Re: [Xenomai-help] xenomai exceptions tracing

Re: [Xenomai-help] xenomai exceptions tracing

[Yigal Goldberger]

Hi Stephane,
Thanks for your help, I used this tool (which is something I've been needing for a long time) and I got the function name .
My architecture is powerpc (freescale 8272)
Does anyine know what this exception means on this platform ?
Best Regards,
Yigal.

--- On Wed, 7/22/09, Fillod Stephane <stephane.fillod@domain.hid> wrote:

> From: Fillod Stephane <stephane.fillod@domain.hid>
> Subject: RE: [Xenomai-help] xenomai exceptions tracing
> To: "Yigal Goldberger" <yigal_gold@domain.hid>, xenomai@xenomai.org
> Date: Wednesday, July 22, 2009, 11:41 AM
> Yigal Goldberger wrote:
> [...]
> > <6>Xenomai: Switching 802Rdr to secondary mode
> after exception #769
> from user-space at 0x10017cec (pid 974)
> 
> I don't know what exception #769 stands for on your arch,
> but have you 
> looked already what code in your program triggers it? This
> command
> may help you, esp. when the program is compiled with debug
> info.
> 
>   ${CROSS_COMPILE}addr2line -f -e 802Rdr 0x10017cec
> 
> HTH
> -- 
> Stephane
> 

[Xenomai-help] xenomai exceptions tracing

[Yigal Goldberger]

Hi all,
I'm using kernel 2.6.29.4 with xenomai .
I'm getting an exception 769 . I'd like to know what this exception means (or hints where to look for it ).

6>MTD_open
<6>MTD_ioctl
<6>MTD_open
<6>MTD_ioctl
<6>Xenomai: Switching 802Rdr to secondary mode after exception #769 from user-sp
ace at 0x10017cec (pid 974)
<6>Xenomai: Posix: closing message queue descriptor 3.
<6>Xenomai: Posix: closing message queue descriptor 4.
<6>Xenomai: Posix: closing message queue descriptor 5.
<6>Xenomai: Posix: closing message queue descriptor 7.
<6>Xenomai: Posix: closing message queue descriptor 8.
<6>Xenomai: Posix: closing message queue descriptor 9.
<6>Xenomai: Posix: closing message queue descriptor 10.
<6>Xenomai: Posix: closing message queue descriptor 11.
<6>Xenomai: Posix: closing message queue descriptor 12.
<6>Xenomai: Posix: closing message queue descriptor 14.
<6>Xenomai: Posix: closing message queue descriptor 15.

Thanks,
Yigal Goldberger.


      

Re: [Xenomai-help] xenomai exceptions tracing

[Gilles Chanteperdrix]

Yigal Goldberger wrote:
> Hi all, I'm using kernel 2.6.29.4 with xenomai . I'm getting an
> exception 769 . I'd like to know what this exception means (or hints
> where to look for it ).

Well the meaning of exceptions is dependent on the platform you are 
using, so we can not answer without more information.

> 
> (...)<6>Xenomai:
> Switching 802Rdr to secondary mode after exception #769 from user-sp 
> ace at 0x10017cec (pid 974) 

However, you can ask to receive a signal at the point where the 
exception happens, doing as in ksrc/skins/native/snippets/sigxcpu.c, 
only using pthread_set_mode_np instead of rt_task_set_mode, if you use 
the posix skin.

Alternatively, you have a way of knowing directly where the exception 
happens by disassembling your example program and look where 0x10017cec is.

P.S: if you do not like the way your message looks when I rewrap it, 
send it correctly wrapped.

-- 
                                           Gilles

Re: [Xenomai-help] xenomai exceptions tracing

[Philippe Gerum]

On Wed, 2009-07-22 at 01:28 -0700, Yigal Goldberger wrote:
> Hi all,
> I'm using kernel 2.6.29.4 with xenomai .
> I'm getting an exception 769 . I'd like to know what this exception means (or hints where to look for it ).
> 

It looks like you are running a ppc system. Those are minor MMU faults,
due to PTE misses that happen on some platforms (83xx and 85xx come to
mind). Those get reported because you turned the nucleus debug switch
on. Nothing bad in your case.

