The latency change that is caused by the underrun can apparently cause a rewind request at a bad time. It seems a bit odd that this crash isn't reported much more often, since it looks like the underrun handling is clearly broken, and underruns are a quite common occurrence. Maybe the rewind request happens only in some rare cases (the code path isn't simple, so it's not easy to figure out).

Thanks for reporting this. I won't promise a quick fix, but this seems like a clear case that can be fixed once someone finds the time to figure out the correct way to handle the underrun (i.e. avoid any rewind requests in the process, or in some other way make sure that there are no rewind requests pending when pa_sink_render_into_full() is called).

Created attachment 114838 [details]
another log

Here is another log with the same assertion.. I suspect that with PA 4.0, it instead resulted in bad playback.

alsa-sink] alsa-sink.c: Latency set to 1.00ms
D: [alsa-sink] alsa-sink.c: hwbuf_unused=383808
D: [alsa-sink] alsa-sink.c: setting avail_min=95977


https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/commit/sound/usb?id=a7d9c0990d5503775784fef7ff44d74d7e3294fd


how do pulseaudio latency related to period size for those usb audio devices ?

can aplay use 1 ms buffer time with hw device of your usb audio devices?


aplay -v --buffer-time=1000 -D hw:0,0 stereo.wav

(In reply to Olivier Crête from comment #2)
> Created attachment 114838 [details]
> another log
> 
> Here is another log with the same assertion.. I suspect that with PA 4.0, it
> instead resulted in bad playback.

any reason to use s32le , seem your combined sink not support 32 bits

http://www.realtek.com.tw/products/productsView.aspx?Langid=1&PFid=28&Level=5&Conn=4&ProdID=284

Primary 16/20/24-bit SPDIF-OUT supports 32k/44.1k/48k/88.2k/96k/192kHz sample rate 
Secondary 16/20/24-bit SPDIF-OUT supports 32k/44.1k/48k/88.2k/96k/192kHz sample rate

do your motherboard have internal spdif out connector and spdif out at rear panel 


post the output of alsa-info.sh 


which sinks are used by the combined sink ?

The combined sink is  HDMI + SPDIF. The board is a Asrock H61TM-ITX, I don't have it here, but the spec page on the asrock website seems to say that there is a single spdif connector, but the connections in this machine are a bit strange, so I can double check the physical connections tomorrow.

I also have no idea why the s32le, it's generated by application code I don't have the source to right now.

http://www.alsa-project.org/db/?f=13394becf97124311d9febe374f62195a4207756

the stream id of two audio output are different (7 and 6) , it is multistreaming of two digital devices
but only one converter support 32 bits

http://www.intel.com/support/motherboards/desktop/sb/CS-034206.htm


The ALC892 provides ten DAC channels that simultaneously support 7.1 channel sound playback, plus 2 channels of independent stereo sound output (multiple streaming) through the front panel stereo outputs.




you can enable by hint using early patching to get device 2 ALC892  alt analog playback for the headphone


 indep_hp (bool): provide the independent headphone PCM stream and
  the corresponding mixer control, if available

https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/tree/Documentation/sound/alsa/HD-Audio.txt



APLAY

**** List of PLAYBACK Hardware Devices ****
card 0: PCH [HDA Intel PCH], device 0: ALC892 Analog [ALC892 Analog]
  Subdevices: 0/1
  Subdevice #0: subdevice #0
card 0: PCH [HDA Intel PCH], device 1: ALC892 Digital [ALC892 Digital]
  Subdevices: 0/1
  Subdevice #0: subdevice #0
card 0: PCH [HDA Intel PCH], device 3: HDMI 0 [HDMI 0]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 0: PCH [HDA Intel PCH], device 7: HDMI 1 [HDMI 1]
  Subdevices: 0/1
  Subdevice #0: subdevice #0
card 0: PCH [HDA Intel PCH], device 8: HDMI 2 [HDMI 2]
  Subdevices: 1/1
  Subdevice #0: subdevice #0



most likely you need to use the format, rate and channel supported by both devices



