Hello,
I recently posted a patch to be able to pass data to
the kernel and make use of it really early.
Use-cases are currently:
Pass modified ACPI tables and make use of these instead of BIOS provided
ones
and (nothing has been done there yet):
Pass CPU microcode updates and flash it really
On Friday 16 December 2011 17:18:37 Jean Pihet wrote:
Hi Thomas,
On Fri, Dec 16, 2011 at 3:35 PM, Thomas Renninger tr...@suse.de wrote:
The last missing manpage for cpupower tools.
Great!
More info about other architecture's sleep state specialities would be
great.
I wonder
The last missing manpage for cpupower tools.
More info about other architecture's sleep state specialities would be great.
Signed-off-by: Thomas Renninger tr...@suse.de
CC: linux-arm-ker...@lists.infradead.org
CC: jean.pi...@newoldbits.com
CC: linux-omap@vger.kernel.org
CC: linux...@lists.linux
On Saturday, August 20, 2011 04:40:09 AM Ming Lei wrote:
Hi,
2011/8/20 Thomas Renninger tr...@suse.de:
On Friday, August 19, 2011 05:04:04 PM tom.leim...@gmail.com wrote:
From: Ming Lei tom.leim...@gmail.com
This patch removes the 'cpu_id' parameter of the cpu_idle
trace point, based
On Friday, August 19, 2011 05:04:04 PM tom.leim...@gmail.com wrote:
From: Ming Lei tom.leim...@gmail.com
This patch removes the 'cpu_id' parameter of the cpu_idle
trace point, based on the ideas below:
- the cpu_id which is passed to trace point is always the current
cpu
Are you sure
Hi,
cpupowerutils is based on the well known cpufrequtils project.
Where do I find it?
---
A git repository is hosted on gitorious:
git://gitorious.org/cpupowerutils/cpupowerutils.git
Be careful, it's not the default, but the cpupowerutils branch!
You can also directly download
cpupowerutils is based on the well known cpufrequtils project.
Forgot to mention:
Big thanks to Dominik Brodowski who already gave this some tests!
Outcome (heavily shortened):
You can add that you have my Signed-off-by / Acked-by to all these
patches
Thomas
--
To unsubscribe from
,
-Original Message-
From: Jean Pihet [mailto:jean.pi...@newoldbits.com]
Sent: Monday, February 07, 2011 9:36 PM
To: Santosh Shilimkar; Tony Lindgren
Cc: Thomas Renninger; linux-omap@vger.kernel.org; Jean Pihet-XID
Subject: Re: [PATCH] perf: add OMAP support for the new power events
Hi
On Wednesday 12 January 2011 07:56:45 Len Brown wrote:
I'm not fond of inventing a new 3-character abbreviation field
for every state because display tools can't handle the existing
16-character name field.
I am also not fond of it, but it's a sane and easy
solution and appropriate for this
On Wednesday 12 January 2011 07:42:59 Len Brown wrote:
I'm happy to see the trace point move up into cpuidle from intel_idle.
If somebody is picking this up in a perf tree,
Acked-by: Len Brown len.br...@intel.com
else I can put it in the idle tree, let me know.
I didn't know you have an
On Wednesday 12 January 2011 23:25:12 Len Brown wrote:
If the display tools can only handle 3 characters,
then why not have them simply use the 1st 3 characters
of the existing name field?
You mean use:
C3 NHM
instead of
NHM-C3
for intel_idle?
Yes.
is there a reason that
On Wednesday 05 January 2011 12:05:18 Jean Pihet wrote:
Hi Paul,
On Tue, Jan 4, 2011 at 7:48 PM, Paul Walmsley p...@pwsan.com wrote:
Hello Jean,
On Tue, 4 Jan 2011, jean.pi...@newoldbits.com wrote:
From: Jean Pihet j-pi...@ti.com
The patch adds the new power management trace
chars or
you can't see much in these pictures.
The name could get used in the title, the introduced abbrevations
inside of the picture and the text must therefore be rather short.
Signed-off-by: Thomas Renninger tr...@suse.de
CC: l...@kernel.org
CC: linux-a...@vger.kernel.org
CC: linux
.
