RMII 100M support

[soypat]

So last byte is processed weirdly, but rest of frame is transmitted via loopback correctly! We have working RMII in TinyGo!

 tinygo flash -target=pico -monitor -scheduler=tasks .
Connected to /dev/ttyACM0. Press Ctrl-C to exit.
=== RMII Loopback Test ===
PHY addr: 1 id: 7 49393 bmsr: 0x0000780d bmcr: 0x00002100
forced linkmode  100M-F
waiting for link...
.link up!
rx: got frame length  24
  MATCH: all bytes sent also received identically! length match: true
rx: got frame length  24
  MATCH: all bytes sent also received identically! length match: true
rx: got frame length  24
  MATCH: all bytes sent also received identically! length match: true
rx: got frame length  24
  MATCH: all bytes sent also received identically! length match: true
rx: got frame length  24
  mismatch, first 24 bytes
rx: 255 255 255 255 255 255 2 0 0 0 0 1 136 181 115 61 0 0 0 5 78 54 246 219 
tx: 255 255 255 255 255 255 2 0 0 0 0 1 136 181 115 61 0 0 0 5 78 54 246 243 
rx: got frame length  24
  MATCH: all bytes sent also received identically! length match: true
rx: got frame length  24
  mismatch, first 24 bytes
rx: 255 255 255 255 255 255 2 0 0 0 0 1 136 181 115 61 0 0 0 7 160 56 151 31 
tx: 255 255 255 255 255 255 2 0 0 0 0 1 136 181 115 61 0 0 0 7 160 56 151 223 
rx: got frame length  24
  MATCH: all bytes sent also received identically! length match: true
rx: got frame length  24
  MATCH: all bytes sent also received identically! length match: true
rx: got frame length  24
  MATCH: all bytes sent also received identically! length match: true
rx: got frame length  24
  MATCH: all bytes sent also received identically! length match: true

[deadprogram]

Is there a specific reason the examples/_rmii directory is named with an underscore?

[soypat]

Is there a specific reason the examples/_rmii directory is named with an underscore?

To exclude it from the Go build system at the top module level. This prevents lneto appearing as a dependency of the pio package. Is there a better way?

Will address the rest of your comments in time. Still trying to figure out a way to go from 8% to ~0% packet loss regarding the PIO program comments

[soypat]

This is working!

[deadprogram]

I will give this a more serious review now. Getting exciting!

[deadprogram]

Commented lines?

[deadprogram]

Amazing work here @soypat 🎆

Now squash/merging!