-END PGP SIGNATURE-
factor out priorities for workqueue.h and sched.h
From: Clark Williams [EMAIL PROTECTED]
This fixes a circular dependency between sched.h and workqueue.h by factoring out the common
defines to a new header which is included by both
Signed-off-by: Clark Williams [EMAIL
. User visable value is in milliseconds but is stored as
jiffies. Setting to 0 (zero) resets to the default (currently 100ms).
Patch against tip/master, currently 3.7-rc3.
Signed-off-by: Clark Williams willi...@redhat.com
---
include/linux/sched.h | 6 +-
kernel/sched.c| 19
, currently 3.7-rc3.
Signed-off-by: Clark Williams willi...@redhat.com
---
include/linux/sched.h | 7 ++-
kernel/sched/core.c | 30 ++
kernel/sched/rt.c | 4 ++--
kernel/sysctl.c | 7 +++
4 files changed, 45 insertions(+), 3 deletions(-)
diff --git
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Paul E. McKenney wrote:
On Fri, Jun 08, 2007 at 03:43:48PM -0400, Steven Rostedt wrote:
On Fri, 2007-06-08 at 12:36 -0700, Paul E. McKenney wrote:
On Fri, Jun 08, 2007 at 11:27:08AM -0400, Steven Rostedt wrote:
The first time I compiled it, I
On Mon, 8 Apr 2013 20:30:12 +0200
Sebastian Andrzej Siewior bige...@linutronix.de wrote:
* Paul Gortmaker | 2013-04-05 19:35:51 [-0400]:
Thanks for the work in putting this together. Just a heads up that the
split queue fails on a patch with no author/date/subject as follows:
[...]
git
On Thu, 7 Feb 2013 20:52:57 +0100
Ingo Molnar mi...@kernel.org wrote:
* Ingo Molnar mi...@kernel.org wrote:
* Ingo Molnar mi...@kernel.org wrote:
Yeah, that looks pretty good.
I'll give it some testing.
Found one build error so far, applied the fix below.
Another
On Fri, 8 Feb 2013 15:18:41 +0100
Ingo Molnar mi...@kernel.org wrote:
* Clark Williams willi...@redhat.com wrote:
On Thu, 7 Feb 2013 20:52:57 +0100
Ingo Molnar mi...@kernel.org wrote:
* Ingo Molnar mi...@kernel.org wrote:
* Ingo Molnar mi...@kernel.org wrote
On Fri, 8 Feb 2013 16:53:17 +0100
Frederic Weisbecker fweis...@gmail.com wrote:
2013/2/7 Christoph Lameter c...@linux.com:
On Thu, 7 Feb 2013, Frederic Weisbecker wrote:
Not with hrtick.
hrtick? Did we not already try that a couple of years back and it turned
out that the overhead of
On Fri, 8 Feb 2013 16:27:26 -0500
Josh Boyer jwbo...@redhat.com wrote:
On Fri, Feb 08, 2013 at 12:45:47PM -0800, Eric W. Biederman wrote:
Josh Boyer jwbo...@redhat.com writes:
Two emails fly past each other in the night
Yep.
My best guess in some dark corner of mock has
On Fri, 08 Feb 2013 14:05:55 -0800
ebied...@xmission.com (Eric W. Biederman) wrote:
Josh Boyer jwbo...@redhat.com writes:
So it looks mock is taking a buggy untested code path and things are not
working as it expected.
Quite possibly, yes. I instrumented the kernel a bit and it is
On Wed, 13 Feb 2013 10:15:12 +0100
Ingo Molnar mi...@kernel.org wrote:
* Namhyung Kim namhy...@kernel.org wrote:
Hi,
On Mon, 11 Feb 2013 10:54:58 +0100, Ingo Molnar wrote:
* Clark Williams willi...@redhat.com wrote:
I figured that was coming. :)
;-)
I'll look
On Sun, 3 Feb 2013 19:57:09 +0100
Ingo Molnar mi...@kernel.org wrote:
* Ingo Molnar mi...@kernel.org wrote:
The first patch looks good, I've applied it.
