Re: [pulseaudio-discuss] ask about passthrough support

2011-09-29 Thread Pierre-Louis Bossart
> I'm confused how it negotiates with underlying alsa device? Does it
> need some changes in alsa driver? E.g. I have an HDMI device, and I
> would like to pass a dts-wav file directly down to this device. Are
> there some commands like 'gst-launch filesrc location=test.wav !
> pulseaudiosink'?

Just to be clear, there's no negotiation with the alsa device for now. I've
been pushing for a kernel patch that makes the ELD information available to
user-space, and I have some code to read it in pulseaudio. We now have to
take the ELD information and set the formats automagically.
For now as Arun said you have to set the formats by hand in PulseAudio so
that they match the capabilities of your receiver.
-Pierre

___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss


Re: [pulseaudio-discuss] ask about passthrough support

2011-09-29 Thread Arun Raghavan
On Thu, 2011-09-29 at 16:46 +0800, Lu Guanqun wrote:
[...]
> Sink #0
>   State: IDLE
>   Name: alsa_output.pci-_00_1b.0.hdmi-stereo
>   Description: Internal Audio Digital Stereo (HDMI)
[...]
>   Formats:
>   pcm

Looks like you didn't set the supports formats on the receiver
(instructions are on the Passthrough page on the wiki). You can also do
a pactl set-sink-formats 0 "pcm; dts-iec61937" for now.

-- Arun

___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss


Re: [pulseaudio-discuss] ask about passthrough support

2011-09-29 Thread Lu Guanqun
On Thu, Sep 29, 2011 at 04:39:39PM +0800, Arun Raghavan wrote:
[...]
> So you're in a mode that supports digital output (S/PDIF or HDMI) and
> the appropriate formats are set on the sink? If yes, then the output of

Which format should I set on the sink to make the passthrough work?

> pactl list and gst-launch output with --gst-debug=pulse*:3 while playing
> would help.

Here's the output just now: (using -v only, I'll use debug level 3 to
give more info).

Setting pipeline to PAUSED ...
Pipeline is PREROLLING ...
/GstPipeline:pipeline0/GstDcaParse:dcaparse0.GstPad:sink: caps = audio/x-dts, 
rate=(int)44100, channels=(int)6, depth=(int)14, endianness=(int)1234, 
framed=(boolean)false
/GstPipeline:pipeline0/GstDcaParse:dcaparse0.GstPad:src: caps = audio/x-dts, 
framed=(boolean)true, rate=(int)44100, channels=(int)6, endianness=(int)1234, 
depth=(int)14, block-size=(int)512, frame-size=(int)2048
/GstPipeline:pipeline0/GstPulseAudioSink:pulseaudiosink0.GstGhostPad:sink: caps 
= audio/x-dts, framed=(boolean)true, rate=(int)44100, channels=(int)6, 
endianness=(int)1234, depth=(int)14, block-size=(int)512, frame-size=(int)2048
/GstPipeline:pipeline0/GstPulseAudioSink:pulseaudiosink0.GstGhostPad:sink.GstProxyPad:proxypad0:
 caps = audio/x-dts, framed=(boolean)true, rate=(int)44100, channels=(int)6, 
endianness=(int)1234, depth=(int)14, block-size=(int)512, frame-size=(int)2048
/GstPipeline:pipeline0/GstPulseAudioSink:pulseaudiosink0/GstDecodeBin2:pulseaudiosink-dbin2/GstTypeFindElement:typefind.GstPad:src:
 caps = audio/x-dts, framed=(boolean)true, rate=(int)44100, channels=(int)6, 
endianness=(int)1234, depth=(int)14, block-size=(int)512, frame-size=(int)2048
/GstPipeline:pipeline0/GstPulseAudioSink:pulseaudiosink0/GstDecodeBin2:pulseaudiosink-dbin2/GstTypeFindElement:typefind.GstPad:sink:
 caps = audio/x-dts, framed=(boolean)true, rate=(int)44100, channels=(int)6, 
endianness=(int)1234, depth=(int)14, block-size=(int)512, frame-size=(int)2048
/GstPipeline:pipeline0/GstPulseAudioSink:pulseaudiosink0/GstDecodeBin2:pulseaudiosink-dbin2.GstGhostPad:sink:
 caps = audio/x-dts, framed=(boolean)true, rate=(int)44100, channels=(int)6, 
endianness=(int)1234, depth=(int)14, block-size=(int)512, frame-size=(int)2048
/GstPipeline:pipeline0/GstPulseAudioSink:pulseaudiosink0/GstDecodeBin2:pulseaudiosink-dbin2.GstGhostPad:sink.GstProxyPad:proxypad1:
 caps = audio/x-dts, framed=(boolean)true, rate=(int)44100, channels=(int)6, 
endianness=(int)1234, depth=(int)14, block-size=(int)512, frame-size=(int)2048
/GstPipeline:pipeline0/GstPulseAudioSink:pulseaudiosink0/GstDecodeBin2:pulseaudiosink-dbin2/GstDcaParse:dcaparse1.GstPad:sink:
 caps = audio/x-dts, framed=(boolean)true, rate=(int)44100, channels=(int)6, 
endianness=(int)1234, depth=(int)14, block-size=(int)512, frame-size=(int)2048
/GstPipeline:pipeline0/GstPulseAudioSink:pulseaudiosink0/GstDecodeBin2:pulseaudiosink-dbin2/GstDcaParse:dcaparse1.GstPad:src:
 caps = audio/x-dts, framed=(boolean)true, rate=(int)44100, channels=(int)6, 
endianness=(int)1234, depth=(int)14, block-size=(int)512, frame-size=(int)2048
/GstPipeline:pipeline0/GstPulseAudioSink:pulseaudiosink0/GstDecodeBin2:pulseaudiosink-dbin2/GstDtsDec:dtsdec0.GstPad:sink:
 caps = audio/x-dts, framed=(boolean)true, rate=(int)44100, channels=(int)6, 
endianness=(int)1234, depth=(int)14, block-size=(int)512, frame-size=(int)2048
/GstPipeline:pipeline0/GstPulseAudioSink:pulseaudiosink0/GstDecodeBin2:pulseaudiosink-dbin2.GstDecodePad:src0:
 caps = audio/x-raw-float, endianness=(int)1234, width=(int)32, 
channels=(int)6, rate=(int)44100, channel-positions=(GstAudioChannelPosition)< 
GST_AUDIO_CHANNEL_POSITION_FRONT_CENTER, GST_AUDIO_CHANNEL_POSITION_FRONT_LEFT, 
GST_AUDIO_CHANNEL_POSITION_FRONT_RIGHT, GST_AUDIO_CHANNEL_POSITION_REAR_LEFT, 
GST_AUDIO_CHANNEL_POSITION_REAR_RIGHT, GST_AUDIO_CHANNEL_POSITION_LFE >
/GstPipeline:pipeline0/GstPulseAudioSink:pulseaudiosink0/GstDecodeBin2:pulseaudiosink-dbin2/GstDtsDec:dtsdec0.GstPad:src:
 caps = audio/x-raw-float, endianness=(int)1234, width=(int)32, 
channels=(int)6, rate=(int)44100, channel-positions=(GstAudioChannelPosition)< 
GST_AUDIO_CHANNEL_POSITION_FRONT_CENTER, GST_AUDIO_CHANNEL_POSITION_FRONT_LEFT, 
GST_AUDIO_CHANNEL_POSITION_FRONT_RIGHT, GST_AUDIO_CHANNEL_POSITION_REAR_LEFT, 
GST_AUDIO_CHANNEL_POSITION_REAR_RIGHT, GST_AUDIO_CHANNEL_POSITION_LFE >
/GstPipeline:pipeline0/GstPulseAudioSink:pulseaudiosink0/GstPulseSink:pulseaudiosink-sink.GstPad:sink:
 caps = audio/x-raw-float, endianness=(int)1234, width=(int)32, 
channels=(int)6, rate=(int)44100, channel-positions=(GstAudioChannelPosition)< 
GST_AUDIO_CHANNEL_POSITION_FRONT_CENTER, GST_AUDIO_CHANNEL_POSITION_FRONT_LEFT, 
GST_AUDIO_CHANNEL_POSITION_FRONT_RIGHT, GST_AUDIO_CHANNEL_POSITION_REAR_LEFT, 
GST_AUDIO_CHANNEL_POSITION_REAR_RIGHT, GST_AUDIO_CHANNEL_POSITION_LFE >
/GstPipeline:pipeline0/GstPulseAudioSink:pulseaudiosink0/GstDecodeBin2:pulseaudiosink-dbin2.GstDecodePad:src0.GstProxyPad:proxypad5:
 caps = audio/x-raw-flo

