From: Charles Steinkuehler <charles@steinkuehler.net>
To: Gilles Chanteperdrix <gilles.chanteperdrix@xenomai.org>
Cc: xenomai@xenomai.org
Subject: Re: [Xenomai] Hung task on Xenomai patched ARM 3.8.13 BeagleBone Kernel
Date: Wed, 16 Oct 2013 06:10:43 -0500 [thread overview]
Message-ID: <525E7433.6070807@steinkuehler.net> (raw)
In-Reply-To: <525E6E17.8020103@xenomai.org>
On 10/16/2013 5:44 AM, Gilles Chanteperdrix wrote:
> It points to several possible problems. So, a question, do you get this
> while a Xenomai task is running? Or on an otherwise idle system?
> Second question: is the Linux timer system running correctly, for
> instance does "sleep 10" sleeps a little bit before returning, or does
> it fail to return too?
In the test case, the system is otherwise idle when this happens. It is
the large amount of "disk" I/O to the SD card that causes the mmcqd
process to hang. Also, the grep -r doesn't fail immediately. It runs
for a while, sometimes getting far enough it finds one or two of the
four matching files in /usr, then the mmcqd task hangs.
Other than this occasional glitch, the kernel and system work normally.
This is the same kernel I am using to run LinuxCNC and I've managed to
get through many overnight 8+ hour prints, and have had uptimes of
several weeks.
I suspect an occasional lost interrupt or maybe a race condition or
something specific to the AM335x MMC driver. Xenomai seems otherwise to
run perfectly well on this hardware.
Note that think I have been seeing this issue occasionally for as long
as I have been testing Xenomain on the BeagleBone. I mostly chalked it
up to a bad SD card I had, but it seems like there's a deeper issue
involved. What changed is I just recently got a report with the very
simple way to trigger the bug using the recursive grep.
--
Charles Steinkuehler
charles@steinkuehler.net
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 261 bytes
Desc: OpenPGP digital signature
URL: <http://www.xenomai.org/pipermail/xenomai/attachments/20131016/6546b01f/attachment.pgp>
next prev parent reply other threads:[~2013-10-16 11:10 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-10-15 22:23 [Xenomai] Hung task on Xenomai patched ARM 3.8.13 BeagleBone Kernel Charles Steinkuehler
2013-10-16 10:44 ` Gilles Chanteperdrix
2013-10-16 11:10 ` Charles Steinkuehler [this message]
2013-10-16 11:13 ` Gilles Chanteperdrix
2013-10-16 11:09 ` Gilles Chanteperdrix
2013-10-16 11:38 ` Charles Steinkuehler
2013-10-16 11:50 ` Gilles Chanteperdrix
2013-10-16 12:08 ` Charles Steinkuehler
2013-10-16 12:28 ` Charles Steinkuehler
2013-10-16 12:44 ` Gilles Chanteperdrix
2013-10-16 12:52 ` Charles Steinkuehler
2013-10-16 15:25 ` Charles Steinkuehler
2013-10-16 15:58 ` Gilles Chanteperdrix
2013-10-16 16:12 ` Charles Steinkuehler
2013-10-16 19:41 ` Charles Steinkuehler
2013-10-17 0:58 ` Charles Steinkuehler
2013-10-17 2:21 ` Charles Steinkuehler
2013-10-17 9:29 ` Gilles Chanteperdrix
2013-10-17 10:24 ` Ralf Roesch
2013-10-17 12:39 ` Charles Steinkuehler
2013-10-17 15:12 ` Ralf Roesch
2013-10-17 18:43 ` Gilles Chanteperdrix
2013-11-04 12:38 ` Charles Steinkuehler
2013-10-16 18:12 ` Ralf Roesch
2013-10-16 12:51 ` Ralf Roesch
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=525E7433.6070807@steinkuehler.net \
--to=charles@steinkuehler.net \
--cc=gilles.chanteperdrix@xenomai.org \
--cc=xenomai@xenomai.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.