On Wed, 10.09.08 20:56, Arthur Pemberton (pemboa@gmail.com) wrote:

> I have a lot of seemingly pulse audio related issues on my F9 desktop.
> Since I really like the idea behind Pulseaudio, I would like to see
> these get sorted out, and would like to know how I can help in terms
> of providing useful information.
>
> Here is my situation
>
> * Sounblaster Live 5.1 PCI -- has full 6 channel audio before, I had
> for a short while after enabling it as per the pulseaudio wiki, but no
> more

Does pavucontrol show the device as having 6 channels? Please make
sure that /etc/pulse/daemon.conf (or ~/.pulse/daemon.conf if you have
that) is configured for 6 channel audio with
"default-sample-channels=6".

Also, an output of "pulseaudio -vvvv" (you might need to stop pa first
by issuing pulseaudio -k first) would be very useful to track down
what is going on.

> * Tvtime -- no audio (doing `padsp sox -r 48000 -w -c 2 -t ossdsp
> /dev/dsp -t ossdsp /dev/dsp` works, but causes noticeable lag)
> * Amarok (xine) -- audio, susceptible to crashes, esp when I scroll to
> fast in Firefox

Xine should work fine. Make sure you are running the newest packages
for F9, then provide crash bt.

> * Kopete -- used to work, mysteriously stopped working
> * Konversation -- never worked
> * Mythfront -- no sound yet (haven't followed
> http://svn.mythtv.org/trac/ticket/3598#comment:16 yet)

I don't know these apps. Don't know much about KDE. Do these also use Xine?

> * V4L tv capture card -- no sound now, used to rely on capturing AUX
> on sound card before

"no sound"? What does that mean?

> * PulseAudio -- seems to average 30% CPU usage when playing music,
> doesn't like me using Firefox too much (seems like a resource issue)

Firefox doesn't use audio. Flash does. Flash is
broken. libflashsupport can fix a few issues but doesn't solve them
entirely.

If PA eats too much CPU this might be caused be the resampler we
use. Consider playinging around with "resample-method" in daemon.conf
and setting it to "trivial".

Lennart

--
Lennart Poettering Red Hat, Inc.
lennart [at] poettering [dot] net ICQ# 11060553
http://0pointer.net/lennart/ GnuPG 0x1A015CC4

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

On Fri, 12.09.08 15:43, Janina Sajka (janina@rednote.net) wrote:

> The problems we have right now are sufficiently sever to be
> showstoppers. At the SpeakupModified.Org we recommend disabling
> pulseaudio. As things stand in F-9, one gets no audio until after a user
> logs in on the GUI. So, how are those who need screen reader support
> supposed to use the a11y features of GDM? As it stands, there seems no
> way to get console audio without that GUI login. Also a nonstarter in
> the screen reader user community.

In Rawhide PA is started on demand (in addition to being started from
gnome-session) and thus should generally work fine on the console too.

> It seems a useful initial step toward resolution is to run pulseaudio
> as a system daemon, via an init script, /sbin/service, /sbin/chkconfig.
> The trade-offs vs the per user model that F-9 employs are probably more
> than acceptable to most a11y users.

Nope. Please don't. This particular problem is solved in Rawhide.

> So, let me simply ask ... Has anyone a working init script for
> pulseaudio as a system daemon? Anything close that we could continue to
> refine as an alternative configuration option for Fedora?

Some distros ship an init script.

Lennart

--
Lennart Poettering Red Hat, Inc.
lennart [at] poettering [dot] net ICQ# 11060553
http://0pointer.net/lennart/ GnuPG 0x1A015CC4

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

On Fri, 12.09.08 22:28, Karel Zak (kzak@redhat.com) wrote:

> > > The problems we have right now are sufficiently sever to be
> > > showstoppers. At the SpeakupModified.Org we recommend disabling
> > > pulseaudio. As things stand in F-9, one gets no audio until after a user
> > > logs in on the GUI. So, how are those who need screen reader support
> > > supposed to use the a11y features of GDM? As it stands, there seems no
> > > way to get console audio without that GUI login. Also a nonstarter in
> > > the screen reader user community.
> >
> > If you want to run terminal applications, can't you just log in to a
> > full screen gnome-terminal?
>
> He wants to use audio before login to X Window. Somehow I don't understand
> how do you want to fix this problem by gnome-terminal...

By shortening the boot time, so that we start X early and have nothing
"before" X. As soon as GDM is up PA is up too.

