mjb/altos/src/drivers, branch tmgps

altos/ao_cc1120: Only call ao_packet_slave_stop when HAS_SLAVE

2018-03-23T23:37:02+00:00

Don't try to disable packet slave mode when the product doesn't support packet slave mode. Signed-off-by: Keith Packard

altos/stmf0: Use double buffering for ChaosKey

2018-01-08T05:44:39+00:00

This improves the USB performance of ChaosKey so that it doesn't NAK during data transfers at all. Signed-off-by: Keith Packard

altos/drivers: Hook up mag sensor for MPU9250

2017-12-11T20:20:25+00:00

Set mag sensor to provide data at 100Hz. Set i2c master to pull mag data at sample rate (200Hz). Signed-off-by: Keith Packard

altos/drivers: Start adding defines to get mag data out of MPU9250

2017-12-05T20:22:34+00:00

Signed-off-by: Keith Packard

altos: Add MPU9250 driver (accel/gyro only for now)

2017-10-12T07:14:30+00:00

This is almost an exact copy of the MPU6000 driver, just a few minor register changes. Signed-off-by: Keith Packard

chaoskey: Wait for input data to stabilize before using it

2017-09-12T20:40:06+00:00

The ADC data takes a while to start working after power on; wait for the range of input values to look reasonable before using the data. Signed-off-by: Keith Packard

altos/rn4678: Add BLE support

2017-06-20T18:19:40+00:00

This just means ignoring the BLE connect status message. Signed-off-by: Keith Packard

altos: Add timeout to RN4678 status message code

2017-06-13T17:34:03+00:00

If we see the start of an RN4678 status message, but then output pauses, assume that this isn't the start of a status message and flush the pending data. Signed-off-by: Keith Packard

altos: HMC5883 output order is X Z Y

2017-06-12T05:54:22+00:00

Re-label everything to have the correct names. This doesn't actually change the code at all, so the eeprom and telemetry is all compatible. Matching changes on the host side will be required to actually process the data correctly, of course. Signed-off-by: Keith Packard

altos: Update sensor data atomically

2017-06-12T05:54:22+00:00

Read data into a temp variable, block interrupts, then update the published value. The bug is easy to see with the HMC5883 which has to byte-swap the output of the chip, and hence can occasionally get caught with the wrong byte order data. Signed-off-by: Keith Packard