Re: snd_pcm_close hangs

[Takashi Iwai <tiwai@suse.de>]

At Mon, 3 Jun 2002 19:48:56 +0200 (CEST),
Tim Goetze wrote:
> 
> unfortunately the bug remains, it only manifests differently.
> 
> after initializing the ice (cap + play are linked):
> 
> [~] cat /proc/asound/ice/pcm0p/sub0/hw_params 
> access: MMAP_INTERLEAVED
> format: S32_LE
> subformat: STD
> channels: 10
> rate: 44100 (44100/1)
> period_size: 128
> buffer_size: 256
> tick_time: 10000
> 
> [~] cat /proc/asound/ice/pcm0p/sub0/sw_params 
> tstamp_mode: NONE
> period_step: 1
> sleep_min: 0
> avail_min: 128
> xfer_align: 128
> start_threshold: 4294967295
> stop_threshold: 256
> silence_threshold: 0
> silence_size: 0
> boundary: 1073741824
> 
> [~] cat /proc/asound/ice/pcm0p/sub0/status 
> state: PREPARED
> trigger_time: 0.000000
> tstamp      : 1023125218.959946
> delay       : 36548963
> avail       : 0
> avail_max   : 0
> -----
> hw_ptr      : 0
> appl_ptr    : 256
 
so far, so good..

> calling snd_pcm_close() takes a few seconds during which i see
> this:
> 
> [~] cat /proc/asound/ice/pcm0p/sub0/status 
> state: DRAINING
> trigger_time: 1023125380.999721
> tstamp      : 1023125381.905192
> delay       : -39746
> avail       : 40002
> avail_max   : 40002
> -----
> hw_ptr      : 40002
> appl_ptr    : 256

the minus delay value is definitely wrong.
something weird goes there...
could you check sw_params during this happens?

 
Takashi

ps.  i'll be at linuxtag, so the next answer will be delayed to the
next week.

_______________________________________________________________

Don't miss the 2002 Sprint PCS Application Developer's Conference
August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm