Re: ALSA: oxygen stable commit difference

[Roman Volkov <rvolkov@v1ros.org>]

В Fri, 15 May 2015 22:40:54 +0200
Clemens Ladisch <clemens@ladisch.de> пишет:

> Jiri Slaby wrote:
> > This:
> > commit 1f91ecc14deea9461aca93273d78871ec4d98fcd
> > Author: Roman Volkov <v1ron@mail.ru>
> > Date:   Fri Jan 24 16:18:11 2014 +0400
> >
> >     ALSA: oxygen: modify adjust_dg_dac_routing function
> >
> > was merged in 3.14 and was also put to several stable trees back in
> > early 2014. But the commit in stable trees is totally different to
> > what is in upstream.
> >
> > 1) why is it a totally different diff?
> 
> The upstream commit was part of a larger reorganization and depended
> on lots of earlier changes.  The OXYGEN_PLAY_MUTE* changes are the
> correct backport for the old driver; the OXYGEN_PLAY_DACx_SOURCE*
> changes would have been meaningless in the old driver.
> 
> It was an error to not mention this in the stable commit message.
> 
> > 2) do we need the one below, which is marked as "Fixes" the above?
> >
> > commit 305564413c99514a14a49215790d96b8d71a6295
> > Author: Clemens Ladisch <clemens@ladisch.de>
> > Date:   Tue Mar 18 09:31:18 2014 +0100
> >
> >     ALSA: oxygen: Xonar DG(X): fix Stereo Upmixing regression
> 
> This reverts exactly those parts that were not backported to -stable,
> so it must not (and cannot) be applied to -stable.
> 
> 
> Regards,
> Clemens

Agree with Clemens, the patches are not totally different. The key
change is the OXYGEN_PLAY_ROUTING register modification. The function
adjust_dg_dac_routing() in the upstream driver is called from
output_select_put() which is the same as output_switch_put() from the
-stable driver.

Regards,
Roman