Re: [pulseaudio-discuss] ask about passthrough support

2011-09-29 Thread Arun Raghavan
On Thu, 2011-09-29 at 16:32 +0800, Lu Guanqun wrote:
> Hi Arun,
> 
> On Thu, Sep 29, 2011 at 04:02:48PM +0800, Arun Raghavan wrote:
> [...]
> > Take a look at http://pulseaudio.org/wiki/Passthrough -- it gives you
> > the basic steps to set things up. I'll try to write up something more
> > comprehensive soon.
> 
> That's great. FYI. I've installed the git version of gstreamer.
> 
> > 
> > The DTS-in-wav case is a bit dicey. In most cases, you should be able to
> > get the data out of wavparse, pass it to dcaparse and pass that on to
> > pulsesink or pulseaudio directly.
> 
> I'm using this command:
> gst-launch filesrc location=test.wav ! wavparse ! dcaparse !
> pulseaudiosink
> But this will utilize the dtsdec plugin, and the data is not
> passthroughed...

So you're in a mode that supports digital output (S/PDIF or HDMI) and
the appropriate formats are set on the sink? If yes, then the output of
pactl list and gst-launch output with --gst-debug=pulse*:3 while playing
would help.

-- Arun

___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss


Re: [pulseaudio-discuss] ask about passthrough support

2011-09-29 Thread Lu Guanqun
Hi Arun,

On Thu, Sep 29, 2011 at 04:02:48PM +0800, Arun Raghavan wrote:
[...]
> Take a look at http://pulseaudio.org/wiki/Passthrough -- it gives you
> the basic steps to set things up. I'll try to write up something more
> comprehensive soon.

That's great. FYI. I've installed the git version of gstreamer.

> 
> The DTS-in-wav case is a bit dicey. In most cases, you should be able to
> get the data out of wavparse, pass it to dcaparse and pass that on to
> pulsesink or pulseaudio directly.

I'm using this command:
gst-launch filesrc location=test.wav ! wavparse ! dcaparse !
pulseaudiosink
But this will utilize the dtsdec plugin, and the data is not
passthroughed...

> 
> However, it is possible that the file is made such that it is already
> padded to the correct size for IEC958 transmission, which is a case we
> do not currently handle. The reason being that pulsesink calls a gst
> utility library to perform IEC 61937 payloading, which isn't possible in
> these sort of files (no space left for the header). We should probably
> just pass this on if the frames are the right size, but at the time, I
> deferred this decision to later since I wasn't sure what to do.
> 
> Regards,
> Arun
> 

