search

arifgursel / pyglet

Automatically exported from code.google.com/p/pyglet
BSD 3-Clause "New" or "Revised" License
0 stars 0 forks source link

ALSA: too many open files #149

Closed GoogleCodeExporter closed 9 years ago

GoogleCodeExporter commented 9 years ago 
What steps will reproduce the problem?
1. Run the noisy.py example
2. When the ball bounces the bellow traceback occurs
3.

Paste in the traceback or error message:
ajones@dill ~/dev/pyglet/examples/noisy $ python noisy.py 
Traceback (most recent call last):
  File "noisy.py", line 85, in <module>
    ball.update(dt)
  File "noisy.py", line 47, in update
    sound.play()
  File "/usr/lib/python2.5/site-packages/pyglet/media/__init__.py", line
377, in play
    player.queue(self)
  File "/usr/lib/python2.5/site-packages/pyglet/media/__init__.py", line
692, in queue
    self._begin_source()
  File "/usr/lib/python2.5/site-packages/pyglet/media/__init__.py", line
774, in _begin_source
    self._create_audio()
  File "/usr/lib/python2.5/site-packages/pyglet/media/__init__.py", line
607, in _create_audio
    self._audio = audio_player_class(source.audio_format)
  File
"/usr/lib/python2.5/site-packages/pyglet/media/drivers/alsa/__init__.py",
line 97, in __init__
    int(self._buffer_time * audio_format.sample_rate)))
  File
"/usr/lib/python2.5/site-packages/pyglet/media/drivers/alsa/__init__.py",
line 56, in check
    raise ALSAException(asound.snd_strerror(err))
pyglet.media.drivers.alsa.ALSAException: Invalid argument

