| Summary: | pulseaudio will continue using high cpu after hibernate | ||
|---|---|---|---|
| Product: | PulseAudio | Reporter: | TOM_Harrison <l12436> |
| Component: | core | Assignee: | pulseaudio-bugs |
| Status: | RESOLVED MOVED | QA Contact: | pulseaudio-bugs |
| Severity: | normal | ||
| Priority: | medium | CC: | lennart, main.haarp |
| Version: | unspecified | ||
| Hardware: | Other | ||
| OS: | All | ||
| Whiteboard: | |||
| i915 platform: | i915 features: | ||
| Attachments: |
settings
output that generate by pulseaudio -vvv detail of log log with pulseaudio, pacmd, aplay |
||
|
Description
TOM_Harrison
2013-10-31 07:59:58 UTC
Please share details about your PA version, and what configuration you have with dmix/dsnoop. Also, if you have sysprof, fire that up when CPU usage is high, take a snapshot of several seconds' worth, and paste the trace because it the problem that only auto-spawn could happened. so it could be not easy to trace. Created attachment 88400 [details]
settings
this is snapshot with sysprof
[/usr/bin/pulseaudio] 0.00% 82.13%
In file /lib/x86_64-linux-gnu/libc-2.17.so 40.41% 40.42%
- - kernel - - 0.00% 0.00%
retint_careful 0.00% 0.00%
_raw_spin_unlock_irq 0.00% 0.00%
No map [/usr/bin/pulseaudio] 0.00% 33.59%
In file [heap] 0.00% 23.29%
ioctl 0.34% 22.48%
- - kernel - - 0.00% 22.14%
system_call_fastpath 0.00% 22.14%
sys_ioctl 0.16% 22.07%
do_vfs_ioctl 0.14% 21.12%
snd_pcm_playback_ioctl 0.02% 20.97%
snd_pcm_playback_ioctl1 0.07% 20.91%
snd_pcm_common_ioctl1 0.08% 20.82%
_raw_read_unlock_irq 20.64% 20.64%
snd_pcm_hwsync 0.08% 0.09%
_raw_read_lock_irq 0.01% 0.01%
snd_pcm_hwsync 0.02% 0.02%
snd_pcm_common_ioctl1 0.03% 0.03%
snd_pcm_playback_ioctl1 0.01% 0.01%
fget_light 0.47% 0.51%
fput 0.19% 0.19%
snd_pcm_playback_ioctl 0.04% 0.04%
__rcu_read_lock 0.02% 0.02%
cap_file_ioctl 0.02% 0.02%
security_file_ioctl 0.00% 0.01%
__rcu_read_unlock 0.01% 0.01%
do_vfs_ioctl 0.03% 0.03%
fput 0.02% 0.02%
fget_light 0.01% 0.01%
In file /usr/lib/x86_64-linux-gnu/libasound.so.2.0.0 0.43% 0.44%
ppoll 0.02% 0.24%
snd_pcm_state 0.04% 0.04%
pa_rtpoll_run 0.02% 0.02%
snd_pcm_hwsync 0.01% 0.01%
pa_asyncmsgq_read_after_poll 0.01% 0.01%
pa_asyncmsgq_write_before_poll 0.01% 0.01%
In file /lib/x86_64-linux-gnu/libc-2.17.so 0.01% 0.01%
pa_fdsem_after_poll 0.01% 0.01%
pa_asyncq_write_after_poll 0.01% 0.01%
gettimeofday 0.00% 0.00%
pa_asyncmsgq_write_after_poll 0.00% 0.00%
snd_pcm_avail_update 0.00% 0.00%
pa_asyncq_write_before_poll 0.00% 0.00%
pa_asyncq_read_before_poll 0.00% 0.00%
In file /lib/x86_64-linux-gnu/libc-2.17.so 8.53% 8.53%
