drath [Mon, 18 Feb 2008 17:59:06 +0000 (17:59 +0000)]
- ignore data aborts during gdb memory read packets by default, and return 0x0 instead
- add configuration option "gdb_report_data_abort <enable|disable> to enable reporting of data aborts during gdb memory read packets
- thanks to Øyvind Harboe for identifying the problem with GDB and data aborts
- fix some warnings in gdb_server.c
ntfreak [Mon, 18 Feb 2008 15:41:38 +0000 (15:41 +0000)]
- fix broken JTAG error handling
- Allow all commands to be executed during config stage
- Help now works for config commands
- make var args handling follow the rules more closely
Thanks Øyvind Harboe
ntfreak [Mon, 18 Feb 2008 15:11:01 +0000 (15:11 +0000)]
- removed a raft of unecessary exit() calls. Issuing a reset will solve these ails.
- now uses jtag_state_pathmove() instead of making assumptions about implementation of jtag_statemove().
- fixed a couple of bugs in timeout handling
- removed megabytes of log output when communication is failing.
- sleep is now 300ms as documented instead of 3000ms
- fixed error path of bulk write
- debug_handler can now be issued during normal operation + has help text.
Thanks Øyvind Harboe
ntfreak [Mon, 18 Feb 2008 14:32:43 +0000 (14:32 +0000)]
- Added a "User:" debug level. These are messages that are intended for the user and are part of normal operation.
- Faster DEBUG/INFO() when they are disabled
- target_read/write_buffer() now uses 16 and 32 bit access for single word aligned requests. Other requests are serviced as quickly as possible.
- *much* faster read/write GDB packets, removing timeout problems.
- GDB read/write packets w/single word aligned 32/16 bit access now use 32/16 bit word access.
- working area can now be changed on the fly. Provides a way to move working area about as MMU is enabled/disabled.
- cleaned up error messages for verify_image.
Thanks Øyvind Harboe
ntfreak [Sat, 16 Feb 2008 15:21:13 +0000 (15:21 +0000)]
- solve lots of problems with stuck GDB connections, making it impossible to connect to GDB, handle timeout of acknowledgement, etc.
- "monitor halt/resume" now works correctly
- "monitor sleep 10000" no longer makes the GDB protocol lock up. There is an error message and the protocol recovers nicely afterwards.
- it's now possible to connect to a target which needs a reset before halt works.
- handle failed memory access more gracefully. Connection is now closed instead of OpenOCD quitting.
- *much* improved handling of 2 second timeout on memory read packets. Especially important w/mouseover evaluation of variables in Eclipse.
- fixed memory leak upon failed memory packet reply.
- 'O' packets w/progress info is no longer sent out randomly.
- faster packet reply code.
- Thanks to Øyvind Harboe for this patch
ntfreak [Thu, 7 Feb 2008 20:20:11 +0000 (20:20 +0000)]
- added synchronous reset patch, Thanks Øyvind Harboe
- added target_init_reset which calls target_process_reset after all drivers have been initialised
mifi [Sat, 2 Feb 2008 13:42:37 +0000 (13:42 +0000)]
- added patch to solve the reset problem with arm9 and cortex which Spen had found.
https://lists.berlios.de/pipermail/openocd-development/2008-January/000729.html
mifi [Sun, 20 Jan 2008 13:52:06 +0000 (13:52 +0000)]
- added patch "remove error handler as planned"
https://lists.berlios.de/pipermail/openocd-development/2008-January/000665.html
(thanks to oyvind Harboe for the patch)
ntfreak [Wed, 9 Jan 2008 15:58:01 +0000 (15:58 +0000)]
- rename flash_init and flash_erase to flash_init_drivers and flash_erase_address_range - stops conflicts with redboot. Thanks Øyvind Harboe
- gdb connection not dropped if we fail to allocate memory in query packets
ntfreak [Sat, 29 Dec 2007 13:51:48 +0000 (13:51 +0000)]
- minimum autoconf 2.59 is now required and verified - due to issues with AS_HELP_STRING
- native win32 now handles WSAECONNRESET - no longer exits openocd
- qCRC packet now works correctly under cygwin (gdb compare-sections command)
- removed __USE_GNU define from gdbserver.c
- gdb qSupported packet is now handled, with this we are able to tell gdb packet size, memory map of target
- added new target script gdb_program_config - called before gdb flash programming
- new gdb server command gdb_memory_map (enable|disable> - default is disable
- new gdb server command gdb_flash_program (enable|disable> - default is disable
- gdb flash programming supported - vFlash packets
- image_elf_read_section now does not clear any remaining data, this was causing the gdb checksum to fail with certain files
- reformat of usbprog.c
- memory leak in command_print fixed
- updated texi doc to include new commands
- added gdb programming section to docs
ntfreak [Thu, 20 Dec 2007 16:19:10 +0000 (16:19 +0000)]
- removed flash write_image - binary compare function has been moved to verify_image command
- minor code reformat and cleanup
- updated docs to include new commands
mifi [Tue, 18 Dec 2007 21:20:28 +0000 (21:20 +0000)]
- added patch for new flash functionality like:
flash verify_image and flash erase_address.
