At Mon, 16 Feb 2009 18:44:21 +0100,
Krzysztof Helt wrote:
>
> On Mon, 16 Feb 2009 10:46:30 +0100
> Takashi Iwai <tiwai@xxxxxxx> wrote:
>
> > At Mon, 16 Feb 2009 07:43:51 +0100,
> > Krzysztof Helt wrote:
> > >
> > > From: Krzysztof Helt <krzysztof.h1@xxxxx>
> > >
> > > Also, the patch adds recognition if the chip is cs4236b+
> > > or earlier part. This unifies drivers for both cs4232
> > > and cs4236+ chips. It allows to use the PnP BIOS
> > > detection for the cs4236+ chips. Previously, only
> > > the snd-cs4232 could be detected by the PnP BIOS.
> > >
> > > The cs4232+ cards reports two separate PnP BIOS ids.
> > >
> > > The patch adds search for the second id to find out
> > > resources assigned to a control port.
> > >
> > > Signed-off-by: Krzysztof Helt <krzysztof.h1@xxxxx>
> >
> > Thanks for the patch.
> > Unfortunately the patch can't be applied to the latest sound git tree.
> > Could you rebase?
> >
>
> I based it at on the latest linux-next tree. The cs4236.c file is the
> same as the latest cs4236.c on the web interface to alsa-kernel tree.
>
> Could you report me error messages about applying the patch?
My bad, it can be applied now. Maybe I did something wrong.
> > > # Toplevel Module Dependency
> > > obj-$(CONFIG_SND_CS4231) += snd-cs4231.o
> > > -obj-$(CONFIG_SND_CS4232) += snd-cs4232.o
> > > +obj-$(CONFIG_SND_CS4232) += snd-cs4232.o snd-cs4236-lib.o
> > > obj-$(CONFIG_SND_CS4236) += snd-cs4236.o snd-cs4236-lib.o
> >
> > Let's remove cs4232 completely, then. Otherwise this will give the
> > duplicated symbols errors. It's fine to rename to cs423x, for
> > example, if you'd like. (And don't forget to change Kconfig entries.)
> >
>
> I would like to leave the snd-cs4236 name as it at the moment. If you
> want to rename it please do it with the separate patch so git system
> can detect renaming (only file name change, no internal changes).
I don't mind to keep the name. I just thought you wanted to rename.
> > In anyway, we can provide module aliases for old module names.
> >
>
> I am adding alias for the snd_cs4232.
> > > + if (chip->hardware & WSS_HW_CS4236B_MASK) {
> > > + snd_wss_free(chip);
> >
> > It's not so nice, but acceptable as a first step for integration.
> >
>
> I could not find anything better that works.
Well, making wss-lib itself a bit more extensible is another option.
But, hey, if it works, I'm not much against it :)
Takashi
_______________________________________________
Alsa-devel mailing list
Alsa-devel@xxxxxxxxxxxxxxxx
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
