Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1277356 > unrolled thread

Debugging problems with SCHED_FIFO and CPU affinity

Started byFabrice Triboix <fabrice.triboix@v-nova.com>
First post2015-11-25 13:50 +0100
Last post2015-11-25 13:50 +0100
Articles 1 — 1 participant

Back to article view | Back to linux.kernel


Contents

  Debugging problems with SCHED_FIFO and CPU affinity Fabrice Triboix <fabrice.triboix@v-nova.com> - 2015-11-25 13:50 +0100

#1277356 — Debugging problems with SCHED_FIFO and CPU affinity

FromFabrice Triboix <fabrice.triboix@v-nova.com>
Date2015-11-25 13:50 +0100
SubjectDebugging problems with SCHED_FIFO and CPU affinity
Message-ID<qyHGi-6l3-29@gated-at.bofh.it>
Dear All,

We are seeing problem with our application, which is based on Linux.
Our application has many threads and we use SCHED_FIFO for threads that must have a low latency. We also use CPU affinity to try to force an even distribution of the load.

When a given event occurs, one of the SCHED_FIFO thread stops for 83ms and this prevents the correct functionning of the application. Fortunately, this bug is 95% reproducible. Interestingly enough, the problem disappears when I do not set any thread priority and CPU affinity (but the application breaks on the long run).

I am trying to investigate the issue with ftrace, with limited success so far, and I would appreciate a bit of help. I am trying to analyse the ftrace stuff using kernelshark, and I suspect a weird interaction between the the kernel and user-land.

My first question is: Is this the right mailing list to get some help on such an issue?

Many thanks for any help!

  Fabrice Triboix





Company registration number: 08121803
Place of registration: England

This email and any attachments are sent in strictest confidence for the sole use of the addressee
and may contain legally privileged, confidential, and proprietary data. If you are not the intended
recipient, please advise the sender by replying promptly to this email and then delete and destroy
this email and any attachments without any further use, copying or forwarding.

Please consider your environmental responsibility before printing this e-mail
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [standalone]


Back to top | Article view | linux.kernel


csiph-web