Node 0x06 [Audio Output] wcaps 0x611: Stereo Digital
  Control: name="IEC958 Playback Con Mask", index=16, device=0
  Control: name="IEC958 Playback Pro Mask", index=16, device=0
  Control: name="IEC958 Playback Default", index=16, device=0
  Control: name="IEC958 Playback Switch", index=16, device=0
  Control: name="IEC958 Default PCM Playback Switch", index=0, device=0
  Device: name="ALC892 Digital", type="SPDIF", device=1
  Converter: stream=7, channel=0
  Digital: Enabled
  Digital category: 0x0
  IEC Coding Type: 0x0
  PCM:
    rates [0x5f0]: 32000 44100 48000 88200 96000 192000
    bits [0xe]: 16 20 24
    formats [0x1]: PCM
  Power states:  D0 D1 D2 D3 EPSS
  Power: setting=D0, actual=D0



 0x03 [Audio Output] wcaps 0x6611: 8-Channels Digital
  Device: name="HDMI 1", type="HDMI", device=7
  Converter: stream=6, channel=0
  Digital: Enabled
  Digital category: 0x0
  IEC Coding Type: 0x0
  PCM:
    rates [0x7f0]: 32000 44100 48000 88200 96000 176400 192000
    bits [0x1e]: 16 20 24 32
    formats [0x5]: PCM AC3
  Power states:  D0 D3 EPSS
  Power: setting=D0, actual=D0

Simple mixer control 'IEC958 Default PCM',0
  Capabilities: pswitch pswitch-joined penum
  Playback channels: Mono
  Mono: Playback [on]

https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/commit/sound/pci/hda?id=9a08160bdbe3148a405f72798f76e2a5d30bd243


Added a new mixer switch to enable/disable the sharing of the default PCM stream with analog and SPDIF outputs.  When "IEC958 Default PCM" switch is on, the PCM stream is routed both to analog and SPDIF outputs.
This is the behavior in the earlier version.

Turning this switch off has a merit for some codecs, though.  Some codec chips don't support 24bit formats for SPDIF but only for analog outputs.
In this case, you can use 24bit format by disabling this switch.



audio output node 0x02 and node 0x06 should use same stream id 



 Node 0x02 [Audio Output] wcaps 0x41d: Stereo Amp-Out
  Control: name="Front Playback Volume", index=0, device=0
    ControlAmp: chs=3, dir=Out, idx=0, ofs=0
  Device: name="ALC892 Analog", type="Audio", device=0
  Amp-Out caps: ofs=0x40, nsteps=0x40, stepsize=0x03, mute=0
  Amp-Out vals:  [0x40 0x40]
  Converter: stream=8, channel=0
  PCM:
    rates [0x560]: 44100 48000 96000 192000
    bits [0xe]: 16 20 24
    formats [0x1]: PCM
  Power states:  D0 D1 D2 D3 EPSS
  Power: setting=D0, actual=D0

seem bug in driver , since audio output 0x02 and nodev0x06 should use same stream id when  'IEC958 Default PCM' is on and device 0 is playing

digital device 1 cannot be open

seem driver don't report error when there are five line out jack at ext rear

bug in parser , hp at node 0x1b is also missng

are you using 17 pins analog surround header ?



6.791557] sound hdaudioC0D0: autoconfig: line_outs=4 (0x1a/0x15/0x16/0x17/0x0) type:line
[    6.791563] sound hdaudioC0D0:    speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[    6.791565] sound hdaudioC0D0:    hp_outs=0 (0x0/0x0/0x0/0x0/0x0)
[    6.791567] sound hdaudioC0D0:    mono: mono_out=0x0
[    6.791569] sound hdaudioC0D0:    dig-out=0x1e/0x0
[    6.791571] sound hdaudioC0D0:    inputs:




Node 0x1a [Pin Complex] wcaps 0x40058f: Stereo Amp-In Amp-Out
  Control: name="Front Playback Switch", index=0, device=0
    ControlAmp: chs=3, dir=Out, idx=0, ofs=0
  Control: name="Line Out Front Jack", index=0, device=0
  Amp-In caps: ofs=0x00, nsteps=0x03, stepsize=0x27, mute=0
  Amp-In vals:  [0x00 0x00]
  Amp-Out caps: ofs=0x00, nsteps=0x00, stepsize=0x00, mute=1
  Amp-Out vals:  [0x00 0x00]
  Pincap 0x00003736: IN OUT Detect Trigger
    Vref caps: HIZ 50 GRD 80 100
  Pin Default 0x01013020: [Jack] Line Out at Ext Rear
    Conn = 1/8, Color = Blue
    DefAssociation = 0x2, Sequence = 0x0
  Pin-ctls: 0x40: OUT VREF_HIZ
  Unsolicited: tag=01, enabled=1
  Power states:  D0 D1 D2 D3 EPSS
  Power: setting=D0, actual=D0
  Connection: 5
     0x0c* 0x0d 0x0e 0x0f 0x26



Node 0x1b [Pin Complex] wcaps 0x40058f: Stereo Amp-In Amp-Out
  Amp-In caps: ofs=0x00, nsteps=0x03, stepsize=0x27, mute=0
  Amp-In vals:  [0x00 0x00]
  Amp-Out caps: ofs=0x00, nsteps=0x00, stepsize=0x00, mute=1
  Amp-Out vals:  [0x80 0x80]
  Pincap 0x0001373e: IN OUT HP EAPD Detect Trigger
    Vref caps: HIZ 50 GRD 80 100
  EAPD 0x2: EAPD
  Pin Default 0x0221411f: [Jack] HP Out at Ext Front
    Conn = 1/8, Color = Green
    DefAssociation = 0x1, Sequence = 0xf
    Misc = NO_PRESENCE
  Pin-ctls: 0x20: IN VREF_HIZ
  Unsolicited: tag=00, enabled=0
  Power states:  D0 D1 D2 D3 EPSS
  Power: setting=D0, actual=D0
  Connection: 5
     0x0c* 0x0d 0x0e 0x0f 0x26

sys/class/sound/hwC0D0/init_pin_configs:
0x11 0x40028008
0x12 0x90a60150
0x14 0x01014010
0x15 0x01011012
0x16 0x01016011
0x17 0x01012014
0x18 0x01a19040
0x19 0x02a19141
0x1a 0x01013020
0x1b 0x0221411f
0x1c 0x411111f0
0x1d 0x4037e629
0x1e 0x90460130
0x1f 0x411111f0

/sys/class/sound/hwC0D0/driver_pin_configs:

/sys/class/sound/hwC0D0/user_pin_configs:
0x11 0x40028008
0x12 0x40f000f0
0x14 0x40f000f0
0x15 0x01011012
0x16 0x01016011
0x17 0x01014413
0x18 0x40f000f0
0x19 0x40f000f0
0x1a 0x01014410
0x1b 0x40f000f0
0x1c 0x411111f0
0x1d 0x4037e629
0x1e 0x014b1180
0x1f 0x411111f0

seem you override pin defaults

but there are no internal speaker connector, hdmi spdif

what are the setting in asrock uefi setup ?


3.4.3  South Bridge Coniguration 

Onboard HD Audio Select [Auto], [Enabled] or [Disabled] for the onboard HD Audio. If you select [Auto], the onboard HD Audio will be disabled when a Sound Card is plugged. 

Front Panel Select [Auto] or [Disabled] for the onboard HD Audio Front Panel. 

External Panel Select [Auto] or [Disabled] for the External Audio Panel.

 Onboard HDMI HD Audio This allows you to enable or disable the Onboard HDMI HD Audio.

  6.713730] snd_hda_intel 0000:00:1b.0: Applying patch firmware 'hda-mmsos.fw'
[    6.722501] snd_hda_intel 0000:00:1b.0: firmware: direct-loading firmware hda-mmsos.fw


any reason to use hda-mmsos.fw ?

seem it try to remove all input pins 

and  disable audio mixer nid using hint

mixer_nid (int): specifies the widget NID of the analog-loopback mixer



but driver still create capture devices without any input pins

ARECORD

**** List of CAPTURE Hardware Devices ****
card 0: PCH [HDA Intel PCH], device 0: ALC892 Analog [ALC892 Analog]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 0: PCH [HDA Intel PCH], device 2: ALC892 Alt Analog [ALC892 Alt Analog]
  Subdevices: 1/1
  Subdevice #0: subdevice #0

https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/commit/sound/pci/hda/hda_auto_parser.c?id=9b7564a64999597844513604df4a206fa4da3b69


the driver already inform the fifth line out was ignored because of different def assosciation with the other four line out pin of the analog surround audio header of you thin mini itx motherboard


you have to file bug report at kernel bugzilla

   to debug this issue, you need to enable DEBUG_TIMING to check the debugged info or call pa_alsa_dump after pulseaudio print "Unerrun!" and check whether the appl_ptr is behind hw_ptr


