altos: include ao_arch_funcs.h at the very end of ao.h

Move it below the definition of the ms5607 init function

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Allow products without MS5607

The define for a missing MS5607 was wrong, so anything using the fancy
multi-sensor data code would break without an MS5607 in place.

Signed-off-by: Keith Packard <keithp@keithp.com>

ao_tools/ao-send-telem: Only start real-time on valid states

Check state to make sure it is < ao_flight_landed to keep invalid
states from switching to real-time playback mode.

Signed-off-by: Keith Packard <keithp@keithp.com>

altoslib: Add in a bunch of java files mising after Mike's cleanups

Signed-off-by: Keith Packard <keithp@keithp.com>

Merge remote-tracking branch 'mjb/altoslib_mjb'

altosui: Use helper functions to access arrays in AltosLib class

These deal with out-of-range values correctly, instead of causing
exceptions that will just break stuff.

Signed-off-by: Keith Packard <keithp@keithp.com>

altosui: Catch errors in state value when saving flight logs

Use AltosLib.state_name() instead of directly accessing the
state_to_string array so that any invalid state values are caught and
replaced with 'invalid' instead of raising an exception.

Signed-off-by: Keith Packard <keithp@keithp.com>

altoslib: move distinct classes to separate files.

Signed-off-by: Mike Beattie <mike@ethernal.org>

altoslib: remove obsolete class

Signed-off-by: Mike Beattie <mike@ethernal.org>

altoslib: move distinct classes to separate files

Signed-off-by: Mike Beattie <mike@ethernal.org>

altoslib: remove duplicate AltosGPSQuery class

Signed-off-by: Mike Beattie <mike@ethernal.org>

altoslib: remove duplicate AltosIdleMonitor class

Signed-off-by: Mike Beattie <mike@ethernal.org>

altoslib: rename AltosIdleRecordTM.java to AltosGPSQuery.java

Signed-off-by: Mike Beattie <mike@ethernal.org>

altoslib: remove dead code

(if object creation fails, an exception will be thrown - not return null)

Signed-off-by: Mike Beattie <mike@ethernal.org>

altoslib: resolve argument/variable ambiguity

(and comment out set_flags() which had no corresponding variable anyway)

Signed-off-by: Mike Beattie <mike@ethernal.org>

altoslib: comment out unused methods

Signed-off-by: Mike Beattie <mike@ethernal.org>

altoslib: access static variables via class, not instance

Signed-off-by: Mike Beattie <mike@ethernal.org>

altoslib: comment out un-used variables

Signed-off-by: Mike Beattie <mike@ethernal.org>

altoslib: Remove un-needed imports

Signed-off-by: Mike Beattie <mike@ethernal.org>

altos: Fix radio slave to run lights in the normal way

These were left in a debug mode, toggling instead of flashing.

Signed-off-by: Keith Packard <keithp@keithp.com>

Revert "first work on PWM input to TeleScience, 'p' command displays timer 1"

PWM bits not suitable for master

This reverts commit e93c6bcc799d76d4ff425815e2601a25e6796229.

Revert "Timer 3 working with slower clock and all 16 bits."

PWM bits not suitable for master

This reverts commit 49b1ff4c614d24977b33cd17b583acc87acff476.

Revert "ICP3 working"

PWM bits not suitable for master

This reverts commit 75d6aa6f798606f1a6c5a46542065dda81e63b2a.

Revert "telescience: correctly calculating rate values with higher resolution"

PWM bits not suitable for master

This reverts commit ada6f2dfc045e77cb9499f20cdec1b4a54ef0db1.

Merge remote-tracking branch 'mjb/freq_menu'

altos: Get mma655x driver limping along

This appears to drive the chip correctly to see values from the accelerometer.

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Fix mma665x pin assigment for mma655x

It's on PE13-PE15, not PA5-PA7

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Add custom panic noise for self-test failures

Make it easier to tell which component is failing self test

Signed-off-by: Keith Packard <keithp@keithp.com>

altoslib: add missing manufacturer parsing for AltosConfigData

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: invert channel/freq ordering

Matches Channel Selector in altosui

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: also display channel numbers

Signed-off-by: Mike Beattie <mike@ethernal.org>

ao-send-telem: fix frequency set command

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: complete frequency change dialog

Also implement Service IPC to action request.

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: Start a hacked-up frequency dialog

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: another .gitignore file

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: ao_cc_spi.h isn't necessary for telelco

The SPI radio defines are in ao_radio_spi.h

Signed-off-by: Keith Packard <keithp@keithp.com>

Merge remote-tracking branch 'mjb/master'

altos: Add a bunch of .gitignore entries

Clean up the git status output

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Shuffle LCO functions around, add telelco first cut

Pull LCO functions shared between LCO UI and command line into
ao_lco_funcs.c.

Import bits for telelco.

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Disable debug printfs and fix pad ignite time for telefire

Signed-off-by: Keith Packard <keithp@keithp.com>

altosdroid: Implement voice just like altosui

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: Move bluetooth check to first task

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: add rssi/serial/flight, and re-work UI ordering

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: rename ambiguous TextView name

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: formatting/whitespace, correct ids

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: rework lat/lon to more common format

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: rename azimuth/altitude to elevation/height respectively.

(Matches altoslib, altosui, altos)

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: move units into code, to match altosui

Signed-off-by: Mike Beattie <mike@ethernal.org>

Merge remote-tracking branch 'mjb/master'

Pull in Mike's fancy new AltosDroid bits

altos: spiradio debug serial is port 1, not port 0

The SPI link uses port 0; it seems like the having the two try to
share the same wires is a bad plan.

