hi there,
at HK connect i learned about the ARM energy probe ( http://www.arm.com/about/newsroom/arm-enables-energy-aware-coding-with-the-a...). Since that we ordered some probes, and I was wondering if someone is working in linaro on enabling power measurements? I would be interested in DS5 or even command line tools to retrieve information, ideally from Panda...
thx
I've been looking a bit at how to get this running under DS-5, but I haven't found much documentation. Best thing I've found so far has been http://www.youtube.com/watch?v=aDStdtopy_g I tried going through this, and even found caiman on my box at /usr/local/DS-5/bin/caiman However, when I try to replicate what they did here, I get a trace ok, but no power data in my graph. You may want to give it a try though, maybe it will work for you.
On Thu, Jul 12, 2012 at 11:51 AM, Dechesne, Nicolas n-dechesne@ti.comwrote:
hi there,
at HK connect i learned about the ARM energy probe ( http://www.arm.com/about/newsroom/arm-enables-energy-aware-coding-with-the-a...). Since that we ordered some probes, and I was wondering if someone is working in linaro on enabling power measurements? I would be interested in DS5 or even command line tools to retrieve information, ideally from Panda...
thx
linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
Hi,
There is documentation at: http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.dui0482h/ch13.ht...
Feel free to let me know if it's useful or not - I didn't write it, but I'm always happy to pass feedback on.
And you could contact ARM support for help (via connect - https://silver.arm.com/).
Regards,
Bernie
From: linaro-dev-bounces@lists.linaro.org [mailto:linaro-dev-bounces@lists.linaro.org] On Behalf Of Paul Larson Sent: 12 July 2012 20:54 To: Dechesne, Nicolas Cc: Linaro Dev Subject: Re: Using ARM Energy probe and PandaBoard
I've been looking a bit at how to get this running under DS-5, but I haven't found much documentation. Best thing I've found so far has been http://www.youtube.com/watch?v=aDStdtopy_g I tried going through this, and even found caiman on my box at /usr/local/DS-5/bin/caiman However, when I try to replicate what they did here, I get a trace ok, but no power data in my graph. You may want to give it a try though, maybe it will work for you. On Thu, Jul 12, 2012 at 11:51 AM, Dechesne, Nicolas <n-dechesne@ti.commailto:n-dechesne@ti.com> wrote: hi there,
at HK connect i learned about the ARM energy probe (http://www.arm.com/about/newsroom/arm-enables-energy-aware-coding-with-the-a...). Since that we ordered some probes, and I was wondering if someone is working in linaro on enabling power measurements? I would be interested in DS5 or even command line tools to retrieve information, ideally from Panda...
thx
_______________________________________________ linaro-dev mailing list linaro-dev@lists.linaro.orgmailto:linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
-- IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.
On Thu, Jul 12, 2012 at 02:54:15PM -0500, Paul Larson wrote:
I've been looking a bit at how to get this running under DS-5, but I haven't found much documentation. Best thing I've found so far has been http://www.youtube.com/watch?v=aDStdtopy_g I tried going through this, and even found caiman on my box at /usr/local/DS-5/bin/caiman
However, when I try to replicate what they did here, I get a trace ok, but no power data in my graph. You may want to give it a try though, maybe it will work for you.
Zach, Alexander and I spent some time trying to get it to work in Hong Kong and got to exactly the same results as you are describing.
IIRC Amit knew how to set it up and get it to produce true results, or am I misremembering?
On 12 Jul 18, Christian Robottom Reis wrote:
On Thu, Jul 12, 2012 at 02:54:15PM -0500, Paul Larson wrote:
I've been looking a bit at how to get this running under DS-5, but I haven't found much documentation. Best thing I've found so far has been http://www.youtube.com/watch?v=aDStdtopy_g I tried going through this, and even found caiman on my box at /usr/local/DS-5/bin/caiman
However, when I try to replicate what they did here, I get a trace ok, but no power data in my graph. You may want to give it a try though, maybe it will work for you.
