[pvrusb2] s5h1411 errors

[Michael Krufky]

On Thu, Aug 27, 2009 at 4:26 PM, Roger<rogerx at sdf.lonestar.org> wrote:
> On Thu, 2009-08-27 at 09:29 -0400, Michael Krufky wrote:
>> On Sun, Aug 9, 2009 at 6:43 PM, Roger<rogerx at sdf.lonestar.org> wrote:
>> > On Sun, 2009-08-09 at 15:14 -0500, Mike Isely wrote:
>> >> About a month ago, I sent out the message below.  Best I can tell is
>> >> that I got a reply from only ONE person.  Is nobody else seeing this?
>> >>
>> >>   -Mike
>> >
>> > Might want to rephrase to, "Anybody have the hvr-1950 using the s5h1411
>> > kernel module and not seeing these errors?"
>> >
>> >
>> > Just for kicks, you want me to send you my firmwares, or send me yours
>> > to verify it's not bad firmwares?  I doubt it though as I believe i used
>> > the perl script from the pvrusb2 tarball, and they do load without
>> > problems on cold boot & hvr-1950 cold boot.
>>
>> This is probably not a real issue.  If you're still seeing this
>> problem, can you post your full dmesg, un edited, containing device
>> usb insertion, repro of the problem, analog streaming *and* digital
>> streaming?
>
> ...
>
> I'm stilling seeing this with git 2.6.31 sources.  I will integrate
> CONFIG_USB_DEBUG & post my results when I get a chance this week (for
> device usb insertion?).  I had both i2c debug & a wealth of kernel
> hacking debug turned on in >2.6.30 kernel hacking lately and didn't get
> any more debug output then a possible circular locking dependency that I
> recently posted.

Actually no, please dont -- I just want to see NORMAL dmesg output
WITHOUT ANY DEBUG enabled.

Also, I am entirely *not* interested in seeing output from your 2.6.31
pre-release kernel.  If you can't reproduce this on a stable kernel,
please say so.  If you're only seeing this on pre-release kernels,
then I can debug this myself.


> Also, digital & analog streaming of the device are completely flaky when
> the computer is rebooted with the hvr-1950 device still having preloaded
> firmware from the last boot.  Or, more precisely put, the device's
> analog & digital devices are completely inoperable, even if their device
> files are shown in /dev.
>
> A cold boot along with power cycling the HVR-1950 has usually been the
> only resolve.
>
>> Regards,
>>
>> Mike Krufky
>
> Just a quick note here, I noticed you're listed within the top five
> developers committing to the Linux Kernel.  This was published by a
> recent Linux Kernel news article detailing top submitters.  A *big*
> thank you, else I'd be stuck complaining about MS Windows a lot more
> then I have done in the past. ;-)

Oh, really??  Top five??  That's so cool!  Can you show me a link so I
can see this for myself?  :-D

I appreciate the thanks, and the recognition -- that's awesome!

Cheers,

Mike Krufky