Daniel Wagner | 24 May 11:41 2016

Realtime Community Event


Christoph and I are organizing a small event (mini conf) at a Austrian
mountain lodge beginning of June. We have some places left. So in case
you are interested to join, please let us know (soon).

I collected all information here (gets updated now and then):


To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in
the body of a message to majordomo <at> vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Brad Hughes | 19 May 02:30 2016

Wrong time displayed on date/time selector custom field

Hi All,

We are using RT 4.2.8 and we have recently started using custom fields
that are date time selectors (ie. they pop up a calendar where you can
select a date and time).

We have encountered a problem where the value in the custom field is
being displayed wrong.

What I believe is happening is that when the date time custom field is
populated with a date/time, that is being stored in RT's mysql
database in local time format. Where as every other time in RTs mysql
database seems to be stored in UTC or GMT time format.

When the RT GUI displays the time stored in custom field of the
date/time selector, it is assuming that it is a UTC or GMT format, and
it's attempting to convert it to local time - which makes it display
incorrect in the RT GUI.

Has anyone encountered this problem and if so what did you do to get around it?

Thank you,

Brad Hughes
To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in
the body of a message to majordomo <at> vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

(Continue reading)

John Kacur | 13 May 15:29 2016

[PATCH] rt-tests: hwlat.txt: smidetect renamed to hwlatdetect

- smidetect was long ago renamed to hwlatdetct, reflect this in hwlat.txt
- Fix a few spelling mistakes, and add a missing option

Signed-off-by: John Kacur <jkacur <at> redhat.com>
 src/hwlatdetect/hwlat.txt | 17 +++++++++--------
 1 file changed, 9 insertions(+), 8 deletions(-)

diff --git a/src/hwlatdetect/hwlat.txt b/src/hwlatdetect/hwlat.txt
index 11dc067c1ee2..4ac79d0a52e3 100644
--- a/src/hwlatdetect/hwlat.txt
+++ b/src/hwlatdetect/hwlat.txt
 <at>  <at>  -45,7 +45,7  <at>  <at>  latency_threshold_us	- minimum latency value to be considered (usecs)
 max_sample_us		- maximum SMI latency spike observed (usecs)
 ms_between_samples	- interval between samples (ms)
 ms_per_sample		- sampling time (ms)
-sample_us		- last sample (usecs). Continously updated, may be used
+sample_us		- last sample (usecs). continuously updated, may be used
 			  to plot graphs or to create histograms
 smi_count		- how many latency spikes have been observed

 <at>  <at>  -76,7 +76,7  <at>  <at>  samples during 5ms every 50ms:
 	# cat ms_per_sample

-After a while, data may be verified to atest the existence of SMI induced
+After a while, data may be verified to test the existence of SMI induced

 	# cat smi_count
(Continue reading)

John Kacur | 13 May 15:11 2016

[PATCH 0/2] rt-tests: Housekeeping

Just doing a little housekeeping before releasing rt-tests-1.0

John Kacur (2):
  rt-tests: Add missing option to hwlatdetect man page
  rt-tests: Housekeeping fix some spelling errors.

 src/backfire/sendme.8         |  2 +-
 src/hwlatdetect/hwlatdetect.8 | 12 ++++++++----
 src/pi_tests/pi_stress.8      |  2 +-
 src/pmqtest/pmqtest.8         |  4 ++--
 src/ptsematest/ptsematest.8   |  6 +++---
 src/signaltest/signaltest.8   |  2 +-
 src/sigwaittest/sigwaittest.8 |  8 ++++----
 src/svsematest/svsematest.8   |  6 +++---
 8 files changed, 23 insertions(+), 19 deletions(-)



To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in
the body of a message to majordomo <at> vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Brian Silverman | 10 May 20:04 2016

[PATCH] Force processes to non-realtime before mm_exit

Without this, a realtime process which has called mlockall exiting
causes large latencies for other realtime processes at the same or
lower priorities. This seems like a fairly common use case too, because
realtime processes generally want their memory locked into RAM.

Signed-off-by: Brian Silverman <brian <at> peloton-tech.com>
 kernel/exit.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/kernel/exit.c b/kernel/exit.c
index a0cf72b..68a97df 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
 <at>  <at>  -730,6 +730,12  <at>  <at>  void do_exit(long code)
 	tsk->exit_code = code;
 	taskstats_exit(tsk, group_dead);

+	if (tsk->policy == SCHED_FIFO || tsk->policy == SCHED_RR) {
+		struct sched_param param = { .sched_priority = 0 };
+		sched_setscheduler_nocheck(current, SCHED_NORMAL, &param);
+	}

 	if (group_dead)


(Continue reading)

Luiz Capitulino | 9 May 16:50 2016

[PATCH RESEND -rt] mm: perform lru_add_drain_all() remotely

lru_add_drain_all() works by scheduling lru_add_drain_cpu() to run
on all CPUs that have non-empty LRU pagevecs and then waiting for
the scheduled work to complete. However, workqueue threads may never
have the chance to run on a CPU that's running a SCHED_FIFO task.
This causes lru_add_drain_all() to block forever.

