This _handler_dg_system_record will be called in a separate thread. If get_dg_system_record is called and clears self.raw_sys_record while you're parsing new data, your data would get corrupted Therefore, it would be best to reduce the scope of self.raw_sys_record where possible. 1. Line 147: Update self.raw_sys_record to a local variable (i.e. raw_sys_record without "self") 2. Line 151: Pass raw_sys_record into _update_dg_system_record_from_fw 3. Return the parsed dg_system_record at the end of _update_dg_system_record_from_fw. Assign it to self.raw_sys_record on line 151. 4. Refactor (rename) _update_dg_system_record_from_fw to _parse_system_record_from_dg
This line caused some issues, we need to remove the log of the application output for now. please change it to by adding & and # before 2>>: $HOME/denali -u & # 2>> $HOME/filesystem.err &
You actually had it with the value of 0.1 ms and I increased it to 0.5 because after running the test multiple times I figured it is not enough time. I will review the waits in the pTx-UF COCO.
Should we send DG modes more often? UI and HD stacks are keying off of these a lot and I worry 1 Hz will not be very responsive. I also worry there may be some modes/states that do not last 1 second and HD may miss it. If we speed this up, we should also speed up the HD modes broadcast as well.