Signed-off-by: Thomas Renninger tr...@suse.de
CC: l...@kernel.org
CC: linux-a...@vger.kernel.org
CC: linux...@lists.linux-foundation.org
CC: Arjan van de Ven ar...@linux.intel.com
CC: Ingo Molnar mi...@elte.hu
CC: linux-ker...@vger.kernel.org
CC: linux-perf-us...@vger.kernel.org
CC: linux-omap
only export C1 and C3.
- everything is fine.
Signed-off-by: Thomas Renninger tr...@suse.de
CC: Robert Schoene robert.scho...@tu-dresden.de
CC: Jean Pihet j-pi...@ti.com
CC: Arjan van de Ven ar...@linux.intel.com
CC: Ingo Molnar mi...@elte.hu
CC: Len Brown l...@kernel.org
CC: Frederic Weisbecker
Argh, forgot guilt refresh on this one.
The changelog could be a bit more detailed by adding:
On Friday 07 January 2011 11:29:49 Thomas Renninger wrote:
Before, power:cpu_idle events were very specific X86 Intel mwait events.
This got fixed with previous patches and cpu_idle events are now
only export C1 and C3.
- everything is fine.
Signed-off-by: Thomas Renninger tr...@suse.de
CC: Robert Schoene robert.scho...@tu-dresden.de
CC: Jean Pihet j-pi...@ti.com
CC: Arjan van de Ven ar...@linux.intel.com
CC: Ingo Molnar mi...@elte.hu
CC: Len Brown l...@kernel.org
CC: linux...@lists.linux
chars or
you can't see much in these pictures.
The name could get used in the title, the introduced abbrivations
inside of the picture and the text must therefore be rather short.
Signed-off-by: Thomas Renninger tr...@suse.de
CC: l...@kernel.org
CC: linux-a...@vger.kernel.org
CC: linux
.
Signed-off-by: Thomas Renninger tr...@suse.de
CC: l...@kernel.org
CC: linux-a...@vger.kernel.org
CC: linux...@lists.linux-foundation.org
CC: Arjan van de Ven ar...@linux.intel.com
CC: Ingo Molnar mi...@elte.hu
CC: linux-ker...@vger.kernel.org
CC: linux-perf-us...@vger.kernel.org
CC: linux-omap
On Thursday 02 December 2010 17:42:27 Thomas Renninger wrote:
Before, power:cpu_idle events were very specific X86 Intel mwait events.
This got fixed with previous patches and cpu_idle events are now thrown by
all cpuidle drivers and can be mapped to the corresponding cpuidle state
in /sys
Hi Robert,
On Monday 01 November 2010 09:11:28 Robert Schöne wrote:
Am Mittwoch, den 27.10.2010, 17:42 +0200 schrieb Thomas Renninger:
Robert: I expect you tested this on a machine with no cpuidle
driver registered?
You're right, there was no idle driver, but the idle process from
Tested with:
acpi_idle and intel_idle as cpuidle drivers.
also tested perf timechart userspace tool without the new interface
and it still works fine with the new kernel changes.
There are quite some issues with the balance of cpu_idle
enter/leave a sleep state, but this was the case already.
-by: Thomas Renninger tr...@suse.de
CC: linux-omap@vger.kernel.org
CC: linux...@lists.linux-foundation.org
CC: linux-trace-us...@vger.kernel.org
CC: Jean Pihet jean.pi...@newoldbits.com
CC: Arjan van de Ven ar...@linux.intel.com
CC: Ingo Molnar mi...@elte.hu
CC: r...@sisk.pl
---
tools/perf/builtin
power_frequency moved to drivers/cpufreq/cpufreq.c which has
to be compiled in, no need to export it.
intel_idle can a be module though...
Signed-off-by: Thomas Renninger tr...@suse.de
CC: linux-omap@vger.kernel.org
CC: linux...@lists.linux-foundation.org
CC: linux-trace-us...@vger.kernel.org
CC
timechart
userspace tool gets adjusted in a separate patch.
