GNU/Linux Audio Mechanics

On Mon, 24 Nov 2003, Andrew A. Gill wrote:

> I'm trying to get glame to run on my DEC Alpha, but I keep
> getting a segfault.  A little poking led me to this line:
>
> 	glade_register_widget(gtk_knob_get_type(), gtk_knob_glade_new, NULL, NULL);
>
> And further, the function gtk_knob_get_type() is returning the
> value of 538227104.  I'm not sure if that's a valid value, but
> it looks as if it's within the uint data range.
>
> Does anyone have any idea?

This could be an alpha problem with libglade? You may want to disable
glade support by, err, changing config.h (HAVE_LIBGLADE).

You're using glame-1.0.1, don't you?

Richard.

--
Richard Guenther <richard dot guenther at uni-tuebingen dot de>
WWW: http://www.tat.physik.uni-tuebingen.de/~rguenth/

On Tue, 25 Nov 2003, Richard Guenther wrote:

> On Mon, 24 Nov 2003, Andrew A. Gill wrote:
>
> > I'm trying to get glame to run on my DEC Alpha, but I keep
> > getting a segfault.  A little poking led me to this line:
> >
> > 	glade_register_widget(gtk_knob_get_type(), gtk_knob_glade_new, NULL, NULL);
> >
> > And further, the function gtk_knob_get_type() is returning the
> > value of 538227104.  I'm not sure if that's a valid value, but
> > it looks as if it's within the uint data range.
> >
> > Does anyone have any idea?
>
> This could be an alpha problem with libglade? You may want to disable
> glade support by, err, changing config.h (HAVE_LIBGLADE).

OK.  Now it doesn't segfault.  But now it complains about the
swap file:

	ERROR: Unable to open swap: No such file or directory

Sigh.  How do I make a swapfile?

> You're using glame-1.0.1, don't you?

Nope. CVS.

-- 
|Andrew A. Gill                       |I posted to Silent-Tristero and|
|<superluser@...>         |all I got was this stupid sig! |
|alt.tv.simpsons CBG-FAQ author       |                               |
|                          (Report all obscene mail to Le Maitre Pots)|
|Nothing here yet: <http://www.needsfoodbadly.com>    Temporary sig: --

TIA.

On Tue, 25 Nov 2003, Andrew A. Gill wrote:

> On Tue, 25 Nov 2003, Richard Guenther wrote:
>
> > On Mon, 24 Nov 2003, Andrew A. Gill wrote:
> >
> > > I'm trying to get glame to run on my DEC Alpha, but I keep
> > > getting a segfault.  A little poking led me to this line:
> > >
> > > 	glade_register_widget(gtk_knob_get_type(), gtk_knob_glade_new, NULL, NULL);
> > >
> > > And further, the function gtk_knob_get_type() is returning the
> > > value of 538227104.  I'm not sure if that's a valid value, but
> > > it looks as if it's within the uint data range.
> > >
> > > Does anyone have any idea?
> >
> > This could be an alpha problem with libglade? You may want to disable
> > glade support by, err, changing config.h (HAVE_LIBGLADE).
>
> OK.  Now it doesn't segfault.  But now it complains about the
> swap file:
>
> 	ERROR: Unable to open swap: No such file or directory
>
> Sigh.  How do I make a swapfile?

Uh. Suppose you're starting glame, not cglame, it should create one for
you - in fact it should pop up the preferences dialog the first time you
start glame and on such errors, too.

For cglame you need to run cglame -c /path/to/swap to create the swapfile.

Hope this helps,

Richard.

> > You're using glame-1.0.1, don't you?
>
> Nope. CVS.
>
> --
> |Andrew A. Gill                       |I posted to Silent-Tristero and|
> |<superluser@...>         |all I got was this stupid sig! |
> |alt.tv.simpsons CBG-FAQ author       |                               |
> |                          (Report all obscene mail to Le Maitre Pots)|
> |Nothing here yet: <http://www.needsfoodbadly.com>    Temporary sig: --
>
> TIA.
>

--
Richard Guenther <richard dot guenther at uni-tuebingen dot de>
WWW: http://www.tat.physik.uni-tuebingen.de/~rguenth/

On Wed, 26 Nov 2003, Richard Guenther wrote:
>
> Uh. Suppose you're starting glame, not cglame, it should create one for
> you - in fact it should pop up the preferences dialog the first time you
> start glame and on such errors, too.

It did, but it didn't generate a swap file.

> For cglame you need to run cglame -c /path/to/swap to create the swapfile.
>
> Hope this helps,

It did.

More or less.  Now I get the following error in glame (looks
like cglame will work):

ALSA lib seq_hw.c:446:(snd_seq_hw_open) open /dev/snd/seq failed: No such
file or directory
open_alsa_sequencer: Error opening ALSA sequencer.
midi_in_alsa_register: Couldn't open alsa midi sequencer
loading /usr/local/share/glame/scripts/glame.scm
/usr/local/share/glame/scripts/resample.scm: uncaught throw to
glame-error: ()
/usr/local/share/glame/scripts/bassboost.scm: uncaught throw to
glame-error: ()
/usr/local/share/glame/scripts/write_mp3.scm: uncaught throw to
glame-error: ()

GLAME testsuite

 Available tests are
   swtest-all        -- swapfile subsystem checks
   test-filter-all   -- filter/plugin checks
   test-gpsm-all     -- gpsm midlayer checks
 Run all tests with (test-glame)

loading /home/superluser/.glamerc
update_preferences: Preferences:
  Swapfile directory /home/superluser/.glameswap
  Undo stack depth is 5
  Audio input plugin audio_in, device "plughw:0,0", rate 44100
  Audio output plugin audio_out, device "plughw:0,0"
  GLAME_WBUFSIZE 1024
  Popup timeout is 200ms
  Mac mode is off
gui_main: path: /home/superluser/.glameswap
WARNING: Unclean swap - running fsck
cluster_cleanup: SWFS_CLUSTER Statistics: 0 gets, 0 lru hits, 0 misses
cluster_cleanup:        0 meta reads, 0 meta writes, 0 data creates
WARNING: Fsck successful
Xlib: sequence lost (0x10000 > 0x29f) in reply type 0x1!
Xlib: unexpected async reply (sequence 0x0)!
Xlib: sequence lost (0x11fff > 0x29f) in reply type 0x8!
Xlib: sequence lost (0x10240 > 0x29f) in reply type 0x3!
Xlib: sequence lost (0x1200a > 0x29f) in reply type 0x60!
Xlib: sequence lost (0x10000 > 0x29f) in reply type 0x0!

Gdk-ERROR **: The program 'glame' received an X Window System error.
This probably reflects a bug in the program.
The error was '234'.
  (Details: serial 8191 error_code 234 request_code 0 minor_code 207)
  (Note to programmers: normally, X errors are reported asynchronously;
   that is, you will receive the error a while after causing it.
   To debug your program, run it with the --sync command line
   option to change this behavior. You can then get a meaningful
   backtrace from your debugger if you break on the gdk_x_error()
function.)
aborting...

-- 
|Andrew A. Gill                       |I posted to Silent-Tristero and|
|<superluser@...>         |all I got was this stupid sig! |
|alt.tv.simpsons CBG-FAQ author       |                               |
|                          (Report all obscene mail to Le Maitre Pots)|
|Nothing here yet: <http://www.needsfoodbadly.com>    Temporary sig: --

Sigh.

On Wed, 26 Nov 2003, Andrew A. Gill wrote:

> On Wed, 26 Nov 2003, Richard Guenther wrote:
> >
> > Uh. Suppose you're starting glame, not cglame, it should create one for
> > you - in fact it should pop up the preferences dialog the first time you
> > start glame and on such errors, too.
>
> It did, but it didn't generate a swap file.

It should, or glame will be terminated afterwards. Hint: go to the
swapfile tab and change location of the swapfile - it's probably way
wrong.

> > For cglame you need to run cglame -c /path/to/swap to create the swapfile.
> >
> > Hope this helps,
>
> It did.
>
> More or less.  Now I get the following error in glame (looks
> like cglame will work):
>
> ALSA lib seq_hw.c:446:(snd_seq_hw_open) open /dev/snd/seq failed: No such
> file or directory
> open_alsa_sequencer: Error opening ALSA sequencer.
> midi_in_alsa_register: Couldn't open alsa midi sequencer
> loading /usr/local/share/glame/scripts/glame.scm
> /usr/local/share/glame/scripts/resample.scm: uncaught throw to
> glame-error: ()
> /usr/local/share/glame/scripts/bassboost.scm: uncaught throw to
> glame-error: ()
> /usr/local/share/glame/scripts/write_mp3.scm: uncaught throw to
> glame-error: ()

These are harmless. You probably dont have libfftw installed.

> GLAME testsuite
>
>  Available tests are
>    swtest-all        -- swapfile subsystem checks
>    test-filter-all   -- filter/plugin checks
>    test-gpsm-all     -- gpsm midlayer checks
>  Run all tests with (test-glame)
>
> loading /home/superluser/.glamerc
> update_preferences: Preferences:
>   Swapfile directory /home/superluser/.glameswap
>   Undo stack depth is 5
>   Audio input plugin audio_in, device "plughw:0,0", rate 44100
>   Audio output plugin audio_out, device "plughw:0,0"
>   GLAME_WBUFSIZE 1024
>   Popup timeout is 200ms
>   Mac mode is off
> gui_main: path: /home/superluser/.glameswap
> WARNING: Unclean swap - running fsck
> cluster_cleanup: SWFS_CLUSTER Statistics: 0 gets, 0 lru hits, 0 misses
> cluster_cleanup:        0 meta reads, 0 meta writes, 0 data creates
> WARNING: Fsck successful
> Xlib: sequence lost (0x10000 > 0x29f) in reply type 0x1!
> Xlib: unexpected async reply (sequence 0x0)!
> Xlib: sequence lost (0x11fff > 0x29f) in reply type 0x8!
> Xlib: sequence lost (0x10240 > 0x29f) in reply type 0x3!
> Xlib: sequence lost (0x1200a > 0x29f) in reply type 0x60!
> Xlib: sequence lost (0x10000 > 0x29f) in reply type 0x0!
>
> Gdk-ERROR **: The program 'glame' received an X Window System error.
> This probably reflects a bug in the program.
> The error was '234'.
>   (Details: serial 8191 error_code 234 request_code 0 minor_code 207)
>   (Note to programmers: normally, X errors are reported asynchronously;
>    that is, you will receive the error a while after causing it.
>    To debug your program, run it with the --sync command line
>    option to change this behavior. You can then get a meaningful
>    backtrace from your debugger if you break on the gdk_x_error()
> function.)
> aborting...

Well - this is of course "interesting" ;) I'd be interested where this
happens, just try to run glame from inside gdb and do a backtrace.
Probably a gnome2 related thing (we didnt yet fully switch over from
gnome1 to gnome2 and use a lot of legacy stuff). Also tell us more about
your system - I suppose its debian unstable?

Maybe you could even track down the problem and fix it 8)

Thanks,

Richard.

--
Richard Guenther <richard dot guenther at uni-tuebingen dot de>
WWW: http://www.tat.physik.uni-tuebingen.de/~rguenth/

On Thu, 27 Nov 2003, Richard Guenther wrote:
>
> Well - this is of course "interesting" ;) I'd be interested where this
> happens, just try to run glame from inside gdb and do a backtrace.
> Probably a gnome2 related thing (we didnt yet fully switch over from
> gnome1 to gnome2 and use a lot of legacy stuff). Also tell us more about
> your system - I suppose its debian unstable?

Debian unstable isn't this unstable.  I'm running off of the
13th CVS snapshot from Xfree86.  They're up to the 16th one, but
I'm waiting for lucky 17.

So I'm pretty sure that the problem there is in XFree.  Still,
the gdb output is attached.

> Maybe you could even track down the problem and fix it 8)

That's always a bunch of fun.  Seriously, I can't tell you how
many times I've tracked down problems that only manifest on
Alpha, only to have people say later on ``Wow!  Our code did
that?''

Incidentally-- new error:

glame> (play "/home/superluser/spatula.wav")
read_file_set_filename: ftype = 4, version = 0
read_file_set_filename: File /home/superluser/spatula.wav: 1 channel(s) 8 bit unsigned at 8000 Hz, framecount 246841, framesize 1.
filter_launch: initting nodes
filter_launch: launching nodes
launch_node: issuing pthread_create for node audio-out
launcher: audio-out launched (pid 26250)
launch_node: issuing pthread_create for node read-file
launcher: read-file launched (pid 26251)
filter_after_init_hook: read-file seems ready for signalling
filter_launch: waiting for nodes to complete initialization
alsa_audio_out_f: Failed to open ALSA audio device: No such file or directory
launcher: audio-out had failure (errstr="Could not open audio device.")
filter_launch: launching waiter
waiter: waiting for network to finish
wait_node: Waiting for audio-out.
filter_launch: all nodes launched.
filter_start: waiting for nodes to complete initialization.
filter_start: Network error. Terminating with result 1.
filter_wait: waiting for waiter to complete
launcher: filter read-file completed.
launcher: audio-out exiting after error
wait_node: Waiting for read-file.
Segmentation fault


-- 
|Andrew A. Gill                       |I posted to Silent-Tristero and|
|<superluser@...>         |all I got was this stupid sig! |
|alt.tv.simpsons CBG-FAQ author       |                               |
|                          (Report all obscene mail to Le Maitre Pots)|
|Nothing here yet: <http://www.needsfoodbadly.com>    Temporary sig: --

blarg.

On Thu, 27 Nov 2003, Andrew A. Gill wrote:

> On Thu, 27 Nov 2003, Richard Guenther wrote:
> >
> > Well - this is of course "interesting" ;) I'd be interested where this
> > happens, just try to run glame from inside gdb and do a backtrace.
> > Probably a gnome2 related thing (we didnt yet fully switch over from
> > gnome1 to gnome2 and use a lot of legacy stuff). Also tell us more about
> > your system - I suppose its debian unstable?
>
> Debian unstable isn't this unstable.  I'm running off of the
> 13th CVS snapshot from Xfree86.  They're up to the 16th one, but
> I'm waiting for lucky 17.
>
> So I'm pretty sure that the problem there is in XFree.  Still,
> the gdb output is attached.