   /* We got a dropout. What a mess! */
        left_to_play = 0;
        underrun = TRUE;

#if 0
        PA_DEBUG_TRAP;
#endif

        if (!u->first && !u->after_rewind)
-          if (pa_log_ratelimit(PA_LOG_INFO))
+          if (pa_log_ratelimit(PA_LOG_INFO)) {
                pa_log_info("Underrun!");
+             pa_alsa_dump(PA_LOG_INFO, u->pcm_handle);
+         }
    }

#ifdef DEBUG_TIMING
    pa_log_debug("%0.2f ms left to play; inc threshold = %0.2f ms; dec threshold = %0.2f ms",
                 (double) pa_bytes_to_usec(left_to_play, &u->sink->sample_spec) / PA_USEC_PER_MSEC,
                 (double) pa_bytes_to_usec(u->watermark_inc_threshold, &u->sink->sample_spec) / PA_USEC_PER_MSEC,
                 (double) pa_bytes_to_usec(u->watermark_dec_threshold, &u->sink->sample_spec) / PA_USEC_PER_MSEC);
#endif

The internal 17 oin onnector is connected to a panel that has 4 separate connections (which we "split" using pulse). And the internal SPDIF connector is also connected.

I didn't setup the fw file, but I think it was to disable other things to prevent noise and enable those that are actually in use.

In the UEFI menu, everything is either on "auto" or "enabled.

When doing the aplay command without pulseaudio running, I get a bunch of underruns.

I'm not sure why the s32le, or the 96000 sinks, it's the application creating those, and I don't have the source code to it right now.

I'm not sure I followed everything, but I should rebuild pulse with DEBUG_TIMING and your patch and post the log?



Here is the "firmware" file:

# ASRock H61TM_ITX (Realtek ALC892)
# Vendor Id, Subsystem Id, Address
# See /proc/asound/card0/codec#0
[codec]
0x10ec0892 0x1849e892 0

# Prevent the AA-loopback from being enabled which can pull in white noise (ACI#380)
[hint]
mixer_nid = 0

# Disable everything except for:
# 0x1e = Main S/PDIF
# 0x1a = Output A (Front)
# 0x16 = Output B (Center/LFE)
# 0x15 = Output C (Rear)
# 0x17 = Output D (Side)
[pincfg]
0x11 0x40028008
0x12 0x40f000f0
0x14 0x40f000f0
0x15 0x01011012
0x16 0x01016011
0x17 0x01014413
0x18 0x40f000f0
0x19 0x40f000f0
0x1a 0x01014410
0x1b 0x40f000f0
0x1c 0x411111f0
0x1d 0x4037e629
0x1e 0x014b1180



And the matching pulse profile:
[General]
auto-profiles = no

[Mapping analog-surround-71]
device-strings = surround71:%f
channel-map = front-left,front-right,rear-left,rear-right,front-center,lfe,side-left,side-right
description = Analog Surround 7.1
paths-output = analog-output
priority = 100
direction = output

[Mapping analog-stereo]
description = Analog Stereo
device-strings = front:%f hw:%f
channel-map = left,right
paths-output = analog-output
priority = 90
direction = output

[Mapping digital-stereo]
device-strings = iec958:%f
channel-map = left,right
description = Digital Stereo
paths-output = iec958-stereo-output
priority = 50
direction = output

[Mapping hdmi-stereo-7]
device-strings = hw:%f,7
channel-map = left,right
description = Digital Stereo (HDMI)
paths-output = hdmi-output-1
priority = 40
direction = output

[Profile output:analog-surround-71+output:digital-stereo+output:hdmi-stereo-7]
description = ASRock
output-mappings = analog-surround-71 digital-stereo hdmi-stereo-7
priority = 100

why do the firmware remove the pink mic jack and green jack ?



do you mean the following  jack state are the status of jacks at the external panel connected to 17 pins analog surround audio header ?