-- 
guanqun
___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss


Re: [pulseaudio-discuss] ask about passthrough support

2011-09-29 Thread Arun Raghavan
Hi Guanqun,

On Thu, 2011-09-29 at 15:52 +0800, Lu Guanqun wrote:
> On Thu, Sep 08, 2011 at 02:09:33AM +0800, Arun Raghavan wrote:
> [...]
> > > 3. To be able to use passthrough, do we still need changes on Gstreamer?
> > > Therefore people can simply use something like 'gst-launch xxx'. And
> > > how's that status?
> > 
> > The base stuff is in git master, and some of it is en route
> > (https://bugzilla.gnome.org/show_bug.cgi?id=657179).
> 
> Hi Arun,
> 
> I've seen that the corresponding patch is already committed in
> gst-plugins-good repository.
> 
> Reading the comments from the file:
> "It transparently takes care of passing compressed format as-is if the
> sink supports it, decoding if necessary, and changes to supported
> formats at runtime."
> 
> I'm confused how it negotiates with underlying alsa device? Does it
> need some changes in alsa driver? E.g. I have an HDMI device, and I
> would like to pass a dts-wav file directly down to this device. Are
> there some commands like 'gst-launch filesrc location=test.wav !
> pulseaudiosink'?

Take a look at http://pulseaudio.org/wiki/Passthrough -- it gives you
the basic steps to set things up. I'll try to write up something more
comprehensive soon.

The DTS-in-wav case is a bit dicey. In most cases, you should be able to
get the data out of wavparse, pass it to dcaparse and pass that on to
pulsesink or pulseaudio directly.

However, it is possible that the file is made such that it is already
padded to the correct size for IEC958 transmission, which is a case we
do not currently handle. The reason being that pulsesink calls a gst
utility library to perform IEC 61937 payloading, which isn't possible in
these sort of files (no space left for the header). We should probably
just pass this on if the frames are the right size, but at the time, I
deferred this decision to later since I wasn't sure what to do.

Regards,
Arun

___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss


Re: [pulseaudio-discuss] ask about passthrough support

2011-09-29 Thread Lu Guanqun
On Thu, Sep 08, 2011 at 02:09:33AM +0800, Arun Raghavan wrote:
[...]
> > 3. To be able to use passthrough, do we still need changes on Gstreamer?
> > Therefore people can simply use something like 'gst-launch xxx'. And
> > how's that status?
> 
> The base stuff is in git master, and some of it is en route
> (https://bugzilla.gnome.org/show_bug.cgi?id=657179).

Hi Arun,

I've seen that the corresponding patch is already committed in
gst-plugins-good repository.

Reading the comments from the file:
"It transparently takes care of passing compressed format as-is if the
sink supports it, decoding if necessary, and changes to supported
formats at runtime."

I'm confused how it negotiates with underlying alsa device? Does it
need some changes in alsa driver? E.g. I have an HDMI device, and I
would like to pass a dts-wav file directly down to this device. Are
there some commands like 'gst-launch filesrc location=test.wav !
pulseaudiosink'?

-- 
guanqun
___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss


Re: [pulseaudio-discuss] ask about passthrough support

2011-09-14 Thread Arun Raghavan
On Thu, 2011-09-15 at 14:03 +0800, Lu Guanqun wrote:
> On Thu, Sep 15, 2011 at 01:58:41PM +0800, Arun Raghavan wrote:
> > On Thu, 2011-09-15 at 11:22 +0800, Lu Guanqun wrote:
> > > On Thu, Sep 08, 2011 at 02:19:41AM +0800, Arun Raghavan wrote:
> > > > On Wed, 2011-09-07 at 11:09 -0700, Arun Raghavan wrote:
> > > > > On Wed, 2011-09-07 at 14:20 +0800, Lu Guanqun wrote:
> > > > > > Hi Arun,
> > > > > > 
> > > > > > I have several things unclear about passthrough support:
> > > > > 
> > > > > I'm talking about this work at Linux Plumbers' Conference on Friday. 
> > > > > If
> > > > > videos are available, I'll link to them -- hopefully that should make
> > > > > the motivation clear.
> > > > 
> > > > Oh, phew, no video -- I'm still happy to answer any questions you
> > > > have. :)
> > > 
> > > Hi Arun,
> > > 
> > > I've a dumb question... On my laptop, when I use pacat --passthrough
> > > , does it mean the passthrough mode is used? Though it's
> > > not the usual case as hardware decoding is not used.
> > 
> > Yes -- all the logic that works with passthrough streams goes via
> > pa_sink_input_is_passthrough(), which you will see checks if the stream
> > has non-PCM data OR has the PASSTHROUGH flag set (which is what
> > --passthrough does).
> 
> That's cool, thanks for the info. If this ALSA device can hard decode,
> then in this scenario, it can play, let's say, MP3 file on the fly. :)

