- Jun 23, 2010
-
-
Dean Camera authored
Split RFCOMM channel signals into Remote and Local pairs. Use memcpy() to copy over data from the command parameters to the RFCOMM response parameters. Ensure that only the valid parameter bytes are sent back in response to MSC commands.
-
Dean Camera authored
Add TEST RFCOMM command handler. Remove the RFCOMM channel UseUIFrame element, as the Bluetooth adaptions to RFCOMM only allow UIH frames to be used.
-
- Jun 22, 2010
-
-
Dean Camera authored
Use a bitmask instead of an array of bools in the SDP UUID List matching algorithm to reduce the compiled code size and stack used.
-
Dean Camera authored
Added new SWAPENDIAN_16() and SWAPENDIAN_32() macros to Common.h for statically initialized variables at compile time.
-
- Jun 16, 2010
-
-
Dean Camera authored
-
- Jun 15, 2010
-
-
Dean Camera authored
Rename RFCOMM_GetFrameDataLength() to RFCOMM_GetVariableFieldValue() as it is multi-purpose, and modify it to move the buffer pointer itself rather than relying on the caller. Make RFCOMM_GetChannelData() return NULL if the required channel DLCI is found, but the channel is closed. Add modem signals (BREAK, DCD, etc.) to the channel state array, and add in a handler for the Modem Status RFCOMM control command.
-
Dean Camera authored
Changed the RFCOMM-specific ItemProtocolChannel_t type to a more generic ItemProtocol_8BitParam_t name, and added a new ItemProtocol_16BitParam_t to the SDP protocol type defines.
-
Dean Camera authored
Add missing BT_* prefix to the Bluetooth stack's channel state enum values.
-
Dean Camera authored
-
- Jun 13, 2010
-
-
Dean Camera authored
Split out the RFCOMM Control Channel command processing code into a seperate set of files for clarity.
-
Dean Camera authored
Rename the incomplete Bluetooth Host demo's ServiceDiscoveryProtocol.c/.h files to SDP.c/.h. Fix compile errors in RFCOMM.c/.h.
-
Dean Camera authored
-
- Jun 12, 2010
-
-
Dean Camera authored
Minor fixes to demos to add in some missing comments, printf() formatting. Make incomplete Bluetooth demo indicate connections visually via the board LEDs.
-
- Jun 07, 2010
-
-
Dean Camera authored
Add stub RFCOMM command handlers. Move out RFCOMM frame length and data pointer calculations to the master RFCOMM frame reception routine, instead of inside each frame type handler function.
-
Dean Camera authored
More RFCOMM layer implementation - fix FCS calculation, add in start of control channel command processing.
-
Dean Camera authored
Make the RFCOMM_SendFrame() function correctly generate frames from the input parameters, change addressing to use the raw DLCI address plus flags.
-
- Jun 06, 2010
-
-
Dean Camera authored
-
Dean Camera authored
-
Dean Camera authored
Add the svn:eol-style property onto new files which where missing it to ensure uniform line endings.
-
- Jun 04, 2010
-
-
Dean Camera authored
Document Bluetooth files with overall file Doxygen comments. Add more initial RFCOMM layer code to receive and respond to SABM packets.
-
- Jun 03, 2010
-
-
Dean Camera authored
-
Dean Camera authored
-
Dean Camera authored
Start implementation of RFCOMM decoding layer. Add new service initialize routines for each Bluetooth service - currently unused, but will eventually reset the service state of RFCOMM when a new connection is made.
-
Dean Camera authored
Delay needed in bluetooth code due to the send function not waiting for the pipe bank to become ready before writing. Add in a call to Pipe_WaitUntilReady() to fix the issue.
-
Dean Camera authored
Extra protocol value in the RFCOMM protocol is the channel number, not the PSM - update typedef name and size. Add delay to the Connection Request signal processing routine - without this, the host will not create the RFCOMM channel correctly. Need to investigate why this delay is needed.
-
Dean Camera authored
Add PSM value to the RFCOMM service so that the host knows what PSM to use on channels when trying to access it.
-
Dean Camera authored
Reduce the size of the attribute search list from 15 ranges to 8 to save RAM.
-
- Jun 02, 2010
-
-
Dean Camera authored
Fix byte ordering of UUIDs in the SDP server - host can now successfully pair to the Bluetooth device and discover the exposed Serial Port Profile RFCOMM service.
-
Dean Camera authored
The SDP UUID lists should be searched and ALL UUIDs matched for a record to be retrieved, not partial matches. Change the SDP code so that the entire list must be matched against a service attribute table's contents before it is returned. Change matching algorithm so that it recursively searches through the entire attribute table, and not just pre-specified sequence attributes. Add browse lists and proper descriptions to the Serial Port service.
-
Dean Camera authored
Add L2CAP service attribute table. Fix up ProtocolDescriptorList attributes, and add macros for used UUIDs to reduce typing and make the code easier to read.
-
- Jun 01, 2010
-
-
Dean Camera authored
-
Dean Camera authored
Remove incorrect Language ID offset attribute from the SDP server - client assumes 0x0100 offset for primary language anyway, and the format of the offset was incorrect anyway.
-
Dean Camera authored
Service Attribute SDP requests should have only one Data Element Sequence container around the single service's retrieved attributes, not two.
-
Dean Camera authored
-
Dean Camera authored
Much faster attribute range lookup - look through short list of attributes and compare to range, rather than looking up by each possible value within the range. Cleanups to SDP code. Add missing RFCOMM language base ID offset attribute. Fix incorrect definition of the SWAPENDIAN_32() macro.
-
Dean Camera authored
-
Dean Camera authored
-
Dean Camera authored
-
- May 31, 2010
-
-
Dean Camera authored
Fix endianness issues in the Service Discovery Protocol service of the incomplete Bluetooth Host demo. Host can now correctly retrieve specific services, however Windows does not appear to be sending the appropriate generic browse class IDs needed to report the virtual serial service.
-
- May 29, 2010
-
-
Dean Camera authored
Partial fix to the Bluetooth SDP code - data should be encoded in big endian, not little endian.
-