'inc_rt_tasks':
kernel/sched_rt.c:290: error: 'struct rq' has no member named 'online'
[...]

Any hints?

Juergen Beisert

-- 
Dipl.-Ing. Juergen Beisert | http://www.pengutronix.de
 Pengutronix - Linux Solutions for Science and Industry
    Handelsregister: Amtsgericht Hildesheim, HRA 2686
         Vertretung Sued/Muenchen, Germany
   Phone: +49-8766-939 228 |  Fax: +49-5121-206917-9

2.6.25.8-rt7

[Steven Rostedt]

We are pleased to announce the 2.6.25.8-rt7 tree, which can be
downloaded from the location:

  http://rt.et.redhat.com/download/

Information on the RT patch can be found at:

  http://rt.wiki.kernel.org/index.php/Main_Page

Changes since 2.6.25.4-rt6

  - ported to 2.6.25.8

  - rwlock pi_lock fixes on waking up multiple readers (Steven Rostedt)

  - revert adaptive locks from RCU (Thomas Gleixner)
     [ for some reason, using RCU causes some latency spikes, this
       may be a sign of something going wrong in RCU itself ]

  - rwlock mutex owner fix for waking up reader with readers held
                                                  (Steven Rostedt)

  - rwlock remove waiter from pi_list on above fix (Steven Rostedt and
                                                    Thomas Gleixner)

  - rwlock replace macro with inline (Thomas Gleixner)

  - rwlock pending namespace updates (Thomas Gleixner)

  - futex fix fault (Thomas Gleixner)

to build a 2.6.25.8-rt7 tree, the following patches should be applied:

  http://www.kernel.org/pub/linux/kernel/v2.6/linux-2.6.25.tar.bz2
  http://kernel.org/pub/linux/kernel/v2.6/patch-2.6.25.8.bz2
  http://rt.et.redhat.com/download/patch-2.6.25.8-rt7.bz2



And like always, my RT version of Matt Mackall's ketchup will get this
for you nicely:

  http://people.redhat.com/srostedt/rt/tools/ketchup-0.9.8-rt3


The broken out patches are also available.


-- Steve