Feature: ESP32-S3

[Dotsch2005]

Is it possible to run the snapclient on an ESP32-S3?

I have changed the sdkconfig target with idf.py set-target esp32s3 and changed the confiuration via idf.py menuconfig.

I am stuck at error: ‘I2S_CLK_SRC_APLL’ undeclared.

I have read so that APLL is not available on the ESP32-S3.

Is there a way to get the ESP32-S3 up and running?

[CarlosDerSeher]

Please try the branch fix_20ms_chunk_size_bug. Latest bug fixes and updates ar in there. They wilI be merged to master soon but there still has to be some testing.

I don't know about S3 but there is a template sdkconfig for S2 which also doesn't have APLL. Maybe you can work your way from there

[Dotsch2005]

Thanks for the feedback.

This has already helped me, but it gets stuck in the idf.py monitor and resets itself.

I think it's somewhere along the line:

--- 0x40380c48: _xt_context_save at <Path>/esp-idf-v5.1.1/components/freertos/FreeRTOS-Kernel/portable/xtensa/xtensa_context.S:217

ESP-ROM:esp32s3-20210327
Build:Mar 27 2021
rst:0x8 (TG1WDT_SYS_RST),boot:0x29 (SPI_FAST_FLASH_BOOT)
Saved PC:0x40380c48
--- 0x40380c48: _xt_context_save at <Path>/esp-idf-v5.1.1/components/freertos/FreeRTOS-Kernel/portable/xtensa/xtensa_context.S:217

SPIWP:0xee
mode:DIO, clock div:1
load:0x3fce3818,len:0x186c
load:0x403c9700,len:0x4
load:0x403c9704,len:0xd50
load:0x403cc700,len:0x30f8
entry 0x403c9918

Do you have any advice?

Thank you very much!

idf_monitor.log

[CarlosDerSeher]

Not sure if this could be a problem here, but in this branch there has been an upgrade to IDF v5.1.5

Also it dies at HAL init, which codec are you using?

[Dotsch2005]

Thank you for your support.

I have updated to version v5.1.5 and started fresh again. Unfortunately, nothing has changed. I have also increased the debug log, but after the HAL: INIT I see nothing.

I assume that the codec means the audio board. I am using the TAS5805M here. (ESP32-S3 Louder)

[Dotsch2005]

I'm sorry, I feel stupid. The GPIO ports have changed between the ESP32 version and the ESP32-S3.

However, it now aborts after the following:

V (1556) partition: Partition table MD5 verified
I (1565) SC: Start codec chip
I (1565) HAL: INIT
W (1568) TAS5805M: Power down pin: 33
W (1592) TAS5805M: Setting to HI Z
W (1602) TAS5805M: Setting to PLAY
I (1602) HAL: codec_hal done
I (1602) AUDIO_BOARD: board-handle done
I (1603) SC: Audio board_init done
I (1607) AUDIO_HAL: Codec mode is 2, Ctrl:1
W (1612) PLAYER: no sync task created?
W (1616) PLAYER: no pcm chunk queue created?
W (1621) PLAYER: no latency buffer semaphore created?
I (1627) PLAYER: deinit player done
ESP-ROM:esp32s3-20210327
Build:Mar 27 2021
rst:0x8 (TG1WDT_SYS_RST),boot:0x29 (SPI_FAST_FLASH_BOOT)
Saved PC:0x403759a1
SPIWP:0xee
mode:DIO, clock div:1
load:0x3fce3818,len:0x18dc
load:0x403c9700,len:0x4
load:0x403c9704,len:0xd50
load:0x403cc700,len:0x3068
entry 0x403c9918
I (26) boot: ESP-IDF v5.1.5 2nd stage bootloader

[CarlosDerSeher]

Is s3 a single core cpu? If so you might need to adjust the assigned core IDs in the task creation functions?

[Dotsch2005]

The Espressif documentation specifies the following processor Xtensa® dual-core 32-bit LX7 microprocessor (with single precision FPU).

The TAS5805M power down pin 33 (should be 17) is also wrong, but can I find the configuration for it?

[CarlosDerSeher]

Search for this line, no sync task created I am pretty sure your problem is related to failed task creation. Player task is hard-coded to core 1. If you change all tasks assigned to core 1 to "no affinity" you are hopefully good to go?!

Edit: Ok, saw your post about cpu architecture just now. So there is a second core indeed...