Signed-off-by: Thomas Renninger tr...@suse.de
Signed-off-by: Arjan van de Ven ar...@linux.intel.com
CC: linux-omap@vger.kernel.org
CC: linux...@lists.linux-foundation.org
CC: linux-trace-us...@vger.kernel.org
CC: Jean Pihet jean.pi
On Thursday 28 October 2010 11:02:59 Thomas Renninger wrote:
Recent changes:
- Adjust state/cpuid to u32 as done in the kernel
Argh, I forgot a guilt refresh...
Below final :) patch also fixes a bug that got introduced
in 2.6.36 already. Will also send a version for stable
inclusion
timechart
userspace tool gets adjusted in a separate patch.
Signed-off-by: Thomas Renninger tr...@suse.de
Signed-off-by: Arjan van de Ven ar...@linux.intel.com
CC: linux-omap@vger.kernel.org
CC: linux...@lists.linux-foundation.org
CC: linux-trace-us...@vger.kernel.org
CC: Jean Pihet jean.pi
On Tuesday 26 October 2010 09:10:20 Ingo Molnar wrote:
* Thomas Renninger tr...@suse.de wrote:
Changes in V2:
- Introduce PWR_EVENT_EXIT instead of 0 to mark non-power state
- Use u32 instead of u64 for cpuid, state which is by far enough
New power trace events
On Tuesday 26 October 2010 13:21:29 Ingo Molnar wrote:
* Jean Pihet jean.pi...@newoldbits.com wrote:
..
+#ifndef _TRACE_POWER_ENUM_
+#define _TRACE_POWER_ENUM_
+enum {
+ POWER_NONE = 0,
+ POWER_CSTATE = 1,
+ POWER_PSTATE = 2,
+};
+#endif
Since we are
On Tuesday 26 October 2010 13:54:22 Ingo Molnar wrote:
* Thomas Renninger tr...@suse.de wrote:
..
As you state above: POWER_NONE does not make sense at all.
The whole thing (type= attribute that vanishes now) is
passed to userspace, but never gets used there because the
same info
On Tuesday 26 October 2010 15:17:43 Thomas Renninger wrote:
On Tuesday 26 October 2010 13:54:22 Ingo Molnar wrote:
* Thomas Renninger tr...@suse.de wrote:
...
If cpuidle is not registered, the events you get are arch specific.
I mean they are arch specific anyway, but with cpuidle you can
cpu_idle events (transition into sleep state and exiting) are
both fired in pm_idle().
Entering sleep state and exiting should always get fired inside
pm_idle() already.
This is a revert of commit c882e0feb937af4e5b991cbd1c
Signed-off-by: Thomas Renninger tr...@suse.de
CC: Robert Schoene
power_frequency moved to drivers/cpufreq/cpufreq.c which has
to be compiled in, no need to export it.
intel_idle can a be module though...
Signed-off-by: Thomas Renninger tr...@suse.de
CC: linux-omap@vger.kernel.org
CC: linux...@lists.linux-foundation.org
CC: linux-trace-us...@vger.kernel.org
CC
be more generic as an interface and support
hardware/software/... events, not only tracepoints, but someone
else could extend that if needed...
Signed-off-by: Thomas Renninger tr...@suse.de
CC: linux-omap@vger.kernel.org
CC: linux...@lists.linux-foundation.org
CC: linux-trace-us...@vger.kernel.org
CC
itself.
perf timechart
userspace tool gets adjusted in a separate patch.
Signed-off-by: Thomas Renninger tr...@suse.de
Signed-off-by: Arjan van de Ven ar...@linux.intel.com
CC: linux-omap@vger.kernel.org
CC: linux...@lists.linux-foundation.org
CC: linux-trace-us...@vger.kernel.org
CC: Jean Pihet
On Tuesday 26 October 2010 08:57:01 pm Rafael J. Wysocki wrote:
On Tuesday, October 26, 2010, Thomas Renninger wrote:
Ok, that's at least generic. Needs the review of Rafael, to determine
whether this state value is all we want to know when we enter suspend.
