I have a Python script that is running as a systemd service. It is using a UART and SPI to communicate with an ARM processor. The process performs a blocking read on the serial port waiting for 4 bytes. Using an oscilloscope I measure the difference between when the serial port packet is transmitted and when the read call returns. Most often this value is around 2 ms which is fine. Sometimes (within 20 seconds to a minute) this value is 500 ms and I've seen it be larger than 1 second. If attempted to use `chrt` to change the static priority and scheduling policy. It reports that the values are set correctly but the large latencies remain.
This is with the 3.8.13-bone62 that I rebuilt to increase the SPI buffer. The buffer size was the only change. Any thoughts on how to improve this? I can probably tolerate up to 100 ms. -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/de2e6ab1-c36f-4bed-92d1-0dac0bd0b1cc%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