In file /usr/lib/x86_64-linux-gnu/libasound.so.2.0.0 0.31% 0.31%
pa_once_begin 0.29% 0.29%
pa_memblock_release 0.24% 0.24%
pa_memblock_acquire 0.20% 0.20%
pa_frame_aligned 0.14% 0.14%
snd_pcm_mmap_begin 0.12% 0.12%
pa_object_ref 0.12% 0.12%
pa_silence_memory 0.11% 0.11%
pa_silence_memchunk 0.07% 0.07%
clock_gettime 0.00% 0.03%
pa_run_once 0.03% 0.03%
snd_pcm_state 0.02% 0.02%
pa_sink_render_into_full 0.02% 0.02%
pa_frame_size 0.01% 0.01%
pa_fdsem_before_poll 0.01% 0.01%
pa_sample_spec_valid 0.01% 0.01%
pa_asyncmsgq_get 0.01% 0.01%
pthread_mutex_unlock 0.00% 0.00%
pa_asyncq_read_before_poll 0.00% 0.00%
pa_ratelimit_test 0.00% 0.00%
In file /usr/lib/libpulsecore-4.0.so 0.00% 0.00%
pa_object_check_type 0.00% 0.00%
pa_asyncq_pop 0.00% 0.00%
In file [heap] 0.00% 1.95%
In file /usr/lib/x86_64-linux-gnu/pulseaudio/libpulsecommon-4.0.so 0.98% 0.98%
In file /usr/lib/pulse-4.0/modules/libalsa-util.so 0.60% 0.63%
In file /usr/lib/x86_64-linux-gnu/libasound.so.2.0.0 0.62% 0.62%
pa_memblock_unref 0.60% 0.60%
In file /usr/lib/libpulsecore-4.0.so 0.17% 0.54%
ioctl 0.00% 0.37%
pa_memblock_new_fixed 0.35% 0.35%
snd_pcm_mmap_commit 0.15% 0.15%
pa_bytes_to_usec 0.14% 0.14%
pa_sink_render_into_full 0.13% 0.13%
pa_memblock_unref_fixed 0.12% 0.12%
_dl_update_slotinfo 0.11% 0.11%
pa_object_unref 0.11% 0.11%
pa_alsa_safe_avail 0.10% 0.10%
pa_flist_push 0.10% 0.10%
update_get_addr 0.10% 0.10%
pa_memblock_get_length 0.09% 0.09%
pa_once_begin 0.08% 0.08%
pa_frame_size 0.08% 0.08%
snd_pcm_format_physical_width 0.08% 0.08%
pa_flist_pop 0.08% 0.08%
pa_alsa_safe_mmap_begin 0.06% 0.06%
pa_hashmap_iterate 0.06% 0.06%
pa_sample_spec_valid 0.05% 0.05%
pa_cvolume_channels_equal_to 0.05% 0.05%
pa_thread_mq_get 0.04% 0.04%
snd_pcm_mmap_begin 0.04% 0.04%
pa_sink_render_into 0.04% 0.04%
__tls_get_addr 0.03% 0.03%
pa_bytes_to_usec_round_up 0.03% 0.03%
pa_run_once 0.02% 0.02%
pa_mempool_block_size_max 0.02% 0.02%
pa_bytes_per_second 0.02% 0.02%
pa_object_check_type 0.02% 0.02%
pa_sink_check_type 0.01% 0.01%
pa_smoother_translate 0.01% 0.01%
snd_pcm_poll_descriptors_revents 0.01% 0.01%
ppoll 0.00% 0.01%
pa_sink_get_requested_latency_within_thread 0.01% 0.01%
pa_rtpoll_item_get_pollfd 0.01% 0.01%
pa_rtpoll_run 0.01% 0.01%
pa_static_mutex_get 0.01% 0.01%
snd_pcm_avail 0.00% 0.00%
pa_rtpoll_set_timer_disabled 0.00% 0.00%
pa_sink_process_input_underruns 0.00% 0.00%
pa_alsa_safe_delay 0.00% 0.00%
pa_asyncq_pop 0.00% 0.00%
snd_pcm_status 0.00% 0.00%
pa_msgobject_check_type 0.00% 0.00%
pa_log_ratelimit 0.00% 0.00%
if using "sudo pm-suspend" will occur same problem this is the backtrace when the pulseaudio stock.