Paste in the output of tools/gl_info.py:
ajones@dill ~/dev/pyglet $ python tools/gl_info.py 
Platform instance is <pyglet.window.xlib.XlibPlatform object at 0x82f45ec>
Display instance is <pyglet.window.xlib.XlibDisplayDevice object at 0xb7ddfa0c>
Screens:
  XlibScreen(screen=0, x=0, y=0, width=1280, height=800, xinerama=0)
Creating default context...
GL attributes:
double_buffer=c_long(1) stereo=c_long(0) buffer_size=c_long(32)
aux_buffers=c_long(4) sample_buffers=c_long(0) samples=c_long(0)
red_size=c_long(8) green_size=c_long(8) blue_size=c_long(8)
alpha_size=c_long(0) depth_size=c_long(24) stencil_size=c_long(8)
accum_red_size=c_long(16) accum_green_size=c_long(16)
accum_blue_size=c_long(16) accum_alpha_size=c_long(16)
GL version: 2.1.0 NVIDIA 96.31
GL vendor: NVIDIA Corporation
GL renderer: GeForce Go 7400/PCI/SSE2
GL extensions:
  GL_ARB_texture_compression GL_NV_point_sprite
  GL_NV_fragment_program_option GL_NV_vertex_array_range2
  GL_EXT_blend_color GL_EXT_blend_subtract GL_EXT_stencil_wrap
  GL_NV_fence GL_EXT_vertex_array GL_IBM_texture_mirrored_repeat
  GL_EXT_texture_compression_s3tc GL_NV_vertex_program2_option
  GL_ARB_depth_texture GL_NV_packed_depth_stencil GL_EXT_Cg_shader
  GL_NV_primitive_restart GL_EXT_texture_object
  GL_NVX_conditional_render GL_NV_texture_compression_vtc
  GL_NV_texture_shader3 GL_NV_texture_shader2 GL_EXT_blend_func_separate
  GL_EXT_texture_cube_map GL_ARB_occlusion_query GL_ARB_transpose_matrix
  GL_ARB_texture_border_clamp GL_EXT_fog_coord GL_NV_depth_clamp
  GL_ARB_texture_cube_map GL_EXT_blend_equation_separate
  GL_ARB_point_parameters GL_EXT_texture_env_dot3
  GL_EXT_multi_draw_arrays GL_HP_occlusion_test
  GL_EXT_texture_env_combine GL_NV_fog_distance GL_ARB_vertex_program
  GL_ARB_texture_env_dot3 GL_EXT_bgra GL_NV_float_buffer
  GL_NV_texture_env_combine4 GL_ARB_texture_env_combine
  GL_ARB_vertex_buffer_object GL_EXT_shadow_funcs
  GL_NV_vertex_array_range GL_NV_texture_expand_normal
  GL_EXT_separate_specular_color GL_EXT_texture_env_add
  GL_EXT_packed_depth_stencil GL_EXT_texture_mirror_clamp
  GL_EXT_timer_query GL_NV_multisample_filter_hint GL_SUN_slice_accum
  GL_ARB_point_sprite GL_EXT_framebuffer_object GL_ARB_vertex_shader
  GL_ARB_half_float_pixel GL_KTX_buffer_region GL_NV_register_combiners
  GL_ARB_draw_buffers GL_ARB_fragment_program_shadow GL_NV_half_float
  GL_NV_fragment_program GL_ATI_texture_mirror_once
  GL_IBM_rasterpos_clip GL_EXT_texture_sRGB GL_EXT_stencil_two_side
  GL_SGIX_shadow GL_NV_texgen_reflection GL_EXT_compiled_vertex_array
  GL_NV_fragment_program2 GL_EXT_point_parameters GL_ARB_texture_float
  GL_ARB_fragment_shader GL_ATI_texture_float GL_NV_texture_shader
  GL_NV_pixel_data_range GL_ARB_imaging GL_ARB_fragment_program
  GL_ARB_shader_objects GL_ARB_shading_language_100 GL_NV_blend_square
  GL_EXT_secondary_color GL_ARB_texture_non_power_of_two
  GL_EXT_texture_lod_bias GL_NV_texture_rectangle
  GL_EXT_pixel_buffer_object GL_EXT_abgr GL_NV_vertex_program
  GL_ARB_texture_rectangle GL_ARB_multisample GL_SGIS_generate_mipmap
  GL_EXT_texture_filter_anisotropic GL_NV_light_max_exponent
  GL_EXT_blend_minmax GL_SGIS_texture_lod GL_NV_vertex_program3
  GL_NV_vertex_program2 GL_SGIX_depth_texture GL_ARB_color_buffer_float
  GL_EXT_texture_edge_clamp GL_ARB_window_pos GL_ARB_shadow
  GL_ARB_texture_mirrored_repeat GL_EXT_gpu_program_parameters
  GL_EXT_texture_lod GL_S3_s3tc GL_EXT_stencil_clear_tag
  GL_NV_vertex_program1_1 GL_EXT_texture3D GL_ARB_multitexture
  GL_NV_copy_depth_to_color GL_ATI_draw_buffers GL_EXT_depth_bounds_test
  GL_EXT_draw_range_elements GL_NV_occlusion_query
  GL_ARB_texture_env_add GL_EXT_packed_pixels GL_EXT_rescale_normal
  GL_ARB_pixel_buffer_object GL_NV_register_combiners2
GLU version: 1.3
GLU extensions:
  GLU_EXT_nurbs_tessellator GLU_EXT_object_space_tess