He already gave an acked
On Monday 25 October 2010 08:54:34 Arjan van de Ven wrote:
On 10/19/2010 4:36 AM, Thomas Renninger wrote:
static void poll_idle(void)
{
- trace_power_start(POWER_CSTATE, 0, smp_processor_id());
local_irq_enable();
while (!need_resched())
cpu_relax
On Monday 25 October 2010 12:04:28 Ingo Molnar wrote:
* Thomas Renninger tr...@suse.de wrote:
New power trace events:
power:processor_idle
power:processor_frequency
power:machine_suspend
C-state/idle accounting events:
power:power_start
power:power_end
are replaced
On Monday 25 October 2010 13:55:25 Ingo Molnar wrote:
* Thomas Renninger tr...@suse.de wrote:
On Monday 25 October 2010 12:04:28 Ingo Molnar wrote:
* Thomas Renninger tr...@suse.de wrote:
New power trace events:
power:processor_idle
power:processor_frequency
On Monday 25 October 2010 15:55:08 Arjan van de Ven wrote:
On 10/25/2010 2:41 AM, Thomas Renninger wrote:
On Monday 25 October 2010 08:54:34 Arjan van de Ven wrote:
On 10/19/2010 4:36 AM, Thomas Renninger wrote:
static void poll_idle(void)
{
- trace_power_start(POWER_CSTATE, 0
On Monday 25 October 2010 16:56:04 Ingo Molnar wrote:
* Arjan van de Ven ar...@linux.intel.com wrote:
On 10/25/2010 7:36 AM, Thomas Renninger wrote:
ok so we have
C0 idle
Ideally this should not be called C0, but expressed
as (#define) POLL_IDLE wherever possible.
In all
@Ingo: Can you queue up 1/3, it's an independent fix.
On Monday 25 October 2010 06:00:17 pm Arjan van de Ven wrote:
On 10/25/2010 8:48 AM, Thomas Renninger wrote:
sure naming is one thing
Yes it should get renamed to not show:
cat /sys/devices/system/cpu/cpu0/cpuidle/state0/name
C0
= field got removed from both, it was never
used and the type is differed by the event type itself.
perf timechart
userspace tool gets adjusted in a separate patch.
Signed-off-by: Thomas Renninger tr...@suse.de
CC: Linus Torvalds torva...@linux-foundation.org
CC: Andrew Morton a...@linux-foundation.org
...
Signed-off-by: Thomas Renninger tr...@suse.de
CC: Linus Torvalds torva...@linux-foundation.org
CC: Andrew Morton a...@linux-foundation.org
CC: Thomas Gleixner t...@linutronix.de
CC: Masami Hiramatsu masami.hiramatsu...@hitachi.com
CC: Frank Eigler f...@redhat.com
CC: Steven Rostedt rost...@goodmis.org
On Sunday 10 October 2010 14:19:28 Ingo Molnar wrote:
* Arjan van de Ven ar...@linux.intel.com wrote:
...
also I have to say that some events are more likely to change than others
function foo in the kernel called is more likely to change than the
processor went to THIS frequency.
is_valid_tracepoint(const char *event_string)
to parse-events.[hc]
which could be more generic as an interface and support
hardware/software/... events, not only tracepoints, but someone
else could extend that if needed...
Signed-off-by: Thomas Renninger tr...@suse.de
CC: Linus Torvalds torva...@linux
power_frequency moved to drivers/cpufreq/cpufreq.c which has
to be compiled in, no need to export it.
intel_idle can a be module though...
Signed-off-by: Thomas Renninger tr...@suse.de
CC: Linus Torvalds torva...@linux-foundation.org
CC: Andrew Morton a...@linux-foundation.org
CC: Thomas
On Thursday 07 October 2010 17:08:25 Mathieu Desnoyers wrote:
[ Adding a few more CCs, since this discussion is about a tracepoint
userspace ABI policy, which is a topic of general interest. ]
...
Yes, sadly this debate running in circles hurts contributors.
Thanks for the summary!
Hi,
On Monday 04 October 2010 17:20:57 Jean Pihet wrote:
Here is a re-spin of the patches after discussion.
what is going to happen here now?
Is this supposed to go through Ingo's tree?
Ingo: do you mind commenting on this.
I see 3 possibilities:
1) Power (or all) perf events are never
On Monday 04 October 2010 05:20:59 pm Jean Pihet wrote:
The patch adds the new power management trace points for
the OMAP architecture.
