[pvrusb2] read attempt on video device leads to error messages

[Mike Isely]

On Fri, 14 Oct 2005, Lars Schillingmann wrote:

> Am Freitag, 14. Oktober 2005 04:15 schrieb Mike Isely:
> > OK, I looked more closely at the log data you sent.  The first real
> > problem starts with an attempt to interact with the saa7115 module (the
> > ioctl() errors before that point aren't really relevant).  The error code
> > coming back is EINVAL, which I believe should never normally happen.  I
> > wonder if there were some incompatible change there somehow.  We need to
> > understand that.  So...
> >
> > 1. Where did the saa7115.ko module you are using come from?  The driver
> > snapshot, ivtv or somewhere else?
>
> The module is in /lib/modules/2.6.13-gentoo-r4/pvrusb2, so i am sure
> it comes with the driver snapshot. I removed it and reinstalled the driver
> just to be more sure.

OK, I just wanted to be sure that was the case.


>
> > 2. What kernel version is this that you are running under?
>
> 2.6.13-gentoo-r4
> i686 Intel(R) Pentium(R) M processor 1700MHz

This is good.  I know there are problems with 2.6.14 and some people have
tried much older kernels (e.g. 2.6.9) with lousy results.  However 2.6.13
should be OK.  I am unfamiliar with any patches that the Gentoo variant
might have.  Anyone else here using Gentoo?  I always work with vanilla
kernels from kernel.org.


> I use the firmware extracted with fwfind.sh.

Based on the log data you sent, the firmware looks good.  You wouldn't
have been able to get very far at all if the 8051 firmware was bad, and
the state of health for the Conexant firmware won't affect the saa7115.


> My gcc Version is: gcc (GCC) 3.4.4 (Gentoo 3.4.4-r1, ssp-3.4.4-1.0, pie-8.7.8)

Should be fine.

I noticed from the log data that your device is multi-standard.  It's
selecting SECAM as a starting point, but that might not be right for your
area.  Use sysfs to fix this if you need to.  I don't *think* that this is
the cause of the trouble because I'd expect different symptoms if it were.
But you may wish to check.

I have to admit I'm pretty stumped by this issue.  The only thing left
that I can think of is that perhaps Hauppauge swapped out this part for
something else?  We can prove / disprove that theory if you take off the
cover and send me a digital picture of the circuit board.  Can you do
that?

I can also look through the code and see where saa7115.ko might be
generating EINVAL errors.  But that may have to wait a day.  I'm
recovering from surgery here and my head is still kind of loopy from the
pain meds.  I think I'm lucky to be lucid enough to reply to this message
:-)

>
> Sorry for the double post before. One smtp-relay i used had a huge delay.

No problem.

  -Mike


-- 
                        |         Mike Isely          |     PGP fingerprint
     Spammers Die!!     |                             | 03 54 43 4D 75 E5 CC 92
                        |   isely @ pobox (dot) com   | 71 16 01 E2 B5 F5 C1 E8
                        |                             |