You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
ttyUSB and Wifi sends data as normal, but also the last value before the "freeze"
blue LED also freezes, either on or off
Menus and buttons are operational during the freeze
The problem appears at some micros() overflow events, which happens about every 71 minutes (2^32 microseconds)
The problem appears inconsistently, it took me 16 hours to recreate it
The problem resolved itself and unfroze the data fetch after 2^32 microseconds in all documented cases to date
No errors are reported via ttyUSB
Symptoms indicate the main loop() executes a delay of 2^32 microseconds or a value close to that.
There is only one place in code where delayMicroseconds() is being called:
I've skimmed over the info in the thread and on the surface it looks, at least in some aspects, to be very similar to the problems I've experienced when trying to build the firmware with latest Espressif323 platform (a bug in PAC library related to I2C). Fix for that bug has made it into master (#22) but not yet into a release.
Can you confirm the bug still happens with firmware built from the latest master that includes the seemingly unrelated fix?
Issue described here in greater detail:
https://forum.odroid.com/viewtopic.php?p=350730#p350730
In short:
Data fetching freezes at some point
Display shows last fetched data
ttyUSB and Wifi sends data as normal, but also the last value before the "freeze"
blue LED also freezes, either on or off
Menus and buttons are operational during the freeze
The problem appears at some micros() overflow events, which happens about every 71 minutes (2^32 microseconds)
The problem appears inconsistently, it took me 16 hours to recreate it
The problem resolved itself and unfroze the data fetch after 2^32 microseconds in all documented cases to date
No errors are reported via ttyUSB
Symptoms indicate the main loop() executes a delay of 2^32 microseconds or a value close to that.
There is only one place in code where delayMicroseconds() is being called:
smartpower3/smartpower3/meas_chan.cpp
Lines 58 to 66 in 933acc1
Yet it seems it should not be possible to ever launch that delay with extreme value.
The text was updated successfully, but these errors were encountered: