NimBLE-Arduino: Stack canary watchpoint triggered (ipc0) - arduino 1_0_5_RC6/v3.3.4-402-gcd59d107b

with the vanilla latest release version of arduino, I’m getting these regularly:

10:26:55.827 readOneSensor - already active reading 2
Guru Meditation Error: Core  0 panic'ed (Unhandled debug exception)
Debug exception reason: Stack canary watchpoint triggered (ipc0) 
Core 0 register dump:
PC      : 0x40094c83  PS      : 0x00060036  A0      : 0x40084a64  A1      : 0x3ffb9a20  
A2      : 0x00000000  A3      : 0x00060023  A4      : 0x00060020  A5      : 0x00000001
A6      : 0x00000020  A7      : 0x00000000  A8      : 0x801c5842  A9      : 0x3ffb9bb0  
A10     : 0x00000008  A11     : 0x40082f30  A12     : 0x3ffb61b4  A13     : 0x00000000
A14     : 0x3ffb616c  A15     : 0x00060b23  SAR     : 0x00000008  EXCCAUSE: 0x00000001  
EXCVADDR: 0x00000000  LBEG    : 0x4000c349  LEND    : 0x4000c36b  LCOUNT  : 0xffffffff

ELF file SHA256: 0000000000000000

Backtrace: 
0x40094c83:0x3ffb9a20 
0x40084a61:0x3ffb9af0 _xt_user_exit
0x40095657:0x3ffb9b10 
0x40095dae:0x3ffb9b30 multi_heap_malloc
0x40084535:0x3ffb9b50 
0x40084566:0x3ffb9b70 
0x40082f35:0x3ffb9b90 _malloc_r
0x4000beaf:0x3ffb9bb0 
0x401c583f:0x3ffb9bd0 btdm_task_post
0x401c58b9:0x3ffb9c00 
0x40181b7d:0x3ffb9c30 async_wakeup_request
0x40181e2d:0x3ffb9c50 esp_vhci_host_send_packet
0x4011eb8c:0x3ffb9c70 ble_hci_trans_hs_acl_tx_on_core_0
0x40087473:0x3ffb9da0 libesp32.a(ipc.o)
0x4009226a:0x3ffb9dc0 (port.o)

Rebooting...
ets Jun  8 2016 00:22:57

rst:0xc (SW_CPU_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT)
configsip: 0, SPIWP:0xee

Although not directly NimBLE related, do you have any advice? (still in an older NimBLE)

br,

About this issue

Original URL
State: closed
Created 3 years ago
Comments: 30 (29 by maintainers)

Most upvoted comments

Damn, no idea how I missed that, thanks for pointing it out!

😃

h2zero on Mar 31, 2021

Thanks, looks like the issue has been pinpointed by another posted here https://github.com/espressif/esp-idf/issues/6358

Looks like you found a workaround for your use at least. I’m curious if you changed the 5 second delay to just a taskYIELD() if it would alleviate this.

h2zero on Jan 22, 2021