- added patch for new parport_write_on_exit command.
Even this patch will fix some memory leaks.
mifi [Mon, 17 Dec 2007 20:52:37 +0000 (20:52 +0000)]
- fixed relocation problems with image loading. Relocation is handled
in a single centralized place, and it now works for binaries.
(thanks to oyvind Harboe)
mifi [Mon, 17 Dec 2007 17:00:24 +0000 (17:00 +0000)]
The changes allows the "script" command to be used from within
config scripts to run sub-config scripts, which was probably the
original intention because "script" is registered as COMMAND_ANY.
This can be useful in e.g. splitting CPU and flash config scripts.
mifi [Sun, 16 Dec 2007 18:58:16 +0000 (18:58 +0000)]
- removed some compiler warnings
- added patch to fix problem in cfi_intel_write_block. In case of
cfi_info->write_algorithm is not NULL, target_code_size was not set.
(thanks to oyvind Harboe for the patch)
mifi [Fri, 14 Dec 2007 19:15:32 +0000 (19:15 +0000)]
Sorry, I have forgotten to add a description for the last version, where
I have added the clear_malloc and fill_malloc functionality. But the default
is the original malloc.
mifi [Fri, 14 Dec 2007 18:06:19 +0000 (18:06 +0000)]
- added patch from Dominic to unlock the ATMEL flash in cfi_read_atmel_pri_ext
- set cfi_info->write_algorithm and cfi_info->erase_check_algorithm to NULL to
get "flash erase_check" command working
mifi [Mon, 10 Dec 2007 19:46:04 +0000 (19:46 +0000)]
- Fixing two compiler warnings
- Reducing stack usage for recursive scripts
- Do not exit on bogus arguments to reset_config. No longer exit the application upon bogus arguments to reset_config, but return errors.
mifi [Sun, 9 Dec 2007 15:36:21 +0000 (15:36 +0000)]
- added patch to display device information as INFO too
- added patch which fixes a crash upon flash write error
- added patch which will improve the reset handling when
SRST is tied to TRST
bodylove [Wed, 5 Dec 2007 13:34:02 +0000 (13:34 +0000)]
Merged rev 215 changes from /branches/xscale-ixp-be into trunk:
- Obvious fixes to big endian type conversion macros
- Fixed obvious typos for byte masks
ntfreak [Wed, 21 Nov 2007 16:37:17 +0000 (16:37 +0000)]
- reset_run now works as expected on cortex-m3
- str9xpec erase status checked on option byte programming
- stm32x flash driver now supports sector protection
- surplus exit removed from flash.c
- openocd.texi documentation added
ntfreak [Mon, 8 Oct 2007 13:12:39 +0000 (13:12 +0000)]
- cleaned up str7, str9 and stm32 flash drivers
- str7 flash driver now checks correct busy bits depending on device
- str9 flash driver now disables ITCM order as per st programming manual
- added str7 disable_jtag command
- added gdb_detach command
- updated arm966e cp15 support
- fix crash on mingw build when enabling target_request debugmsgs
ntfreak [Mon, 1 Oct 2007 08:31:30 +0000 (08:31 +0000)]
- flash autoerase <on|off> cmd added, default is off - flash banks are calculated and erased prior to write (flash write_image only)
- corrected array overrun in armv7m.c
- corrected breakpoint memory allocation bug
- image read now uses fgets, vastly improves reading of large files
- improved hex file reading, support for Linear Address Record added
- Fixed display of sector sizes in flash.c
- Clean up, remove unused variables and code in armv7, cortex_m3 and stellaris code
- Move restore_context from cortex_m3 to armv7m
- Updated halt handling for cortex_m3
Patch by Michael Schwingen that
- adds support for the Altium universal JTAG cable
- adds support for "wiggler2" cable (basically a wiggler with added LED, documentation coming soon)
- adds LED support. The LED is turned on during data transfer - works fine on Altium and wiggler2.
- adds PORT_EXIT pattern that is written to port when exiting, in order to turn off power on cables that get their power from parallel port data lines
- move port writes (with the system-specific ifdefs) to one central function
- increased image cache size to 2KB (might require more adaptive cache handling, e.g. LRU)
drath [Sat, 25 Aug 2007 09:59:42 +0000 (09:59 +0000)]
- added support for setting JTAG frequency on ASIX PRESTO (thanks to Pavel Chromy)
- usbprog update (thanks to Benedikt Sauter)
- added embeddedice_send and _handshake functions (thanks to Pavel Chromy)
- added support for 4, 8 and 16 bit ports to etb.c
projects / openocd / log