Lennart

--
Lennart Poettering Red Hat, Inc.
lennart [at] poettering [dot] net ICQ# 11060553
http://0pointer.net/lennart/ GnuPG 0x1A015CC4

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

On Fri, 12.09.08 15:11, Chris Weyl (cweyl@alumni.drew.edu) wrote:

> >> So, let me simply ask ... Has anyone a working init script for
> >> pulseaudio as a system daemon? Anything close that we could continue to
> >> refine as an alternative configuration option for Fedora?
> >
> > We're going to be removing the legacy non-X system consoles by default
> > in the long run.
>
> On a related note, here we get to my pet bug, too: 444172. I often
> use my desktop with one X session to the machine itself, and another X
> session remotely logged in to my work laptop via XDMCP. However, the
> per-user model means that not only can I not use PA on the laptop to
> play over the desktop's speakers, but I can't continue to play
> anything through the first X session.
>
> https://bugzilla.redhat.com/show_bug.cgi?id=444172
>
> (It's filed under ConsoleKit, as that seemed to be where the problem
> was. I'm more than happy to reassign it if that makes sense )

We make sure that a user that is not active on a screen gets no access
to audio device so that he cannot wiretap what you say or hear. (this
works only to a certain degree ince we still lack revoke() in the kernel)

Now, if I correctly understood your problem, then your XDMCP session
on the second vt will run as local gdm user or similar. We thus make
sure that only the gdm user has access.

I see the problem you are experiencing. On the other hand I would say
PA/CK behave correctly here.

I think the proper fix is to start PA locally on the XDMCP vt, so that
your remote audio is actually forwarded to a PA server that is
attached to your X11 server. Not sure about the security implications
however...

Lennart

--
Lennart Poettering Red Hat, Inc.
lennart [at] poettering [dot] net ICQ# 11060553
http://0pointer.net/lennart/ GnuPG 0x1A015CC4

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

On Mon, 15.09.08 16:07, Dominik 'Rathann' Mierzejewski (dominik@greysector.net) wrote:

> Do you mean Open Source Software or Open Sound System? In case of OSS,
> it's realy a shame, because it was (and still is) a great piece of software
> with nice API and doesn't require any external libraries like ALSA.
> But you can't compare console/X to OSS/ALSA. The latter provide functionality

I must correct you: the OSS API sucks. And ALSA is certainly a far
greater piece of software than OSS ever was, and among the reasons is
precisely the fact that it is a proper library instead of some fucked up kernel
interface based on ioctls().

Everyone hates ioctl()s. The kernel people do. The userspace people
too. An API for application usage that is based around ioctl()s is
thus mandatorily a big failure.

Lennart

--
Lennart Poettering Red Hat, Inc.
lennart [at] poettering [dot] net ICQ# 11060553
http://0pointer.net/lennart/ GnuPG 0x1A015CC4

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

On Mon, Sep 22, 2008 at 6:13 PM, Lennart Poettering
<mzerqung@0pointer.de> wrote:
> On Wed, 10.09.08 20:56, Arthur Pemberton (pemboa@gmail.com) wrote:
>
>> I have a lot of seemingly pulse audio related issues on my F9 desktop.
>> Since I really like the idea behind Pulseaudio, I would like to see
>> these get sorted out, and would like to know how I can help in terms
>> of providing useful information.
>>
>> Here is my situation
>>
>> * Sounblaster Live 5.1 PCI -- has full 6 channel audio before, I had
>> for a short while after enabling it as per the pulseaudio wiki, but no
>> more
>
> Does pavucontrol show the device as having 6 channels? Please make
> sure that /etc/pulse/daemon.conf (or ~/.pulse/daemon.conf if you have
> that) is configured for 6 channel audio with
> "default-sample-channels=6".
>
> Also, an output of "pulseaudio -vvvv" (you might need to stop pa first
> by issuing pulseaudio -k first) would be very useful to track down
> what is going on.
>
>> * Tvtime -- no audio (doing `padsp sox -r 48000 -w -c 2 -t ossdsp
>> /dev/dsp -t ossdsp /dev/dsp` works, but causes noticeable lag)
>> * Amarok (xine) -- audio, susceptible to crashes, esp when I scroll to
>> fast in Firefox
>
> Xine should work fine. Make sure you are running the newest packages
> for F9, then provide crash bt.
>
>> * Kopete -- used to work, mysteriously stopped working
>> * Konversation -- never worked
>> * Mythfront -- no sound yet (haven't followed
>> http://svn.mythtv.org/trac/ticket/3598#comment:16 yet)
>
> I don't know these apps. Don't know much about KDE. Do these also use Xine?
>
>> * V4L tv capture card -- no sound now, used to rely on capturing AUX
>> on sound card before
>
> "no sound"? What does that mean?
>
>> * PulseAudio -- seems to average 30% CPU usage when playing music,
>> doesn't like me using Firefox too much (seems like a resource issue)
>
> Firefox doesn't use audio. Flash does. Flash is
> broken. libflashsupport can fix a few issues but doesn't solve them
> entirely.
>
> If PA eats too much CPU this might be caused be the resampler we
> use. Consider playinging around with "resample-method" in daemon.conf
> and setting it to "trivial".
>
> Lennart


