![]() Power Supply: USB (custom HW has external power)Į (116633) task_wdt: Task watchdog got triggered.Compiler version (run xtensa-esp32-elf-gcc -version to find it): 2020r1.Kit version (for WroverKit/PicoKit/DevKitC): v4.Development Kit: ESP32-DevKitC (but also custom hardware with ethernet).I noticed it in my actual application code. It also occurs under normal load conditions, but not as often. It's a pretty intensive load, but that way the problem can be reproduced quickly. That, plus it seems stable (so far) with dupldst. The handshake is using nonblocking sockets and resets the WDT, so I don't think it should ever exceed the 60s timeout. ![]() Perhaps it would be a good idea to also supply newlib and sdk libs compiled with dupldst? ![]() No other changes were made, and the effect on stability is obvious. Seems to be MUCH more stable with dupldst rather than memw workaround (not sure if it's completely stable though - too soon to tell). CONFIG_MBEDTLS_HARDWARE_xxx are disabled, but it doesn't really matter. ![]() #Pale moon homepage code#The code always ends up causing a WDT timeout sooner or later (timeout set to 60s) in UNICORE mode. Code to reproduce is based on a modified https_mbedtls example from esp-idf: ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |