1.20.0rc13 guru meditation

[poesel]

Hi,

I'm getting a guru meditation on my GPy. Not always but repeatable.

boot.py has:

import pycom

pycom.heartbeat(False)

For main.py this is enough:

from network import Bluetooth
import time
import struct

The fault is:

Guru Meditation Error: Core  0 panic'ed (Cache disabled but cached memory region accessed)
Core 0 register dump:
PC      : 0x401da75c  PS      : 0x00060c34  A0      : 0x800921e8  A1      : 0x3ffc0970  
A2      : 0x00000001  A3      : 0x00000001  A4      : 0xb33fffff  A5      : 0x0000cdcd  
A6      : 0x3ffbdafc  A7      : 0x00000000  A8      : 0x8008c97e  A9      : 0x00100000  
A10     : 0x11640000  A11     : 0x00000001  A12     : 0x00000013  A13     : 0x00000000  
A14     : 0x3ffc48fc  A15     : 0x00000002  SAR     : 0x0000000e  EXCCAUSE: 0x00000007  
EXCVADDR: 0x00000000  LBEG    : 0x00000000  LEND    : 0x00000000  LCOUNT  : 0x00000000  
Core 0 was running in ISR context:
EPC1    : 0x40093ee7  EPC2    : 0x00000000  EPC3    : 0x00000000  EPC4    : 0x401da75c

Backtrace: 0x401da75c:0x3ffc0970 0x400921e5:0x3ffc0990 0x400929d7:0x3ffc09b0 0x4009271e:0x3ffc09d0 0x4009304f:0x3ffc09f0 0x4008394d:0x3ffc0a10 0x4008a44d:0x3ffd44a0 0x40081efb:0x3ffd44c0

and then a core dump (which I can provide if its useful for anyone.

History:
I'm working on some BLE stuff when I suddenly had problems uploading (pymakr in VSC). I then upgraded the firmware to 1.20.1r1 with LittleFS to no avail. Did the 'erase_all' & firmware upload several times to no success. Finally went back to 1.20.0 and here I am.

I'm not sure if this is a software or hardware problem. And I don't know how to proceed from here.
Thanks.

[andreas]

Dear @poesel,

@poesel said in 1.20.0rc13 guru meditation:

Guru Meditation Error: Core 0 panic'ed (Cache disabled but cached memory region accessed)

A hardware bug of the ESP32 might bite you here [1]. You might want to try [2], compiled with #define CONFIG_FREERTOS_UNICORE 1. However, we don't really know if this really makes any difference, it's just trial-and-error for now.

Additionally, you might want to try to stop your program and the BLE subsystem before uploading.

With kind regards,
Andreas.

[1] https://community.hiveeyes.org/t/random-memory-corruption-faults-on-esp32-wrover-rev-1-and-rev-2-when-running-in-dual-core-mode/2515
[2] https://packages.hiveeyes.org/hiveeyes/foss/pycom/vanilla/GPy-1.20.1.r1-0.1.0-vanilla-unicore.tar.gz

[andreas]

Dear @poesel,

following [1] thoroughly, we brewed another wave of firmware images. So, we would also like to recommend to try [2] and will be happy to hear back from you about this.

With kind regards,
Andreas.

[1] https://github.com/espressif/esp-idf/issues/2892
[2] https://packages.hiveeyes.org/hiveeyes/foss/pycom/vanilla/GPy-1.20.1.r1-0.2.0-vanilla-psram-fix.tar.gz

[andreas]

Dear @poesel,

we found through [1] that the panic keeps popping up. Sorry for that!

@poesel said in 1.20.0rc13 guru meditation:

from network import Bluetooth
import time
import struct

Does it really occasionally happen with just that code? We see that you can reliably reproduce the panic using your code you shared at [2]. Unfortunately, we don't have the other device which advertises itself as a ‘HEN’ node so we can't reproduce it on our workbench.

We're just curious about the background and scope of that very problem.

With kind regards,
Andreas.

[1] https://community.hiveeyes.org/t/investigating-core-panics-with-ble-on-pycom-devices/2715
[2] https://github.com/poesel/BLE-uPy-Pycom/blob/master/Pycom/main.py

[andreas]

Dear @poesel,

you might want to check out the most recent release of our custom build [1] when coming back to this.

With kind regards,
Andreas.

[1] https://packages.hiveeyes.org/hiveeyes/foss/pycom/vanilla/GPy-1.20.1.r1-0.6.0-vanilla-dragonfly.tar.gz

[psrebrny]

@andreas I also got this type of error, when executing my code.

G01, GPy with firmware version 1.20.1.r1

Guru Meditation Error: Core 1 panic'ed (Cache disabled but cached memory region accessed)
Guru Meditation Error: Core 1 panic'ed (IllegalInstruction). Exception was unhandled.
Memory dump at 0x401ffee4: bad00bad bad00bad bad00bad

[andreas]

Dear @psrebrny,

@psrebrny said in 1.20.0rc13 guru meditation:

I also got this type of error, when executing my code.

Did you have the chance to run your code on our custom build [1] already and do you observe any difference?

With kind regards,
Andreas.

[1] https://packages.hiveeyes.org/hiveeyes/foss/pycom/vanilla/GPy-1.20.1.r1-0.6.0-vanilla-dragonfly.tar.gz