UPDATE dmasound to 2.4.3p3

UPDATE dmasound to 2.4.3p3

[Iain Sandoe]

hi all,

I've updated the dmasound (_all) patch to 2.4.3p3.

The only real change is in the way the Makefile is done (this patch should
also apply cleanly to Paulus rsync tree - but I didn't test that yet ;-)

I have another 4 changes in the pipeline - but this will keep anyone who's
using them from having to mess with merging.

against 2.4.3.p3 BK as of 16:00 GMT/UTC today.

ftp://ftp.penguinppc.org/users/iain/kernel/2_4_x/2.4.3p3-dmasound/dmasnd-2.4
.3p3_all.diff.bz2

ciao,
Iain.

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: UPDATE dmasound to 2.4.3p3

[Bill Fink]

Iain,

How much of an effort would it be to add support for the
SNDCTL_DSP_GETIPTR/SNDCTL_DSP_GETOPTR ioctls to the dmasound
driver?  I am working on porting xine to PPC and it requires
the GETOPTR ioctl to provide timing information to synchronize
the audio and video.  As a temporary kludge, I managed to get
the 0.3.7 version of xine to work by using the real time clock
to provide pseudo timing information just based on the 44.1 KHz
audio sampling rate, but this is obviously sub-optimal at best,
and may be more problematic with the newer 0.4.01 version of
xine that just added true audio/video synchronization support.

					-Thanks in advance

					-Bill


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: UPDATE dmasound to 2.4.3p3

[Iain Sandoe]

Hi Bill,

> How much of an effort would it be to add support for the
> SNDCTL_DSP_GETIPTR/SNDCTL_DSP_GETOPTR ioctls to the dmasound
> driver?  I am working on porting xine to PPC and it requires
> the GETOPTR ioctl to provide timing information to synchronize
> the audio and video.  As a temporary kludge, I managed to get
> the 0.3.7 version of xine to work by using the real time clock
> to provide pseudo timing information just based on the 44.1 KHz
> audio sampling rate, but this is obviously sub-optimal at best,
> and may be more problematic with the newer 0.4.01 version of
> xine that just added true audio/video synchronization support.

I discussed the xine issues briefly with Henry Worth and...

(1) I think I can fix the bug that makes the dmasound driver claim to be
able to do 48K (when it can't) quickly (as in prob. today).

(2) I'll look at the GETXPTR stuff.... later on.

(3) Implementing mmio doesn't appeal to me until we can lose the translation
code completely.  I might do a "CONFIG_LEAN" version soon-ish.

so - let's see what happens later today ;-)

ciao,
Iain.

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/