#0 0x00007f546ac8203f in __GI_ppoll (fds=<optimized out>, nfds=<optimized out>, timeout=<optimized out>,
sigmask=sigmask@entry=0x0) at ../sysdeps/unix/sysv/linux/ppoll.c:57
#1 0x00007f546befe8e7 in ppoll (__ss=0x0, __timeout=<optimized out>, __nfds=<optimized out>, __fds=<optimized out>)
at /usr/include/x86_64-linux-gnu/bits/poll2.h:77
#2 pa_mainloop_poll (m=m@entry=0x98f2d0) at pulse/mainloop.c:858
#3 0x00007f546befeede in pa_mainloop_iterate (m=m@entry=0x98f2d0, block=block@entry=1, retval=retval@entry=0x7ffffc21ff58)
at pulse/mainloop.c:932
#4 0x00007f546befef90 in pa_mainloop_run (m=0x98f2d0, retval=0x7ffffc21ff58) at pulse/mainloop.c:950
#5 0x0000000000406e82 in main ()
> pulseaudio with alsa dmix and dsnoop. >load-module module-alsa-sink device=dmixer >load-module module-alsa-source device=dsnooper why do you use dmix and dsnoop ? #load-module module-udev-detect tsched=0 do pulseaudio really work when you disable udev-detect ? you have to provide pulseaudio verbose log my system has install pulseaudio and alsa together, and i could let it work together dmix and dsnoop just a settings let the pulseaudio could work like a alsa plugin without udev detect, the pulse audio just do not auto detect the device. but i could manual setting the pulseaudio to let it work. i change pulseaudio like a alsa plugin. to let the alsa and pulse work together. this settings i have using for a long time. but the stock problem is recently. there is a problem. the stock problem is happened when wake up. i do not know that if i could provide the verbose log the alsa work with pulseaudio i found it on the internet. so i do not know that it is the original pulseaudio could do or it just a trick Created attachment 91205 [details]
output that generate by pulseaudio -vvv
it is strange that pulseaudio scheduling delay of 5.247 seconds ? I: [alsa-sink-ALC272 Analog] alsa-sink.c: Scheduling delay of 5247.71ms > 7.06ms, you might want to investigate this to improve latency... I: [alsa-source-ALC272 Analog] alsa-source.c: Scheduling delay of 5246.44ms, you might want to investigate this to improve latency... D: [alsa-sink-ALC272 Analog] alsa-util.c: Got POLLERR from ALSA D: [alsa-sink-ALC272 Analog] alsa-util.c: Got POLLOUT from ALSA D: [alsa-sink-ALC272 Analog] alsa-util.c: PCM state is SUSPENDED I: [alsa-sink-ALC272 Analog] (alsa-lib)pcm_hw.c: SNDRV_PCM_IOCTL_RESUME failed (-38) D: [alsa-source-ALC272 Analog] alsa-util.c: Got POLLERR from ALSA D: [alsa-source-ALC272 Analog] alsa-util.c: Got POLLIN from ALSA D: [alsa-source-ALC272 Analog] alsa-util.c: PCM state is SUSPENDED I: [alsa-source-ALC272 Analog] (alsa-lib)pcm_hw.c: SNDRV_PCM_IOCTL_RESUME failed (-38) I: [alsa-sink-ALC272 Analog] alsa-sink.c: Starting playback. D: [alsa-sink-ALC272 Analog] alsa-sink.c: Cutting sleep time for the initial iterations by half. I: [alsa-source-ALC272 Analog] alsa-source.c: Starting capture. I: [alsa-sink-ALC272 Analog] alsa-sink.c: Scheduling delay of 5386.48ms > 7.06ms, you might want to investigate this to improve latency... D: [alsa-sink-ALC272 Analog] alsa-util.c: Got POLLERR from ALSA D: [alsa-sink-ALC272 Analog] alsa-util.c: Got POLLOUT from ALSA D: [alsa-sink-ALC272 Analog] alsa-util.c: PCM state is SUSPENDED I: [alsa-sink-ALC272 Analog] (alsa-lib)pcm_hw.c: SNDRV_PCM_IOCTL_RESUME failed (-38) I: [alsa-sink-ALC272 Analog] alsa-sink.c: Starting playback. you have to use --log-time D: [pulseaudio] alsa-util.c: appl_ptr : 0 D: [pulseaudio] alsa-util.c: hw_ptr : 0 D: [alsa-source-ALC272 Analog] alsa-source.c: Thread starting up D: [alsa-source-ALC272 Analog] core-util.c: SCHED_RR worked. I: [alsa-source-ALC272 Analog] core-util.c: Successfully enabled SCHED_RR scheduling for thread, with priority 5. I: [alsa-source-ALC272 Analog] alsa-source.c: Starting capture. D: [pulseaudio] main.c: Got org.PulseAudio1! D: [pulseaudio] main.c: Got org.pulseaudio.Server! I: [pulseaudio] main.c: 幕後程å¼å•Ÿå‹•å®Œæˆã€‚ D: [pulseaudio] bluetooth-util.c: dbus: interface=org.freedesktop.DBus, path=/org/freedesktop/DBus, member=NameAcquired E: [alsa-source-ALC272 Analog] alsa-util.c: snd_pcm_avail_delay() returned strange values: delay 0 is less than avail 8. E: [alsa-source-ALC272 Analog] alsa-util.c: Most likely this is a bug in the ALSA driver 'snd_hda_intel'. Please report this issue to the ALSA developers. E: [alsa-source-ALC272 Analog] alsa-util.c: snd_pcm_dump(): E: [alsa-source-ALC272 Analog] alsa-util.c: Direct Snoop PCM E: [alsa-source-ALC272 Analog] alsa-util.c: Its setup is: E: [alsa-source-ALC272 Analog] alsa-util.c: stream : CAPTURE E: [alsa-source-ALC272 Analog] alsa-util.c: access : MMAP_INTERLEAVED E: [alsa-source-ALC272 Analog] alsa-util.c: format : S32_LE E: [alsa-source-ALC272 Analog] alsa-util.c: subformat : STD E: [alsa-source-ALC272 Analog] alsa-util.c: channels : 2 E: [alsa-source-ALC272 Analog] alsa-util.c: rate : 48000 E: [alsa-source-ALC272 Analog] alsa-util.c: exact rate : 48000 (48000/1) E: [alsa-source-ALC272 Analog] alsa-util.c: msbits : 32 E: [alsa-source-ALC272 Analog] alsa-util.c: buffer_size : 8192 E: [alsa-source-ALC272 Analog] alsa-util.c: period_size : 1024 E: [alsa-source-ALC272 Analog] alsa-util.c: period_time : 21333 E: [alsa-source-ALC272 Analog] alsa-util.c: tstamp_mode : ENABLE E: [alsa-source-ALC272 Analog] alsa-util.c: period_step : 1 E: [alsa-source-ALC272 Analog] alsa-util.c: avail_min : 7310 E: [alsa-source-ALC272 Analog] alsa-util.c: period_event : 0 E: [alsa-source-ALC272 Analog] alsa-util.c: start_threshold : -1 E: [alsa-source-ALC272 Analog] alsa-util.c: stop_threshold : 4611686018427387904 E: [alsa-source-ALC272 Analog] alsa-util.c: silence_threshold: 0 E: [alsa-source-ALC272 Analog] alsa-util.c: silence_size : 0 E: [alsa-source-ALC272 Analog] alsa-util.c: boundary : 4611686018427387904 E: [alsa-source-ALC272 Analog] alsa-util.c: Hardware PCM card 0 'HDA Intel MID' device 0 subdevice 0 E: [alsa-source-ALC272 Analog] alsa-util.c: Its setup is: E: [alsa-source-ALC272 Analog] alsa-util.c: stream : CAPTURE E: [alsa-source-ALC272 Analog] alsa-util.c: access : MMAP_INTERLEAVED E: [alsa-source-ALC272 Analog] alsa-util.c: format : S32_LE E: [alsa-source-ALC272 Analog] alsa-util.c: subformat : STD E: [alsa-source-ALC272 Analog] alsa-util.c: channels : 2 E: [alsa-source-ALC272 Analog] alsa-util.c: rate : 48000 E: [alsa-source-ALC272 Analog] alsa-util.c: exact rate : 48000 (48000/1) E: [alsa-source-ALC272 Analog] alsa-util.c: msbits : 32 E: [alsa-source-ALC272 Analog] alsa-util.c: buffer_size : 8192 E: [alsa-source-ALC272 Analog] alsa-util.c: period_size : 1024 E: [alsa-source-ALC272 Analog] alsa-util.c: period_time : 21333 E: [alsa-source-ALC272 Analog] alsa-util.c: tstamp_mode : ENABLE E: [alsa-source-ALC272 Analog] alsa-util.c: period_step : 1 E: [alsa-source-ALC272 Analog] alsa-util.c: avail_min : 1024 E: [alsa-source-ALC272 Analog] alsa-util.c: period_event : 0 E: [alsa-source-ALC272 Analog] alsa-util.c: start_threshold : 1 E: [alsa-source-ALC272 Analog] alsa-util.c: stop_threshold : 4611686018427387904 E: [alsa-source-ALC272 Analog] alsa-util.c: silence_threshold: 0 E: [alsa-source-ALC272 Analog] alsa-util.c: silence_size : 0 E: [alsa-source-ALC272 Analog] alsa-util.c: boundary : 4611686018427387904 E: [alsa-source-ALC272 Analog] alsa-util.c: appl_ptr : 0 E: [alsa-source-ALC272 Analog] alsa-util.c: hw_ptr : 16424 the sound card alreay capture (hw_ptr is at 16424) but pulseaudio still not read anything (appl_ptr = 0 ) I: [alsa-sink-ALC272 Analog] alsa-sink.c: Scheduling delay of 5247.71ms > 7.06ms, you might want to investigate this to improve latency... I: [alsa-source-ALC272 Analog] alsa-source.c: Scheduling delay of 5246.44ms, you might want to investigate this to improve latency... Created attachment 91668 [details]
detail of log
do supsend and resume work as expected when you aplay -Dplughw:0,0 any.wav aplay -Dplug:dmix:0 any.wav Created attachment 91736 [details]
log with pulseaudio, pacmd, aplay
-- GitLab Migration Automatic Message -- This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity. You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/pulseaudio/pulseaudio/issues/330. |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.