On Mon, 2011-01-17 at 16:30 +0100, Wong Sheng Chao wrote:
> You can try booting in text mode, and set the vga mode to have a lot of
> lines, then run the test in text mode, to at least see if the reason of
> the freeze you observe is a kernel oops.
> 
> I tried booting it in text mode and the program still freezes after i 
> execute the user-space program. There were no error messages.
> 
> If you do not see anything, then enable all the I-pipe and Xenomai
> debugs, and try again.
> 
> How do I enable I-pipe and Xenomai debugs? Do I need to set these 
> options during kernel configuration time? Will enabling I-pipe let me 
> know if it's a kernel-oops?

Before anything else, you should really start testing the return code of
your syscalls, specifically for rt_task_wait_period(), and break from
your while() loop if something went wrong.

> 
> Thanks again for your help!
> 
> Regards
> Wong
> 
> 
> 
> On 01/17/2011 02:34 PM, Gilles Chanteperdrix wrote:
> > Wong Sheng Chao wrote:
> >> On 01/16/2011 12:03 AM, Gilles Chanteperdrix wrote:
> >>> Wong Sheng Chao wrote:
> >>>> Hi
> >>>>
> >>>> I'm a newbie in real time programming and I recently came to know of
> >>>> Xenomai. After patching the Xenomai to the linux kernel, I ran some
> >>>> examples that I found from the Internet to get a better understanding of
> >>>> the periodic task. Everything seems to run fine when I start two tasks
> >>>> with period of 1s and 2s, but the system freezes when I added more tasks
> >>>> ( more than 2 tasks). The period is at 1s, 2s and 3s so i think the
> >>>> processor has more than enough time to process the tasks.
> >>>>
> >>>> I also read that the periodic mode is emulated by a software driver
> >>>> which uses one-shot mode programming. Thus i did not enable periodic
> >>>> timing when I compile the Linux kernel, is this the cause of the problem?
> >>>>
> >>>> I'm using a system with Core i7, ubuntu 10.04, linux kernel 2.6.32.15,
> >>>> xenomai 2.5.4
> >>>>
> >>>> Do let me know what I did wrong in the code, thanks in advance!!
> >>> You almost send us all the information we ask on this page:
> >>> http://www.xenomai.org/index.php/Request_for_information
> >>>
> >>> We lack:
> >>> - the version of the Adeos patch you use
> >>> - the kernel logs from the boot up to the bug itself.
> >>>
> >> - the version of the Adeos patch you use
> >>
> >> I'm using the Adeos patch 2.6.32.15-x86-2.7-01.patch
> >>
> >> - the kernel logs from the boot up to the bug itself.
> >>
> >> I was not sure how to do this, so I google online and it seems that
> >> there are two ways of doing this, through serial port or ethernet.
> >> However I have only one linux system setup. Any suggestion on how I can
> >> provide the kernel logs?
> >>
> >> After the kernel freezes, I reboot the system and went to /var/log to
> >> check the kern.log file at the time of the bug, but no logs were made on
> >> that particular time, is netconsole or serial port the only method of
> >> troubleshooting?
> > You can try booting in text mode, and set the vga mode to have a lot of
> > lines, then run the test in text mode, to at least see if the reason of
> > the freeze you observe is a kernel oops.
> >
> > If you do not see anything, then enable all the I-pipe and Xenomai
> > debugs, and try again.
> >
> 
> 
> _______________________________________________
> Xenomai-help mailing list
> [email protected]
> https://mail.gna.org/listinfo/xenomai-help

-- 
Philippe.



_______________________________________________
Xenomai-help mailing list
[email protected]
https://mail.gna.org/listinfo/xenomai-help

Reply via email to