[CarlosDerSeher]

You can change the pinout from menuconfig audio hal

[CarlosDerSeher]

normal boot log should look like this, so you are good until it fails...

I (375) SC: Start codec chip
I (376) HAL: INIT
I (376) ADAU1961: Power ON CODEC
I (376) ADAU1961: Looking for a adau1961 chip at address 0x70
I (381) ADAU1961: Found a adau1961 chip at address 0x70
I (391) HAL: codec_hal done
I (391) AUDIO_BOARD: board-handle done
I (395) SC: Audio board_init done
I (399) AUDIO_HAL: Codec mode is 2, Ctrl:1
W (405) PLAYER: no sync task created?
W (408) PLAYER: no pcm chunk queue created?
W (413) PLAYER: no latency buffer semaphore created?
I (418) PLAYER: deinit player done
I (423) PLAYER: player_setup_i2s: dma_buf_len is 100, dma_buf_count is 22, sample rate: 44100, bits: 16
I (433) PLAYER: enable initial sync timer
I (437) PLAYER: Start player_task
I (441) PLAYER: init player done
I (441) PLAYER: started sync task

seems like it fails somewhere in player_setup_i2s

[luar123]

Just tried it here:

• I think no sync task created is normal at startup because init_player() calls deinit_player() first.
• Check your port config. I got similar errors initially when some i2s pins where not valid.
• I a coming further, but then it crashes with a stack overflow:

 (9049) NETF: Setup mdns
I (9052) dspProc: dsp_processor_init: init done
I (9056) OTA: idf.py build ; curl snapclient.local:8032 --data-binary @- < build/snapclient.bin
I (9066) SC: try connecting to static configuration 192.168.x.x:1704
I (9073) main_task: Returned from app_main()
I (9079) SC: netconn connected
I (9083) SC: netconn sent hello message

***ERROR*** A stack overflow in task IDLE1 has been detected.


Backtrace: 0x4037590a:0x3fca1f40 0x4037c10d:0x3fca1f60 0x4037c8a2:0x3fca1f80 0x4037d101:0x3fca2000 0x4037c1c4:0x3fca2020 0x4037c1ba:0x00060023 |<-CORRUPTED
--- 0x4037590a: panic_abort at esp-idf-v5.1.5/components/esp_system/panic.c:466
0x4037c10d: esp_system_abort at esp-idf-v5.1.5/components/esp_system/port/esp_system_chip.c:84
0x4037c8a2: vApplicationStackOverflowHook at esp-idf-v5.1.5/components/freertos/FreeRTOS-Kernel/portable/xtensa/port.c:582
0x4037d101: vTaskSwitchContext at esp-idf-v5.1.5/components/freertos/FreeRTOS-Kernel/tasks.c:3841
0x4037c1c4: _frxt_dispatch at esp-idf-v5.1.5/components/freertos/FreeRTOS-Kernel/portable/xtensa/portasm.S:450
0x4037c1ba: _frxt_int_exit at esp-idf-v5.1.5/components/freertos/FreeRTOS-Kernel/portable/xtensa/portasm.S:245

[CarlosDerSeher]

@luar123 you are right. I'll change these warnings to verbose logs in a future commit, as they are somewhat missleading. Changed it locally on my dev machine for now.

Did you try increasing CONFIG_FREERTOS_IDLE_TASK_STACKSIZE ?

[luar123]

It is working now, had to increase the idle stack size. Did not connect a dac, but logs look ok.
@Dotsch2005: Try this sdkconfig
You need to change wifi/server/audio board settings.

@CarlosDerSeher: Might be a good idea to include sdkconfig.default files instead of full sdkconfigs. I created one based on my working esp32 setting with idf.py save-defconfig and then switched to esp32s3 with set-target.

[CarlosDerSeher]

@luar123 oh didn't know about this. What exactly does this idf.py save-defconfig do?

[luar123]

It creates a sdkconfig.defaults file that includes all non-default values from your sdkconfig file. It is never changed by idf but is applied when no sdkconfig file is present, like when running set-target.
See https://docs.espressif.com/projects/esp-idf/en/stable/esp32/api-reference/kconfig.html#using-sdkconfig-defaults

[CarlosDerSeher]

This sounds like something for the future commit list too. Thanks

[luar123]

Seems we could even split it in multiple files, one for base configs (maybe target dependent) and one for the different audio board settings: https://docs.espressif.com/projects/esp-idf/en/stable/esp32/api-guides/build-system.html#custom-sdkconfig-defaults

[CarlosDerSeher]

This would have been my next question :) as the IDLE task stack size was never a problem on ESP32

[CarlosDerSeher]

So a file sdkconfig.defaults.esp32s3 should look like this

CONFIG_FREERTOS_IDLE_TASK_STACKSIZE=1024

[luar123]

I have tried only the default size 1536.