This commit solves this problem by changing lru_add_drain_all()
to drain the LRU pagevecs of remote CPUs. This is done by grabbing
swapvec_lock and calling lru_add_drain_cpu().

PS: This is based on an idea and initial implementation by
    Rik van Riel.

Signed-off-by: Rik van Riel <riel <at> redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino <at> redhat.com>
 include/linux/locallock.h | 31 +++++++++++++++++++++++++++++++
 mm/swap.c                 | 35 +++++++++++++++++++++++++----------
 2 files changed, 56 insertions(+), 10 deletions(-)

diff --git a/include/linux/locallock.h b/include/linux/locallock.h
index 6fe5928..f4cd691 100644
--- a/include/linux/locallock.h
+++ b/include/linux/locallock.h
 <at>  <at>  -104,6 +104,17  <at>  <at>  static inline void __local_unlock(struct local_irq_lock *lv)
 		put_local_var(lvar);				\
 	} while (0)

+#define local_lock_other_cpu(lvar, cpu)                         \
+	do {                                                    \
(Continue reading)

Roosen Henri | 6 May 11:23 2016

WARNING: complate_all() with 7 waiters at system boot

Hi all,

My x86_64 development PC with kernel 4.4.7-rt16 throws the following warning during system boot:

[   12.606217] ------------[ cut here ]------------
[   12.606220] WARNING: CPU: 5 PID: 872 at kernel/sched/swait.c:48 complete_all+0x2a/0x40()
[   12.606220] complate_all() with 7 waiters
[   12.606243] Modules linked in: arc4 ecb md4 nls_utf8 cifs dns_resolver fscache ftdi_sio usbserial
joydev hid_logitech_hidpp hid_logitech_dj hid_generic usbhid usb_storage hid intel_rapl iosf_mbi
evdev iTCO_wdt iTCO_vendor_support intel_powerclamp coretemp kvm_intel kvm irqbypass
crct10dif_pclmul crc32_pclmul jitterentropy_rng sha256_ssse3 sha256_generic hmac drbg ansi_cprng
aesni_intel aes_x86_64 lrw gf128mul glue_helper ablk_helper cryptd serio_raw pcspkr nouveau mxm_wmi
tpm_infineon wmi ttm snd_hda_codec_hdmi drm_kms_helper drm i2c_i801 i2c_algo_bit i2c_core
snd_hda_codec_realtek snd_hda_codec_generic battery snd_hda_intel snd_hda_codec fujitsu_laptop
snd_hda_core snd_hwdep snd_pcm video snd_timer tpm_tis mei_me snd loop lpc_ich tpm soundcore mfd_core
mei shpchp button processor fuse 
[   12.606252]  ppdev lp parport autofs4 ext4 crc16 mbcache jbd2 sg sr_mod cdrom sd_mod ahci libahci
crc32c_intel libata xhci_pci ehci_pci psmouse scsi_mod xhci_hcd e1000e ehci_hcd ptp usbcore pps_core
usb_common fan thermal
[   12.606253] CPU: 5 PID: 872 Comm: cryptomgr_probe Not tainted 4.4.7-rt16 #2
[   12.606254] Hardware name: FUJITSU CELSIUS W530/D3227-A1, BIOS V4.6.5.4 R1.34.0 for D3227-A1x 01/08/2015
[   12.606256]  0000000000000000 ffffffff812e0e0e ffff88021142be30 ffffffff81731e02
[   12.606257]  ffffffff81073f81 ffff88003614fd30 ffff88021142be80 0000000000000246
[   12.606258]  ffff880212206400 0000000000000000 ffffffff8107401c ffffffff81726fb0
[   12.606258] Call Trace:
[   12.606262]  [<ffffffff812e0e0e>] ? dump_stack+0x5c/0x7e
[   12.606265]  [<ffffffff81073f81>] ? warn_slowpath_common+0x81/0xd0
[   12.606266]  [<ffffffff8107401c>] ? warn_slowpath_fmt+0x4c/0x50
[   12.606267]  [<ffffffff810b5849>] ? swake_up_all_locked+0x29/0x80
(Continue reading)

Uwe Kleine-König | 27 Apr 08:51 2016

[PATCH rt-tests] manpages: fix typos found by lintian

This fixes these spelling errors:

	mutiple -> multiple
	occuring -> occurring
	specifed -> specified
	synchonized -> synchronized

BTW, I wonder if the manpages are using dos line endings on purpose.

 src/hwlatdetect/hwlatdetect.8 | 4 ++--
 src/pi_tests/README           | 2 +-
 src/pmqtest/pmqtest.8         | 2 +-
 src/ptsematest/ptsematest.8   | 2 +-
 src/sigwaittest/sigwaittest.8 | 4 ++--
 src/svsematest/svsematest.8   | 2 +-
 6 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/src/hwlatdetect/hwlatdetect.8 b/src/hwlatdetect/hwlatdetect.8
index 2904e5fd8fde..6f5501445ce1 100644
--- a/src/hwlatdetect/hwlatdetect.8
+++ b/src/hwlatdetect/hwlatdetect.8
 <at>  <at>  -38,7 +38,7  <at>  <at>  written for use by the "RT" patch since the Real Time kernel is highly
 latency sensitive. 