I've skipped this one after all, as it sometimes generates:
include/linux/sched/sysctl.h:108:41: warning: ‘struct signal_struct’
On Mon, 4 Feb 2013 15:58:26 +0100 (CET)
Thomas Gleixner t...@linutronix.de wrote:
Dear RT Folks,
I'm pleased to announce the 3.6.11-rt26 release.
Changes since 3.6.11-rt25:
1) Fix the RT highmem implementation on x86
2) Support highmem + RT on ARM
3) Fix an one off error
On Mon, 4 Feb 2013 09:54:38 -0600
Clark Williams willi...@redhat.com wrote:
On Mon, 4 Feb 2013 15:58:26 +0100 (CET)
Thomas Gleixner t...@linutronix.de wrote:
Dear RT Folks,
I'm pleased to announce the 3.6.11-rt26 release.
Changes since 3.6.11-rt25:
1) Fix the RT highmem
.
Clark Williams (3):
sched: move sched.h sysctl bits into separate header
sched/rt: add a tuning knob to allow changing SCHED_RR timeslice
sched/rt: move rt specific bits into new header file
block/blk-exec.c | 1 +
drivers/spi/spi.c | 2 +-
drivers
Add a /proc/sys/kernel scheduler knob named sched_rr_timeslice_ms
that allows global changing of the SCHED_RR timeslice value. User
visable value is in milliseconds but is stored as jiffies. Setting
to 0 (zero) resets to the default (currently 100ms).
Signed-off-by: Clark Williams willi
Move the sysctl-related bits from include/linux/sched.h into
a new file: include/linux/sched/sysctl.h. Then update source
files requiring access to those bits by including the new
header file.
Signed-off-by: Clark Williams willi...@redhat.com
---
block/blk-exec.c | 1 +
include
move rt scheduler definitions out of include/linux/sched.h into
new file include/linux/sched/rt.h
Signed-off-by: Clark Williams willi...@redhat.com
---
drivers/spi/spi.c | 2 +-
drivers/staging/csr/bh.c | 2 +-
drivers/staging/csr/unifi_sme.c | 2 +-
drivers/tty
On Tue, 29 Jan 2013 09:12:45 +0800
Li Zefan lize...@huawei.com wrote:
-extern void sched_autogroup_create_attach(struct task_struct *p);
-extern void sched_autogroup_detach(struct task_struct *p);
-extern void sched_autogroup_fork(struct signal_struct *sig);
-extern void
On Tue, 29 Jan 2013 15:05:43 +0900
Namhyung Kim namhy...@kernel.org wrote:
Hi Clark,
On Mon, 28 Jan 2013 14:31:20 -0600, Clark Williams wrote:
Move all the scheduler sysctl-related bits out of include/linux/sched.h
into a new file include/linux/sched_sysctl.h.
Signed-off-by: Clark
dir,
things like NO_HZ and CGROUPS but wanted to get this in place first.
Clark Williams (2):
sched: create new scheduler include dir and move sysctl defs there
sched/rt: add a tuning knob to allow changing SCHED_RR timeslice
include/linux/sched.h| 108
Add a /proc/sys/kernel scheduler knob named sched_rr_timeslice_ms
that allows global changing of the SCHED_RR timeslice value. User
visable value is in milliseconds but is stored as jiffies. Setting
to 0 (zero) resets to the default (currently 100ms).
Signed-off-by: Clark Williams willi
Create a new scheduler include directory, include/linux/sched, then
move all the sysctl bits from include/linus/sched.h into new file
include/linux/sched/sysctl.h.
Signed-off-by: Clark Williams willi...@redhat.com
---
include/linux/sched.h| 108
is if we should move them back into uaccess.h as
static inlines (like the upstream versions). I've done that in the
attached patch and it seems to work fine. What I'm not sure of are the
performance implications of calling migrate_{disable,enable} from a
static inline.
Thoughts?
Signed-off-by: Clark
On Thu, 24 Jan 2013 17:59:55 +0100
Ingo Molnar mi...@kernel.org wrote:
* Clark Williams willi...@redhat.com wrote:
This version stores the user-input value in a separate
location from the jiffies values used by the scheduler, to
prevent a race condition.
