strtol crashes

[aep]

• esp32
• at commit 977aba6

void main(void)
{
   strtol("10",0, 10);
}

[00:00:00.475,000] <err> os:  ** FATAL EXCEPTION
[00:00:00.475,000] <err> os:  ** CPU 0 EXCCAUSE 28 (load prohibited)
[00:00:00.475,000] <err> os:  **  PC 0x4000be94 VADDR (nil)
[00:00:00.475,000] <err> os:  **  PS 0x60a20
[00:00:00.475,000] <err> os:  **    (INTLEVEL:0 EXCM: 0 UM:1 RING:0 WOE:1 OWB:10 CALLINC:2)
[00:00:00.475,000] <err> os:  **  A0 0x80056822  SP 0x3ffe5ca0  A2 0x3f4001e8  A3 0x3ffe5cc0
[00:00:00.475,000] <err> os:  **  A4 0x3ffe5ca0  A5 0x4  A6 0x144  A7 (nil)
[00:00:00.475,000] <err> os:  **  A8 (nil)  A9 0x3ffe5bd0 A10 0x13 A11 0x25
[00:00:00.475,000] <err> os:  ** A12 0x3f400525 A13 0x13 A14 0x3ffe5ca0 A15 0x8
[00:00:00.475,000] <err> os:  ** LBEG 0x400014fd LEND 0x4000150d LCOUNT 0xffffffff
[00:00:00.475,000] <err> os:  ** SAR 0x4
[00:00:00.484,000] <err> os: >>> ZEPHYR FATAL ERROR 0: CPU exception on CPU 0
[00:00:00.484,000] <err> os: Current thread: 0x3ffb0cf8 (unknown)
[00:00:00.672,000] <err> os: Halting system

[aep]

this looks like a compiler issue. renaming the function works fine.
possibly it's being replaced with a built-in?

compiler was installed with

aep@stark: ~/zephyrproject/zephyr west espressif install
=== downloading ESP-IDF tools..
Installing tools: xtensa-esp32-elf
Installing xtensa-esp32-elf@esp-2020r3-8.4.0
Downloading xtensa-esp32-elf-gcc8_4_0-esp-2020r3-linux-amd64.tar.gz to /home/aep/.espressif/dist/xtensa-esp32-elf-gcc8_4_0-esp-2020r3-linux-amd64.tar.gz.tmp
Done
Extracting /home/aep/.espressif/dist/xtensa-esp32-elf-gcc8_4_0-esp-2020r3-linux-amd64.tar.gz to /home/aep/.espressif/tools/xtensa-esp32-elf/esp-2020r3-8.4.0
=== downloading ESP-IDF tools completed

[sylvioalves]

Hi @aep, yes, just comment the L56 below:

-T${CMAKE_CURRENT_SOURCE_DIR}/../components/esp_rom/esp32/ld/esp32.rom.newlib-funcs.ld

from file
zephyr/modules/hal/espressif/zephyr/CMakeLists.txt

I will submit this soon.

[aep]

thanks @sylvioalves

that fixed it.

btw is esp32 actually supported? "priority: low" for a fairly base level bug doesnt give me much confidence i should use this at all.

[sylvioalves]

We have started supporting ESP32 and upcoming Espressif's chips a few months ago. Regarding this specific issue you reported, it was a regression from BT support added recently.