On Thu, May 5, 2016 at 9:29 PM, Mark Brown wrote:
> On Thu, May 05, 2016 at 09:01:05PM +0530, Amit Kucheria wrote:
>> On Thu, May 5, 2016 at 5:15 PM, Marcin Juszkiewicz
>
>> > Solution for existing SoCs is usually adding 1MB of SPI flash during design
>> > phase of
On Thu, May 5, 2016 at 5:15 PM, Marcin Juszkiewicz
wrote:
> Solution for existing SoCs is usually adding 1MB of SPI flash during design
> phase of device and store boot loader(s) there. But it is so expensive
> someone would say when it is in 10-30 cents range...
>
> Even 96boards CE specificat
was the missing VBlank support in the DRM
> driver.
>
> Br,
>
> Zoltan
>
> On 2 March 2016 at 10:53, Amit Kucheria wrote:
>>
>> Hi all,
>>
>> For the upcoming 16.03 release of the reference platform, we've been
>> working on a unified kernel
Hi all,
For the upcoming 16.03 release of the reference platform, we've been
working on a unified kernel that works on Hikey, Dragonboard 410c and
also enables enterprise HW that we have access to (D02, Overdrive, APM
XGene).
That kernel is now pushed to github[1] to the 96b/releases/2016.03
bran
On Mon, Apr 20, 2015 at 10:03 PM, Lisa Nguyen wrote:
> Update the README file with bugs, questions, and contribution
> sections and fix the format for readability.
>
> Signed-off-by: Lisa Nguyen
Feel free to add my ack and commit.
> ---
> Changes in v2:
> * Removed references to LAVA
> * Rewrit
On Wed, Apr 15, 2015 at 3:32 AM, Lisa Nguyen wrote:
> Update the README file with bugs, questions, and contribution
> sections and fix the format for readability.
>
> Signed-off-by: Lisa Nguyen
> ---
> README | 46 +++---
> 1 file changed, 39 insertions(
On Fri, Apr 3, 2015 at 1:09 AM, Larry Bassel wrote:
> Add test which verifies capacity calculation on arm
> architecture platforms.
>
> Since there isn't yet a similar calculation done
> for arm64 platforms, this architecture isn't supported
> by this test (at least as of now).
Rephrase to "This
On Thu, Feb 19, 2015 at 5:44 AM, Lisa Nguyen wrote:
> The check function already determines if a subtest fails, passes, or
> skips. No need to check for the return value of the command.
>
> Signed-off-by: Lisa Nguyen
> ---
> cpufreq/cpufreq_08.sh |3 ---
> cpufreq/cpufreq_09.sh |6 --
On Tue, Mar 24, 2015 at 1:58 PM, Vincent Guittot
wrote:
> On 24 March 2015 at 07:34, Amit Kucheria wrote:
>> On Wed, Mar 11, 2015 at 1:08 AM, Larry Bassel
>> wrote:
>>>> root@linaro-developer:~/pm-qa/cputopology# ./cputopology_03.sh
>>>> cputopology
Larry,
Could you also please attach the output of the script on your test platform?
Regards,
Amit
On Mon, Mar 9, 2015 at 3:55 AM, Larry Bassel wrote:
> Add test which verifies capacity calculation on arm
> architecture platforms.
>
> Since there isn't yet a similar calculation done
> for arm64
On Wed, Mar 11, 2015 at 1:08 AM, Larry Bassel wrote:
> On 10 Mar 15 12:30, Lisa Nguyen wrote:
>> On 3 March 2015 at 13:02, Larry Bassel wrote:
>> >
>> > Add test which checks and prints scheduler domain flags.
>> >
>> > Signed-off-by: Larry Bassel
>> > ---
>> > cputopology/cputopology_03.sh |
On Thu, Mar 5, 2015 at 3:49 PM, Vincent Guittot
wrote:
> On 19 November 2014 at 14:14, pi-cheng.chen wrote:
>> This scripts strips all comments in the input JSON file, modifies the
>> parameters according to commnad line arguments, and write the content to a
>> new
>> JSON file.