[CarlosDerSeher]

So I decreased this myself at some point while optimizing RAM usage? Totally forgot about this

[Dotsch2005]

Thank you very much CarlosDerSeher and luar123

I have now tried a few things and was able to solve the first problem. After help from anabolyc I was able to get the board to run.

The problem here was with the fix_20ms_chunk_size_bug and the CustomBoard TAS5805M. I used the TAS5805M component from features/s3-support. With this it starts and connects to the snapcast server.

Now I can get the board to work with WiFi. However, when I start streaming, the following error occurs:

I (2661) OTA: idf.py build ; curl snapclient.local:8032 --data-binary @- < build/snapclient.bin
I (2671) SC: try connecting to static configuration 192.168.1.100:1704
I (2681) main_task: Returned from app_main()
I (2691) SC: netconn connected
I (2691) SC: netconn sent hello message
I (2861) SC: Buffer length:  1000
I (2861) SC: Latency:        0
I (2861) SC: Mute:           0
I (2861) SC: Setting volume: 72
I (2861) dspProc: Set volume to 0.720000
I (2871) dspProc: Set volume to 0.720000
I (2871) SC: pcm sampleformat: 48000:16:2
I (3071) SC: latency buffer full
I (986831) dspProc: got new setting for dspfEQBassTreble
W (986841) PLAYER: pcm chunk queue not created
W (986861) PLAYER: pcm chunk queue not created
I (986871) PLAYER: player_setup_i2s: dma_buf_len is 100, dma_buf_count is 22, sample rate: 48000, bits: 16
W (986871) PLAYER: no pcm chunk queue created?
I (986871) PLAYER: created new queue with 48
I (986881) PLAYER: snapserver config changed, buffer 1000ms, chunk 960 frames, sample rate 48000, ch 2, bits 16 mute 0 latency 0
Guru Meditation Error: Core  1 panic'ed (LoadProhibited). Exception was unhandled.

Core  1 register dump:
PC      : 0x4202a675  PS      : 0x00060830  A0      : 0x803816fc  A1      : 0x3fcb3d80
--- 0x4202a675: player_task at D:/Dev/Github/CarlosDerSeher/snapclient/components/lightsnapcast/player.c:1371

A2      : 0x00000000  A3      : 0x00000000  A4      : 0x3c111f94  A5      : 0x3fcd8460
A6      : 0x00000001  A7      : 0x3fcb3d80  A8      : 0x8202a672  A9      : 0x3fcb3d40
A10     : 0x00000000  A11     : 0x3fcb3da4  A12     : 0x3fca91e4  A13     : 0x0000000a  
A14     : 0x00000001  A15     : 0x00000f00  SAR     : 0x00000019  EXCCAUSE: 0x0000001c
EXCVADDR: 0x0000000c  LBEG    : 0x40056f5c  LEND    : 0x40056f72  LCOUNT  : 0x00000000
--- 0x40056f5c: memcpy in ROM
0x40056f72: memcpy in ROM



Backtrace: 0x4202a672:0x3fcb3d80 0x403816f9:0x3fcb3ed0
--- 0x4202a672: player_task at D:/Dev/Github/CarlosDerSeher/snapclient/components/lightsnapcast/player.c:1371
0x403816f9: vPortTaskWrapper at D:/Apps/esp/v5.1.5/esp-idf/components/freertos/FreeRTOS-Kernel/portable/xtensa/port.c:162





ELF file SHA256: 5187c5dd295e0e57

Rebooting...

The ESP32S3 has a direct JTAG driver. I have tried to debug from here with VSCode and the ESP-IDF/OpenOCD. Unfortunately I can't get a working debug environment to run.

[luar123]

Did you enable PSRAM (I think most/all S3 modules have PSRAM)? Otherwise 1000ms buffer length is too much. You would need to reduce it in the snapserver config to ~700ms (see readme).