Subject: [PATCH v2
On Fri, 25 Jan 2013 09:19:16 +0100
Ingo Molnar mi...@kernel.org wrote:
* Clark Williams willi...@redhat.com wrote:
On Thu, 24 Jan 2013 17:59:55 +0100
Ingo Molnar mi...@kernel.org wrote:
* Clark Williams willi...@redhat.com wrote:
This version stores the user-input value
On Fri, 25 Jan 2013 19:36:10 +0100
Ingo Molnar mi...@kernel.org wrote:
* Clark Williams willi...@redhat.com wrote:
Shouldn't this be in kernel/sched/sched.h instead of the
(already too large) linux/sched.h?
I don't think that will work be cause kernel/sysctl.c needs
.
These patches are against tip/master.
Clark Williams (2):
sched: move scheduler sysctl bits into dedicated header file
sched/rt: add a tuning knob to allow changing RR timeslice
include/linux/sched.h| 102 +--
include/linux/sched_sysctl.h | 126
Move all the scheduler sysctl-related bits out of include/linux/sched.h
into a new file include/linux/sched_sysctl.h.
Signed-off-by: Clark Williams willi...@redhat.com
---
include/linux/sched.h| 102 +---
include/linux/sched_sysctl.h | 121
Add a /proc/sys/kernel scheduler knob named sched_rr_timeslice_ms
that allows global changing of the SCHED_RR timeslice value. User
visable value is in milliseconds but is stored as jiffies. Setting
to 0 (zero) resets to the default (currently 100ms).
Signed-off-by: Clark Williams willi
On Wed, 23 Jan 2013 15:09:38 -0500
Steven Rostedt rost...@goodmis.org wrote:
On Thu, 2012-11-08 at 14:27 -0600, Clark Williams wrote:
Subject: [PATCH] sched: add a tuning knob to allow changing RR timeslice
User wanted a facility simliar to the ability on Solaris to adjust
the SCHED_RR
Signed-off-by: Clark Williams clark.willi...@gmail.com
---
include/linux/sched/rt.h | 6 ++
include/linux/sched/sysctl.h | 6 --
2 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/include/linux/sched/rt.h b/include/linux/sched/rt.h
index 94e19ea..440434d 100644
Move the various scheduler clock definitions out of include/linux/sched.h
into a new file include/linux/sched/clock.h.
Signed-off-by: Clark Williams clark.willi...@gmail.com
---
arch/arm/kernel/sched_clock.c | 2 +-
arch/x86/kernel/cpu/amd.c | 2 +-
arch/x86/kernel/cpu/intel.c | 2
Ingo,
These two patches continue the sched.h reorganizatin. The first is a simple
move of the RR_TIMESLICE definition from sysctl to rt.h.
The second moves the clock-related items from sched.h into sched/clock.h
Tested against allmodconfigs for x86_64, arm, mips, alpha and ia64.
Clark
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Ingo Molnar wrote:
ok - lets try it. Clark: does this impact the set_kthread_prio utility?
I've changed IRQ 123 to IRQ-123 to make pidof friends work better.
Yes it does, but I believe I can fix that. Most of the logic is in awk
and I believe
On Fri, 07 Dec 2012 18:56:15 -0500
Steven Rostedt rost...@goodmis.org wrote:
I've been debugging large latencies on a 40 core box and found a major
cause due to the thundering herd like grab of the rq lock due to the
pull_rt_task() logic.
Basically, if a large number of CPUs were to lower
retry;
}
}
Thomas,
I had to add this to get -rt24 to compile:
commit cb25d4b84a6d8f155134f6d9aa2997a452ab1dca
Author: Clark Williams willi...@redhat.com
Date: Fri Dec 21 09:22:26 2012 -0600
[RT] pull in include/linux/delay.h to get cpu_chill()
Signed-off
Sebastian,
I needed this for my laptop on 3.12, so tweaked it to apply properly.
Clark
commit 7b1573eaa0f176731bc231b249598b6e4850babf
Author: Clark Williams clark.willi...@gmail.com
Date: Tue Nov 12 12:17:07 2013 -0600
net: iwlwifi: request only a threaded handler for interrupts
On Tue, 12 Nov 2013 13:16:59 -0600
Clark Williams willi...@redhat.com wrote:
Sebastian,
I needed this for my laptop on 3.12, so tweaked it to apply properly.
