aboutsummaryrefslogtreecommitdiff
path: root/Debugger/debug_comm.S
AgeCommit message (Collapse)Author
2011-07-06update comment on nak message processingTat-Chee Wan (USM)
2011-07-06the previous transmitted message checksum must be cleared before retransmittingTat-Chee Wan (USM)
The previous transmitted message contained a checksum suffix. It must be cleared before retransmission in order to use dbg__putDebugMsg() to retransmit.
2011-07-06attempt to fix gdb nak retransmit endless loopTat-Chee Wan (USM)
When GDB sends a NAK (-) but does not receive what it expects, it will keep trying to get a valid response. We just try to retransmit the last message in this case, hopefully GDB will be satisfied.
2011-06-22more refactoring, moved platform specific includes to debug_runlooptasks.hTat-Chee Wan (USM)
2011-06-22added message overrun protectionTat-Chee Wan (USM)
2011-06-21exported debug_outcommbuf and debug_incommbuf symbolsTat-Chee Wan (USM)
debug_runlooptasks.S need to access some symbols, so export them
2011-06-21moved platform specific code from debug_comm to debug_runlooptasksTat-Chee Wan (USM)
Reimplement debugger message handling code Moved platform specific code from debug_comm.S to debug_runlooptasks.S
2011-06-20declared variables and functions as globals for inter-module accessTat-Chee Wan (USM)
2011-06-17fix return status check for nx_usb_data_writtenTat-Chee Wan (USM)
The C return value for True is not identical to that used by ARMDEBUG. Check for False instead.
2011-06-16fix condition check in _dbg__sendCommMsg due to c vs. armdebug encodingTat-Chee Wan (USM)
True is encoded as #1 by C Compiler, while ARMDEBUG uses #0xFFFFFFFF. Changed logic test to avoid this problem.
2011-06-16revamped nxos debugger invocation interfaceTat-Chee Wan (USM)
2011-06-14moved hex conversion routines to its own source fileTat-Chee Wan (USM)
2011-06-06work in progress, cleanup nxos support codeTat-Chee Wan (USM)
Zero copy fantom message handling, don't reset fantom_message buffer until message has been examined.
2011-06-05use fantom_init to reset incoming message bufferTat-Chee Wan (USM)
2011-06-04work in progress to integrate fantom message filtering for nxosTat-Chee Wan (USM)
Filter Fantom Messages in NxOS usb_isr().
2011-05-27made variables needed by both nxt and nxos unconditionalTat-Chee Wan (USM)
2011-05-23don't reset nxtcommchannel in _dbg__comm_readbuf_reset since it is called ↵Tat-Chee Wan (USM)
after each received message
2011-04-21reenable nxos test compilation, added comment regarding buffer usageTC Wan
2011-04-15modified communications module to handle bluetooth comms for nxt firmwareTat-Chee Wan (USM)
2011-04-01update armdebug to handle ctrl-c messagesTat-Chee Wan (USM)
2011-04-01refactored message ack or nak routines to always send a reply to any gdb messageTat-Chee Wan (USM)
For GDB commands such as Continue or Step, no reply is expected. However, the GDB server needs to receive a reply to avoid hanging the read() command on the Host side. Consequently these 2 commands now send an Ack (no message) to keep the GDB server happy.
2011-03-28split out internal constants from debug_stub.hTat-Chee Wan (USM)
2011-03-24ascii2hex_varlen_be should not autoincrement buffer pointerTat-Chee Wan (USM)
2011-03-24rewrite char2hex to avoid trashing register r1Tat-Chee Wan (USM)
2011-03-24fix incoming message packet acknowlegement parsing logicTat-Chee Wan (USM)
2011-03-23cannot use teq for negative one constant value comparisonTC Wan
2011-03-23variable length big endian hex value conversion routine for gdb param parsingTC Wan
2011-03-23cannot use rol, covert to ror insteadTC Wan
2011-03-23rewrite multibyte ascii to hex conversion routines to take care of little ↵TC Wan
endian issues
2011-03-23work in progressTat-Chee Wan (USM)
2011-03-23changed output routines to use little endian as defaultTat-Chee Wan (USM)
2011-03-23fix char2hex to allow lower case ascii hex charsTat-Chee Wan (USM)
2011-03-22clarified comments regarding usb and gdb message formatsTat-Chee Wan (USM)
2011-03-22rework message handling logic to take care of packet acknowledgements sent ↵Tat-Chee Wan (USM)
by gdb server
2011-03-18second attempt at determining whether we are in debug modeTat-Chee Wan (USM)
Checking CPSR does not seem to be reliable in cCommHandleDebug which runs in Thumb mode. Modified to use explicit mode variable for keeping track of our operating mode (normal vs. debug).
2011-03-18we should only invoke bkpt in ccommhandledebug in normal operation modeTat-Chee Wan (USM)
Before invoking Breakpoint in cCommHandleDebug, we need to check if we're already in Abort (Debug) mode. Refactored header files to allow access to Mode Bit definitions from other files.
2011-03-15buffer pointer variable not accessible from debug_stub moduleTat-Chee Wan (USM)
Rewrite parameter passing for dbg__getDebugMsg to pass buffer pointer as return variable, to avoid accessing private variables from the debug_comm module.
2011-03-15misc bugfixes for gdb message parsing and error reportingTat-Chee Wan (USM)
Fixed gdb message parsing to access correct buffer address Fixed byte2ascii conversion routine
2011-02-21forgot to update _dbg__usb_readbuf_reset routine callsTat-Chee Wan (USM)
2011-02-21we can't use teq for true value since it is 32-bitsTat-Chee Wan (USM)
2011-02-21more refactory to support nxt firmware, cleanup codeTat-Chee Wan (USM)
2011-02-21checkin refactored code for nxt firmware supportTat-Chee Wan (USM)
2011-02-14made _copy_msg_to_usbbuf conditional since we need different logic for nxt ↵Tat-Chee Wan (USM)
firmware
2011-02-14initial nxt firmware support hooksTat-Chee Wan (USM)
2011-02-11added preliminary run loop separationTat-Chee Wan (USM)
2011-01-18implemented get/set registers and insert/remove breakpoints commandsTat-Chee Wan (USM)
2011-01-18fixing command decoder after changes made to comm routinesTat-Chee Wan (USM)
2011-01-12implement usb message fragmentation and send logicTat-Chee Wan (USM)
2011-01-12checkin wipTC Wan
2011-01-12checkin wipTC Wan