> 6>MTD_open
> <6>MTD_ioctl
> <6>MTD_open
> <6>MTD_ioctl
> <6>Xenomai: Switching 802Rdr to secondary mode after exception #769 from user-sp
> ace at 0x10017cec (pid 974)
> <6>Xenomai: Posix: closing message queue descriptor 3.
> <6>Xenomai: Posix: closing message queue descriptor 4.
> <6>Xenomai: Posix: closing message queue descriptor 5.
> <6>Xenomai: Posix: closing message queue descriptor 7.
> <6>Xenomai: Posix: closing message queue descriptor 8.
> <6>Xenomai: Posix: closing message queue descriptor 9.
> <6>Xenomai: Posix: closing message queue descriptor 10.
> <6>Xenomai: Posix: closing message queue descriptor 11.
> <6>Xenomai: Posix: closing message queue descriptor 12.
> <6>Xenomai: Posix: closing message queue descriptor 14.
> <6>Xenomai: Posix: closing message queue descriptor 15.
> 
> Thanks,
> Yigal Goldberger.
> 
> 
>       
> 
> _______________________________________________
> Xenomai-help mailing list
> Xenomai-help@domain.hid
> https://mail.gna.org/listinfo/xenomai-help
-- 
Philippe.

Re: [Xenomai-help] xenomai exceptions tracing

[Yigal Goldberger]

Hi Philippe,
Thanks for replying .
Indeed it's a powerpc . it is a mpc8272 freescale powerpc family cpu.
I printed the kmsg log and that's what I printed in this email below ,
from the kmsg log I see that the next message after this exception is
from where the application is being shut down .
Could this be a response to this exception , or is it just by chance printed straight after ?
Thanks,
Yigal. 

--- On Wed, 7/22/09, Philippe Gerum <rpm@xenomai.org> wrote:

> From: Philippe Gerum <rpm@xenomai.org>
> Subject: Re: [Xenomai-help] xenomai exceptions tracing
> To: "Yigal Goldberger" <yigal_gold@domain.hid>
> Cc: xenomai@xenomai.org
> Date: Wednesday, July 22, 2009, 12:07 PM
> On Wed, 2009-07-22 at 01:28 -0700,
> Yigal Goldberger wrote:
> > Hi all,
> > I'm using kernel 2.6.29.4 with xenomai .
> > I'm getting an exception 769 . I'd like to know what
> this exception means (or hints where to look for it ).
> > 
> 
> It looks like you are running a ppc system. Those are minor
> MMU faults,
> due to PTE misses that happen on some platforms (83xx and
> 85xx come to
> mind). Those get reported because you turned the nucleus
> debug switch
> on. Nothing bad in your case.
> 
> > 6>MTD_open
> > <6>MTD_ioctl
> > <6>MTD_open
> > <6>MTD_ioctl
> > <6>Xenomai: Switching 802Rdr to secondary mode
> after exception #769 from user-sp
> > ace at 0x10017cec (pid 974)
> > <6>Xenomai: Posix: closing message queue
> descriptor 3.
> > <6>Xenomai: Posix: closing message queue
> descriptor 4.
> > <6>Xenomai: Posix: closing message queue
> descriptor 5.
> > <6>Xenomai: Posix: closing message queue
> descriptor 7.
> > <6>Xenomai: Posix: closing message queue
> descriptor 8.
> > <6>Xenomai: Posix: closing message queue
> descriptor 9.
> > <6>Xenomai: Posix: closing message queue
> descriptor 10.
> > <6>Xenomai: Posix: closing message queue
> descriptor 11.
> > <6>Xenomai: Posix: closing message queue
> descriptor 12.
> > <6>Xenomai: Posix: closing message queue
> descriptor 14.
> > <6>Xenomai: Posix: closing message queue
> descriptor 15.
> > 
> > Thanks,
> > Yigal Goldberger.
> > 
> > 
> >       
> > 
> > _______________________________________________
> > Xenomai-help mailing list
> > Xenomai-help@domain.hid
> > https://mail.gna.org/listinfo/xenomai-help
> -- 
> Philippe.
> 
> 
> 