>>
>> Signed-off
On Tue, Feb 3, 2015 at 3:14 AM, Lisa Nguyen wrote:
> On 1 February 2015 at 22:00, Amit Kucheria wrote:
>> On Mon, Feb 2, 2015 at 7:34 AM, Lisa Nguyen wrote:
>>> Set the sampling_rate to zero if the sampling_rate attribute cannot
>>> be found in either directories
On Tue, Feb 3, 2015 at 6:52 AM, Lisa Nguyen wrote:
> My previous patchset[1] had too many types of patches (cleanups,
> bug fixes, additions, etc.), so it will be broken up much further
> to make it easy on the eyes :)
>
> This particular set focuses on renaming existing sanity scripts,
> adding a
On Mon, Feb 2, 2015 at 11:40 AM, Amit Kucheria wrote:
> Hi Lisa,
>
> On Mon, Feb 2, 2015 at 7:34 AM, Lisa Nguyen wrote:
>> Changes in v2:
>> - Reorganized previous patchset and recreated patches based on
>> logical change
>> - Rewritten logic to find certain att
Hi Lisa,
On Mon, Feb 2, 2015 at 7:34 AM, Lisa Nguyen wrote:
> Changes in v2:
> - Reorganized previous patchset and recreated patches based on
> logical change
> - Rewritten logic to find certain attributes in sysfs
>
> [Let's try again...]
Looks much better and easier to review. Still a few more
On Mon, Feb 2, 2015 at 7:34 AM, Lisa Nguyen wrote:
> With this new change, the final result for the cpufreq_09
> script will always show up and be counted in LAVA.
Why?
> Signed-off-by: Lisa Nguyen
> ---
> cpufreq/cpufreq_09.sh |5 ++---
> 1 file changed, 2 insertions(+), 3 deletions(-)
>
On Mon, Feb 2, 2015 at 7:34 AM, Lisa Nguyen wrote:
> Before searching for specific governors, check to see if the
> scaling_available_governors attribute exists first. It is to add
> more error handling.
This doesn't seem to handle interactive governor that is usually the
default on Android.
> S
On Mon, Feb 2, 2015 at 7:34 AM, Lisa Nguyen wrote:
> Set the sampling_rate to zero if the sampling_rate attribute cannot
> be found in either directories listed in the if condition. This will
> also resolve minor bugs where the sampling_rate is missing.
>
> Signed-off-by: Lisa Nguyen
> ---
> inc
On Mon, Feb 2, 2015 at 7:34 AM, Lisa Nguyen wrote:
> In PM-QA, there are programs written in C (e.g. cpuidle_killer,
> uevent_reader) that run in the background while some scripts are
> being.
>
> However, remove the extra slash and period when it's not necessary
> and create variables to store th
On Tue, Jan 20, 2015 at 6:06 AM, Lisa Nguyen wrote:
> Update the Android makefiles in cputopology and cpuhotplug
> since a few subtests have been added or renamed. Remove
> unnecessary whitespace detected by checkpatch as well.
Have you ensured that the test spec[1] corresponds to the current
scr
On Tue, Jan 20, 2015 at 6:06 AM, Lisa Nguyen wrote:
> Rename z_sanity_check.sh script to cpuhotplug_09.sh to give it a
> more meaningful name and to check if all cpus are back online after
> running previous cpuhotplug tests.
>
> Signed-off-by: Lisa Nguyen
> ---
> cpuhotplug/cpuhotplug_09.sh |
On Tue, Jan 20, 2015 at 6:06 AM, Lisa Nguyen wrote:
> [ It's been long overdue and PM-QA gets a makeover... ]
Indeed.
> This patchset primarily fixes Linaro Bug# 772
> (https://bugs.linaro.org/show_bug.cgi?id=772) to remove
> bashisms in all PM-QA scripts. These scripts will run on
> Android, Op
On Tue, Dec 2, 2014 at 12:57 PM, Viresh Kumar wrote:
> On 2 December 2014 at 12:51, pi-cheng.chen wrote:
>> Add 2 new kind of event for running a memory or a io bounded load.
>> "mem" name for a load is memory bounded, and "iorun" name for a load is io
>> bounded. The default file to be written t
Hi Ivan,
Our main developer was on holidays this week. We'll review these next week.
Thanks for the patches.
