v1.0.2

Release of P2 Octo Serial port driver

Update: isp_mem_strings.spin2 fixed use of now reserved work in latest Spin2 def. (v37),

• Services 1 to 8 ports from single cog running Pasm2
• Tested initial configuration: 115,200 bps, N81
• Validated with serial decoding logic analyzer
• Tested Raspberry Pi <-> P2 - two serial pairs
• Tested P2 <-> P2 - six more serial pairs
• Added new method PUB txBytesEmpty(portHandle) : nbrCharsAvail for determining when new message will fit entirely in tx queue
• Now tested P2 <-> P2 all 8 channels with CRC validation on all messages. No data loss or corruption at 115,200
• Service rate per port is ~860nSec suggesting we likely can run > 2Mbaud on all 8-ports (but need to verify)

Known Issues

• We need more characterization now that it's working so well.
• We saw one issue with the 2nd serial port connected to a test RPi 4. The second serial port started the first transmission with a leading 0x00 byte. The cause has not been found.
• When testing P2 <-> P2 with many ports we saw no such issues as is the same with the first of the two serial lines connected to the RPi

Full Changelog: v1.0.1...v1.0.2

v1.0.1

Release of P2 Octo Serial port driver

(Update: same as v1.0.0 but disabled use of Logic Analyzer Characterization Pins 4, 6, 8-10)

• Services 1 to 8 ports from single cog running Pasm2
• Tested initial configuration: 115,200 bps, N81
• Validated with serial decoding logic analyzer
• Tested Raspberry Pi <-> P2 - two serial pairs
• Tested P2 <-> P2 - six more serial pairs

Known Issues

• We need to characterize this now that it's mostly working
• We saw one issue with the 2nd serial port connected to a test RPi 4. The second serial port started the first transmission with a leading 0x00 byte. The cause has not been found.
• When testing P2 <-> P2 with many ports we saw no such issues as is the same with the first of the two serial lines connected to the RPi

v1.0.0

Initial release of P2 Octo Serial port driver

• Services 1 to 8 ports from single cog running Pasm2
• Tested initial configuration: 115,200 bps, N81
• Validated with serial decoding logic analyzer
• Tested Raspberry Pi <-> P2 - two serial pairs
• Tested P2 <-> P2 - six more serial pairs

Known Issues

• WARNING (totally my fault) I accidentally left the logic analysis outputs active in this release! I'm posting a next release without these enabled.
• We need to characterize this now that it's mostly working
• We saw one issue with the 2nd serial port connected to a test RPi 4. The second serial port started the first transmission with a leading 0x00 byte. The cause has not been found.
• When testing P2 <-> P2 with many ports we saw no such issues as is the same with the first of the two serial lines connected to the RPi

FYI- My LA Instrumentation outputs:

• Pins 8,9,10 are the port address [0-7] being serviced
• Pin 4 is high while servicing the receive side of a port
• Pin 5 is high while servicing the transmit side of a port