On Thu, Feb 25, 2010 at 08:18:15AM +0100, Oliver Hartkopp wrote:
> Kurt Van Dijck wrote:
> > On Wed, Feb 24, 2010 at 04:50:15PM +0100, Oliver Hartkopp wrote:
> >> Kurt Van Dijck wrote:
> >>
> >>> This is my first iteration to make softing_cs driver compile under
> >>> linux-2.6.33 kernel.
> >>> Please not it just compiles. I had no way to verify actual operation
> >>> yet.
> >>>
> >>> The pcmcia_request_window() is postponed a little, as some flags must be
> >>> set before this call (I believe). It is a small change, but with some
> >>> noise in the patch.
> >>>
> >>> Oliver (or anyone else with a softing pcmcia card),
> >>> are you able to verify this with a net-next-2.6 kernel?
> >>> Otherwise, I'll have to see if I get it working.
> >> I'll check it tomorrow morning as i don't have a softing card at home.
> > In the meanwhile, I managed to pull net-next-2.6 (huge download).
> > I'm compiling too, but no guarantee it will run on my PC yet.
> >> Especially i'm interested if i do get the same lockdep warning at card
> >> insertion i've currently seen on the latest PEAK SVN snapshot and the
> >> ems_pcmcia driver (see below).
> >>
> >> If so, there is a general problem that emerged in 2.6.32 ...
> > interesting ...
>
> Hi Kurt,
>
> looks good and works so far with receiving on both channels. I currently
I got my net-next-2.6 running, but no luck for softing driver. I suspect
my interrupt chips are different.
> have some problems with sending on the CAN with the softing driver ...
> Can you check that on your hardware?
>
> Btw. i get the same lockdep kernel warning with the Softing driver.
> So it looks like a general problem with PCMCIA & netdevices.
I get no lockdep warning (I get some output on cpu-freq stuff, so it's
in there), strange.
looking at your trace, I do not think it's in generic netdev stuff. It
during firmware load?
>
> The other PCMCIA hardware it could get here were only PCMCIA ethernet
> cards that siad they are PCI cards.
>
Kurt
_______________________________________________
Socketcan-core mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/socketcan-core
