aboutsummaryrefslogtreecommitdiff
path: root/src/drivers/px4io
Commit message (Collapse)AuthorAgeFilesLines
* px4io: don't try the px4io serial interface on FMUv1Andrew Tridgell2013-08-021-0/+2
| | | | this caused px4io start to fail on FMUv1
* Symbol cleanup for servo vs. battery voltageLorenz Meier2013-07-261-3/+3
|
* Fix handling of register read operation errors.px4dev2013-07-171-13/+17
|
* Add additional file name optionsLorenz Meier2013-07-171-2/+4
|
* Fixed compile error due to bad mergeLorenz Meier2013-07-161-3/+3
|
* Merge branch 'master' of https://github.com/PX4/Firmware into fmuv2_bringuppx4dev2013-07-142-15/+124
|\
| * Don't leave RX in bind mode on console open failJean Cyr2013-07-111-2/+2
| | | | | | | | | | Don't leave RX in bind mode in the unlikely eventuality that console open fails
| * General cleanup of /dev/px4io and /dev/px4fmuJean Cyr2013-07-091-5/+5
| | | | | | | | | | | | | | - Use distinct common symbols for px4io and px4fmu device files, and use instead of hardcoded filenames - Use common symbols defining px4io bits consistently between px4fmu and px4io builds.
| * Minor px4io optimizationJean Cyr2013-07-071-18/+9
| | | | | | | | | | | | Since this module creates the PX4IO object and that the IOCTL function doesn't use the file descriptor parameter, there is no need to invoke IOCTL via the filesystem since we can call it directly.
| * Prevent RELAY1 control via IOCTL if DSM bind feature is enabledJean Cyr2013-07-071-14/+45
| |
| * Merge remote-tracking branch 'upstream/master'Jean Cyr2013-07-071-1/+8
| |\
| | * Turn off loggingpx4dev2013-06-161-4/+4
| | |
| | * Set the serial port speed before trying to talk to IOpx4dev2013-06-161-3/+10
| | |
| * | Support binding DSM2 and DSMX satellite receiversJean Cyr2013-07-051-5/+91
| |/ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The px4io bind command allows you to put a DSM satellite receiver into bind mode. Since this feature requires that the dsm VCC line (red wire) be cut and routed through relay one, it is not enabled by default in order not to affect those not using a DSM satellite receiver or wising to use relay one for other purposes. NOTE: Binding DSM2 satellites in 11-bit mode is not supported due to potential bug in some DSM2 receiver streams when in 11-bit mode. Furthermore the px4io software folds 11 bit data down to 10 bits so there is no resolution advantage to to 11-bit mode. To enable the feature the RC_RL1_DSM_VCC parameter must be set to a non zero value from the console, or using QGroundControl: param set RC_RL1_DSM_VCC 1 From the console you can initiate DSM bind mode with: uorb start param set RC_RL1_DSM_VCC 1 px4io start px4io bind dsm2 For binding a DSMX satellite to a DSMX transmitter you would instead use: px4io bind dsmx Your receiver module should start a rapid flash and you can follow the normal binding sequence of your transmitter. Note: The value of parameter RC_RL1_DSM_VCC defaults to 0, so none of this will have any effect on an unmodified DSM receiver connection. For this feature to work, the power wire (red) must be cut and each side connected to a terminal on relay1 of the px4io board. This has been tested using Spektrum as well as Hobby King 'Orange' DSM satellite receivers. Both px4fmu and px4io images are updated.
* | Don't build interface drivers we don't have config for.px4dev2013-07-142-8/+8
| |
* | Use common, board-type-agnostic code to allocate the PX4IO interface.px4dev2013-07-141-34/+32
| |
* | Direct-access device functions return errors directly.px4dev2013-07-144-156/+79
| | | | | | | | Move to using ::init rather than ::probe in keeping with device changes.
* | rename the px4io serial perf counters so it's clearer what they belong topx4dev2013-07-131-10/+10
| |
* | Build the px4io interfaces on top of the Device direct-access API.px4dev2013-07-133-144/+134
| |
* | Make px4io driver filenames less ambiguous.px4dev2013-07-134-3/+3
| |
* | Pull v2 pieces up to build with the mergepx4dev2013-07-073-7/+6
| |
* | Tweak protocol register assignments and add new registers to accommodate ↵px4dev2013-07-061-18/+35
| | | | | | | | differences in IOv2.
