* [Xenomai] RTcfg module waits forever after "rtcfg rteth0 server" has been called
@ 2015-06-18 12:50 fre go
2015-06-18 13:39 ` Gilles Chanteperdrix
0 siblings, 1 reply; 2+ messages in thread
From: fre go @ 2015-06-18 12:50 UTC (permalink / raw)
To: xenomai
I'm currently working with Xenomai 3 (first rc4, now git head) with Kernel
3.16.7 and encountered a problem with both versions concerning rtnet: When
I try to start rtnet with "rtnet start" the call stalls and sometimes even
locks the kernel. I could trace this back to the "rtcfg rteth0 server" and
further investigation via kgdb debugging via virtual machine showed that
the call to "rtdm_event_signal" in rtnet_rtpc.c seems to be not answered.
Although the kernel should be waiting interruptible most of the time ctrl+c
does nothing. Sometimes it does what it's supposed to (interrupt) but after
another call to "rtnet start" the kernel often fails with a oops or even a
panic). Could it be the kernel which fails here and not xenomai? I'm
currently out of ideas how to tackle this problem on my own. Maybe somebody
could point me into the right direction.
My dev machine is running on KVM and it as been tested with xenomai 2.6.4
and RTnet 0.9.13 git head, both with different nics (e1000, rtl8139) so I
can rule out possible KVM or Qemu issues.
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [Xenomai] RTcfg module waits forever after "rtcfg rteth0 server" has been called
2015-06-18 12:50 [Xenomai] RTcfg module waits forever after "rtcfg rteth0 server" has been called fre go
@ 2015-06-18 13:39 ` Gilles Chanteperdrix
0 siblings, 0 replies; 2+ messages in thread
From: Gilles Chanteperdrix @ 2015-06-18 13:39 UTC (permalink / raw)
To: fre go; +Cc: xenomai
fre go wrote:
> I'm currently working with Xenomai 3 (first rc4, now git head) with Kernel
> 3.16.7 and encountered a problem with both versions concerning rtnet: When
> I try to start rtnet with "rtnet start" the call stalls and sometimes even
> locks the kernel.
Yes, it has been reported already, rtcfg is broken in rtnet over xeno 3.
> I could trace this back to the "rtcfg rteth0 server" and
> further investigation via kgdb debugging via virtual machine showed that
> the call to "rtdm_event_signal" in rtnet_rtpc.c seems to be not answered.
As is usual with debug tools what you see here is a symptom of the
problem, not the problem itself. If you want to solve this, you need to
dig deeper and understand the code. A debug tool can not do that for you.
> Although the kernel should be waiting interruptible most of the time
> ctrl+c
> does nothing. Sometimes it does what it's supposed to (interrupt) but
> after
> another call to "rtnet start" the kernel often fails with a oops or even a
> panic). Could it be the kernel which fails here and not xenomai? I'm
> currently out of ideas how to tackle this problem on my own. Maybe
> somebody
> could point me into the right direction.
It is probably not Xenomai, but rather the port of rtnet it contains.
>
> My dev machine is running on KVM and it as been tested with xenomai 2.6.4
> and RTnet 0.9.13 git head, both with different nics (e1000, rtl8139) so I
> can rule out possible KVM or Qemu issues.
I do not understand. Does it work or does it no with 2.6? I guess it does.
--
Gilles.
https://click-hack.org
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2015-06-18 13:39 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-06-18 12:50 [Xenomai] RTcfg module waits forever after "rtcfg rteth0 server" has been called fre go
2015-06-18 13:39 ` Gilles Chanteperdrix
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.