Context is XlibGLContext()
GLX is direct
GLX server vendor: NVIDIA Corporation
GLX server version: 1.3
GLX server extensions:
  GLX_EXT_visual_info GLX_EXT_visual_rating GLX_SGIX_fbconfig
  GLX_SGIX_pbuffer GLX_SGI_video_sync GLX_SGI_swap_control
  GLX_EXT_texture_from_pixmap GLX_ARB_multisample GLX_NV_float_buffer
  GLX_ARB_fbconfig_float
GLX client vendor: NVIDIA Corporation
GLX client version: 1.4
GLX client extensions:
  GLX_ARB_get_proc_address GLX_ARB_multisample GLX_EXT_visual_info
  GLX_EXT_visual_rating GLX_EXT_import_context GLX_SGI_video_sync
  GLX_NV_swap_group GLX_NV_video_out GLX_SGIX_fbconfig GLX_SGIX_pbuffer
  GLX_SGI_swap_control GLX_NV_float_buffer GLX_ARB_fbconfig_float
  GLX_EXT_fbconfig_packed_float GLX_EXT_texture_from_pixmap
GLX extensions:
  GLX_EXT_visual_info GLX_EXT_visual_rating GLX_SGIX_fbconfig
  GLX_SGIX_pbuffer GLX_SGI_video_sync GLX_SGI_swap_control
  GLX_EXT_texture_from_pixmap GLX_ARB_multisample GLX_NV_float_buffer
  GLX_ARB_fbconfig_float GLX_ARB_get_proc_address

Any additional info (platform/language/hardware) thaOS: Ubuntu 7.04
$ uname -a
Linux dill 2.6.20-16-generic #2 SMP Sun Sep 23 19:50:39 UTC 2007 i686 GNU/Linux
Harware: Sony VAIO VGN-SZ28GPt may be relevant?

Original issue reported on code.google.com by andy.lar...@gmail.com on 2 Oct 2007 at 12:50

GoogleCodeExporter commented 9 years ago 
Can't replicate it here (I have a Creative Audigy 4); I suspect there's a 
hardware
limitation on buffer size.

See if you can get past the error (audio will be broken) by changing the value 
of 

  int(self._buffer_time * audio_format.sample_rate)

at pyglet/media/drivers/alsa/__init__.py:97.  I imagine the number needs to be 
either
smaller, larger, or aligned to some number (for example, a multiple of 4).

If this is the case, we need to find the appropriate ALSA magic that gives the
constraints.

Original comment by Alex.Hol...@gmail.com on 2 Oct 2007 at 1:02

GoogleCodeExporter commented 9 years ago 
8192 gets passed it. I get about 4-6 ball bounce sounds before I get the 
following trace.
4096 also works but only get 1 bounce. 
16384 gives the original trace.

Traceback (most recent call last):
  File "noisy.py", line 85, in <module>
    ball.update(dt)
  File "noisy.py", line 47, in update
    sound.play()
  File "/home/ajones/dev/pyglet/pyglet/media/__init__.py", line 377, in play
    player.queue(self)
  File "/home/ajones/dev/pyglet/pyglet/media/__init__.py", line 692, in queue
    self._begin_source()
  File "/home/ajones/dev/pyglet/pyglet/media/__init__.py", line 775, in _begin_source
    self._fill_audio()
  File "/home/ajones/dev/pyglet/pyglet/media/__init__.py", line 628, in _fill_audio
    self._audio.write(audio_data)
  File "/home/ajones/dev/pyglet/pyglet/media/drivers/alsa/__init__.py", line 146, in
write
    check(asound.snd_pcm_delay(self.pcm, delay))
  File "/home/ajones/dev/pyglet/pyglet/media/drivers/alsa/__init__.py", line 56, in check
    raise ALSAException(asound.snd_strerror(err))
pyglet.media.drivers.alsa.ALSAException: Broken pipe

Original comment by andy.lar...@gmail.com on 2 Oct 2007 at 1:45

GoogleCodeExporter commented 9 years ago 
Also the value I was getting for int(self._buffer_time * 
audio_format.sample_rate)
was 13230

Original comment by andy.lar...@gmail.com on 2 Oct 2007 at 1:48

GoogleCodeExporter commented 9 years ago 
This is the error that I was getting too (see
http://groups.google.com/group/pyglet-users/web/sound-under-ubuntu-feisty for 
details
on switching to openal). I can have a dig around when I get home tonight and 
see if I
can shed more light.

Original comment by anthony....@gmail.com on 3 Oct 2007 at 12:34

GoogleCodeExporter commented 9 years ago 
My error is actually on the line above (95), but they may be related. I put the
following debug statements into pyglet/media/drivers/alsa/__init__.py:95:

        print "Sample rate is:", audio_format.sample_rate
        print asound.snd_pcm_hw_params_set_rate(self.pcm, self.hwparams,
audio_format.sample_rate, 0)
        check(asound.snd_pcm_hw_params_set_rate(self.pcm, self.hwparams,
            audio_format.sample_rate, 0))

And got the following output:
Sample rate is: 44100
-22
Traceback (most recent call last):
  File "./test.py", line 6, in <module>
    bullet_sound.play()
  File "/usr/lib/python2.5/site-packages/pyglet/media/__init__.py", line 377, in play
    player.queue(self)
  File "/usr/lib/python2.5/site-packages/pyglet/media/__init__.py", line 692, in queue
    self._begin_source()
  File "/usr/lib/python2.5/site-packages/pyglet/media/__init__.py", line 774, in
_begin_source
    self._create_audio()
  File "/usr/lib/python2.5/site-packages/pyglet/media/__init__.py", line 607, in
_create_audio
    self._audio = audio_player_class(source.audio_format)
  File "/usr/lib/python2.5/site-packages/pyglet/media/drivers/alsa/__init__.py", line
97, in __init__
    audio_format.sample_rate, 0))
  File "/usr/lib/python2.5/site-packages/pyglet/media/drivers/alsa/__init__.py", line
56, in check
    raise ALSAException(asound.snd_strerror(err))
pyglet.media.drivers.alsa.ALSAException: Invalid argument

Other factors:
This is against the latest release (r1308), running on Ubuntu Feisty.
My sound card is also quite old, so your earlier hunch about buffer size may be 
right:
00:08.0 Multimedia audio controller: Ensoniq 5880 AudioPCI (rev 02)
        Subsystem: Ensoniq Creative SoundBlaster AudioPCI 128
        Flags: bus master, slow devsel, latency 64, IRQ 11
        I/O ports at e800 [size=64]
        Capabilities: [dc] Power Management version 1

Hope that helps track down what's going on -- let me know if you need any more 
info.

Original comment by anthony....@gmail.com on 3 Oct 2007 at 1:27

GoogleCodeExporter commented 9 years ago 
I've added some commented experimental code to r1309 to address both the buffer 
size
and the sample rate problems.  Before starting, set an environment variable
PYGLET_DEBUG_MEDIA to 1; this will print some things to stdout as well as 
create an
alsa.log file.

Try commenting/uncommenting code in pyglet/media/drivers/alsa/__init__.py around
lines 97 and 122 according to the comments and log what happens.  

Apologies for the experimental nature of this... experiment!

Original comment by Alex.Hol...@gmail.com on 3 Oct 2007 at 3:15

GoogleCodeExporter commented 9 years ago 
Well, the change seems to work for me (I can hear sound with ALSA), although 
from the
output and looking at the new code I'm not sure what it's changing:

anthony@anthony:~/pyglet/pyglet/examples/astraea$ python test.py 
Requested sample rate 44100
Actual sample rate 44100
Requested buffer size 13230
Actual buffer size 13230

That's the same sample rate that I had before, but
asound.snd_pcm_hw_params_set_rate_near(...) obviously isn't returning an error 
code.
I've played around with the commented out code, but both
asound.snd_pcm_hw_params_set_rate and 
asound.snd_pcm_hw_params_set_rate_resample both
cause errors.

I've attached my test file - I've been running it from within the astraea 
directory.

Original comment by anthony....@gmail.com on 4 Oct 2007 at 9:14

Attachments:

GoogleCodeExporter commented 9 years ago 
Ok, sounds promising.  Can you attach the alsa.log file (from the set_rate_near 
run) as well?

