No functionality changed. Clean up only.
Make early init code (Reset_Handler) use the existing irq stack
rather than the hard-coded one.
Remove (now) unused references to the hard-coded stack.
Removed the link-time check for a minimal task stack in RAM since
FreeRTOS allocates user stacks from the heap which is already reserved
in the .bss section (as symbol xHeap).
git-svn-id: svn://svn.openpilot.org/OpenPilot/trunk@756 ebee16cc-31ac-478f-84a7-5cbb03baadba
Split the floss-jtag.cfg file into separate versions
for OP and AHRS.
Push AHRS onto non-default ports for gdb, tcl and
telnet.
Update the AHRS gdb setup script to point at the new
gdb port.
Add (commented out) example sytax to support distinguishing
between multiple floss-jtag boards that don't have serial
numbers. Uses the usb bus address of each device as the
selector. See this patch posted to the openocd mailing list
for how to add this functionality to openocd:
http://lists.berlios.de/pipermail/openocd-development/2010-June/015785.html
git-svn-id: svn://svn.openpilot.org/OpenPilot/trunk@755 ebee16cc-31ac-478f-84a7-5cbb03baadba
New map widget is able to display maps from: GoogleMap,GoogleSatellite,GoogleLabels,GoogleTerrain,GoogleHybrid,GoogleMapChina,GoogleSatelliteChina,GoogleLabelsChina,GoogleTerrainChina,GoogleHybridChina, OpenStreetMap,OpenStreetOsm,OpenStreetMapSurfer,OpenStreetMapSurferTerrain,YahooMap,YahooSatellite,YahooLabels,YahooHybrid,BingMap,BingSatellite,BingHybrid, ArcGIS_Map,ArcGIS_Satellite,ArcGIS_ShadedRelief,ArcGIS_Terrain,ArcGIS_MapsLT_Map,
ArcGIS_MapsLT_OrtoFoto,ArcGIS_MapsLT_Map_Labels,ArcGIS_MapsLT_Map_Hybrid, PergoTurkeyMap,SigPacSpainMap,GoogleMapKorea,GoogleSatelliteKorea, GoogleLabelsKorea,GoogleHybridKorea,YandexMapRu
git-svn-id: svn://svn.openpilot.org/OpenPilot/trunk@753 ebee16cc-31ac-478f-84a7-5cbb03baadba
QT Creator windows binaries are compiled with MSVC and the plugin would not be compatible when built with mingw so I opted not to go down that road.
Current test project is "finaltest".
Zoom, Pan, Rotate, OpenGL choice, geodecoding, map reloading... all working.
git-svn-id: svn://svn.openpilot.org/OpenPilot/trunk@752 ebee16cc-31ac-478f-84a7-5cbb03baadba
before
"**** Compiling C :" ../PiOS/STM32F10x/pios_sys.c "->" Build/pios_sys.o
after
**** Compiling C : ../PiOS/STM32F10x/pios_sys.c -> Build/pios_sys.o
git-svn-id: svn://svn.openpilot.org/OpenPilot/trunk@731 ebee16cc-31ac-478f-84a7-5cbb03baadba
TODO: make fonts look nicer - if anyone can recommend a font chooser ?
git-svn-id: svn://svn.openpilot.org/OpenPilot/trunk@727 ebee16cc-31ac-478f-84a7-5cbb03baadba
Stac, please let me know. Yeah, I should do this in a patch, but this change should do no harm as it keeps Windows working and Linux in theory. It is cosmetic for the list of ports.
git-svn-id: svn://svn.openpilot.org/OpenPilot/trunk@721 ebee16cc-31ac-478f-84a7-5cbb03baadba
This change adds '/dev' to the beginning of the portName
on Linux. This doesn't seem quite right but using physName
breaks serial port functionality on Windows.
git-svn-id: svn://svn.openpilot.org/OpenPilot/trunk@714 ebee16cc-31ac-478f-84a7-5cbb03baadba
On Linux, port.portName is the simple name of the serial port
like 'ttyS0'. The physical name of the device is '/dev/ttyS0'.
This value is stored in port.physName. This bug doesn't get
in the way on Windows since 'COM1' is both the simple name as
well as the physical name. On Linux, the distinction is
important.
git-svn-id: svn://svn.openpilot.org/OpenPilot/trunk@710 ebee16cc-31ac-478f-84a7-5cbb03baadba
nmeaProcess was attempting to null-terminate the
NMEA sentence but was not considering that the
preceeding loop may have looped beyond the end
of the packet buffer.
git-svn-id: svn://svn.openpilot.org/OpenPilot/trunk@709 ebee16cc-31ac-478f-84a7-5cbb03baadba