[CarlosDerSeher]

The problem here was with the fix_20ms_chunk_size_bug and the CustomBoard TAS5805M. I used the TAS5805M component from features/s3-support. With this it starts and connects to the snapcast server.

You probably have a wrong pin out configured then? I never tried this DAC myself, but others reported it works from customBoard.

--- 0x4202a675: player_task at D:/Dev/Github/CarlosDerSeher/snapclient/components/lightsnapcast/player.c:1371

What's the code at this line?

[Dotsch2005]

Did you enable PSRAM (I think most/all S3 modules have PSRAM)? Otherwise 1000ms buffer length is too much. You would need to reduce it in the snapserver config to ~700ms (see readme).

Thanks for the hint, I have changed it. In your example, the PSRAM was not activated. I have activated the PSRAM in my current configuration (8MB - ESP32-S3-WROOM-N16R8).

The problem here was with the fix_20ms_chunk_size_bug and the CustomBoard TAS5805M. I used the TAS5805M component from features/s3-support. With this it starts and connects to the snapcast server.

You probably have a wrong pin out configured then? I never tried this DAC myself, but others reported it works from customBoard.

--- 0x4202a675: player_task at D:/Dev/Github/CarlosDerSeher/snapclient/components/lightsnapcast/player.c:1371

What's the code at this line?

In the current branch fix_20ms_chunk_size_bug it is line 1351 (old 1371)

fragment = chnk->fragment;
p_payload = fragment->payload;
size = fragment->size;

I think that chnk->fragment is empty and therefore no memcopy can be made.

[CarlosDerSeher]

@Dotsch2005 what's your server config? mainly chunk size. Which codec do you use? PCM I see from your log? You could try enabling this line and maybe add chnk->fragment and fragment->payload pointers to this print just to see if they are NULL

[Dotsch2005]

Yes, the codec is set to PCM in the Snapcast Config. Here is the config:

#######################################################################################
#     ______                                                                          #
#    / _____)                                                                         #
#   ( (____   ____   _____  ____    ___  _____   ____  _   _  _____   ____            #
#    \____ \ |  _ \ (____ ||  _ \  /___)| ___ | / ___)| | | || ___ | / ___)           #
#    _____) )| | | |/ ___ || |_| ||___ || ____|| |     \ V / | ____|| |               #
#   (______/ |_| |_|\_____||  __/ (___/ |_____)|_|      \_/  |_____)|_|               #
#                          |_|                                                        #
#                                                                                     #
#  Snapserver config file                                                             #
#                                                                                     #
#  Source: https://github.com/badaix/snapcast/blob/develop/server/etc/snapserver.conf #                                                                          #
#                                                                                     #
#######################################################################################

# default values are commented
# uncomment and edit to change them

# Settings can be overwritten on command line with:
#  "--<section>.<name>=<value>", e.g. --server.threads=4


# General server settings #####################################################
#
[server]
# Number of additional worker threads to use
# - For values < 0 the number of threads will be 2 (on single and dual cores)
#   or 4 (for quad and more cores)
# - 0 will utilize just the processes main thread and might cause audio drops
#   in case there are a couple of longer running tasks, such as encoding
#   multiple audio streams
#threads = -1

# the pid file when running as daemon
#pidfile = /var/run/snapserver/pid

# the user to run as when daemonized
#user = snapserver
# the group to run as when daemonized
#group = snapserver

# directory where persistent data is stored (server.json)
# if empty, data dir will be
#  - "/var/lib/snapserver/" when running as daemon
#  - "$HOME/.config/snapserver/" when not running as daemon
datadir = /data/

#
###############################################################################


# HTTP RPC ####################################################################
#
[http]
# enable HTTP Json RPC (HTTP POST and websockets)
enabled = true

# address to listen on, can be specified multiple times
# use "0.0.0.0" to bind to any IPv4 address or :: to bind to any IPv6 address
# or "127.0.0.1" or "::1" to bind to localhost IPv4 or IPv6, respectively
# use the address of a specific network interface to just listen to and accept
# connections from that interface
bind_to_address = 0.0.0.0

# which port the server should listen to
port = 1780

# serve a website from the doc_root location
# disabled if commented or empty
doc_root = /usr/share/snapweb

# Hostname or IP under which clients can reach this host
# used to serve cached cover art
# use <hostname> as placeholder for your actual host name
#host = <hostname>

#
###############################################################################


# TCP RPC #####################################################################
#
[tcp]
# enable TCP Json RPC
enabled = true