* | Optimise the RC input fetch for <= 9ch transmitters; this eliminates one ↵px4dev2013-07-061-22/+22
| | | | | | | | read per cycle from IO in the common case.
* | Crank up the test speed for 'px4io iftest 1'px4dev2013-07-061-2/+1
| |
* | Move the common definitions for the PX4IO serial protocol into the shared ↵px4dev2013-07-061-79/+2
| | | | | | | | header.
* | Abort the px4io worker task if subscribing to the required ORB topics fails.px4dev2013-07-061-0/+9
| |
* | Add retry-on-error for non-protocol errors.px4dev2013-07-051-52/+98
| | | | | | | | Add more performance counters; run test #1 faster.
* | Enable handling for short-packet reception on FMU using the line-idle ↵px4dev2013-07-051-58/+66
| | | | | | | | interrupt from the UART. Enable short packets at both ends.
* | More link performance counters.px4dev2013-07-051-15/+33
| |
* | Be more consistent with the packet format definition.px4dev2013-07-051-6/+12
| | | | | | | | Free perf counters in ~PX4IO_serial
* | Eliminate the TD DMA callback; we don't need to know that it's completed.px4dev2013-07-051-57/+18
| | | | | | | | Fix abort behaviour on timeouts, now we don't wedge after the first one.
* | Check packet CRCs and count errors; don't reject packets yet.px4dev2013-07-051-1/+12
| |
* | Add an 8-bit CRC to each transmitted packet.px4dev2013-07-051-4/+56
| |
* | Encode the packet type and result in the unused high bits of the word count.px4dev2013-07-051-8/+18
| |
* | Always send and expect a reply for every message.px4dev2013-07-051-33/+27
| |
* | Use the NuttX built-in crc32, it works fine.px4dev2013-07-051-46/+4
| |
* | Switch to the 'normal' way of doing register accessors.px4dev2013-07-041-57/+82
| | | | | | | | Be more aggressive en/disabling DMA in the UART, since ST say you should.
* | Add PX4IOv2 support to the uploader.px4dev2013-07-041-4/+33
| |
* | Add DMA error handling.px4dev2013-07-041-60/+225
| | | | | | | | | | Add serial error handling. Add short-packet-reception handling (this may allow us to send/receive shorter packets… needs testing).
* | Add a test hook for the PX4IO interface. Wire up some simple tests for the ↵px4dev2013-07-034-128/+210
| | | | | | | | | | | | serial interface. Signal quality looks good at 1.5Mbps. Transmit timing is ~450µs per packet, ~20µs packet-to-packet delay spinning in a loop transmitting.
* | Rework the FMU<->IO connection to use a simple fixed-size DMA packet; this ↵px4dev2013-07-031-153/+190
| | | | | | | | | | | | should let us reduce overall latency and bump the bitrate up. Will still require some tuning.
* | More implementation for the serial side on IO; fix a couple of bugs on the ↵px4dev2013-06-281-16/+7
| | | | | | | | | | | | FMU side. Still needs serial init and some more testing/config on the FMU side, but closer to being ready to test.
* | Checkpoint: interface abstraction for px4io driverpx4dev2013-06-255-52/+666
|/
* Hotfix: Allow the IO mixer loading to load larger mixers, fix up the px4io ↵Lorenz Meier2013-06-071-2/+19
| | | | test command to allow a clean exit
* Better error handling for too large argumentsLorenz Meier2013-06-041-3/+8
|
* Set unknown channels to zero, since centering them is a slightly dangerous guessLorenz Meier2013-05-291-1/+2
|
* Set default failsafe value to 0 of mixerLorenz Meier2013-05-291-8/+21
|
* Slightly reworked IO internal failsafe, added command to activate it (px4io ↵Lorenz Meier2013-05-281-7/+45
| | | | failsafe), does not parse commandline arguments yet
* Hotfix: Changed alarms back to what they originally were designed for: Traps ↵Lorenz Meier2013-05-161-0/+3
| | | | to later see if condition was once violated. Currente status can be read through the status flags
* arming: added PWM_SERVO_SET_ARM_OK and PWM_SERVO_CLEAR_ARM_OKAndrew Tridgell2013-05-161-0/+10
| | | | | these new ioctls allow for the flight code to tell the IO board that arming can proceed