Regards,
Amit
On Fri, Oct 31, 2014 at 7:50 PM, Ivan T. Ivanov wrote:
> Hi,
>
> On Fri, 2014-10-24 at 17:45 +0300, Ivan T. Ivanov wrote:
>> Hi,
>>
>> Here small set of patches which fix:
Applied with a fix to check for bad filenames. See earlier commit in the tree.
On Fri, Aug 22, 2014 at 1:58 PM, pi-cheng.chen wrote:
> Currently the serial terminal connected to the boards running idlestat are
> restricted to be at least 80 characters wide to output the report. Otherwise
> idlest
On Fri, Aug 8, 2014 at 7:02 PM, Sanjay Singh Rawat
wrote:
>
> On Friday 08 August 2014 10:43 AM, Amit Kucheria wrote:
>>
>> On Fri, Aug 8, 2014 at 7:05 AM, Lisa Nguyen
>> wrote:
>>>
>>> Create a global variable that will look for number of cpus
>>&
On Thu, Aug 7, 2014 at 8:27 PM, Juri Lelli wrote:
> Hi Amit,
>
> On 06/08/14 12:35, Amit Kucheria wrote:
>> Hi Juri,
>>
>> On Wed, Aug 6, 2014 at 3:38 PM, Juri Lelli wrote:
>>> Hi Vincent,
>>>
>>> I'm actually adding Chris to Cc: as
On Fri, Aug 8, 2014 at 7:05 AM, Lisa Nguyen wrote:
> Create a global variable that will look for number of cpus
> for a platform and use it in multiple functions to make the
> code simplier to read.
Agree with the patch.
> Signed-off-by: Lisa Nguyen
> ---
> diff --git a/cpuhotplug/cpuhotplug_s
Hi Juri,
On Wed, Aug 6, 2014 at 3:38 PM, Juri Lelli wrote:
> Hi Vincent,
>
> I'm actually adding Chris to Cc: as he is also part of the
> original discussion about rt-app.
>
> On 05/08/14 22:45, Vincent Guittot wrote:
>> Hi Juri,
>>
>> Here is the pull request for the changes that create a worklo
On Tue, Aug 5, 2014 at 1:39 PM, Sanjay Singh Rawat
wrote:
> - add the governor sampling rate with the frequency transition latency to
> make sure tansition happens.
> - bug 77
Replace with complete URL: https://bugs.linaro.org/show_bug.cgi?id=77
to make git changelogs much more useful.
> Signed-
igned-off-by: Lisa Nguyen
Acked-by: Amit Kucheria
Please apply this to the PM-QA git tree to allow Milosz to take it for a spin.
Regards,
Amit
> ---
> Changes from v3 to v4:
> - Remove =~ symbol and rewrote test condition in
> check_cpuhotplug_files() function
>
> Chan
On Thu, Jul 3, 2014 at 3:38 PM, Jon Medhurst (Tixy) wrote:
> On Thu, 2014-07-03 at 14:59 +0530, Amit Kucheria wrote:
>> On Thu, Jul 3, 2014 at 2:22 PM, Jon Medhurst (Tixy) wrote:
>> > On Thu, 2014-07-03 at 10:24 +0530, Amit Kucheria wrote:
>> >> On Thu, Jul 3
On Thu, Jul 3, 2014 at 2:22 PM, Jon Medhurst (Tixy) wrote:
> On Thu, 2014-07-03 at 10:24 +0530, Amit Kucheria wrote:
>> On Thu, Jul 3, 2014 at 7:37 AM, Lisa Nguyen wrote:
> [...]
>> > diff --git a/include/functions.sh b/include/functions.sh
>> > index 6d75e34..d44
On Thu, Jul 3, 2014 at 7:37 AM, Lisa Nguyen wrote:
> By default, cpuhotplug does not run on cpu0. However, if
> we want to allow cpuhotplug to run on cpu0, then we must
> pass the parameter, hotplug_allow_cpu0=1, to make check.
>
> Signed-off-by: Lisa Nguyen
> ---
> Changes from v2 to v3:
> - Ren
uot; tab and powertop version
> shows 2.6.1.
>
>
>
> On Tue, Jul 1, 2014 at 6:47 PM, Sergey Senozhatsky
> wrote:
>> On (07/01/14 17:54), Amit Kucheria wrote:
>>> Date: Tue, 1 Jul 2014 17:54:15 +0530
>>> From: Amit Kucheria
>>> To: Mohammad Meraju
Mohammad,
This fix should go upstream. cc'ing the powertop list.
