GitLab

Restore closeReadingPipe merge for mysensors

Restore closeReadingPipe to be compatible with mysensors

- Enabled use of hardware CS1 pin
- Removed custom_cs code for software driven CSN
- Modified BCM driver to allow use of hardware SPI pins not in use while
SPI is active.
- Updated constructor in examples

Added support for hardware SPI on RPI per jscrane fork:
- Fixed a few bugs with single SPI transfers
- Aligned with optimized RF24 library fork
- Updated bmc2835 files per
http://www.airspayce.com/mikem/bcm2835/index.html
- Merged bcm files with changes from
https://github.com/farconada/RF24Network
- Added example files
- Added setup info to readme

- fixed mistake with changes to setPALevel prior to pushing

- Using status register is faster to read TX FIFO status - Zephyrr (GitHub)
- Fixed typo on FLUSH_RX cmd
- Added support for pwr levels on SI24R1 chips with smaller code

- Updated some example files and tests
- Small changes to minimize code size

For nRF24L01+ to go from power down mode to TX or RX mode it must first
pass through stand-by mode.
There must be a delay of Tpd2stby (see Table 16.) after the nRF24L01+
leaves power down mode before
the CEis set high.

5ms is the maximum defined in the v1.0 datasheet

- Added new function: enableDynamicAck()  to enable single-write NOACK
payloads (merge closer with gcopeland fork)
- Added single NOACK writes for write(), writeFast(), startWrite() and
startFastWrite()
- powerUp() now checks to see if radio is powered on. No delay if
powered up.
- radio must still be powered up manually if powered down manually
- use setRetries function in begin()
- set CE low on powerDown
- updated readme

ATTiny support pulled in from https://github.com/jscrane/RF24
- SPI support for ATTiny is included with the library. Do NOT include
SPI.h with tiny.

Pin info etc can be found at:
http://programmablehardware.blogspot.ca/2013/09/rf24-with-attiny84.html

ATTiny 85 Example:
RF24 radio(4,3);    // CE: 4  CS: 3

Merge improved write_payload() and read_payload()  - looks good and passes my testing.

Restructured based on needing or not needing blank padding

Found that Due would send multiple payloads or fail in writing in some
cases. Change ensures single-payload transmission using write().
- Updated pingpair_dyn.ino example sketch to conform with available()
changes
- Slight doc update

- SPI transfer fix to read every available byte in the payload with
payloads less than 32 bytes with dynamic payload size disabled.

- Tested and now fully functional on Arduino Due
- Added support for the extended SPI functions: brings Due performance
up to par with Due managing the CSN   pin and settings
- Corrected issues with printDetails() on Due
- Changed added variables to staticly defined types: ie: uint16_t
instead of unsigned int
- I think this will work on all ARM based boards that use the Arduino
libs
- Added maskIRQ function - to be used in RF24Network sleep mode
- Changed csn(int mode) to csn(bool mode)
- Updated GettingStarted_CallResponse.ino example: Capture start time
AFTER printing to serial so measurement is more accurate

- Found I could create a hang condition on startup with some code.
a: Now power up the radio on begin() to simplify things
b: Increase delay to 5ms for powerUp, although it should only take 2ms
tops, this prevents the hanging condition.

- Initial support for Due and ATTiny boards
- Fix for 0 timeout period causing hang
- Remove unnecessary delays from startWrite
- Cleaned up/Updated example files
- Optimization is nearing completion

- For use in noisy or low signal scenarios
- Allows user specified timeout period of extended duration
- Modified txStandBy() and writeBlocking() functions to rely on a user
defined timeout period, automatic retries and payload reUse

for use in wireless audio sketch, needed to move startWrite() to a
public function

txStandBy() now either works in two modes. The main mode will block
until success or fail. Using txStandBy(1) will block and retry forever
until succesful. Using a while loop with the function is no longer
required. See the updated class reference at
http://tmrh20.github.io/RF24/class_r_f24.html

Added back original startWrite function for when users want to use
manual methods. The library uses startFastWrite for its internal
functions. Provides compatibility with prev. code while still using the
new methods for everything else.

- txStandBy was hanging if the last of the data timed out. Now clears
the MAX_RT flag if timed out
- Finished updating example file Transfer.ino for demonstrating
half-rate transfers using the FIFO buffers

Realized that the normal write command should always drop to standby-I
mode immediately since the FIFO cannot be filled using this method, and
TX will be complete.

Updated GettingStarted, pingpair, added medium speed version of transfer
rate test

See http://tmrh20.github.io/RF24/class_r_f24.html for the class
documenation.

Realised that data would be lost if dropping to standby or power down
mode before the FIFO is empty.
txStandby now returns a boolean value to indicate whether the
transmission is complete. If there is data in the TX FIFO the radio will
not drop out of TX mode. eg: while(  !txStandBy()  ){}

- put the CE HIGH, LOW changes in the cmd for non-internal use
- Put the clearing of MAX_RT in the reUse cmd for non-internal use
- Removed unneeded delay from writeBlocking()

- WriteFast allows the FIFO buffers to be filled up before returning 0
- The current payload will be dropped, and the previous payload resent
- The user should just keep trying to send the same data if a 0 is
received

- Changed writeClear to the default radio.write() option
- Changed write() to writeBlock() - uses the auto-retry functionality of
the chip and will block forever until successful

- advisory of non-compliance with current documentation

Changes:

- Enable PTX at radio begin, in standby-I mode
- Enable PTX at radio stopListening(); in standy-I mode

- Radio startListening
- removed delay. What is there to wait for when the radio will just be
listening anyway?

- Radio powerUp
- radio will need to be powered up after configuration
- radio will not power down unless instructed to
- need to issue radio.txStandBy(); when finished transmitting to engage
standby-I mode

- Added txStandBy() to drop to standBy-I mode if not switching to RX
mode

- Radio Write: Sped up
- radio previously did not use standby mode effectively, after TX
sending of each packet, the radio is
powered down introducing a 1.5ms delay in addition to the 130us delay
from standby.
- No observe-tx code, a simple NOP cmd is sent to get the status info
- No millis() timeout, hanging condition should only occur because of
faulty code
- Simplified whatHappened() code
- Flush TX FIFO if MAX_RT met as lib does not support re-transmit
instead of flush every time.
- Remove ack size monitoring
- No powerdown needed,

- Added writeClear() function for writes that need the interrupt flags
cleared (ack payloads, etc). Same as write() otherwise

- Radio startWrite: Sped up
- Enable PTX mode is now done in stopListening() code. Enter TX mode by
writing CE high.
This brings the data transfer capabilities up to a reasonable level
- No need to power up, already in Standby-I mode
- No need for delayMicroseconds() in write functon.
- No need for delay after writing CE high
- Does not enter Standby-I mode be default. It is up to the MCU to call
the new txStandBy() function to enter standby-I mode from PTX if not
switching back to PRX

- Radio available:
- Now checks the FIFO buffer instead of the interrupt flag (no missed
data)
- Moved clearing of interrupt flags to Read()

- Radio read:
- No longer returns any value - use available to check for more data

- Radio isAckPayloadAvailable() now checks the FIFO directly

- Radio getDynamicPayloadSize() - Per the datasheet, added fix to flush
corrupted packets

Updated led_remote example to use correct var on line 172

Line 172 was using button_pins instead of led_pins.  This caused leds to be lit dim when set to HIGH.  Changed to correct var.

Fix bug introduced in the move to PROGMEM.  Props to @basilrx for finding this.  Also added a test case to demonstrate the bug, and guard against more like it in the future.