Clark
So of course I sent the wrong patch:
commit 49d487614d56bc10969dd0dcbce709825fc06d0e
Author: Clark Williams clark.willi
a named fifo to
be created so that an external program can read run statistics while
the run is in process. This one is early days yet so any suggestions,
corrections, flames, etc that can improve this are welcome.
Here's the changelog:
* Fri Nov 15 2013 Clark Williams willi...@redhat.com - 0.86-1
On Mon, 29 Apr 2013 22:12:02 +0200
Sebastian Andrzej Siewior bige...@linutronix.de wrote:
- suspend / resume seems to program program the timer wrong and wait
ages until it continues.
It has to be something we're doing when we apply RT to v3.8.x, since
v3.8.x suspends/resumes with
On Tue, 30 Apr 2013 19:09:48 +0200
Sebastian Andrzej Siewior bige...@linutronix.de wrote:
* Clark Williams | 2013-04-29 16:19:25 [-0500]:
On Mon, 29 Apr 2013 22:12:02 +0200
Sebastian Andrzej Siewior bige...@linutronix.de wrote:
- suspend / resume seems to program program the timer
On Tue, 30 Apr 2013 14:18:24 -0500
Clark Williams willi...@redhat.com wrote:
On Tue, 30 Apr 2013 19:09:48 +0200
Sebastian Andrzej Siewior bige...@linutronix.de wrote:
* Clark Williams | 2013-04-29 16:19:25 [-0500]:
On Mon, 29 Apr 2013 22:12:02 +0200
Sebastian Andrzej Siewior bige
Move the softlockup watchdog and related prototypes into a separate
header and update source that requires them.
Done against tip/master.
Tested against x86_64 allmodconfig.
Signed-off-by: Clark Williams clark.willi...@gmail.com
---
arch/x86/xen/smp.c | 2 ++
drivers/ide/ide
On Thu, 25 Apr 2013 17:09:12 +0200
Sebastian Andrzej Siewior bige...@linutronix.de wrote:
On 04/08/2013 10:25 PM, Clark Williams wrote:
BTW I imported -rt onto v3.8.6 with no hiccups from 'git
quiltimport' other than what you just fixed. Not much runtime on it
but it booted without
On Thu, 25 Apr 2013 13:14:59 -0500
Clark Williams willi...@redhat.com wrote:
I notice that nouveau is in there and yes, my laptop uses nouveau as
well. First thing I'll try is some of the pm-suspend quirks regarding
graphics devices.
If that doesn't change anything I'll try swapping out
On Thu, 12 Dec 2013 16:47:19 +0200
Ciprian Barbu ciprian.ba...@linaro.org wrote:
Hello,
While playing around with hackbench I discovered that I would sometimes
get an enormous time reported, even if the run time would be less than a
second or so. The problem was that the struct timeval
I just pushed v0.88 of rt-tests up to kernel.org. This release has a
set of fixes for ARM systems from:
Uwe Kleine-König u.kleine-koe...@pengutronix.de:
- cyclictest: Fix tracemark output when the latency threshold is hit
on ARM
Gary S. Robertson gary.robert...@linaro.org:
- cyclictest:
On Wed, 9 Apr 2014 15:19:22 -0400
Steven Rostedt rost...@goodmis.org wrote:
This patch is built on top of the two other patches that I posted
earlier, which should not be as controversial.
If you have any benchmark on large machines I would be very happy if
you could test this patch
Verion 0.89 of rt-tests has been pushed to the git repository on
kernel.org:
git://git.kernel.org/pub/scm/linux/kernel/git/clrkwllms/rt-tests.git
tag: v0.89
This version fixes a problem with hwlatdetect on distros
where /usr/bin/python is linked to python3.
In addition it cleans up a leftover
the lock is actually an rt_mutex).