Regards,
Amit
On Thu, Jun 26, 2014 at 12:42 PM, Mohammad Merajul Islam Molla
wrote:
> Fixes some resource leaks detected by valgrind and coverity scan.
>
>
> diff --git a/src/devices/ahci.cpp b/src/devices/ahci.cpp
> index ac06460
Mohammad,
This fix should go upstream. cc'ing the powertop list.
Regards,
Amit
On Thu, Jun 26, 2014 at 12:27 PM, Mohammad Merajul Islam Molla
wrote:
> powertop creates temporary file in /tmp folder during calibration.
> Simply closing does not remove these files.
> Instead use unlink to remove
Some nits below:
On Tue, Jul 1, 2014 at 5:35 AM, Lisa Nguyen wrote:
> Run the cpuhotplug test suite on all cpus including cpu0. If we wish
> to skip cpu0, pass the parameter, hotplug_allow_cpu0=1, to make check
> like in these examples:
Umm, this description is the opposite of what the default c
On Wed, Jun 25, 2014 at 9:48 AM, Lisa Nguyen wrote:
> Pass hotplug_cpu_start=0 to make check like
> these examples below:
>
> sudo make -C cpuhotplug hotplug_cpu_start=0 check
> sudo make hotplug_cpu_start=0 check
>
> to enable cpuhotplug to run on cpu0 to N.
This description doesn't match the be
On Wed, Jun 25, 2014 at 8:21 PM, Sanjay Singh Rawat
wrote:
> Milosz,
>
> Yes i meant that. Hmm, i think i am missing the understanding here. So now
> all cpu are hotpluggable?
Technically any cpu can be hotpluggable. But you need to have atleast
one plugged in at anytime else there won't be a ker
Can you try changing your workload?
On Fri, Jun 20, 2014 at 4:57 PM, Salman Ahmed wrote:
> I tried 10 times running 15 seconds and out of that 3 times I got output
> without any P-state information.
>
> --
> Thanks,
> Salman Ahmed
>
>
>
> On Friday, 20 June 2014,
On Fri, Jun 20, 2014 at 4:01 PM, Salman Ahmed wrote:
> Hello Amit,
>
> Actually, I ran it on the same machine with the same command again and these
> time the output has frequency information-
>
> $ sudo ./idlestat --trace -t 5 -f /tmp/myoutput -- top
>
> Log is 4.999547 secs long with 710 events
On Fri, Jun 20, 2014 at 9:47 AM, Salman Ahmed wrote:
> Hello Amit,
>
> I have pulled changes and run with your suggested following command
> sudo ./idlestat --trace -t 5 -f /tmp/myoutput -- top
>
> But still it came up with the output without any frequency information.
What platform are you runni
Hi Salman,
On Thu, Jun 19, 2014 at 1:36 PM, Salman Ahmed wrote:
> Hello,
> Running top command with idlestat tool doesn't show any frequency
> information sometimes. I have run the program with the following
> command -
> # sudo idlestat --trace -t 5 -f /tmp/myoutput /usr/bin/top
> and it ended u
Thanks for the fixes Mohammad.
On Mon, Jun 9, 2014 at 8:57 AM, Mohammad Merajul Islam Molla
wrote:
> Patch to fix some minor issues in idlestat tool -
>
>- use USEC_PER_SEC instead of hardcoded value
>- replace 'f' with 'stderr'
>- check for root only if idlestat is run in trace mode
Thanks. Pushed out to git.
On Tue, Jun 3, 2014 at 6:08 PM, Daniel Lezcano
wrote:
> On 06/02/2014 02:49 PM, Amit Kucheria wrote:
>>
>> From: Amit Kucheria
>>
>> More cleanups to the command-line options to make idlestat easier to use
>> and
>> separa
From: Amit Kucheria
More cleanups to the command-line options to make idlestat easier to use and
separate the two modes: tracing and reporting.