Re: [Xenomai-help] xenomai exceptions tracing

[Philippe Gerum]

On Wed, 2009-07-22 at 02:28 -0700, Yigal Goldberger wrote:
> Hi Philippe,
> Thanks for replying .
> Indeed it's a powerpc . it is a mpc8272 freescale powerpc family cpu.
> I printed the kmsg log and that's what I printed in this email below ,
> from the kmsg log I see that the next message after this exception is
> from where the application is being shut down .
> Could this be a response to this exception , or is it just by chance printed straight after ?

This kind of exception is handled internally between the linux kernel
and the Xenomai core, it would not stop the application. Xenomai would
only kill a kernel-based RT thread dereferencing a bad memory area (with
a non-ambiguous message in the klog), which is certainly not the case
here. Linux would kill a userland thread doing the same (RT / non-RT,
regardless) with a signal.

This is not what happens here. You may want to enable CONFIG_PRINTK_TIME
to have timestamps on your kernel messages, in order to confirm this.

> Thanks,
> Yigal. 
> 
> --- On Wed, 7/22/09, Philippe Gerum <rpm@xenomai.org> wrote:
> 
> > From: Philippe Gerum <rpm@xenomai.org>
> > Subject: Re: [Xenomai-help] xenomai exceptions tracing
> > To: "Yigal Goldberger" <yigal_gold@domain.hid>
> > Cc: xenomai@xenomai.org
> > Date: Wednesday, July 22, 2009, 12:07 PM
> > On Wed, 2009-07-22 at 01:28 -0700,
> > Yigal Goldberger wrote:
> > > Hi all,
> > > I'm using kernel 2.6.29.4 with xenomai .
> > > I'm getting an exception 769 . I'd like to know what
> > this exception means (or hints where to look for it ).
> > > 
> > 
> > It looks like you are running a ppc system. Those are minor
> > MMU faults,
> > due to PTE misses that happen on some platforms (83xx and
> > 85xx come to
> > mind). Those get reported because you turned the nucleus
> > debug switch
> > on. Nothing bad in your case.
> > 
> > > 6>MTD_open
> > > <6>MTD_ioctl
> > > <6>MTD_open
> > > <6>MTD_ioctl
> > > <6>Xenomai: Switching 802Rdr to secondary mode
> > after exception #769 from user-sp
> > > ace at 0x10017cec (pid 974)
> > > <6>Xenomai: Posix: closing message queue
> > descriptor 3.
> > > <6>Xenomai: Posix: closing message queue
> > descriptor 4.
> > > <6>Xenomai: Posix: closing message queue
> > descriptor 5.
> > > <6>Xenomai: Posix: closing message queue
> > descriptor 7.
> > > <6>Xenomai: Posix: closing message queue
> > descriptor 8.
> > > <6>Xenomai: Posix: closing message queue
> > descriptor 9.
> > > <6>Xenomai: Posix: closing message queue
> > descriptor 10.
> > > <6>Xenomai: Posix: closing message queue
> > descriptor 11.
> > > <6>Xenomai: Posix: closing message queue
> > descriptor 12.
> > > <6>Xenomai: Posix: closing message queue
> > descriptor 14.
> > > <6>Xenomai: Posix: closing message queue
> > descriptor 15.
> > > 
> > > Thanks,
> > > Yigal Goldberger.
> > > 
> > > 
> > >       
> > > 
> > > _______________________________________________
> > > Xenomai-help mailing list
> > > Xenomai-help@domain.hid
> > > https://mail.gna.org/listinfo/xenomai-help
> > -- 
> > Philippe.
> > 
> > 
> > 
> 
> 
>       
-- 
Philippe.