Umm, the gdb output is not exactly helpful - the stack seems corrupt.

> > Maybe you could even track down the problem and fix it 8)
>
> That's always a bunch of fun.  Seriously, I can't tell you how
> many times I've tracked down problems that only manifest on
> Alpha, only to have people say later on ``Wow!  Our code did
> that?''
>
> Incidentally-- new error:
>
> glame> (play "/home/superluser/spatula.wav")
> read_file_set_filename: ftype = 4, version = 0
> read_file_set_filename: File /home/superluser/spatula.wav: 1 channel(s) 8 bit unsigned at 8000 Hz, framecount 246841, framesize 1.
> filter_launch: initting nodes
> filter_launch: launching nodes
> launch_node: issuing pthread_create for node audio-out
> launcher: audio-out launched (pid 26250)
> launch_node: issuing pthread_create for node read-file
> launcher: read-file launched (pid 26251)
> filter_after_init_hook: read-file seems ready for signalling
> filter_launch: waiting for nodes to complete initialization
> alsa_audio_out_f: Failed to open ALSA audio device: No such file or directory
> launcher: audio-out had failure (errstr="Could not open audio device.")
> filter_launch: launching waiter
> waiter: waiting for network to finish
> wait_node: Waiting for audio-out.
> filter_launch: all nodes launched.
> filter_start: waiting for nodes to complete initialization.
> filter_start: Network error. Terminating with result 1.
> filter_wait: waiting for waiter to complete
> launcher: filter read-file completed.
> launcher: audio-out exiting after error
> wait_node: Waiting for read-file.
> Segmentation fault

As you are using experimental... you're using a 2.6 kernel, too? Then you
need to disable NPTL or glame will fall on its face. I didnt have time to
debug that one, unfortunately.

Richard.

On Thu, 27 Nov 2003, Richard Guenther wrote:

> On Thu, 27 Nov 2003, Andrew A. Gill wrote:
>
> > On Thu, 27 Nov 2003, Richard Guenther wrote:
> > >
> > > Well - this is of course "interesting" ;) I'd be interested where this
> > > happens, just try to run glame from inside gdb and do a backtrace.
> > > Probably a gnome2 related thing (we didnt yet fully switch over from
> > > gnome1 to gnome2 and use a lot of legacy stuff). Also tell us more about
> > > your system - I suppose its debian unstable?
> >
> > Debian unstable isn't this unstable.  I'm running off of the
> > 13th CVS snapshot from Xfree86.  They're up to the 16th one, but
> > I'm waiting for lucky 17.
> >
> > So I'm pretty sure that the problem there is in XFree.  Still,
> > the gdb output is attached.
>
> Umm, the gdb output is not exactly helpful - the stack seems corrupt.

Didn't expect it to be, I'm pretty sure it's a bug in XFree.

> As you are using experimental... you're using a 2.6 kernel, too? Then you
> need to disable NPTL or glame will fall on its face. I didnt have time to
> debug that one, unfortunately.

I'm using a 2.4 kernel.  No NPTL.

-- 
|Andrew A. Gill                       |I posted to Silent-Tristero and|
|<superluser@...>         |all I got was this stupid sig! |
|alt.tv.simpsons CBG-FAQ author       |                               |
|                          (Report all obscene mail to Le Maitre Pots)|
|Nothing here yet: <http://www.needsfoodbadly.com>    Temporary sig: --

Sorry.

On Thu, 27 Nov 2003, Andrew A. Gill wrote:

> I'm using a 2.4 kernel.  No NPTL.

So it shouldn't segfault and I'd be interested in seeing the backtrace for
this, too (if its going with an ok stack).

Thanks,

Richard.

On Thu, 27 Nov 2003, Richard Guenther wrote:

> On Thu, 27 Nov 2003, Andrew A. Gill wrote:
>
> > I'm using a 2.4 kernel.  No NPTL.
>
> So it shouldn't segfault and I'd be interested in seeing the backtrace for
> this, too (if its going with an ok stack).

(The last gdb output that I sent was a set of excerpts, by the
way.  I really don't think that that was anything that you
should have been concerned with.  I really do think that it's
just a problem with X that's probably been corrected in newer
versions.)

But--back to the question at hand.  Log attached.

-- 
|Andrew A. Gill                       |I posted to Silent-Tristero and|
|<superluser@...>         |all I got was this stupid sig! |
|alt.tv.simpsons CBG-FAQ author       |                               |
|                          (Report all obscene mail to Le Maitre Pots)|
|Nothing here yet: <http://www.needsfoodbadly.com>    Temporary sig: --

TIA.