[linux-dvb] About fixes into Mantis branch. Was: Asus Tiger can't find any channels

Marko Ristola marko.ristola at kolumbus.fi
Wed Oct 10 13:17:59 CEST 2007

Hi Manu,

I have had an own Mantis branch version that is based on v4l-dvb
and not based on jusst.de. My card has worked there. I have also
done Suspend/Resume development on that branch as you know.

So I'd like to concentrate on developing and maintaining a version
based on v4l-dvb. That way the change set against
the main HG branch is minimal. Also I get all compile
problem solves and other fixes from there.
Also it should be easier to create a patch based on
main v4l-dvb so that all related changes could be merged there,
or fixed/improved and merged.

If the current Mantis branch changes would be merged into v4l-dvb,
you could concentrate more on other maybe more
important projects. You would just sometimes pull
v4l-dvb changes into your Mantis branch and if some bug is found in v4l-dvb,
you could fix it first into Mantis branch, or then you could send
a fix into v4l-dvb directly and pull the changes back.
That way if some patch would break a Mantis driver on v4l-dvb, that patch
would not be accepted into v4l-dvb. Instead the patch would be corrected 
(maybe a Mantis driver changes in the process) and the Mantis driver 
wouldn't break, although it would need testing to be sure. Thus there 
would be
less maintenance burden for you.

The merge from Mantis branch into v4l-dvb wouldn't have to be done
by you alone, you know. All merge requirements should be met though.
(good enough coding style, working features, copyrights, robustness,
shouldn't break the work of others, what else?). If I would do part of
the work, I would learn a lot from you and others during the process.

What do you think?

Best Regards,
Marko Ristola

Marko Ristola wrote:
> Hi Manu,
> I might be able to make a patch this week or next week:
> I'm on a holiday now.
> Anssi got his card working now. He set Kaffeine to use INVERSION=0.
> At least scanning started to work.
> The inversion swap heuristics in tda1004x.c obfuscates though
> the actual hardware inversion setting that he used on success.
> He did also some modprobe changes. Thus I'm can't be
> absolutely sure what actually helped.
> Regards,
> Marko
> Manu Abraham wrote:
>> Marko Ristola wrote:
>>> My DVB-C PCI card uses cu1216.c (Manu's branch).
>>> That file has it's own inversion heuristics currently,
>>> but unfortunately the heuristics tries repeatedly INVERSION=0 and never
>>> inversion=1 with INVERSION_AUTO. I'd like that you Manu would accept a
>>> fix for that
>>> some day.
>> Sure. Just send a patch that can be applied against http://jusst.de/hg/v4l-dvb/
>> Manu
> _______________________________________________
> linux-dvb mailing list
> linux-dvb at linuxtv.org
> http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb

More information about the linux-dvb mailing list