Signed-off-by: Clark Williams willi...@redhat.com
---
include/linux/rwsem_rt.h | 5 +
1 file changed, 5 insertions(+)
diff --git a/include/linux/rwsem_rt.h b/include/linux/rwsem_rt.h
index e94d945c5844..77035d7abc04 100644
--- a/include/linux/rwsem_rt.h
+++ b
rcutree.kthread_prio to be used on
the kernel command line at boot to set a new value
(rcutree.kthread_prio=N).
Signed-off-by: Clark Williams clark.willi...@gmail.com
---
Documentation/kernel-parameters.txt | 6 ++
init/Kconfig| 23 ---
kernel/rcu/tree_plugin.h
On Tue, 16 Sep 2014 13:41:56 -0700
Paul E. McKenney paul...@linux.vnet.ibm.com wrote:
On Mon, Sep 15, 2014 at 10:05:24AM -0500, Clark Williams wrote:
Paul,
Here's the result of our conversation Friday, regarding the per-CPU and
boost priorities:
Rename CONFIG_RCU_BOOST_PRIO
are built, but if one wants
to build with non-NPTL toolchain just add HAVE_NPTL=no in command line
or modify HAVE_NPTL variable right in Makefile and execute make.
Signed-off-by: Alexey Brodkin abrod...@synopsys.com
Cc: Vineet Gupta vgu...@synopsys.com
Cc: Clark Williams clark.willi
On Tue, 17 Feb 2015 11:09:11 +0100
Sebastian Andrzej Siewior bige...@linutronix.de wrote:
* Clark Williams | 2014-08-27 16:09:28 [-0500]:
Thomas,
The latest btrfs code uses rwsem_is_contended() in the function
caching_thread(). On RT systems, include/linux/rwsem.h is replaced
On Mon, 20 Apr 2015 14:15:04 -0700
Paul E. McKenney paul...@linux.vnet.ibm.com wrote:
On Mon, Apr 20, 2015 at 04:40:49PM -0400, Steven Rostedt wrote:
On Mon, Apr 20, 2015 at 10:09:03AM -0700, Paul E. McKenney wrote:
The sysfs knob might be nice, but as far as I know nobody has been
On Sat, 18 Apr 2015 19:05:42 -0700
Paul E. McKenney paul...@linux.vnet.ibm.com wrote:
Real-time priority to use for RCU worker threads
(RCU_KTHREAD_PRIO) [0] (NEW)
Indeed, Linus complained about this one. ;-)
:-) Yes, it's an essentially unanswerable question.
On Mon, 20 Apr 2015 10:09:03 -0700
Paul E. McKenney paul...@linux.vnet.ibm.com wrote:
On Mon, Apr 20, 2015 at 11:35:54AM -0500, Clark Williams wrote:
On Sat, 18 Apr 2015 19:05:42 -0700
Paul E. McKenney paul...@linux.vnet.ibm.com wrote:
Real-time priority to use for RCU worker
On Tue, 26 May 2015 11:19:24 -0400
Steven Rostedt rost...@goodmis.org wrote:
On Tue, 26 May 2015 08:48:02 -0500
Clark Williams willi...@redhat.com wrote:
Change the WARN_ON to WARN_ON_NORT
Do we have a WARN_ON_NORT? I see a WARN_ON_NONRT, but not a
WARN_ON_NORT. Does this compile
the interrupt
is actually handled in a threaded handler and not actual interrupt
context).
Ifdef it out for PREEMPT_RT_FULL.
Signed-off-by: Clark Williams willi...@redhat.com
---
drivers/gpu/drm/i915/intel_display.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/gpu/drm/i915
On Tue, 26 May 2015 09:38:32 -0400
Steven Rostedt rost...@goodmis.org wrote:
On Tue, 26 May 2015 08:34:49 -0500
Clark Williams willi...@redhat.com wrote:
diff --git a/drivers/gpu/drm/i915/intel_display.c
b/drivers/gpu/drm/i915/intel_display.c
index f75173c20f47..f532b22f56ba 100644
Sebastian,
Below is a traceback I hit while running 4.1.3-rt3 on my Lenovo T530.
I was doing my normal, play music, copy files over the lan, do compiles,
do email, etc., so I I can't really point you at a reproducer. The graphics
system stayed up somewhat but the actual trace I hit scrolled off.