Hi Lennart,

I appreciate your assistance, but I regret to say that I removed
PulseAudio this past weekend. It relatively easy, and my system almost
shamefully more functional now. I have since discovered at least one
feature that I did not did not even know existed as PulseAudio wasn't
functioning properly. This situation is made even more unfortunate as
I have recently discovered that current version on Microsoft Windows
has PulseAudio like features working just fine.

If there is a specific test that I can run, I'll be happy to do so,
but I cannot really run PulseAudio full time as is, I enjoy my music
collection too much for this.

Arthur Pemberton


--
Fedora 9 : sulphur is good for the skin
( www.pembo13.com )

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

On Tuesday, 23 September 2008 at 01:39, Lennart Poettering wrote:
> On Mon, 15.09.08 16:07, Dominik 'Rathann' Mierzejewski (dominik@greysector.net) wrote:
>
> > Do you mean Open Source Software or Open Sound System? In case of OSS,
> > it's realy a shame, because it was (and still is) a great piece of software
> > with nice API and doesn't require any external libraries like ALSA.
> > But you can't compare console/X to OSS/ALSA. The latter provide functionality
>
> I must correct you: the OSS API sucks. And ALSA is certainly a far
> greater piece of software than OSS ever was, and among the reasons is
> precisely the fact that it is a proper library instead of some fucked up kernel
> interface based on ioctls().
>
> Everyone hates ioctl()s. The kernel people do. The userspace people
> too. An API for application usage that is based around ioctl()s is
> thus mandatorily a big failure.

The people whose opinion I value disagree. I have no strong opinion
of my own, because I never wrote code to interface with either ALSA
or OSS.

Regards,
R.

--
Fedora http://fedoraproject.org/wiki/User:Rathann
RPMFusion http://rpmfusion.org | MPlayer http://mplayerhq.hu
"Faith manages."
-- Delenn to Lennier in Babylon 5:"Confessions and Lamentations"

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

On Tue, Sep 23, 2008 at 01:39:24AM +0200, Lennart Poettering wrote:
> Everyone hates ioctl()s. The kernel people do. The userspace people

Not true. Ioctl has a very important place in things.

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

On Tue, 23.09.08 10:54, Dominik 'Rathann' Mierzejewski (dominik@greysector.net) wrote:

> > > Do you mean Open Source Software or Open Sound System? In case of OSS,
> > > it's realy a shame, because it was (and still is) a great piece of software
> > > with nice API and doesn't require any external libraries like ALSA.
> > > But you can't compare console/X to OSS/ALSA. The latter provide functionality
> >
> > I must correct you: the OSS API sucks. And ALSA is certainly a far
> > greater piece of software than OSS ever was, and among the reasons is
> > precisely the fact that it is a proper library instead of some fucked up kernel
> > interface based on ioctls().
> >
> > Everyone hates ioctl()s. The kernel people do. The userspace people
> > too. An API for application usage that is based around ioctl()s is
> > thus mandatorily a big failure.
>
> The people whose opinion I value disagree. I have no strong opinion
> of my own, because I never wrote code to interface with either ALSA
> or OSS.

Oh my. So you know someone who thinks that ioctl()s are ingenious API
design? You probably should choose your friends more carefully, then. ;-)

Lennart

--
Lennart Poettering Red Hat, Inc.
lennart [at] poettering [dot] net ICQ# 11060553
http://0pointer.net/lennart/ GnuPG 0x1A015CC4

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list

On Tue, 23.09.08 08:09, Alan Cox (alan@redhat.com) wrote:

>
> On Tue, Sep 23, 2008 at 01:39:24AM +0200, Lennart Poettering wrote:
> > Everyone hates ioctl()s. The kernel people do. The userspace people
>
> Not true. Ioctl has a very important place in things.

Sure they have. But its an awkward interface. Everything multiplexed
via a single syscall. Unreadable/unpronouncable names. No type-safety,
not even size-safety. No proper signatures. ioctl()s are awful.

Lennart

--
Lennart Poettering Red Hat, Inc.
lennart [at] poettering [dot] net ICQ# 11060553
http://0pointer.net/lennart/ GnuPG 0x1A015CC4

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list