Re: [Xenomai-help] Xenomai and PEAK can bus card problem

[Gilles Chanteperdrix <gilles.chanteperdrix@xenomai.org>]

Peter Pastor wrote:
> Hey Jan, Wolfgang, Gilles,
> 
> After actually trying to reroute the interrupt by cutting and jumping
> the pins on the PCI card we figured that there was actually a 64bit PCI
> slot free. Also, according to wikipedia...  "PCI-X 1.0b and PCI-X 2.0
> are backward compatible with some PCI standards." Thus, we simply stuck
> the card in that slot and got assigned an interrupt that is not shared
> by other devices. Well, after rerouting things back to the initial
> configuration on the card, we were able to load the rtcan modules and
> communicate on the CAN bus !!
> 
> THANKS again for all the help and please let me know if you want me to
> run some tests. We are running xenomai 2.6.31 with xenomai 2.5.5.1 and
> thelinux_2.6.31-14.48.diff.gz ubuntu patch and the
> adeos-ipipe-2.6.31.8-x86-2.4-09.patch on a 64bit dell precision
> workstation T7500.
> 
> The only problem remaining is the fact that we get latency problems the
> moment we start an opengl window but as mentioned in another thread,
> there is nothing Xenomai can do about. Also, the MSI workaround is
> enabled but fails. I hope that this does not cause problems for our
> purposes of using Xenomai.

Hi Peter,

good news. Moving the card in the slot was what I was among the things I
was thinking about when I suggested you to look at the FAQ. So,
apparently, I was not precise enough. The same goes for the X issue,
when I told you to look at the TROUBLESHOOTING guide, the
TROUBLESHOOTING guide suggests to use "NoAccel" in the X server
configuration. Did you try that?

Regards.

-- 
                                                                Gilles.