[pxc] Re: PXC strangeness on Mandriva 2006.0

[M Bianchi]

On Mon, Jul 10, 2006 at 02:14:37PM -0400, M Bianchi wrote:
> On Sun, Jul 09, 2006 at 05:32:19PM -0400, M Bianchi wrote:
> > I've been having trouble bringing up 2  PXC200AL  framegrabbers up on an
> >  Intel D949GNT  mother board with Celeron 2.53GHz processor under
> >  Mandriva 2006.0 (kernel 2.6.12-12mdk)  and  pxc-snapshot-051124 .
> > 	:
> I still have the problem that the pxc_live and pxc_show applications do not
> show any image when started after a reboot, BUT do work after
> 		modprobe bttv
> 		modprobe -r bttv tvaudio
> 	:

I have some more insight to this issue.

First, I said the  modprobe  command lines incorrectly.  They should be
 		modprobe bt878
 		modprobe -r bt878 tvaudio

I notices something very strange.  When the machine first boots and the  bt878
related modudules are prevented from being loaded,  lspci(8)  reports the
framegrabbers:
	PCI slot 06:00.0	IRQ 10
	PCI slot 06:01.0	IRQ  9

and  cat /proc/interrupts  does not list them.  IRQ 9 is associated with the
 apci  module; IRQ 10 is not listed.

After  modprobe bt878  the IRQs become(!):
	PCI slot 06:00.0	IRQ 21
	PCI slot 06:01.0	IRQ 22	(yes, the IRQ order changes)

and  cat /proc/interrupts  lists them and there associated modules.
I've never seen the IRQs change like that!
Note that  /proc/interrupts  reports the IRQs even though no application is
running.

After
	modprobe -r bt878 tvaudio
	pxc_load buffers=3

lspci(8) lists the IRQs as 22 and 21, lsmod(8) lists the pxc module as loaded,
but  /proc/interrupts  does not list the IRQs.

Run  pxc_live  and then  /proc/interrupts  lists the IRQ for the running
framegrabber;  stop  pxc_live  and  /proc/interrupts  does not.


There is a  setpci(8)  but I have not been able to convince myself that
changing the  INTERRUPT_LINE  field is changing the IRQ.  Changing
 INTERRUPT_LINE  causes byte  3c  (reported by  lspci -x ) to change BUT does
not change the IRQ reported by  lspci -v .

Clearly something in the bt878/bttv code is changing the IRQ, but i have not
yet tried to find it.  Also, without having another means to assign the IRQ
myself, I am not convinced that the IRQ is what is preventing  pxc.ko  from
returning data to the applications like  pxc_live .

Any suggestions?


Since there is now a workaround to get my applications to work, this is less
urgent than before, but I would like to understand what is preventing  pxc.ko
from working without the  bt878  module hack.

-- 
 Mike Bianchi
 Foveal Systems

 973 822-2085	call to arrange Fax

 MBianchi@Foveal.com
 http://www.AutoAuditorium.com
 http://www.FovealMounts.com