: Clark Williams willi...@redhat.com
Date: Fri, 7 Aug 2015 15:07:30 -0500
Subject: [PATCH] hwlat_detector: update stats code to record when outer
interval exceeds threshold
The hwlat_detector polls the system clock (either ktime_get() or
trace_clock_local())
source looking for intervals between
On Sun, 5 Jun 2016 08:16:58 -0700
Alison Chaiken wrote:
> Steven Rostedt suggests in reference to "[PATCH][RT] netpoll: Always
> take poll_lock when doing polling"
> >> [ Alison, can you try this patch ]
>
> Sebastian follows up:
> >Alison, did you try it?
>
>
On Mon, 3 Aug 2015 14:27:00 -0500
Clark Williams <willi...@redhat.com> wrote:
> Sebastian,
>
> Below is a traceback I hit while running 4.1.3-rt3 on my Lenovo T530.
>
> I was doing my normal, play music, copy files over the lan, do compiles,
> do email, etc., so I
RT has dropped support of rcu_bh, comment out in rcutorture.
Signed-off-by: Clark Williams <willi...@redhat.com>
---
kernel/rcu/rcutorture.c | 7 +++
1 file changed, 7 insertions(+)
diff --git a/kernel/rcu/rcutorture.c b/kernel/rcu/rcutorture.c
index d89328e260df..5bb3364a6284
On Mon, 15 Feb 2016 22:53:58 -0600
Clark Williams <willi...@redhat.com> wrote:
> On Sat, 13 Feb 2016 00:47:08 +0100
> Sebastian Andrzej Siewior <bige...@linutronix.de> wrote:
>
> > - There is a useless rcu_bh thread which has been deactivated.
> >
&
rcutorture.c wanted to test the
(now ifdef'ed out) rcu_bh routines. This fixes the compile issue but is
untested for actually *running* rcutorture. I'll try that on a test box
tomorrow.
Clark
From a835ff3aad772520d34548dc336925788a649b17 Mon Sep 17 00:00:00 2001
From: Clark Williams <will
I'm hitting the WARN_ON(wakes > 2) in $SUBJECT when resuming from suspend on my
laptop (quad-core i7 with HT on). Looks like the warning gets hit 36 times on
resume. E.g.:
Call Trace:
[] dump_stack+0x65/0x85
[] warn_slowpath_common+0x82/0xd0
[] warn_slowpath_null+0x1a/0x20
[]
On Fri, 1 Apr 2016 12:33:18 +0200
Sebastian Andrzej Siewior <bige...@linutronix.de> wrote:
> * Thomas Gleixner | 2016-03-14 09:49:52 [+0100]:
>
> >On Sun, 13 Mar 2016, Clark Williams wrote:
> >
> >> I'm hitting the WARN_ON(wakes > 2) in $SUBJECT when resu
On Wed, 30 Mar 2016 12:22:51 +0200
Sebastian Andrzej Siewior <bige...@linutronix.de> wrote:
> * Thomas Gleixner | 2016-03-14 09:49:52 [+0100]:
>
> >On Sun, 13 Mar 2016, Clark Williams wrote:
> >
> >> I'm hitting the WARN_ON(wakes > 2) in $SUBJECT when resu
the marker to the trace buffers and then write a "0\n" to
the tracing_on file to turn off tracing, otherwise we lose the information
immediately prior to the point where we hit the latency.
Signed-off-by: Clark Williams <willi...@redhat.com>
---
src/cyclictest/c
On Tue, 3 May 2016 15:56:44 -0400
Luiz Capitulino <lcapitul...@redhat.com> wrote:
> On Tue, 3 May 2016 12:59:53 -0500
> Clark Williams <willi...@redhat.com> wrote:
>
> > John,
> >
> > This patch is against the devel/v0.98 branch. It turns off tracing i
Revert "vtime: Split lock and seqcount"
>
> which is what you want, correct?
> >
> > Signed-off-by: Rik van Riel <r...@redhat.com>
> > Signed-off-by: Clark Williams <willi...@redhat.com>
>
> Sebastian
>
Yes, that's it. I just want
with isolcpus/rcu_nocbs/nohz_full cpus. No ill effects seen.
