I'm drafting the blueprint [1], and need your favors to collect LTP (Linux Test Project [2]) result on boards that Linaro supports with Natty kernel running on.
The example steps of the testing (I did on i.mx51 babbage with Beta-2 linaro-n-developer) are documented on Whiteboard of [1]. Please help run the test on the boards you have except i.mx51, and send me the result. I appreciate the effort.
[1] https://blueprints.launchpad.net/linux-linaro/+spec/linaro-kernel-o-ras-fix-... [2] http://ltp.sourceforge.net/
W dniu 02.05.2011 10:59, Shawn Guo pisze:
I'm drafting the blueprint [1], and need your favors to collect LTP (Linux Test Project [2]) result on boards that Linaro supports with Natty kernel running on.
Have you seen abrek, launch-control and lava?
You can run LTP via abrek on any board, submit results to launch-control, automate the whole process with lava and poke at the results to create reports with launch-control dashboard.
Best regards ZK
On Mon, May 02, 2011 at 11:00:48AM +0200, Zygmunt Krynicki wrote:
W dniu 02.05.2011 10:59, Shawn Guo pisze:
I'm drafting the blueprint [1], and need your favors to collect LTP (Linux Test Project [2]) result on boards that Linaro supports with Natty kernel running on.
Have you seen abrek, launch-control and lava?
You can run LTP via abrek on any board, submit results to launch-control, automate the whole process with lava and poke at the results to create reports with launch-control dashboard.
Great. That's the info I'm looking for. Since Validation team has LTP in Linaro test pool, I would utilize this great infrastructural support to make my lift easier.
With the info given by Paul Larson, I have got abrek ltp running on my setup, and will try launch-control/lava out later for collecting results on other boards. Thanks, both.
http://validation.linaro.org/launch-control/dashboard/test-runs/fccb906e-730... the results from a recent run I did on panda in the validation lab. I do restrict the default set of tests in ltp to skip a few of the nastier IO ones that run ok on fast machines with sata disks, but on our dev boards just needlessly pound the sd cards, typically long enough to make you think the system is just completely unresponsive.
-Paul Larson
On Mon, May 2, 2011 at 3:59 AM, Shawn Guo shawn.guo@freescale.com wrote:
I'm drafting the blueprint [1], and need your favors to collect LTP (Linux Test Project [2]) result on boards that Linaro supports with Natty kernel running on.
The example steps of the testing (I did on i.mx51 babbage with Beta-2 linaro-n-developer) are documented on Whiteboard of [1]. Please help run the test on the boards you have except i.mx51, and send me the result. I appreciate the effort.
[1] https://blueprints.launchpad.net/linux-linaro/+spec/linaro-kernel-o-ras-fix-... [2] http://ltp.sourceforge.net/
-- Regards, Shawn
linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
Hi Paul,
Since LTP is being used by Validation team, I would really like to link my effort with your team.
On Mon, May 02, 2011 at 07:30:39AM -0500, Paul Larson wrote:
http://validation.linaro.org/launch-control/dashboard/test-runs/fccb906e-730... the results from a recent run I did on panda in the validation lab. I do restrict the default set of tests in ltp to skip a few of the nastier IO
I'm seeing 'runltp -f syscalls' in ltp.py. So you are saying test set 'syscalls' is the default one?
ones that run ok on fast machines with sata disks, but on our dev boards just needlessly pound the sd cards, typically long enough to make you think the system is just completely unresponsive.
I noticed you have great experience with LTP, so you might be the right person to help me draft blueprint [1]. One thing unclear to me is the test scenario/set we should focus on. Is 'syscalls' good enough for us to address in 11.10 cycle?
[1] https://blueprints.launchpad.net/linux-linaro/+spec/linaro-kernel-o-ras-fix-...
-
Paul Larson -
Shawn Guo -
Zygmunt Krynicki