Regards,
Amit
Amit Kucheria (6):
Use the correct version in the log file
Print more details when fopen fails
kick the dump option off to use -z
From: Amit Kucheria
Signed-off-by: Amit Kucheria
---
idlestat.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/idlestat.c b/idlestat.c
index 9228618..9d88ee9 100644
--- a/idlestat.c
+++ b/idlestat.c
@@ -1085,7 +1085,7 @@ static int idlestat_store(const char *path
From: Amit Kucheria
Useful for tracking down why the following command fails:
sudo ./idlestat -o /tmp/newfile
where, newfile doesn't already exist
Signed-off-by: Amit Kucheria
---
idlestat.c | 11 +++
1 file changed, 7 insertions(+), 4 deletions(-)
diff --git a/idlesta
From: Amit Kucheria
Signed-off-by: Amit Kucheria
---
idlestat.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/idlestat.c b/idlestat.c
index a5619a9..787d7b0 100644
--- a/idlestat.c
+++ b/idlestat.c
@@ -46,7 +46,7 @@
#include "list.h"
#include "topolog
From: Amit Kucheria
idlestat can take a pre-created tracefile as input or create one itself. The
-o option is confusing for an input file. Change it to -f in preparation for a
new option to signify what the file name is to be used for.
Signed-off-by: Amit Kucheria
---
README | 4
From: Amit Kucheria
Signed-off-by: Amit Kucheria
---
idlestat.c | 12 ++--
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/idlestat.c b/idlestat.c
index 998f62d..233d554 100644
--- a/idlestat.c
+++ b/idlestat.c
@@ -948,7 +948,7 @@ struct cpuidle_cstates
From: Amit Kucheria
--trace will create a local trace and use that for the report
--import will import an existing trace
Signed-off-by: Amit Kucheria
---
README | 10 --
idlestat.c | 36 +++-
2 files changed, 35 insertions(+), 11 deletions(-)
diff
Hi Mohammad,
Yes, I found and fixed that bug in my local tree. Will push it out later today.
In the meanwhile, please use the -t option (it is compulsory) to take
traces on the local machine.
Regards,
Amit
On Mon, Jun 2, 2014 at 2:03 PM, Mohammad Merajul Islam Molla
wrote:
> Hi
>
> I was tryin
Thank for the review. I've made the requested change and pushed all
these fixes with your Ack.
On Fri, May 30, 2014 at 1:13 PM, Daniel Lezcano
wrote:
> On 05/29/2014 12:17 AM, Amit Kucheria wrote:
>>
>> Change the flag field to NULL while we're at it
>>
Signed-off-by: Amit Kucheria
---
idlestat.c | 24
1 file changed, 12 insertions(+), 12 deletions(-)
diff --git a/idlestat.c b/idlestat.c
index 27000aa..8e3cec0 100644
--- a/idlestat.c
+++ b/idlestat.c
@@ -957,11 +957,11 @@ static void version(const char *cmd)
static
Change the flag field to NULL while we're at it
Signed-off-by: Amit Kucheria
---
idlestat.c | 21 ++---
1 file changed, 10 insertions(+), 11 deletions(-)
diff --git a/idlestat.c b/idlestat.c
index 96fc1f2..c4a405a 100644
--- a/idlestat.c
+++ b/idlestat.c
@@ -955,17 +
Switch the dump option (which might be used infrequently) to use -m
Signed-off-by: Amit Kucheria
---
idlestat.c | 14 +++---
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/idlestat.c b/idlestat.c
index 48e4166..a3b9363 100644
--- a/idlestat.c
+++ b/idlestat.c
@@ -946,7
Signed-off-by: Amit Kucheria
---
README | 7 +--
idlestat.c | 11 ---
2 files changed, 13 insertions(+), 5 deletions(-)
diff --git a/README b/README
index 3a8eb49..24485c5 100644
--- a/README
+++ b/README
@@ -28,14 +28,17 @@ Requirements
1. It must be run as root in order to
commit 1e80e5eaf71b606405cbdf6751aa7446533d8b3a (idlestat: get ride of
trace-cmd) seems to have missed adding this
Signed-off-by: Amit Kucheria
---
idlestat.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/idlestat.c b/idlestat.c
index 8e3cec0..7de9c1d 100644
--- a/idlestat.c
+++ b
Hi,
These are a few cleanups around idlestat commandline options while I was
working on reformating the report.