The trace points are for cpuidle, cpufreq (DVFS), the clocks
changes (enable, disable, set_rate) and the power domains
transitions.
...
diff --git
On Wednesday 22 September 2010 19:06:54 Arjan van de Ven wrote:
On 9/22/2010 9:43 AM, Peter Zijlstra wrote:
On Wed, 2010-09-22 at 09:32 -0700, Arjan van de Ven wrote:
What are the apps that are using it? I know about builtin-timechart,
pytimechart. Is powertop using this as well?
Hi,
I had a quick look at this and it's amazing how broken
the whole power event tracing interfaces are.
It's not your fault, Jean, they always were and adding your stuff is
fine.
Some questions, maybe I've overseen something:
Why does this event:
DEFINE_EVENT(power, power_frequency,
exist and
On Friday 17 September 2010 16:05:51 Jean Pihet wrote:
Hi Thomas,
On Fri, Sep 17, 2010 at 3:08 PM, Thomas Renninger tr...@suse.de wrote:
Hi,
I had a quick look at this and it's amazing how broken
the whole power event tracing interfaces are.
It's not your fault, Jean, they always
On Friday 17 September 2010 16:24:59 Ingo Molnar wrote:
* Jean Pihet jean.pi...@newoldbits.com wrote:
Apropos documentation..., are the power trace events documented
somewhere?
No. We need something like Documentation/trace/events-kmem.txt. I
can write that with for the new power
, but should get used for any P- (processor frequency),
C- (processor sleep), T- (processor throttling), or S- (machine sleep)
state.
Since it's used in cpufreq.c which must be compiled in
and not in acpi-cpufreq.c anymore there is no need to export it
for modules.
Signed-off-by: Thomas Renninger tr
use of it, even if compiled as module.
Signed-off-by: Thomas Renninger tr...@suse.de
---
arch/x86/kernel/process.c |9 +++--
drivers/cpuidle/cpuidle.c |3 +++
drivers/idle/intel_idle.c |3 +++
kernel/trace/power-traces.c |2 +-
4 files changed, 14 insertions(+), 3
power-trace: Add x86 ACPI S- (machine sleep) state events.
Signed-off-by: Thomas Renninger tr...@suse.de
---
drivers/acpi/sleep.c | 11 ++-
1 file changed, 10 insertions(+), 1 deletion(-)
Index: linux-2.6.35-master/drivers/acpi/sleep.c
DO NOT APPLY THIS ONE!!!
The others should go into a mainline tree if Jean is ok with them.
This one does not work, due to some include dependencies or whatever
else I can't see right now.
The idea: Provide old trace power interfaces via .config option to not break
existing
On Friday 17 September 2010 18:24:12 Ingo Molnar wrote:
* Thomas Renninger tr...@suse.de wrote:
On Friday 17 September 2010 16:24:59 Ingo Molnar wrote:
[ You dont even have to document it, as good code is self-explanatory ;-) ]
I recently posted a patch exporting some things through /sys
Hi Mike,
On Thursday 20 May 2010 08:42:21 pm Mike Chan wrote:
v2:
Rebased off of Thomas Renninger's patch for cgroups_cpuacct refactoring,
thanks.
A general comment:
I don't know much about the cgroup stuff.
I am also not sure how exactly power can be measured on this arch based on
frequency
CONFIG_CGROUP_CPUACCT set on x86_64.
Signed-off-by: Thomas Renninger tr...@suse.de
CC: linux-ker...@vger.kernel.org
CC: m...@android.com
CC: men...@google.com
CC: l...@cn.fujitsu.com
CC: contain...@lists.linux-foundation.org
CC: mi...@elte.hu
CC: pet...@infradead.org
---
diff --git a/include/linux
On Wednesday 19 May 2010 15:11:03 Nishanth Menon wrote:
Mike Chan had written, on 05/18/2010 08:30 PM, the following:
Specify new power field in struct omap_opp, which is
power exported in milliWatt.
...
+ totalpower += cpuacct_freq[i] * mpu_opps[i + 1].power;
...
+ unsigned
62 matches
Mail list logo