Original comment by Alex.Hol...@gmail.com on 4 Oct 2007 at 9:22

GoogleCodeExporter commented 9 years ago 
I updated to r1309 and now I'm getting this:
ajones@dill ~/dev/pyglet/examples/noisy $ python noisy.py 
Traceback (most recent call last):
  File "noisy.py", line 85, in <module>
    ball.update(dt)
  File "noisy.py", line 47, in update
    sound.play()
  File "/home/ajones/dev/pyglet/pyglet/media/__init__.py", line 377, in play
    player.queue(self)
  File "/home/ajones/dev/pyglet/pyglet/media/__init__.py", line 692, in queue
    self._begin_source()
  File "/home/ajones/dev/pyglet/pyglet/media/__init__.py", line 774, in _begin_source
    self._create_audio()
  File "/home/ajones/dev/pyglet/pyglet/media/__init__.py", line 607, in _create_audio
    self._audio = audio_player_class(source.audio_format)
  File "/home/ajones/dev/pyglet/pyglet/media/drivers/alsa/__init__.py", line 85, in
__init__
    asound.SND_PCM_NONBLOCK))
  File "/home/ajones/dev/pyglet/pyglet/media/drivers/alsa/__init__.py", line 58, in check
    raise ALSAException(asound.snd_strerror(err))
pyglet.media.drivers.alsa.ALSAException: Device or resource busy
python: pcm.c:695: snd_pcm_close: Assertion `pcm' failed.
Aborted (core dumped)

Original comment by andy.lar...@gmail.com on 4 Oct 2007 at 9:37

GoogleCodeExporter commented 9 years ago 
Attached alsa.log file

Original comment by anthony....@gmail.com on 4 Oct 2007 at 9:56

Attachments:

GoogleCodeExporter commented 9 years ago 
Hmm, spoke too soon. I get the same error message that Larry's getting. It 
seems to
happen if more than one sound tries to play at the same time. It's repeatable - 
I can
trigger it by either adding lots of balls, or waiting until the ball hits the 
corner
(strikes two walls in quick succession).

Here's my traceback:

anthony@anthony:~/pyglet/pyglet/examples/noisy$ python noisy.py 
Traceback (most recent call last):
  File "noisy.py", line 85, in <module>
    ball.update(dt)
  File "noisy.py", line 44, in update
    sound.play()
  File "/usr/lib/python2.5/site-packages/pyglet/media/__init__.py", line 377, in play
    player.queue(self)
  File "/usr/lib/python2.5/site-packages/pyglet/media/__init__.py", line 692, in queue
    self._begin_source()
  File "/usr/lib/python2.5/site-packages/pyglet/media/__init__.py", line 774, in
_begin_source
    self._create_audio()
  File "/usr/lib/python2.5/site-packages/pyglet/media/__init__.py", line 607, in
_create_audio
    self._audio = audio_player_class(source.audio_format)
  File "/usr/lib/python2.5/site-packages/pyglet/media/drivers/alsa/__init__.py", line
85, in __init__
    asound.SND_PCM_NONBLOCK))
  File "/usr/lib/python2.5/site-packages/pyglet/media/drivers/alsa/__init__.py", line
58, in check
    raise ALSAException(asound.snd_strerror(err))
pyglet.media.drivers.alsa.ALSAException: Device or resource busy
python: pcm.c:695: snd_pcm_close: Assertion `pcm' failed.
Aborted (core dumped)

Original comment by anthony....@gmail.com on 4 Oct 2007 at 10:15

GoogleCodeExporter commented 9 years ago 
Ok, Anthony's alsa.log shows that the sample rate that's actually being used is 
44101
instead of the 44100 that we requested.  So it seems that using *_near() is a 
good
solution that fixes this problem (at least).

Device/resource busy at that point (line 85) can only occur when there are no 
more
audio buffers.  I was under the impression that the plug:front device creates
software-mixed buffers when there are no more hardware buffers available (it's 
clear
that both of your audio cards have only a single audio buffer) -- it seems I was
mistaken.  (Incidentally, I would guess that you're unable to play sound from 
two
ALSA applications at a time?)

Andy: you should at least be able to play a single file now, am I correct? 
(Since
this new failure is occurring at a point in the source before any of the 
changes I
introduced).

Unless either of you has any additional information, I will look into 
suppressing the
error when the buffer cannot be created (thus muting sounds when there's 
insufficient
resources) and "locking-in" the changes to use *_near().

Original comment by Alex.Hol...@gmail.com on 4 Oct 2007 at 10:40

GoogleCodeExporter commented 9 years ago 
> (Incidentally, I would guess that you're unable to play sound 
>  from two ALSA applications at a time?)

That seems to be the case - If you have a good way of determining that for sure 
(and
it's worthwhile), I can try it.

Original comment by anthony....@gmail.com on 4 Oct 2007 at 11:41

GoogleCodeExporter commented 9 years ago 
You could run aplay from two different terminals.

Original comment by Alex.Hol...@gmail.com on 4 Oct 2007 at 12:03

GoogleCodeExporter commented 9 years ago 
I've run the following command in three terminal windows at once:

for ITEM in `ls`; do aplay res/27504.wav; done

And got multiple channels all going at once.

I tried to do the same thing with pyglet (see test.py), I get the same "python:
pcm.c:695: snd_pcm_close: Assertion `pcm' failed" error if I try and run the 
sounds
too close together.

If I run test.py in a second window while the first is running, I also get the 
same
Assertion error.

Original comment by anthony....@gmail.com on 5 Oct 2007 at 12:00

Attachments:

GoogleCodeExporter commented 9 years ago 
Ok, try replacing "plug:front" on line 62 with "default" (this is the default 
device
alsaplayer uses, according to the latest source).

Original comment by Alex.Hol...@gmail.com on 5 Oct 2007 at 2:04

GoogleCodeExporter commented 9 years ago 
Can you let me know if replacing "plug:front" with "default" makes any 
difference in playback of multiple audio 
sources?

Original comment by Alex.Hol...@gmail.com on 23 Oct 2007 at 1:58

GoogleCodeExporter commented 9 years ago 
Sorry, I have been a bit quiet - mostly RL issues -- standard excuse, I know :)

Changing to 'default' allows me to play multiple sounds - I have three going at 
the
moment. I'll leave them running for the next little while and see if any of them
explode, but they've been going for a minute or two so far.

The sounds aren't 'syncronised', in that if I play two test.py scripts they'll 
go in
and out of sync, but I don't think that's such a big deal :)

Original comment by anthony....@gmail.com on 25 Oct 2007 at 9:45

GoogleCodeExporter commented 9 years ago 
Ok, I fixed up the bug in the previous test.py version (don't load your media 
over
and over again in the main game loop) which seems to make it more stable. 
However,
after a while I get the error below - 'Too many open files'. I've put in a 
counter to
check when it happens - I've got 508 for the first run, and I'll run it again 
and see
if that changes.

anthony@anthony:~/pyglet/pyglet-old/examples/astraea$ ./test.py 
Playing sound...
ALSA lib pcm_dmix.c:914:(snd_pcm_dmix_open) unable to open slave
Error in sys.excepthook:
Traceback (most recent call last):
  File "/var/lib/python-support/python2.5/apport_python_hook.py", line 30, in
apport_excepthook
    import apport.report, apport.fileutils
  File "/var/lib/python-support/python2.5/apport/__init__.py", line 1, in <module>
    from apport.report import Report
ImportError: No module named report

Original exception was:
Traceback (most recent call last):
  File "./test.py", line 15, in <module>