Regards,
Amit
Amit Kucheria (9):
'verbose' option isn't used anywhere, remove it
Sort options in alphabetic order of the single-char options
Adding missing '
Adding ':' as first character of optstring to getopt_long allows separating
error message for invalid option and invalid option arguments
Signed-off-by: Amit Kucheria
---
idlestat.c | 15 ++-
1 file changed, 10 insertions(+), 5 deletions(-)
diff --git a/idlestat.c b/
Signed-off-by: Amit Kucheria
---
idlestat.c | 16
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/idlestat.c b/idlestat.c
index 7de9c1d..48e4166 100644
--- a/idlestat.c
+++ b/idlestat.c
@@ -956,14 +956,14 @@ static void version(const char *cmd)
}
static struct
Signed-off-by: Amit Kucheria
---
idlestat.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/idlestat.c b/idlestat.c
index 3b4bd17..27000aa 100644
--- a/idlestat.c
+++ b/idlestat.c
@@ -960,7 +960,6 @@ static struct option long_options[] = {
{ "iterations",
Signed-off-by: Amit Kucheria
---
idlestat.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/idlestat.c b/idlestat.c
index 849b6dd..00b2032 100644
--- a/idlestat.c
+++ b/idlestat.c
@@ -959,7 +959,7 @@ static void version(const char *cmd)
printf("%s version
Hi Mohammad,
Thanks for investigating. Care to send this patch as a diff?
Also, we'll need to come up with a better solution that to add the
"Samsung" string in there before we submit this upstream.
Regards,
Amit
On Tue, May 6, 2014 at 11:06 AM, Mohammad Merajul Islam Molla
wrote:
> Hello,
>
>
On Wed, Apr 23, 2014 at 5:11 PM, Sanjay Singh Rawat wrote:
> Signed-off-by: Sanjay Singh Rawat
> ---
> - This will help in catching issue like
> https://lkml.org/lkml/2014/3/25/122
>
> - Tested on my x86 dev m/c
> Output:
> running hotplug stress on cpu : cpu1
> pass: cpu hotplug count = 200 dur
Thanks for the patch Mohammad.
Sanjay, I thought we detect terminal size at startup. Can we change this
dynamically?
On Mon, Apr 21, 2014 at 1:44 PM, Mohammad Merajul Islam Molla <
meraj.eni...@gmail.com> wrote:
> Hello,
>
> This is for launchpad bug 1307896 (
> https://bugs.launchpad.net/linar
Hi Daniel,
Have you noticed this on any platform yet with this test?
Regards,
Amit
On Tue, Apr 8, 2014 at 4:05 PM, Daniel Lezcano wrote:
> As pointed by Lorenzo, when a cpu powers down, the L1 cache must be flushed
> before, otherwise:
>
> * data cachelines are not empty and the other cpu may
On Thu, Oct 17, 2013 at 6:39 AM, Christoffer Dall
wrote:
> Hi all,
>
> [apologies in advance for the spam nature of this message]
>
> I have been using a tool for remote collaboration and problem solving
> that I find perfect in the context of remote collaboration. It allows
> you to very easily
On Tue, Oct 1, 2013 at 7:53 PM, Dave Pigott wrote:
>
> On 1 Oct 2013, at 14:50, Sanjay Singh Rawat wrote:
>
>> On Tuesday 01 October 2013 06:45 PM, Gaurang Shastri wrote:
>>> Hi,
>>>
>>> Do you mean to say something like lm-sensor kind of utility?? May be
>>> from this package you will get to kno
On Thu, May 30, 2013 at 7:36 AM, Andy Green wrote:
> Hi -
>
> We're using one kernel binary with BL Switcher enabled in config, but able
> to work on SoC without Big Little.
>
> This is OK except where the BL patches touch the PMU driver. It makes an
> assumption about BL configured == in use whi
e.g. interrupt
> disabling, locks). How does SCHED_DEADLINE handle this?
