[mythtv-users] I2C RW failure, with experimental 80e driver...
dheitmueller at kernellabs.com
Wed May 11 13:35:08 UTC 2011
On Wed, May 11, 2011 at 9:14 AM, Martin, Charles <cmart at dolby.com> wrote:
> Hi Devin -
> Thanks for the fast response. I entirely believe you on being busy and all.
> I will admit it's a case of varying mileage when I try experimental code against an older kernel, on a STB to boot!
> I will enable all the I2C options and see if that will fix it.
> I don't recall any message of any firmwares loading up with this particular device (nothing called from request_firmware() ) in the USB stack.
> I did see one quirky thing where the frontend id was assigned some garbage value. I put a memset statement in the drx39xyj_attach() routine, but
> I'm not sure if that's the correct thing to do (because the state->fe field seems to be ptr too).
> Actually, I was trying to test ClearQAM reception when I tried it, so that's good to know. However, if you need help getting this up and running,
> I could probably help, if you can give me access to chip datasheets and so forth, or give me pointers on how to proceed. I'll code and test, and you can
> tell me what to do essentially ;-)
> I can be reached at wraith_dart-at-yahoo-dot-com as well.
> Cheers, C.
You're right - I forgot that the driver in question currently still
has the firmware embedded, It will need to be moved out to a separate
binary in order to get upstream, but I didn't get that far in the
driver port. Again, my bad for just not having poked at the code in
almost a year.
If I can make a suggestion: you may find it useful to start your
process on a current Ubuntu or Fedora x86 box. That will allow you to
learn how the various applications and drivers work, after which point
you can then move to the embedded platform. It seems that you've
combined learning about how LinuxTV and MythTV work at the same time
as trying to make it work on an obscure platform, so it would be extra
difficult to understand whether your issues are with the platform
specifically or just your general lack of knowledge as to how the
devices and applications are supposed to work.
I cannot make available any datasheets as everything is under NDA.
That said though, this should be pretty straightforward to debug if
you have any kernel driver experience (the problem appears to be
related to general i2c communications).
Devin J. Heitmueller - Kernel Labs
More information about the mythtv-users