Signed-off-by: Rik van Riel <r...@redhat.com>
Signed-off-by: Clark Williams <willi...@redhat.com>
---
kernel/sched/cputime.c | 18 --
1 file changed, 18 deletions(-)
diff --git a/kernel/sched/cputime.c b/
quot; when trying to place a measurement thread on an isolated
cpu.
This patch removes the wrapper function which uses libnuma cpumask parsing
functions and instead uses the parser function we wrote for when libnuma is not
available.
Signed-off-by: Clark Williams <willi...@redhat.com>
---
On Thu, 4 Aug 2016 11:30:33 -0400
Steven Rostedt wrote:
> Note, I'm currently working on adding code to detect NMIs to this as
> well. And perhaps even tracing SMI counters. Just to show what caused
> the latency, as latency isn't measured by the counters (that I know of).
>
On Mon, 7 Nov 2016 09:17:55 +0100
Daniel Bristot de Oliveira wrote:
> The rt throttling mechanism prevents the starvation of non-real-time
> tasks by CPU intensive real-time tasks. In terms of percentage,
> the default behavior allows real-time tasks to run up to 95% of a
>
On Mon, 7 Nov 2016 13:30:46 -0500
Steven Rostedt <rost...@goodmis.org> wrote:
> On Mon, 7 Nov 2016 12:22:21 -0600
> Clark Williams <willi...@redhat.com> wrote:
>
> > I'm still reviewing the patch, but I have to wonder why bother with making
> > it a scheduler
usb_hcd_irq(0, hcd);
> - local_irq_enable();
> + local_irq_enable_nort();
>
> /* Note: dev_set_drvdata must be called while holding the rwsem */
> if (dev->class == CL_EHCI) {
> --
> 2.11.0
>
> --
> To unsubscribe from this list: send the line &
From 8ea8311b75a40bdea03e7f8228a0578b6367e9d1 Mon Sep 17 00:00:00 2001
From: Clark Williams <willi...@redhat.com>
Date: Mon, 20 Nov 2017 14:26:12 -0600
Subject: [PATCH] [rt] sched/rt: fix panic in double_lock_balance with
simplified IPI RT balancing
I was testing 4.14-rt1 on a large
18:30:18 [+0200], To Clark Williams wrote:
> > This is the minimum to get this working on RT splat free. There is one
> > memory deallocation with irqs off which should work on RT in its current
> > way.
> > Once this and the on_each_cpu() invocation, I was wondering if…
&
Joe,
This looks interesting. Do you have a git repo where I can pull the
source?
Clark
On Tue, 20 Nov 2018 12:29:00 -0500
Steven Rostedt wrote:
> [ Adding Clark and John who manage the rt-tests repo ]
>
> -- Steve
>
> On Mon, 19 Nov 2018 19:46:23 +
> Joe Korty wrote:
>
> > Hi Julia &
kernel but is problematic
on an RT kernel where spin locks are converted to rt_mutex_t, which can sleep.
Convert the quarantine_lock to a raw spinlock. The usage of quarantine_lock
is confined to quarantine.c and the work performed while the lock is held is
limited.
Signed-off-by: Clark Williams
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Ingo Molnar wrote:
>
> ok - lets try it. Clark: does this impact the set_kthread_prio utility?
> I've changed "IRQ 123" to "IRQ-123" to make pidof & friends work better.
Yes it does, but I believe I can fix that. Most of the logic is in awk
and I
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Paul E. McKenney wrote:
> On Fri, Jun 08, 2007 at 03:43:48PM -0400, Steven Rostedt wrote:
>> On Fri, 2007-06-08 at 12:36 -0700, Paul E. McKenney wrote:
>>> On Fri, Jun 08, 2007 at 11:27:08AM -0400, Steven Rostedt wrote:
The first time I compiled
or out priorities for workqueue.h and sched.h
From: Clark Williams <[EMAIL PROTECTED]>
This fixes a circular dependency between sched.h and workqueue.h by factoring out the common
defines to a new header which is included by both
Signed-off-by: Clark Williams <[EMAIL PROTECTED]>
---
inc
On Mon, 8 Apr 2013 20:30:12 +0200
Sebastian Andrzej Siewior wrote:
> * Paul Gortmaker | 2013-04-05 19:35:51 [-0400]:
>
> >Thanks for the work in putting this together. Just a heads up that the
> >split queue fails on a patch with no author/date/subject as follows:
> >
> >[...]