# address to listen on, can be specified multiple times
# use "0.0.0.0" to bind to any IPv4 address or :: to bind to any IPv6 address
# or "127.0.0.1" or "::1" to bind to localhost IPv4 or IPv6, respectively
# use the address of a specific network interface to just listen to and accept
# connections from that interface
bind_to_address = 0.0.0.0

# which port the server should listen to
port = 1705
#
###############################################################################


# Stream settings #############################################################
#
[stream]
# address to listen on, can be specified multiple times
# use "0.0.0.0" to bind to any IPv4 address or :: to bind to any IPv6 address
# or "127.0.0.1" or "::1" to bind to localhost IPv4 or IPv6, respectively
# use the address of a specific network interface to just listen to and accept
# connections from that interface
bind_to_address = 0.0.0.0

# which port the server should listen to
port = 1704

# source URI of the PCM input stream, can be configured multiple times
# The following notation is used in this paragraph:
#  <angle brackets>: the whole expression must be replaced with your specific setting
# [square brackets]: the whole expression is optional and can be left out
# [key=value]: if you leave this option out, "value" will be the default for "key"
#
# Format: TYPE://host/path?name=<name>[&codec=<codec>][&sampleformat=<sampleformat>][&chunk_ms=<chunk ms>][&controlscript=<control script filename>[&controlscriptparams=<control script command line arguments>]]
#  parameters have the form "key=value", they are concatenated with an "&" character
#  parameter "name" is mandatory for all sources, while codec, sampleformat and chunk_ms are optional
#  and will override the default codec, sampleformat or chunk_ms settings
# Non blocking sources support the dryout_ms parameter: when no new data is read from the source, send silence to the clients
# Available types are:
# pipe: pipe:///<path/to/pipe>?name=<name>[&mode=create][&dryout_ms=2000], mode can be "create" or "read"
# librespot: librespot:///<path/to/librespot>?name=<name>[&dryout_ms=2000][&username=<my username>&password=<my password>][&devicename=Snapcast][&bitrate=320][&wd_timeout=7800][&volume=100][&onevent=""][&nomalize=false][&autoplay=false][&params=<generic librepsot process arguments>]
#  note that you need to have the librespot binary on your machine
#  sampleformat will be set to "44100:16:2"
# file: file:///<path/to/PCM/file>?name=<name>
# process: process:///<path/to/process>?name=<name>[&dryout_ms=2000][&wd_timeout=0][&log_stderr=false][&params=<process arguments>]
# airplay: airplay:///<path/to/airplay>?name=<name>[&dryout_ms=2000][&port=5000]
#  note that you need to have the airplay binary on your machine
#  sampleformat will be set to "44100:16:2"
# tcp server: tcp://<listen IP, e.g. 127.0.0.1>:<port>?name=<name>[&mode=server]
# tcp client: tcp://<server IP, e.g. 127.0.0.1>:<port>?name=<name>&mode=client
# alsa: alsa:///?name=<name>&device=<alsa device>[&send_silence=false][&idle_threshold=100][&silence_threshold_percent=0.0]
# meta: meta:///<name of source#1>/<name of source#2>/.../<name of source#N>?name=<name>
source = librespot://librespot?name=LibreSpot&bitrate=320&sampleformat=44100:16:2
# source = airplay:///usr/bin/shairport-sync?name=shairport[&dryout_ms=2000&devicename=Snapcast&port=5000
source = pipe:////audio/snapcast_fifo?name=Mopidy

# Default sample format: <sample rate>:<bits per sample>:<channels>
sampleformat = 48000:16:2

# Default transport codec
# (flac|ogg|opus|pcm)[:options]
# Start Snapserver with "--stream:codec=<codec>:?" to get codec specific options
codec = pcm

# Default source stream read chunk size [ms].
# The server will continously read this number of milliseconds from the source into buffer and pass this buffer to the encoder.
# The encoded buffer is sent to the clients. Some codecs have a higher latency and will need more data, e.g. Flac will need ~26ms chunks
#chunk_ms = 20

# Buffer [ms]

# The end-to-end latency, from capturing a sample on the server until the sample is played-out on the client
buffer = 700

# Send audio to muted clients
send_to_muted = false

# Streaming client options ####################################################
#
[streaming_client]

# Volume assigned to new snapclients [percent]
# Defaults to 100 if unset
initial_volume = 100
#
###############################################################################