Zach, Alexander and I spent some time trying to get it to work in Hong Kong and got to exactly the same results as you are describing.
IIRC Amit knew how to set it up and get it to produce true results, or am I misremembering?
We've been able to get it working with DS-5 just fine. Perhaps Vincent will be able to guide you through the process since he is in a better timezone. He's vingu on irc.
/Amit
On Wed, 2012-07-18 at 17:01 -0300, Christian Robottom Reis wrote:
On Thu, Jul 12, 2012 at 02:54:15PM -0500, Paul Larson wrote:
I've been looking a bit at how to get this running under DS-5, but I haven't found much documentation. Best thing I've found so far has been http://www.youtube.com/watch?v=aDStdtopy_g I tried going through this, and even found caiman on my box at /usr/local/DS-5/bin/caiman
However, when I try to replicate what they did here, I get a trace ok, but no power data in my graph. You may want to give it a try though, maybe it will work for you.
Zach, Alexander and I spent some time trying to get it to work in Hong Kong and got to exactly the same results as you are describing.
If I was to guess I'd say you hit a device detection bug in caiman. In 99% cases it can be easily worked around (providing you use 5.10 release) by enforcing the device node name in the cayman configuration line.
So, click "Capture options" button (the cogwheel like icon, the middle one of three) and make sure that the "Energy Capture" input contains
<DS-5-path>/bin/caiman -d /dev/ttyACM0
Good luck!
Paweł
On Thu, Jul 19, 2012 at 5:04 AM, Pawel Moll pawel.moll@arm.com wrote:
On Wed, 2012-07-18 at 17:01 -0300, Christian Robottom Reis wrote:
On Thu, Jul 12, 2012 at 02:54:15PM -0500, Paul Larson wrote:
I've been looking a bit at how to get this running under DS-5, but I haven't found much documentation. Best thing I've found so far has
been
http://www.youtube.com/watch?v=aDStdtopy_g I tried going through this, and even found caiman on my box at /usr/local/DS-5/bin/caiman
However, when I try to replicate what they did here, I get a trace ok,
but
no power data in my graph. You may want to give it a try though,
maybe it
will work for you.
Zach, Alexander and I spent some time trying to get it to work in Hong Kong and got to exactly the same results as you are describing.
If I was to guess I'd say you hit a device detection bug in caiman. In 99% cases it can be easily worked around (providing you use 5.10 release) by enforcing the device node name in the cayman configuration line.
So, click "Capture options" button (the cogwheel like icon, the middle one of three) and make sure that the "Energy Capture" input contains
<DS-5-path>/bin/caiman -d /dev/ttyACM0I have 5.10 (build 1389)
I tried putting this in the energy capture field: /usr/local/DS-5/bin/caiman -d /dev/ttyACM0 (I also see /usr/local/DS-5/sw/energy_probe/caiman, but the one in /usr/local/DS-5/bin seems to be the script that sets things up for it and calls the latter)
I tried this, but still not getting anything. The only things I see in the timeline after running the capture are: Linux Scheduler Instruction Disk IO Memory Network
Also, fwiw, I tried running caiman from the command line, even with specifying -d /dev/ttyACM0 and I get no output. Not sure if I'm supposed to get anything though. All it does is create an empty file called "0000000000" Thanks, Paul Larson
On Thu, 2012-07-19 at 19:44 +0100, Paul Larson wrote:
So, click "Capture options" button (the cogwheel like icon, the middle one of three) and make sure that the "Energy Capture" input contains <DS-5-path>/bin/caiman -d /dev/ttyACM0I have 5.10 (build 1389) I tried putting this in the energy capture field: /usr/local/DS-5/bin/caiman -d /dev/ttyACM0 (I also see /usr/local/DS-5/sw/energy_probe/caiman, but the one in /usr/local/DS-5/bin seems to be the script that sets things up for it and calls the latter)
Ok, last thing to check would be the /dev/ttyACM0 permissions... Can you actually do "cat /dev/ttyACM0"? It's a serial device, so on Ubuntu it will be owned by "dialout" group.