> >git am of
Sebastian,
I needed this for my laptop on 3.12, so tweaked it to apply properly.
Clark
commit 7b1573eaa0f176731bc231b249598b6e4850babf
Author: Clark Williams
Date: Tue Nov 12 12:17:07 2013 -0600
net: iwlwifi: request only a threaded handler for interrupts
On RT the trans_pcie
On Tue, 12 Nov 2013 13:16:59 -0600
Clark Williams wrote:
> Sebastian,
>
> I needed this for my laptop on 3.12, so tweaked it to apply properly.
>
> Clark
>
So of course I sent the wrong patch:
commit 49d487614d56bc10969dd0dcbce709825fc06d0e
Author: Clark Williams
Date:
a named fifo to
be created so that an external program can read run statistics while
the run is in process. This one is early days yet so any suggestions,
corrections, flames, etc that can improve this are welcome.
Here's the changelog:
* Fri Nov 15 2013 Clark Williams - 0.86-1
- cyclictest: allow
On Thu, 12 Dec 2013 16:47:19 +0200
Ciprian Barbu wrote:
> Hello,
>
> While playing around with hackbench I discovered that I would sometimes
> get an enormous time reported, even if the run time would be less than a
> second or so. The problem was that the struct timeval start was not
>
.
Clark Williams (3):
sched: move sched.h sysctl bits into separate header
sched/rt: add a tuning knob to allow changing SCHED_RR timeslice
sched/rt: move rt specific bits into new header file
block/blk-exec.c | 1 +
drivers/spi/spi.c | 2 +-
drivers
Add a /proc/sys/kernel scheduler knob named sched_rr_timeslice_ms
that allows global changing of the SCHED_RR timeslice value. User
visable value is in milliseconds but is stored as jiffies. Setting
to 0 (zero) resets to the default (currently 100ms).
Signed-off-by: Clark Williams
---
include
Move the sysctl-related bits from include/linux/sched.h into
a new file: include/linux/sched/sysctl.h. Then update source
files requiring access to those bits by including the new
header file.
Signed-off-by: Clark Williams
---
block/blk-exec.c | 1 +
include/linux/sched.h
move rt scheduler definitions out of include/linux/sched.h into
new file include/linux/sched/rt.h
Signed-off-by: Clark Williams
---
drivers/spi/spi.c | 2 +-
drivers/staging/csr/bh.c | 2 +-
drivers/staging/csr/unifi_sme.c | 2 +-
drivers/tty/sysrq.c
On Thu, 7 Feb 2013 20:52:57 +0100
Ingo Molnar wrote:
>
> * Ingo Molnar wrote:
>
> >
> > * Ingo Molnar wrote:
> >
> > > Yeah, that looks pretty good.
> > >
> > > I'll give it some testing.
> >
> > Found one build error so far, applied the fix below.
>
> Another one was in NOMMU mode
On Fri, 8 Feb 2013 15:18:41 +0100
Ingo Molnar wrote:
>
> * Clark Williams wrote:
>
> > On Thu, 7 Feb 2013 20:52:57 +0100
> > Ingo Molnar wrote:
> >
> > >
> > > * Ingo Molnar wrote:
> > >
> > > >
> > &
On Fri, 8 Feb 2013 16:53:17 +0100
Frederic Weisbecker wrote:
> 2013/2/7 Christoph Lameter :
> > On Thu, 7 Feb 2013, Frederic Weisbecker wrote:
> >
> >> Not with hrtick.
> >
> > hrtick? Did we not already try that a couple of years back and it turned
> > out that the overhead of constantly
dir,
things like NO_HZ and CGROUPS but wanted to get this in place first.
Clark Williams (2):
sched: create new scheduler include dir and move sysctl defs there
sched/rt: add a tuning knob to allow changing SCHED_RR timeslice
include/linux/sched.h| 108
1 - 100 of 231 matches
Mail list logo