>
> -- Ola
>
> Ola Liljedahl, Networking System Architect, ARM
> Telephone: +46 706 866 373Skype: ola.liljedahl
>
>
> -Original Message-
> From: Juri Lelli [mailto:juri.le...@gmail.com]
&g
On Mon, May 20, 2013 at 5:48 PM, Juri Lelli wrote:
> On 05/20/2013 12:27 PM, Ola Liljedahl wrote:
>>
>> Predictable response times for specific threads. In the use cases I
>> am aware of, these threads would probably belong to the same process
>> and run core-affine on a specific core where as few
On Mon, May 20, 2013 at 3:45 PM, Viresh Kumar wrote:
> On 20 May 2013 15:42, Amit Kucheria wrote:
>> Do you also have a tree based on linux-linaro where the reference MP
>> patches are removed and this branch applied so Naresh and Vincent can
>> continue their benchmarking
>
>
>
> -- Ola
>
>
>
> Ola Liljedahl, Networking System Architect, ARM
>
> Telephone: +46 706 866 373Skype: ola.liljedahl
>
>
>
> From: Amit Kucheria [mailto:amit.kuche...@linaro.org]
> Sent: 19 May 2013 11:51
> To: Mark Orvek
> Cc: Patrick M
Viresh,
Thanks!
Do you also have a tree based on linux-linaro where the reference MP
patches are removed and this branch applied so Naresh and Vincent can
continue their benchmarking experiments? Or will we wait until Andrey
has finalised the linux-linaro tree before you create this?
Regards,
Am
nchmarking interesting to
>>> interpret.
>>> In addition there are discussions about adding zero overhead linux (ZOL)
>>> like features.
>>>
>>> Mike
>>>
>>>
>>>
>>> On Friday, May 17, 2013 6:08:20 AM UTC-4, David Rusling
Sanjay,
This deserves to be posted to LAKML and even LKML I think.
/Amit
On Sun, May 19, 2013 at 12:38 AM, Sanjay Singh Rawat
wrote:
> Various SoCs with arm processors define their own platform hotplug related
> functions. These functions can be defined as part of the core cpu hotplug
> functio
Hi all,
As part of our investigations into the Linux scheduler we've
interacted with Juri Lelli at the University of Pisa (cc'ed) who is
part of a group that is working on a DEADLINE scheduler[1] for
Linux[2].
While we're coming at this from a power managment angle[3], I suspect
that LEG and LNG
Sebastian,
Could you help Sanjay with verification of this feature?
Regards,
Amit
On Sun, May 5, 2013 at 1:57 AM, Sanjay Singh Rawat
wrote:
> - use the clk_summary file of CCF for showing clocks
> - add option to show active or all clocks
>
> Signed-off-by: Sanjay Singh Rawat
> ---
> clocks.
On Thu, May 2, 2013 at 12:03 AM, Nicolas Pitre wrote:
> I'm delighted to announce that the Linaro Technical Steering Committee
> has approved the release of the big.LITTLE in-kernel switcher (IKS) code
> to the public.
Good stuff Nico.
> So here it is:
>
> Branch big.LITTLE-IKS-snapshot of
> git
cc'ing linaro-dev
On Tue, Apr 23, 2013 at 5:52 AM, zoran markovic
wrote:
> Hi,
> Working as a newbie in the PMWG, I noticed I'm not able to resume my
> pandaboard-es with the latest 3.9 kernel from Linus (configuration file
> omap2plus_defconfig). Suspend/resume appears to work with the Linaro 12
Hi Sanjay,
Both patches look OK. You should mention the bug number in your commit
message to show that it solves an open bug.
Regards,
Amit
On Mon, Apr 8, 2013 at 5:39 PM, Sanjay Singh Rawat
wrote:
> pmqa should be on a RW permission partition. Need to change the path
> in lava pm-qa script too
On Mon, Mar 25, 2013 at 5:37 PM, Sanjay Singh Rawat
wrote:
> From: Daniel Lezcano
>
> test checks for topology attributes of cpus
>
> Signed-off-by: Sanjay Singh Rawat
Ack modulo one comment below.
> ---
> cputopology/Android.mk | 18 ++
> cputopology/Makefile
Sanjay, Hongbo,
Nack to removing all the tests. SCHED_MC still exists but it deals
only with CPU topology (powersavings support under SCHED_MC was
removed from kernel in shaid 8e7fbcbc)
I just looked through these and test 02 and 03 check for topology
files which is desirable to keep.
So I sugge
On Sun, Mar 10, 2013 at 8:38 AM, Viresh Kumar wrote:
> Hi Guys,
>
> Below are hangout upstreams of Scheduler Internals by Vincent Guittot
> done in LCA13.