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: SPI radio - use 1->0 for 'done' and 0->1 for 'ready'

This changes how the SPI radio protocol uses the interrupt
line. Instead of a pulse indicating operation done, this now uses a 0
value for done and a 1 value for ready. The key distinction is that
the master can tell when the slave is waiting for the next command
instead of hoping that it got done 'soon enough'.

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: fix ao_pad debug output for query command

Igniter status is an array these days.

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Use updated pad protocol for lco commands

Stop using the older single-channel protocol and switch to the new
multi-channel protocol

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Note changes to configured AES key for SPI radio convenience

Keep a sequence number to mark when the AES key is changed so that the
radio code can avoid sending the key before every CMAC radio operation.

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Wait for IN dma complete on slave SPI send

SPI send double buffered, so the DMA completes one byte too early. Use
the recv DMA to know when the SPI transfer is complete.

Signed-off-by: Keith Packard <keithp@keithp.com>

altosdroid: initial attempt at a UI.

Signed-off-by: Mike Beattie <mike@ethernal.org>

altos: Oops. forgot ao_data.c

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Add spiradio Makefile

git add doesn't add Makefile by default.

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Mostly working SPI radio link

This includes long delays to avoid overrunning the cc1111 input,
otherwise it works pretty well. The delays mean that we can't capture
the reply to a cmac command though, so more work is needed.

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Include the radio_cmac debug commands in telefire

Just temporary debugging

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Explicitly erase memory in STM eeprom driver.

This seems to make the STM32L152 happier

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Track protocol changes to ao_pad debug messages

Make the debug output build again.

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Stop providing debug commands in seven-segment driver

The radio_cmac debug commands use the same letter, and this code works now...

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Move radio_cmac test funcs from ao_lco_cmd.c to new file

These are useful in the firing node for testing as well, so move them
to be shared.

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: When sharing radio DMA for AES, use it for in instead of out

We look at the out_done value, but not the in_done value; if we use
the radio DMA for out, we would have to use ao_radio_dma_done to check
for completion. This way, we can ignore that value and use the
existing ao_aes_dma_out_done value.

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Fix ao_delay function and move from per-chip code to ao_task.c

ao_delay hasn't been chip-specific for a long time, and it had a bug
in not calling ao_clear_alarm.

Signed-off-by: Keith Packard <keithp@keithp.com>

ao-stmload: Always round up load amount to 4 byte boundary

The flashing code doesn't deal with partial writes.

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Enable STM SYSCFG when routing EXTI

The EXTI routing information is in the syscfg unit, so that needs to
be powered up or writes to its registers will be lost.

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: No need to initialize EXTI priorities at startup time

They all get set to the correct value when enabled.

Signed-off-by: Keith Packard <keithp@keithp.com>

Merge branch 'master' of ssh://git.gag.com/scm/git/fw/altos

Conflicts:
debian/control

libelf-dev added to build deps

altosdroid: update Makefile.am for source file changes.

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: Add Dumper class for testing

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: add timer to stop service

* Stops when no UI clients, and no bluetooth connection remains

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: do service start/bind/unbind in start/stop, not create/destroy.

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: move methods around

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: tidy up old messages

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: stop sending device name, just send config data

* Kinda complicated, but ultimately more sensible
* Just send the config data as an arg to MSG_CONNECTED
* keep retrying connection till we actually get config data

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: whitespace

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: override add_reply() to add android based debugging

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: more logical name for connection_lost() method.

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: rework ConnectThread

* Start input_thread after connection, otherwise it's painful to kill
  on connect failure.

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: fix a connection retry having a null pointer

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: fix double call of stopAltosBluetooth()

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: Miscellaneous comments/debug/etc cleanup

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: miscellaneous cleanup

* Copyright info
* whitespace
* comments
* unused imports

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: Add passing of Device Config at connect

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: Add new "TelemetryReader" class to handle Telemetry

* Add MSG_TELEMETRY messages to both AltosDroid and TelemetryService
  to handle passing of AltosState object all the way back to the UI.
* Remove linkedblockinglist from TelemetryService
* (MSG_TELEMETRY is a rename of MSG_INCOMING_TELEM in AltosDroid)
* commented code in case statement inside AltosDroind - won't work with
  the objects it is currently passed.
* Add new "MSG_DEVCONFIG" message to AltosDroid - allows TelemetryService
  to pass information about the connected device back to the UI.

Signed-off-by: Mike Beattie <mike@ethernal.org>

altosdroid: create connected() method

Signed-off-by: Mike Beattie <mike@ethernal.org>

altos: Add spiradio product

Implements the SPI radio protocol

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: sdcdb rc file for telefire

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Add SPI linked radio API

Forward the necessary radio functions over the SPI link

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Clean up radio APIs

Move api to ao_radio_cmac.h include file.
Expose ao_radio_test as standard API.

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Build telelco and spiradio when possible

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Expose a signed version of the tick count

Useful when doing time comparisons.

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Specify the LCD duty cycle for stm-demo

This is the demo for the large 7-segment displays and needs static drive

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Test multiple quadrature devices. Export quadrature count.

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Seven segment display driver

Hex numbers and the decimal point.

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: More SPI slave changes for cc1111 driver

Don't enable DMA in the other direction when doing slave transfers.

Signed-off-by: Keith Packard <keithp@keithp.com>

altos: Add SPI slave get/put macros to cc1111

Theese don't try to drive the chip select line

Signed-off-by: Keith Packard <keithp@keithp.com>