Also, fwiw, I tried running caiman from the command line, even with specifying -d /dev/ttyACM0 and I get no output. Not sure if I'm supposed to get anything though. All it does is create an empty file called "0000000000"
Yeah, it doesn't actually start talking to the probe before Streamline connects to it...
Paweł
On 20 July 2012 02:44, Paul Larson paul.larson@linaro.org wrote:
On Thu, Jul 19, 2012 at 5:04 AM, Pawel Moll pawel.moll@arm.com wrote:
On Wed, 2012-07-18 at 17:01 -0300, Christian Robottom Reis wrote:
On Thu, Jul 12, 2012 at 02:54:15PM -0500, Paul Larson wrote:
I've been looking a bit at how to get this running under DS-5, but I haven't found much documentation. Best thing I've found so far has
been
http://www.youtube.com/watch?v=aDStdtopy_g I tried going through this, and even found caiman on my box at /usr/local/DS-5/bin/caiman
However, when I try to replicate what they did here, I get a trace
ok, but
no power data in my graph. You may want to give it a try though,
maybe it
will work for you.
Zach, Alexander and I spent some time trying to get it to work in Hong Kong and got to exactly the same results as you are describing.
If I was to guess I'd say you hit a device detection bug in caiman. In 99% cases it can be easily worked around (providing you use 5.10 release) by enforcing the device node name in the cayman configuration line.
So, click "Capture options" button (the cogwheel like icon, the middle one of three) and make sure that the "Energy Capture" input contains
<DS-5-path>/bin/caiman -d /dev/ttyACM0I have 5.10 (build 1389)
I tried putting this in the energy capture field: /usr/local/DS-5/bin/caiman -d /dev/ttyACM0 (I also see /usr/local/DS-5/sw/energy_probe/caiman, but the one in /usr/local/DS-5/bin seems to be the script that sets things up for it and calls the latter)
I tried this, but still not getting anything. The only things I see in the timeline after running the capture are: Linux Scheduler Instruction Disk IO Memory Network
Also, fwiw, I tried running caiman from the command line, even with specifying -d /dev/ttyACM0 and I get no output. Not sure if I'm supposed to get anything though. All it does is create an empty file called "0000000000" Thanks, Paul Larson
Hi all, I am facing the same problem as Paul's, is there any solution? Thanks.
linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
On Fri, 2012-09-14 at 11:44 +0100, Hongbo Zhang wrote:
Hi all, I am facing the same problem as Paul's, is there any solution?
There's a couple things you should try:
1. Upgrade to 5.11 2. Check /dev/ttyACM0 permissions (try to do "cat /dev/ttyACM0" - don't expect any data out of it, just see if you get any error) 3. Check warnings in the capture directory, eg. ~/Documents/Streamline/Capture_C02.apc/energy/warnings.xml (and if there is anything inside, do let me know).
Cheers!
Paweł
On 12 July 2012 11:51, Dechesne, Nicolas n-dechesne@ti.com wrote:
hi there,
at HK connect i learned about the ARM energy probe (http://www.arm.com/about/newsroom/arm-enables-energy-aware-coding-with-the-a...). Since that we ordered some probes, and I was wondering if someone is working in linaro on enabling power measurements? I would be interested in DS5 or even command line tools to retrieve information, ideally from Panda...
We got an NI PXIe-4154 working. Its compatible with DS-5. Here's a video of me using it on Panda:
https://plus.google.com/104422661029399872488/posts/53H7S5sMXN6
thx
linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
-
Amit Kucheria -
Bernard Ogden -
Christian Robottom Reis -
Dechesne, Nicolas -
Hongbo Zhang -
Paul Larson -
Pawel Moll -
Zach Pfeffer