four line out jacks were plugged 

control.18 {
		iface CARD
		name 'Line Out Front Jack'
		value true
		comment {
			access read
			type BOOLEAN
			count 1
		}
	}
	control.19 {
		iface CARD
		name 'Line Out Surround Jack'
		value true
		comment {
			access read
			type BOOLEAN
			count 1
		}
	}
	control.20 {
		iface CARD
		name 'Line Out CLFE Jack'
		value true
		comment {
			access read
			type BOOLEAN
			count 1
		}
	}
	control.21 {
		iface CARD
		name 'Line Out Side Jack'
		value true
		comment {
			access read
			type BOOLEAN
			count 1
		}
	}

your alsa info indicate that the three devices are open/playing with different streams


this may mean that you may need to check whether the three streams does not exceed the maixmum  bandwidth as you are using 96000Hz and s32le 

do this error also happen when you change the dedault sample rate to 48000Hz and s16le ?


APLAY

**** List of PLAYBACK Hardware Devices ****
card 0: PCH [HDA Intel PCH], device 0: ALC892 Analog [ALC892 Analog]
  Subdevices: 0/1
  Subdevice #0: subdevice #0
card 0: PCH [HDA Intel PCH], device 1: ALC892 Digital [ALC892 Digital]
  Subdevices: 0/1
  Subdevice #0: subdevice #0
card 0: PCH [HDA Intel PCH], device 7: HDMI 1 [HDMI 1]
  Subdevices: 0/1
  Subdevice #0: subdevice #0

The only "visible" jacks on the machine are the 4 stereo outputs connected from the 17-pin, the sdpif and the hdmi out.

The analog part is opened at 48khz s16le if I read this correctly? Then software remapped to this stupid 96khz s32le ?

pactl list short sinks
0	alsa_output.pci-0000_00_1b.0.analog-surround-71	module-alsa-card.c	s16le 8ch 48000Hz	RUNNING
1	alsa_output.pci-0000_00_1b.0.digital-stereo	module-alsa-card.c	s16le 2ch 44100Hz	SUSPENDED
2	alsa_output.pci-0000_00_1b.0.hdmi-stereo-7	module-alsa-card.c	s16le 2ch 44100Hz	SUSPENDED
3	null	module-null-sink.c	s16le 2ch 44100Hz	IDLE
4	digital-96000_s32le._.alsa_output.pci-0000_00_1b.0.digital-stereo	module-alsa-sink.c	s32le 2ch 96000Hz	RUNNING
5	hdmi-96000_s32le._.alsa_output.pci-0000_00_1b.0.digital-stereo	module-alsa-sink.c	s32le 2ch 48000Hz	RUNNING
6	96000_s32le._.alsa_output.pci-0000_00_1b.0.digital-stereo	module-combine-sink.c	s32le 2ch 96000Hz	IDLE
7	96000_s32le._.analog-stereo-a	module-remap-sink.c	s32le 2ch 96000Hz	RUNNING
8	96000_s32le._.analog-stereo-b	module-remap-sink.c	s32le 2ch 96000Hz	RUNNING
9	96000_s32le._.analog-stereo-c	module-remap-sink.c	s32le 2ch 96000Hz	RUNNING
10	96000_s32le._.analog-stereo-d	module-remap-sink.c	s32le 2ch 96000Hz	RUNNING

pactl list source


do any available port in alc892 capture source ?

it is bug of alsa driver to create analog capture device and alt analog capture device when there is no input pin complex (e.g. internal mic) ?


ARECORD

**** List of CAPTURE Hardware Devices ****
card 0: PCH [HDA Intel PCH], device 0: ALC892 Analog [ALC892 Analog]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 0: PCH [HDA Intel PCH], device 2: ALC892 Alt Analog [ALC892 Alt Analog]
  Subdevices: 1/1
  Subdevice #0: subdevice #0




do your aio have internal mic , digital mic  , webcam mic ?

Fixing the rate and format to what alsa_output.pci-0000_00_1b.0.analog-surround-71 is setup for does not change anything.

(In reply to Nicolas Dufresne from comment #21)
> Fixing the rate and format to what
> alsa_output.pci-0000_00_1b.0.analog-surround-71 is setup for does not change
> anything.

you have to provide stacktrace , pulseaudio verbose log and output of alsa-info.sh

-- 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/207.