# Logging options #############################################################
#
[logging]

# log sink [null,system,stdout,stderr,file:<filename>]
# when left empty: if running as daemon "system" else "stdout"
#sink =

# log filter <tag>:<level>[,<tag>:<level>]*
# with tag = * or <log tag> and level = [trace,debug,info,notice,warning,error,fatal]
#filter = *:info
#
###############################################################################

I have changed the code as follows:

if (chnk == NULL) {
  if (pcmChkQHdl != NULL) {
    ret = xQueueReceive(pcmChkQHdl, &chnk, pdMS_TO_TICKS(100));
    if (ret != pdFAIL) {
        ESP_LOGI(TAG, "got pcm chunk with size %d",
                chnk->fragment->size);
    }
  }
} else {
  ESP_LOGW(TAG, "got pcm chunk with size %d",
  chnk->fragment->size);
}

ESP_LOGI(TAG, "chnk object at %p:", (void *)chnk);
ESP_LOGI(TAG, "fragment is %s", (chnk->fragment) ? "not NULL" : "NULL");

fragment = chnk->fragment;
p_payload = fragment->payload;
size = fragment->size;

ESP_ERROR_CHECK(
    i2s_channel_preload_data(tx_chan, p_payload, size, &written));

Error on Line player.c:1352 is:
ESP_LOGI(TAG, "fragment is %s", (chnk->fragment) ? "not NULL" : "NULL");

Log:

I (2881) dspProc: Set volume to 1.000000
I (2881) SC: pcm sampleformat: 48000:16:2
I (3081) SC: latency buffer full
I (3171) dspProc: got new setting for dspfEQBassTreble
W (3171) PLAYER: pcm chunk queue not created
W (3201) PLAYER: pcm chunk queue not created
W (3231) PLAYER: pcm chunk queue not created
W (3251) PLAYER: pcm chunk queue not created
W (3261) PLAYER: pcm chunk queue not created
I (3271) PLAYER: player_setup_i2s: dma_buf_len is 100, dma_buf_count is 22, sample rate: 48000, bits: 16
W (3271) PLAYER: no pcm chunk queue created?
I (3271) PLAYER: created new queue with 33
I (3281) PLAYER: snapserver config changed, buffer 700ms, chunk 960 frames, sample rate 48000, ch 2, bits 16 mute 0 latency 0
W (3301) PLAYER: got pcm chunk with size 3840
I (3301) PLAYER: chnk object at 0x3fcb35f0:
I (3301) PLAYER: fragment is not NULL
I (3411) PLAYER: chnk object at 0x0:
Guru Meditation Error: Core  1 panic'ed (LoadProhibited). Exception was unhandled.

Core  1 register dump:
PC      : 0x4202b1cc  PS      : 0x00060c30  A0      : 0x803816fc  A1      : 0x3fcb3fc0  
--- 0x4202b1cc: player_task at D:/Dev/Github/CarlosDerSeher/snapclient/components/lightsnapcast/player.c:1352 (discriminator 13)

A2      : 0x00000000  A3      : 0x3c111f94  A4      : 0x3fcdd984  A5      : 0x00000001
A6      : 0x00000000  A7      : 0x3fcb3fc0  A8      : 0x8202b1ca  A9      : 0x3fcb3fa0  
A10     : 0x00000d53  A11     : 0x3c0e0348  A12     : 0x3c0e0b20  A13     : 0x00000d53
A14     : 0x3c0e0348  A15     : 0x00000000  SAR     : 0x00000004  EXCCAUSE: 0x0000001c
EXCVADDR: 0x0000000c  LBEG    : 0x400556d5  LEND    : 0x400556e5  LCOUNT  : 0xfffffffe
--- 0x400556d5: strlen in ROM
0x400556e5: strlen in ROM



Backtrace: 0x4202b1c9:0x3fcb3fc0 0x403816f9:0x3fcb4110
--- 0x4202b1c9: player_task at D:/Dev/Github/CarlosDerSeher/snapclient/components/lightsnapcast/player.c:1352 (discriminator 13)
0x403816f9: vPortTaskWrapper at D:/Apps/esp/v5.1.5/esp-idf/components/freertos/FreeRTOS-Kernel/portable/xtensa/port.c:162





ELF file SHA256: ad0965b5a1141297

Rebooting...

The problem is that the chnk object is not filled.

What does this function do after the while (1) {?