This moves the netdev/netlink namespace out of drivers and into their
own packages. Also, defines netdev and netlink as L3/L4 and L2 OSI
layers, respectively. Move some L3 functionality from netlink to
netdev (GetIPAddr).
For netlink, add ConnectParams for NetConnect to pass in L2 connection
parameters (ssid, pass, auth_type, etc). Also adds connection mode
(STA, AP, etc).
For netlink, add SendEth and RecvEthFunc funcs to handle L2 send/recv of
Ethernet pkts.
Tested with a rp2040.
TODO: add the ability to set the absolute humidity for more accurate
sensor details. The formula for that is rather complex, so I've left
this as a future addition.
Replace DrawRGBBitmap8 with DrawBitmap, following the change in the
previous commit.
This improves performance from 86fps to 100fps! I didn't investigate
why, but I suspect it's because it now needs only a single store instead
of two to update a pixel.
I wrote this for the PineTime, and all available sensors (accelerometer,
step counter, temperature sensor) do work.
This commit also includes two "configuration files", that actually
appear to be firmware files to run on the accelerometer for special
features like step counting.
I'm not sure where they originally came for, and I don't know the
copyright status of them. However, Bosch has open-sourced the BMA423
driver which includes a similar binary blob and the InfiniTime and
Wasp-OS projects have been shipping these blobs without issues, so I
think it's reasonably safe to include these binary blobs directly in the
source.
This is implemented in inline assembly using machine.CPUFrequency() to
know how long a single CPU cycle takes. As long as it is called with a
constant duration, it should be fully inlined and all values can be
const-propagated resulting in very tight inline assembly.
For example, when I convert i2csoft to use this delay function, the
entire delay function compiles to something like this:
8784: movs r6, #100
8786: mov r0, r6
8788: nop
878a: nop
878c: nop
878e: nop
8790: nop
8792: subs r0, #1
8794: bne 0x8788
That means that all the math to calculate the number of cycles is
entirely optimized away (in this case, to 100 loops).
I ran the example on a few boards to see how well it works:
| board | 100ms wait | CPU core
|-----------------------|------------|------
| microbit | 121.6ms | Cortex-M0 so it has 12% overhead
| circuitplay-express | 100.1ms | Cortex-M0+ so it is cycle accurate
| pico | 100.2ms | Cortex-M0+
| pyportal | 100.3ms | Cortex-M4
| circuitplay-bluefruit | 125.8ms | Cortex-M4
| esp8266 | 125.1ms |
This shows that there is some loop overhead because of conservative
estimates, but note that even though there may be a 25% overhead, the
actual overhead per `delay.Sleep()` call is very small. It should be
good enough for software I2C at least, and can potentially be improved
in the future.
- Remove all default loraconf default initialisation in lorawan examples (Region Settings takes care of this now)
- Remove retries while receiving JoinAccept and increase LoraRX to 10s (no need to wait more than 10s)
- Add constants for RegionSetting default frequencies
- Standardize all lora default configurations in lora examples
- Add new AT+LW=NET,(ON|OFF) command in atcmd example
Scott Feldman
SoftTacos
Thomas Richner
Ayke van Laethem
deadprogram
Christian Ege
Thomas
fchiesadoc
Patricio Whittingslow
Daniel Esteban
ivoszz
Daniel Esteban
Neil Davis
Olivier Fauchon
Olivier Fauchon