>
> We have got another version of this recording that is done by some
> other cameras, but
> its size was 30 GB and so hard to upstream. In ca
On Thu, Feb 28, 2013 at 4:56 PM, Amit Kucheria wrote:
> Hi,
>
> Please find a link[1] to some of the things we plan to discuss and
> work on in Hong Kong next week.
>
> If you're interested in some of these topics and are attending in
> person, please come and say hell
Hi,
Please find a link[1] to some of the things we plan to discuss and
work on in Hong Kong next week.
If you're interested in some of these topics and are attending in
person, please come and say hello.
See you in Hong Kong!
Regards,
Amit
---
PMWG Te
On Thu, Feb 21, 2013 at 4:27 PM, Viresh Kumar wrote:
> On 18 February 2013 23:50, Serge Broslavsky
> wrote:
>> On 18.02.13 12:20 -0500, Nicolas Pitre wrote:
>>> For the record, I think what Peter is suggesting is reasonable. There
>>> must be better ways for people to find out what Linaro is par
On Thu, Feb 21, 2013 at 11:24 AM, Botao Sun wrote:
> Calendar Week 8, 2013: Here is test result summary for Linux Linaro ubuntu
> Quantal image on following boards:
>
> 1) ARM Versatile Express A9;
> 2) Samsung Arndale;
> 3) TI Panda 4430;
> 4) TI Panda 4460;
> 5) ST Ericsson Snowball.
>
> Synopsi
On Tue, Feb 19, 2013 at 10:11 AM, Sanjay Singh Rawat
wrote:
> Add functionality to record the overall and current runtime cpuidle
> statistics.
I'd like the output to be something list this:
[state1]: [number of times entered]: [average residency]: [std.
deviation(residency)]: [variance]
[state2
On Wed, Feb 20, 2013 at 10:35 AM, Viresh Kumar wrote:
> Hi Guys,
>
> I really don't know whom to direct this mail to and hence the wide spread.
>
> Problem: When we send a mail to kernel mailing lists with linaro-dev
> or linaro-kernel
> in cc, and we get replies to those mails, sometimes the mail
Since Peter and I were discussing this on IRC that led to this email,
here is a bit of context on the current behaviour:
On Mon, Feb 18, 2013 at 8:16 PM, Peter Maydell wrote:
> Hi; does anybody else think it would be a good idea to move all
> the kernel patch email traffic off linaro-dev and onto
On Thu, Feb 7, 2013 at 3:59 PM, Sanjay Singh Rawat
wrote:
> For common clock framework directory and attributes are changed,
> adapt to the new structure.
What version of the kernel did these changes happen in? It would be
useful to note that in the commit message here.
> Signed-off-by: Sanjay S
On Wed, Feb 6, 2013 at 3:28 PM, Viresh Kumar wrote:
> On 5 February 2013 21:51, Viresh Kumar wrote:
>> commit 15b5548c9ccfb8088270f7574710d9d67edfe33b
>> Author: Viresh Kumar
>> Date: Tue Feb 5 21:29:05 2013 +0530
>>
>> cpufreq: Make governors directory sysfs location based on
>> have_mult
> Hi -
>
> We added a bunch of thermal-related bits (different from the thermal bits
> that Linaro PM guys got upstream) to the tilt- kernels. It's support for
> their thermal sensor and TI's "thermal framework" which monitors it and
> messes with the cpu frequency limit.
:) Upstreamed and seen
On Mon, Jan 28, 2013 at 11:38 AM, Hongbo Zhang wrote:
> This script aims at testing all the trip points are crossed or not, since the
> trip points are usually very high and can be reached only in extreme
> conditions
> like high environment temerature etc, in normal conditions they cannot be all
On Fri, Jan 25, 2013 at 9:08 AM, Sanjay Singh Rawat
wrote:
> On Thursday 24 January 2013 11:24 PM, Mike Turquette wrote:
>>
>> Quoting Sanjay Singh Rawat (2013-01-24 08:01:33)
You're missing a changelog describing why this patch is required. :)
>>>
>>> Signed-off-by: Sanjay Singh Rawat
>>
>>
>
1 - 100 of 399 matches
Mail list logo