Yes, if it's payloaded in IEC61937 format.

-- Arun

___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss


Re: [pulseaudio-discuss] ask about passthrough support

2011-09-14 Thread Lu Guanqun
On Thu, Sep 15, 2011 at 01:58:41PM +0800, Arun Raghavan wrote:
> On Thu, 2011-09-15 at 11:22 +0800, Lu Guanqun wrote:
> > On Thu, Sep 08, 2011 at 02:19:41AM +0800, Arun Raghavan wrote:
> > > On Wed, 2011-09-07 at 11:09 -0700, Arun Raghavan wrote:
> > > > On Wed, 2011-09-07 at 14:20 +0800, Lu Guanqun wrote:
> > > > > Hi Arun,
> > > > > 
> > > > > I have several things unclear about passthrough support:
> > > > 
> > > > I'm talking about this work at Linux Plumbers' Conference on Friday. If
> > > > videos are available, I'll link to them -- hopefully that should make
> > > > the motivation clear.
> > > 
> > > Oh, phew, no video -- I'm still happy to answer any questions you
> > > have. :)
> > 
> > Hi Arun,
> > 
> > I've a dumb question... On my laptop, when I use pacat --passthrough
> > , does it mean the passthrough mode is used? Though it's
> > not the usual case as hardware decoding is not used.
> 
> Yes -- all the logic that works with passthrough streams goes via
> pa_sink_input_is_passthrough(), which you will see checks if the stream
> has non-PCM data OR has the PASSTHROUGH flag set (which is what
> --passthrough does).

That's cool, thanks for the info. If this ALSA device can hard decode,
then in this scenario, it can play, let's say, MP3 file on the fly. :)

-- 
guanqun
___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss


Re: [pulseaudio-discuss] ask about passthrough support

2011-09-14 Thread Arun Raghavan
On Fri, 2011-09-09 at 09:54 +0100, Colin Guthrie wrote:
> 'Twas brillig, and Arun Raghavan at 07/09/11 19:19 did gyre and gimble:
> > On Wed, 2011-09-07 at 11:09 -0700, Arun Raghavan wrote:
> >> On Wed, 2011-09-07 at 14:20 +0800, Lu Guanqun wrote:
> >>> Hi Arun,
> >>>
> >>> I have several things unclear about passthrough support:
> >>
> >> I'm talking about this work at Linux Plumbers' Conference on Friday. If
> >> videos are available, I'll link to them -- hopefully that should make
> >> the motivation clear.
> > 
> > Oh, phew, no video -- I'm still happy to answer any questions you
> > have. :)
> 
> That sucks that there is no video it's a shame for those of us
> genuinely interested but who cannot attend (tell them to bring LPC to
> Europe!)
> 
> Please badger them to get things filmed as when people involved in the
> "scene" can't easily exchange knowledge it's a real drawback :(

IME, the main problem with videos is the cost (in terms of time or
money) of editing and putting things up. Didn't happen to run into the
organisers, but will badger them if I ever do. :)

In the mean time, I've uploaded my slides to:

http://arunraghavan.net/downloads/talks/passthrough-LPC.pdf

Sorry, the PDF version is quite ugly because it's exported from
Pinpoint, which is alpha-quality at the moment.