 SMIs are usually not serviced by the Linux kernel, which typically does not
-even know that they are occuring. SMIs are instead are set up by BIOS code
+even know that they are occurring. SMIs are instead are set up by BIOS code
 and are serviced by BIOS code, usually for "critical" events such as
 management of thermal sensors and fans. Sometimes though, SMIs are used for
(Continue reading)

John Kacur | 25 Apr 16:17 2016

[PATCH] rt-tests: Update the cyclictest man page

From 7e007ac70a83f62583d25b10ad2df0430306bf32 Mon Sep 17 00:00:00 2001
From: John Kacur <jkacur <at> redhat.com>
Date: Mon, 25 Apr 2016 16:08:28 +0200
Subject: [PATCH] rt-tests: Update the cyclictest man page

We have done a good job keeping cyclictest help up-to-date, but we have
not been actively updating the man page.

There were many missing options, as well as options that don't exist
anymore. To be more precise, some short form options had been removed,
but there are still corresponding long-version options.

This patch contains a considerable amount of updates to the man-page to
keep it in sync, and in some cases it also updates the cyclictest help
where needed. For example, the cyclictest help options were not always
in the proper order. In some cases I changed the wording to make it more
closely correspond to the man page

Signed-off-by: John Kacur <jkacur <at> redhat.com>
 src/cyclictest/cyclictest.8 | 169 ++++++++++++++++++++++++--------------------
 src/cyclictest/cyclictest.c |  35 ++++-----
 2 files changed, 110 insertions(+), 94 deletions(-)

diff --git a/src/cyclictest/cyclictest.8 b/src/cyclictest/cyclictest.8
index 437e7a68a74f..120dae23860e 100644
--- a/src/cyclictest/cyclictest.8
+++ b/src/cyclictest/cyclictest.8
 <at>  <at>  -1,5 +1,5  <at>  <at> 
 .\"                                      Hey, EMACS: -*- nroff -*-
(Continue reading)

helix250@libero.it | 25 Apr 13:44 2016

Kernel RT questions

I do not know if this is the right mailing list to ask this question, 
if this is not, please apologize and tell me where I could ask for this 

I am trying to do my first tests with PREEMPT_RT on my BeagleBoneBlack board.
I am testing it with cyclictest and in the meanwhile I am copying a big file 
from /home/roor to /tmp to stress the system. 

I have two options at the moment, a standard kernel and a PREEMPT one.
I've got the kernels from https://github.com/beagleboard/linux.git using two 
different branch/tag as described below.

The result of my tests seems to indicate a not RT kernel and any RT 
My questions are:
- How can I detect during runtime that my kernel was properly configured for 
RT performances?
- Should I see any message in dmesg or at boot related to "Ingo Molnar"?
- Are the performances detected above with cyclictest the correct way to 
measure RT performances?

* Standard version

root <at> beaglebone:~# uname -a
Linux beaglebone 4.1.6-yocto-standard #2 PREEMPT Thu Dec 3 16:13:24 CET 2015 
armv7l GNU/Linux

root <at> beaglebone:~# cat /proc/config.gz | gunzip | grep PREEMPT
(Continue reading)

John Kacur | 21 Apr 23:13 2016

[PATCH] rt-tests: Add man page for rt-migrate-test

From 02d556361267f8335b2417bbb62ef6a498da112c Mon Sep 17 00:00:00 2001
From: John Kacur <jkacur <at> redhat.com>
Date: Thu, 21 Apr 2016 23:10:26 +0200
Subject: [PATCH] rt-tests: Add man page for rt-migrate-test

Added a man page for rt-migrate-test
Fairly minimal, based on the output from the --help option

Signed-off-by: John Kacur <jkacur <at> redhat.com>
 src/rt-migrate-test/rt-migrate-test.8 | 56 +++++++++++++++++++++++++++++++++++
 1 file changed, 56 insertions(+)
 create mode 100644 src/rt-migrate-test/rt-migrate-test.8

diff --git a/src/rt-migrate-test/rt-migrate-test.8 b/src/rt-migrate-test/rt-migrate-test.8
new file mode 100644
index 000000000000..8af50e54291d
--- /dev/null
+++ b/src/rt-migrate-test/rt-migrate-test.8
 <at>  <at>  -0,0 +1,56  <at>  <at> 
+.TH RT-MIGRATE-TEST 8 "April 21, 2016"
+.\" Please adjust this date whenever editing this manpage
+rt-migrate-test \- real-time task migration program
+.B rt-migrate-test 
+.RI "[\-ceh] [\-p prio] [\-r time] [\-s time] [\-m time] [\-l loops] [nr_tasks]
+Test real-time multiprocessor scheduling of tasks to ensure the highest priority tasks are running on
(Continue reading)