  File "/usr/lib/python2.5/site-packages/pyglet/media/__init__.py", line 377, in play
    player.queue(self)
  File "/usr/lib/python2.5/site-packages/pyglet/media/__init__.py", line 695, in queue
    self._begin_source()
  File "/usr/lib/python2.5/site-packages/pyglet/media/__init__.py", line 784, in
_begin_source
    self._create_audio()
  File "/usr/lib/python2.5/site-packages/pyglet/media/__init__.py", line 610, in
_create_audio
    self._audio = audio_player_class(source.audio_format)
  File "/usr/lib/python2.5/site-packages/pyglet/media/drivers/alsa/__init__.py", line
85, in __init__
    asound.SND_PCM_NONBLOCK))
  File "/usr/lib/python2.5/site-packages/pyglet/media/drivers/alsa/__init__.py", line
58, in check
    raise ALSAException(asound.snd_strerror(err))
pyglet.media.drivers.alsa.ALSAException: Too many open files
python: pcm.c:695: snd_pcm_close: Assertion `pcm' failed.
Aborted (core dumped)
anthony@anthony:~/pyglet/pyglet-old/examples/astraea$

Original comment by anthony....@gmail.com on 25 Oct 2007 at 10:33

GoogleCodeExporter commented 9 years ago 
Happened at 508 again, so I guess there's something not being closed properly in
ALSA/pyglet? I've attached my test script.

...
504
505
506
507
ALSA lib pcm_dmix.c:914:(snd_pcm_dmix_open) unable to open slave
Error in sys.excepthook:
Traceback (most recent call last):
...

Original comment by anthony....@gmail.com on 25 Oct 2007 at 10:48

Attachments:

GoogleCodeExporter commented 9 years ago 
OK, "default" replaces "plug:front" in r1329.  I can't reproduce the "Too many 
open
files" error.

Original comment by Alex.Hol...@gmail.com on 26 Oct 2007 at 1:29

GoogleCodeExporter commented 9 years ago 
Only remaining error is the "too many open files".  I suspect this is an ALSA 
bug,
we're closing all the PCM handles we open.

The ALSA driver is extremely substandard compared to OpenAL and DirectSound.  
We have
no way to synchronise multiple sounds or to adjust the playback volume or pan. 
There's nothing we can do about this without introducing a fast array module 
(e.g.
numpy) for doing software audio mixing.  I will make notes in the documentation
explaining the trade-off between using OpenAL and ALSA on Linux.

Original comment by Alex.Hol...@gmail.com on 26 Oct 2007 at 2:21

GoogleCodeExporter commented 9 years ago 
I also get the following error in r1355 with my Audigy4 after shooting lots of 
things
in astraea:

python: pcm_params.c:2276: snd_pcm_hw_refine: Assertion `pcm && params' failed.
Abort

Original comment by Alex.Hol...@gmail.com on 4 Nov 2007 at 8:42

GoogleCodeExporter commented 9 years ago 
I get:

python2.5: pcm_params.c:2276: snd_pcm_hw_refine: Assertion `pcm && params' 
failed.
Aborted

running noisy or astraea after snd_pcm_open has been called 5 times more oftem 
than
snd_pcm_close.
Using alsa-lib-1.0.15-33.el4 and  Multimedia audio controller: VIA 
Technologies, Inc.
VT8233/A/8235/8237 AC97 Audio Controller (rev 60)

Original comment by elmar.me...@gmail.com on 23 Dec 2007 at 10:56

GoogleCodeExporter commented 9 years ago 
Thanks for the report.  You should have no problems with audio if you install 
OpenAL (I am recommending this 
to all Linux users now, as the ALSA problems seem insurmountable).

Original comment by Alex.Hol...@gmail.com on 23 Dec 2007 at 11:58

GoogleCodeExporter commented 9 years ago 
I'm closing this bug since it is rather old and we need a clean list to follow 
new issues. Can you retest the problem with a recent version and add the issue 
again?

Original comment by winstonw...@gmail.com on 22 Jun 2012 at 9:51