Cheers,
Arun

___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss


Re: [pulseaudio-discuss] ask about passthrough support

2011-09-14 Thread Arun Raghavan
On Thu, 2011-09-15 at 11:22 +0800, Lu Guanqun wrote:
> On Thu, Sep 08, 2011 at 02:19:41AM +0800, Arun Raghavan wrote:
> > On Wed, 2011-09-07 at 11:09 -0700, Arun Raghavan wrote:
> > > On Wed, 2011-09-07 at 14:20 +0800, Lu Guanqun wrote:
> > > > Hi Arun,
> > > > 
> > > > I have several things unclear about passthrough support:
> > > 
> > > I'm talking about this work at Linux Plumbers' Conference on Friday. If
> > > videos are available, I'll link to them -- hopefully that should make
> > > the motivation clear.
> > 
> > Oh, phew, no video -- I'm still happy to answer any questions you
> > have. :)
> 
> Hi Arun,
> 
> I've a dumb question... On my laptop, when I use pacat --passthrough
> , does it mean the passthrough mode is used? Though it's
> not the usual case as hardware decoding is not used.

Yes -- all the logic that works with passthrough streams goes via
pa_sink_input_is_passthrough(), which you will see checks if the stream
has non-PCM data OR has the PASSTHROUGH flag set (which is what
--passthrough does).

Cheers,
Arun

___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss


Re: [pulseaudio-discuss] ask about passthrough support

2011-09-14 Thread Lu Guanqun
On Thu, Sep 08, 2011 at 02:19:41AM +0800, Arun Raghavan wrote:
> On Wed, 2011-09-07 at 11:09 -0700, Arun Raghavan wrote:
> > On Wed, 2011-09-07 at 14:20 +0800, Lu Guanqun wrote:
> > > Hi Arun,
> > > 
> > > I have several things unclear about passthrough support:
> > 
> > I'm talking about this work at Linux Plumbers' Conference on Friday. If
> > videos are available, I'll link to them -- hopefully that should make
> > the motivation clear.
> 
> Oh, phew, no video -- I'm still happy to answer any questions you
> have. :)

Hi Arun,

I've a dumb question... On my laptop, when I use pacat --passthrough
, does it mean the passthrough mode is used? Though it's
not the usual case as hardware decoding is not used.

-- 
guanqun
___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss


Re: [pulseaudio-discuss] ask about passthrough support

2011-09-09 Thread Colin Guthrie
'Twas brillig, and Arun Raghavan at 07/09/11 19:19 did gyre and gimble:
> On Wed, 2011-09-07 at 11:09 -0700, Arun Raghavan wrote:
>> On Wed, 2011-09-07 at 14:20 +0800, Lu Guanqun wrote:
>>> Hi Arun,
>>>
>>> I have several things unclear about passthrough support:
>>
>> I'm talking about this work at Linux Plumbers' Conference on Friday. If
>> videos are available, I'll link to them -- hopefully that should make
>> the motivation clear.
> 
> Oh, phew, no video -- I'm still happy to answer any questions you
> have. :)

That sucks that there is no video it's a shame for those of us
genuinely interested but who cannot attend (tell them to bring LPC to
Europe!)

Please badger them to get things filmed as when people involved in the
"scene" can't easily exchange knowledge it's a real drawback :(

Col


-- 

Colin Guthrie
gmane(at)colin.guthr.ie
http://colin.guthr.ie/

Day Job:
  Tribalogic Limited [http://www.tribalogic.net/]
Open Source:
  Mageia Contributor [http://www.mageia.org/]
  PulseAudio Hacker [http://www.pulseaudio.org/]
  Trac Hacker [http://trac.edgewall.org/]
___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss


Re: [pulseaudio-discuss] ask about passthrough support

2011-09-08 Thread Lu Guanqun
On Thu, Sep 08, 2011 at 02:19:41AM +0800, Arun Raghavan wrote:
> On Wed, 2011-09-07 at 11:09 -0700, Arun Raghavan wrote:
> > On Wed, 2011-09-07 at 14:20 +0800, Lu Guanqun wrote:
> > > Hi Arun,
> > > 
> > > I have several things unclear about passthrough support:
> > 
> > I'm talking about this work at Linux Plumbers' Conference on Friday. If
> > videos are available, I'll link to them -- hopefully that should make
> > the motivation clear.
> 
> Oh, phew, no video -- I'm still happy to answer any questions you
> have. :)

Thanks Arun for your info. At least there'll be slides published?

