params: ITERATION: "1000" TARGET: "host" # Sets the amount of data to send in each message DATASIZE: "100" # How many messages each sender/receiver pair should send LOOPS: "100" # Defines how many groups of senders and receivers should be started GRPS: "10" # Defines how many file descriptors each child should use. Note that the # effective number will be twice the amount you set here, as the sender # and receiver children will each open the given amount of file descriptors. FDS: "20" # Sends the data via a pipe instead of the socket (default) PIPE: "false" # Each sender/receiver child will be a POSIX thread of the parent. # The default is to run in process mode (-P) THREADS: "false"

run: steps: - cd ./automated/linux/hackbench/ - "./hackbench.sh -i ${ITERATION} -t ${TARGET} -s ${DATASIZE} -l ${LOOPS} -g ${GRPS} -f ${FDS} -p ${PIPE} -T ${THREADS}" - ../../utils/send-to-lava.sh ./output/result.txt

Hello Tim,

Thank you for this pointer. Shows a lot. I am not so versatile in .yaml, and I need a bit different example.

This is what I need, actually, or similar, so I can store some data as output from the cyclictest command, while I need for the hackbench the following: while true; do hackbench >/dev/null; done

Namely, the ideal example in .yaml will be the following (as /bin/bash script): while true; do hackbench >/dev/null; done & cyclictest -p 80 -n -m -S -h2000

(also included in this email the providers of this good example, might they can help me in no time)

Thank you, Zoran _______

On Wed, Mar 21, 2018 at 3:25 AM, Tim Orling ticotimo@gmail.com wrote:

Rather than trying to run V1 jobs which are obsolete, perhaps you should be asking a different question: Is there an existing LAVA job for hackbench? https://git.linaro.org/qa/test-definitions.git/tree/automated/linux/hackbenc...

On Mon, Mar 19, 2018 at 9:16 AM Zoran S zoran.stojsavljevic.de@gmail.com wrote:

...

Hello,

I have another problem/ignorance with Lava, so I need a bit of push here. I did execute cyclic test (timer interrupt measurements, time of reaction), using as application hackbench, which introduces tremendous application load (in the peak around 300 loads per 1m/3m/15m) on the core.

So, poking around, I found the following JSON script:

https://github.com/kernelci/lava-ci/blob/master/templates/cyclictest/generic...

Since i know (at least this is my understanding) that the Lava V2 works as a pipeline, so the test jobs must be broken to some blocks (deploy, boot and test).

I would like to a bit change this JSON script to get hackbench as load, but much more important is how I can use this JSON script as Lava test job?

I guess, I need to incorporate this script into some yaml format, as explained/described above... Am I correct?

How to do this? Any description, example... Advice?

Thank you, Zoran _______

---

Lava-users mailing list Lava-users@lists.linaro.org https://lists.linaro.org/mailman/listinfo/lava-users