-- 
guanqun
___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss


Re: [pulseaudio-discuss] ask about passthrough support

2011-09-07 Thread Pierre-Louis Bossart

> > 1. Is it only useful when there's hardware decoding functionality
> > available? Can I simply test out passthrough with the current ALSA sound
> > card (some emulations possible)?
> 
> Pretty much. The whole idea is that you're pushing out compressed audio,
> so something needs to decode it -- an a/v receiver that does AC3/DTS, or
> a Bluetooth headset that supports MP3 decode.

To build on Arun's comment, the infrastructure is in place at the
PulseAudio level for HDMI/SPDIF and BT. Once the compressed API is
available in the kernel, we'll enable DSP-based decoding as well.
Ironically this was the initial goal but we took the long road to avoid
having to write platform-specific PulseAudio sinks/sources.
-Pierre

___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss


Re: [pulseaudio-discuss] ask about passthrough support

2011-09-07 Thread Arun Raghavan
On Wed, 2011-09-07 at 11:09 -0700, Arun Raghavan wrote:
> On Wed, 2011-09-07 at 14:20 +0800, Lu Guanqun wrote:
> > Hi Arun,
> > 
> > I have several things unclear about passthrough support:
> 
> I'm talking about this work at Linux Plumbers' Conference on Friday. If
> videos are available, I'll link to them -- hopefully that should make
> the motivation clear.

Oh, phew, no video -- I'm still happy to answer any questions you
have. :)

-- Arun

___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss


Re: [pulseaudio-discuss] ask about passthrough support

2011-09-07 Thread Arun Raghavan
On Wed, 2011-09-07 at 14:20 +0800, Lu Guanqun wrote:
> Hi Arun,
> 
> I have several things unclear about passthrough support:

I'm talking about this work at Linux Plumbers' Conference on Friday. If
videos are available, I'll link to them -- hopefully that should make
the motivation clear.

> 1. Is it only useful when there's hardware decoding functionality
> available? Can I simply test out passthrough with the current ALSA sound
> card (some emulations possible)?

Pretty much. The whole idea is that you're pushing out compressed audio,
so something needs to decode it -- an a/v receiver that does AC3/DTS, or
a Bluetooth headset that supports MP3 decode.

> 2. As right now, ALSA only have PCM interface, how do we pass the
> encoded data down? Do we have to use some other driver specific
> interfaces? Vinod, recently posted some patches to add compression
> interface for ALSA. Is that needed?

No, that's not needed. We currently do IEC61937 payloading to make 

> 3. To be able to use passthrough, do we still need changes on Gstreamer?
> Therefore people can simply use something like 'gst-launch xxx'. And
> how's that status?

The base stuff is in git master, and some of it is en route
(https://bugzilla.gnome.org/show_bug.cgi?id=657179).

Regards,
Arun

___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss


[pulseaudio-discuss] ask about passthrough support

2011-09-06 Thread Lu Guanqun
Hi Arun,

I have several things unclear about passthrough support:

1. Is it only useful when there's hardware decoding functionality
available? Can I simply test out passthrough with the current ALSA sound
card (some emulations possible)?

2. As right now, ALSA only have PCM interface, how do we pass the
encoded data down? Do we have to use some other driver specific
interfaces? Vinod, recently posted some patches to add compression
interface for ALSA. Is that needed?

3. To be able to use passthrough, do we still need changes on Gstreamer?
Therefore people can simply use something like 'gst-launch xxx'. And
how's that status?

-- 
guanqun
___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss