Menu
▾
▴
[openocd-svn] Main OpenOCD repository annotated tag, v0.2.0, created. v0.2.0
|
From: David B. <dbr...@us...> - 2009-10-21 08:51:35
|
This is an automated email from the git hooks/post-receive script. It was
generated because a ref change was pushed to the repository containing
the project "Main OpenOCD repository".
The annotated tag, v0.2.0 has been created
at 772a80f0d6065f45efdf4efd9fc80982c0a42080 (tag)
tagging 64d998e584e0140fff2ead4dfe09f2d14b43c89e (commit)
tagged by David Brownell
on Tue Jul 14 09:00:00 2009 -0700
- Log -----------------------------------------------------------------
Milestone for 0.2.0 source release.
bodylove (12):
Merged rev 215 changes from /branches/xscale-ixp-be into trunk:
Changed openocd version info creation to be more useful.
- XScale DebugHandler code is now installed into
- Added PKGLIBDIR define to C options. Allows to store
- Enabled URL keyword expansion for enhanced version
- Stripped trailing spaces (patch preparation)
- Eleminated leading tabs/white space
- Eleminated leading tabs/white space
- Eleminated leading tabs/white space
- Synced code with branch
[MERGE] Merged all remaining changes between trunk and xscale big-endian branch.
- Added example config and startup-script for a XScale IXP42x system
drath (109):
- prepare OpenOCD for branching, created ./trunk/
- fixed some spelling errors (thanks to Andrew Dyer)
- the 'help' command now takes an optional argument to display help only on a certain command (thanks to Andrew Dyer for this enhancement)
- keep additional information for decoded instructions
- added support for AT91SAM7A3 flash (patch from andre renaud, thanks)
- added "version" command (patch from John Hartman, thanks)
- changed use of bzero (deprecated) to memset (thanks to Spen for pointing this out)
- fixed bug in Thumb sw breakpoint handling (thanks to Spen for this patch)
- Added support for native MinGW builds (thanks to Spencer Oliver and Michael Fischer) - you still need to install GiveIO (not part of OpenOCD)
- added configurable delays after reset lines get deasserted. useful if reset circuitry keeps lines asserted for too long.
- allow building for MinGW using either -mno-cygwin or the MinGW gcc
- fixed a minor problem with the GDB server that could drop the first packet (non-fatal)
- fixed jtag_n[st]rst_delay. time is now miliseconds (as documented), not microseconds
- added support for FreeBSD ppi (parallel port access similar to /dev/parport on linux) (thanks to Darius for this patch)
- added support for the Signalyzer USB->JTAG dongle (www.signalyzer.com)
- updated configuration examples, installation instructions and README (including list of supported JTAG interfaces)
- fixed bug in .cfg files. Thanks to Michael Fischer for noticing this.
- renamed jtag_interface_t.support_statemove to jtag_interface_t.support_pathmove (it is used to indicate jtag_add_pathmove support)
- updated version info in openocd.c
- made bp command more verbose
- endianess fixes everywhere but in the flash code. flashing might still be broken on big-endian targets and/or hosts
- added missing AT91RM9200 files
- made lpc2000 code endianness safe (support big-endian hosts, target is always little)
- added debug output for D/I FSR and FAR (arm920t)
- integrated patch from Magnus Lundin that fixes at91sam7 flash timing bugs and possible endianness problems (big endian hosts)
- free working area used to store flash write buffer
- set the T bit in the CPSR when the core was executing in Thumb state. Mask the T bit from the CPSR for all CPSR-writing operations
- allow writes to second flash bank on STR71x devices (cleaned up sector list building)
- str9x flash support (Thanks to Spencer Oliver)
- fixed regression in gdb_server.c (Thanks to Michael Fischer for finding these bugs)
- added support for American Microsystem's M5960 (FT2232 based USB JTAG interface)
- allow FT2232 devices to be opened by serial number instead of device description ('ft2232_serial <serial>' command)
- fixed endianness handling in arm7_9_read_core_reg (thanks to Magnus Lundin for reporting this)
- several small fixes (thanks to Magnus Lundin and Spencer Oliver)
- correctly enter debug state on a "soft_reset_halt" command
- added support for the oocd-link (http://www.joernonline.de/dw/doku.php?id=en:projects:oocdlink)
- added a PLD (programmable logic device) subsystem for FPGA, CPLD etc. configuration
- checked in missing configure.in changes for GW16012 support
- disabled FT2232 debug output (that code should be fairly stable)
- fix incorrect parsing of whitespace in command.c (thanks to Magnus Lundin)
- Added support for ARM926EJ-S based cores
- disabled JTAG debug output
- disabled use of single-step bit for EmbeddedICE version 6 cores
- merged several changes from XScale
- add missing parentheses around macro parameters (thanks to Matthias Bauch for noticing this bug and providing a fix)
- fixed libftdi initialization
- reworked file i/o. every fileaccess (target, flash, nand, in future configuration, too) should now go through the fileio subsystem
- added missing files for last commit
- fixed arm926 cp15 command bug (thanks to Vincent Palatin for this patch)
- merged XScale branch back into trunk
- added "arm9tdmi vector_catch ['all'|'none'|'vec1 [vec2 [...]]']" command
- added support for Turtelizer 2 (USB JTAG interface) (Thanks to Michael Fischer for this patch)
- disabled excessive debug output in jtag.h
- correctly mask out bits that aren't part of a copied buffer
- allow multiple USB vendor and product ids when searching for connected FT2232 based JTAG interfaces (thanks to Werner Almesberger for this patch)
- explicitly disable monitor mode on ARM7/9 targets
- added support for error handlers to JTAG scan commands (jtag_[plain_][ir|dr]_scan)
- improved ETB trace output
- further work on ETB decoding (not yet functional)
- split fileio handling into fileio part and image handling
- add missing image handling files
- add missing file src/target/trace.h
- reworked image handling to support multiple sections (tested with ihex file containing gaps)
- add support for ELF images (thanks to Vincent Palatin for this patch)
- added manpage for OpenOCD (thanks to Uwe Hermann)
- merged support for Cortex-M3 from cortex-m3 branch (thanks to Magnus Lundin)
- added stellaris.[ch] missing from Cortex-M3 merge
- added support for pseudo image type "mem", currently only used for etm analysis: "etm image <target#> <dummy base address> mem"
- added support for Intel/Marvel PXA27x (XScale) targets
- fixed endianness helper macros (thanks to obilix and wiml for finding and fixing this bug)
- added support for Asix Presto JTAG interface (thanks to Pavel Chromy and Asix for making this addition possible)
- reformat src/jtag/bitq.c (thanks to Pavel Chromy)
- documentation fixes (thanks to Uwe Hermann)
- restrict direct parallel port access to x86 platforms (thanks to Vincent Palatin)
- calculate cycles since last executed instruction when cycle-accurate tracing is enabled
- add missing configure.in change from last checkin
- cleaned up cycle counting in ETM analysis
- renamed M5960 USB JTAG to "flyswatter"
- changed mingw host detection to *-mingw instead of *-*-mingw to work with i586-mingw32msvc-gcc
- bumped version information
- reworked presto.c to allow use of either FTD2XX or libftdi (libftdi not functional yet). Configure option changed from --enable-presto to
- corrected automake script
- fixed several bugs in flash writing code (thanks to Pavel Chromy)
- update jtag_speed setting when changing it during runtime with a FT2232 based interface
- renamed "xscale dump_trace_buffer" to "xscale dump_trace" and added code for it
- update file size when writing to a fileio
- correctly initialize start address for XScale trace buffer decodes in fill-once mode
- added support for setting JTAG frequency on ASIX PRESTO (thanks to Pavel Chromy)
Patch by Michael Schwingen that
- fixed ETM configuration register decoding
- use correct SCAN_N check value (disabled by default)
- added tms470 flash support (thanks to Chris Kilgour for this patch)
- fixed 'make distcheck' (thanks to Theodore A. Roth for this patch
- ignore data aborts during gdb memory read packets by default, and return 0x0 instead
- Fix problems with stuck telnet sessions. Thanks to Ãyvind Harboe for this patch.
- add support for the majority of the Samsung ARM SoC family, S3C2410, S3C2412, S3C2413, S3C2440 and S3C2443 (thanks to Ben Dooks for this patch)
- Use NAND extended geometry information (thanks to Ben Dooks for this patch)
- fix for feroceon CP15 register access (thanks to Niolas Pitre for this patch)
- fixes possible crash when GDB connection is closed while target is running due to log callback sending messages to connection which does not
- concretize JTAG state transition rules (previously implied behaviour is now documented)
- convert all files to unix line-ending
- fix bug in ft2232 pathmove (thanks to Michael Bruck for the patch)
- fix typo in ep93xx jtag driver to allow OpenOCD compilation on ARM (thanks to Uwe Hermann for the patch)
- fix line-endings
- Added coding style to README
- fix warnings during configure cause by ecosboard. default to no, if host cpu isn't arm.
- Flash auto-erase is disabled by default
- fixed xsvf_add_statemove()
- fix a off-by-one error in the buffer read/write code that checks for a address wrap
duane (103):
Test checkin from commandline
jtag newtap change & huge manual update
Fix for Hiroshi Ito discovery of mis-aligned memory allocation
Manual tweak based on feedback from mailing list the iar dongle works
Added Joern Kaipf link for oocdlink
Test of commit email from duane
Typos in config files found by Sergey Lapin
Fix from Kees Jongenburger
Fix from Kees Jongenburger (Part Duex)
More fixes thanks to Kees Jongenburger
Updates and fixes from Kees Jongenburger
Typos found by miceal catudal
Build Permutations with ftd2xx and libftdi addressed. Also added a new se of regression makefiles to build openocd in multiple ways
Renamed build.tests to build.test1
Added documentation for build test case1
From Dirk Behme - Further docu fixes
Warning removal from lou...@fi...
Added dongle VSLLINK - from Simon Qian
Quoteify the FT2232 description list, and a missed fix from Dirk Behme about jtag tapisenabled
From Lou...@fi...
Missed support for without --exec-prefix and ftd2xx stuff
Some systems report linux as host, others linux-gnu... grrr
From Dirk Behme - another set of typos
Added note to user manual for packagers of OpenOCD
Tweaks - old crusty code with some warnings & errors - minor stuff
Missed the svn add on earlier commit, duh
Warnings cleanup ... finish up earlier commit
Added another test build configuration
Updates from SimonQian
Added HostOS variable
Accept/create both A and Non-A ft2232 based descriptions
Patch from Dimitar Dimitrov adding support for Olimex ARM-JTAG-EW
Added VID/PID pair to olimex-jtag-tiny-a the non-a version already has the vid pid
Commands: reg, profile, ocd_mem2array, ocd_array2mem, fast_load, etc only work *IF* there is an actual target
Added -endstate to irscan and drscan to support beagleboard (omap3530)
Commit OMAP3530 and TI_BEAGLEBOARD config files from Kees, Dick, Derk, and others
Added eol-style props
Remove warning
commit test from Duane
printf() warning fixes due to uint32_t change-over
C99 Type updates, include inttypes.h - it is catagorically required
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
C99 printf() -Werror fixes
Add a growable sprintf memory buffer library
Add quick target - full cygwin builds take a long long long time, this shortens the edit/build/debug cycle
Add support for ATMEL AT91SAM3U - CortexM3 Family
Fixes from Oleksandr Tymoshenko "go...@bl..."
Switch to strotk() grr....
Change from alloca() to malloc()
Add Breakpoint/Watchpoint unique ID to help debug hardware debug register leakage
Refactor code, create target_state_name()
A bit more log detail about connections comming and going
Remove extra newline from debug log message
Simple warning fix
kc8apf (94):
AXM0432 layout for FDTI provided by Alan Carvalho de Assis (Freescale)
Allow -expected-id to be specified multiple times when creating a jtag tap
Change tap_state naming to be consistent with SVF documentation.
Clean up references to old tap_state names
Add JTAG tap events for enable/disable
- Update tap_state_strings to correspond to SVF state names
- rlink interface support from Lou Deluxe <lou...@fi...>
Fix distcheck failure
Fix to install manpage as part of 'make install' by Uwe Hermann <uw...@he...>
Fix guess-rev.sh where the builtin echo doesn't support -n
Fix handling of thumb disassembly in armv4_5 disassemble. Courtesy of Adam Dybkowski <ady...@wp...>
SVF player courtesy of Simon Qian <sim...@Si...>
imx21 config provided by Alan Carvalho de Assis <ac...@gm...>
Fix support for ADuC702x flash. Courtesy of Michael Ashton <da...@gt...>
Ignore src/svf/Makefile.in
Fix 'make maintainer-clean' courtesy of Zach Welch <zw...@su...>
Fixes for handling release versions
Add axm0432 interface config courtesy of Alan Carvalho de Assis <ac...@gm...>
AT91SAM9260 and Olimex changes by Dean Glazeski <dn...@gm...>
Fix for incorrect filename in include for at91sam9260 in unknown-board-atmel-at91sam9260.cfg
Add udev rules to contrib. Courtesy of Uwe Hermann <uw...@he...>
vsllink support for stable clocks by Simon Qian <sim...@Si...>
Add TRST support to xsvf tools. Courtesy of Dick Hollenbeck <di...@so...>
Search path fixes for MinGW builds. Courtesy of Dimitar Dimitrov <di...@gm...>
Fixes for svf player by Simon Qian <sim...@Si...>
xvsf player fixes by Dick Hollenbeck <di...@so...>
Add uncrustify config file and helper script
Doxygen config file
- Cable driver helper API courtesy of Dick Hollenbeck <di...@so...>
A few more usages of TRUE and FALSE to remove
Use C89/C99/C++ compliant boolean types
refactor jlink_execute_queue courtesy of Zach Welch <zw...@su...>
fix jtag run/idle state name courtesy of Jeff Williams <je...@ga...> and Zach Welch <zw...@su...>
Fix use of wrong format conversion for size_t (%zu instead of %u)
Add D2XX support for OS X courtesy of Piotr Esden-Tempski <pi...@es...>
Shrink JLink buffer sizes to specified 2KB courtesy of Jeff Williams <je...@ga...> and Zach Welch <zw...@su...>
Fix environ declaration for non-glibc systems
Whitespace cleanup from David Brownell <da...@pa...>
ETM/ETB documentation from David Brownell <da...@pa...>
Consolidate target selection code into single get_target() that handles both names and numbers. Provided by David Brownell <da...@pa...>
Fix logically inverted comment
Relocate documentation on working area to better align with use of new syntax. Provided by David Brownell <da...@pa...>
JTAG state table updates (short table still disabled). Provided by Dick Hollenbeck <di...@so...>
PATCHES updates from David Brownell <da...@pa...>
ftdi_set_interface correctness by Strontium <str...@gm...>
Add jtag_queue_command() by Michael Bruck <mb...@di...>
Add jtag_queue_command() by Michael Bruck <mb...@di...> [2/8]
Add jtag_queue_command() by Michael Bruck <mb...@di...> [3/8]
Add jtag_queue_command() by Michael Bruck <mb...@di...> [4/8]
Add jtag_queue_command() by Michael Bruck <mb...@di...> [5/8]
Add jtag_queue_command() by Michael Bruck <mb...@di...> [6/8]
Add jtag_queue_command() by Michael Bruck <mb...@di...> [7/8]
Add jtag_queue_command() by Michael Bruck <mb...@di...> [8/8]
Change last_comand_pointer to last_command_pointer by Michael Bruck <mb...@di...>
Fix fallout from r1818
FT2232 support for non-7-cycle state moves by Dick Hollenbeck <di...@so...>
JLink support for non-7-cycle state moves by Dick Hollenbeck <di...@so...>
Enable non-7-cycle state table for FT2232 and JLink
Author: Michael Bruck <mb...@di...>
Author: Michael Bruck <mb...@di...>
Author: Michael Bruck <mb...@di...>
Author: Michael Bruck <mb...@di...>
Author: Michael Bruck <mb...@di...>
Author: Michael Bruck <mb...@di...>
Author: Michael Bruck <mb...@di...>
Author: Michael Bruck <mb...@di...>
Author: Michael Bruck <mb...@di...>
Author: Michael Bruck <mb...@di...>
Author: Spencer Oliver <sp...@sp...>
Author: Michael Bruck <mb...@di...>
Author: Michael Bruck <mb...@di...>
Author: Michael Bruck <mb...@di...>
Author: Michael Bruck <mb...@di...>
Author: Michael Bruck <mb...@di...>
Author: Michael Bruck <mb...@di...>
Author: Michael Bruck <mb...@di...>
Author: Michael Bruck <mb...@di...>
Author: Michael Bruck <mb...@di...>
Author: Michael Bruck <mb...@di...>
Author: Holger Schurig <hs...@ma...>
Author: Ãyvind Harboe <oyv...@zy...>
Author: Thomas Kindler <ma...@t-...>
Author: David Brownell <da...@pa...>
Author: Michael Bruck <mb...@di...>
Author: Rick Altherr <kc...@kc...>
Author: David Brownell <da...@pa...>
Author: Raúl Sánchez Siles <rsa...@in...>
Author: Raúl Sánchez Siles <rsa...@in...>
Author: Raúl Sánchez Siles <rsa...@in...>
Author: Raúl Sánchez Siles <rsa...@in...>
Author: Simon Qian <sim...@Si...>
Author: Nicolas Pitre <ni...@ca...>
Author: Nicolas Pitre <ni...@ca...>
Fix eol-style on guess-rev.sh
mifi (80):
- added patch to display device information as INFO too
- Fixing two compiler warnings
- added patch from Dominic to unlock the ATMEL flash in cfi_read_atmel_pri_ext
git-svn-id: svn://svn.berlios.de/openocd/trunk@228 b42882b7-edfa-0310-969c-e2dbd0fdcd60
Sorry, I have forgotten to add a description for the last version, where
Corrected typo (to many brackets)
- added the patch CFI-BE-Fixes-Blockwrite-Support.diff
- initialize num_sectors and sectors in handle_flash_bank_command.
- change error message for image_open error
- removed some compiler warnings
The changes allows the "script" command to be used from within
- added patch to corrected argument of image_open in handle_flash_write_image_command
- fixed relocation problems with image loading. Relocation is handled
- only some cosmetic changes, convert \r\n to unix
- added patch for new flash functionality like:
- this patch allows OpenOCD running under eCos to access files via tftp
- corrected working_area of LPC2129 script
- added gdb flash fixes patch
- added patch for faster start of OpenOCD
- added patch "remove error handler as planned"
- added patch to fix brocken STR9 line
- added patch to fix crash in load_image on corrupt elf file or out of memory.
- added patch to check some malloc problems.
- added autoprobe functionality
- added patch to solve the reset problem with arm9 and cortex which Spen had found.
- added patch to solve problem with AT91SAM9260 (dirty register)
- added patch to access cp15 register (XScale)
- added patch to change pathmove handling
- added starting point for development on a ngw100 eval board
- removed "starting point" for ngw100, will add it again if it is working
- added patch to Improving progress/error output for telnet & GDB monitor
- added patch to stop DEBUG messages to be forwarded to telnet/gdb.
- added patch which fixed big/small endian problems on non-x86 host CPUs.
- added patch to make single-stepping more resilient
- added patch to solve problem with AT91SAM7SE MCU have 3, rather than just 2 GPNVM bits.
- cosmetic changes to print out version without line number/time
- added patch to remove count and time information from log_printf
- added a test document as a starting point
- added STR710 example which I used for Eclipse debug testing
- added test result for str710, and test description
- added str912 test example, and test result
- added lpc2148 test example, and test result
- added lpc2294 test example, and test result
- added sam7s256 test example, and test result
- added sam7x256 test example, and test result
- added at91r40008 test example, and test result
- updated svn:eol-style prop to native for AT91R40008
- added example for testing the JTAG speed with a FT2232 device
- removed warnings "xxxxx" might be used uninitialized in this function (arm_simulator.c)
- added test result for r423
- added str912/str710_program.script
- added new test results
- added test results
- remove AC_C_VARARRAYS and set check of autoconf back to 2.59
- added testing binaries
- added test results for r592
- corrected typo in cfg file
- smoketest for r651
- added support for ft2232/jtag_khz
- corrected rounding in ft2232_khz
- added/renamed interfaces, remove jtag_speed from the interface part
- added patch from uwe hermann, thanks for the hint.
- added arm-usb-ocd.cfg and signalyzer.cfg to the interface
- added smoketest result for r657
- replaced JLINK_xxxx command defines by EMU_CMD_ from the
- removed obsolete cmd JLINK_TAP_SEQUENCE_COMMAND, use EMU_CMD_HW_JTAG instead
- corrected copy/paste type and renamed jlink_usb_read_result to jlink_usb_read_emu_result
- swap processing of reset handling. First srst and than trst.
- comment out usb_set_altinterface, because it is not working under Mac OS X.
- changed some buffer defines to increase performance
- renamed EMU_CMD_HW_JTAG to EMU_CMD_HW_JTAG3
- added check for max. value of jtag_speed in ft2232_khz.
- added target and event script for the eir-sam7se512 target
The following patches was applied:
Added functionality to support jtag_khz for the jlink.
Partially fix USBprog and JLink adapters on XScale target
Added jtag_nsrst_delay 200 and jtag_ntrst_delay 200 to the LPC2294 target.
Change the setting for the sam7se512 and sam7x256
Added the options calc_checksum to the flash driver.
Set default for want_ftd2xx_highspeed to "no" instead of "maybe"
mlu (25):
Added patch for Luminar Micors Fury class devices (Joe Kroesche)
- trying to remove a breakpoint with target running should not exit(-1) from OpenOCD
- Support for AT91SAM7xx512 chips
- Support for AT91SAM7xx512 chips
- added ft2232_latency command
- Fixed display of sector sizes in flash.c
Updated list of LMI partnames
Changed at91sam7_ecr working area to a valid SRAM adress for the AT91SAM7S/X target family
Cortex-M3 cleanup and performance patch
Slight improvement in run_algorithm register restore.
Solve problem with single stepping.
Improved (for humans) error reporting for flash programming errors.
Fix str_to_buf problem, thanks Andy Chenee
Break ft2232_execute_quie into smaller functions, follows restructure of jlink.c
Fix ft2232 TX buffer overflow
Corrected statement order
Added arm_adi_v5.c/h, to replace cortex_swjdp.c/h. Better conformance to ARM Debug Interface rev 5 documentation and remoed code specific to the Cortex-M3 targets.
Changed armv7m and cortexm3 to use nev arm_adi_v5 instead of cortex_swjdp.
Deleted depreciated files ( new versions are arm_adi_v5.c/h )
Added dap baseaddr and dap apid commands
Correctedout buffer size and missing jlink_tap_init() call.
More error logging for DAP errors
Clear FLASH_SR error flags after flash errors to avoid reset befor further flash operations.
Small changes that might improve stability.
Added (BUILD_JLINK==1) condition to us new tables with JLink
ntfreak (297):
- added support for ST flashlink cable
- added support for Hitex comstick.
-- corrected embedded ice definitions
- added mingw elf patches from Vincent Palatin
- str9xpec driver was using stricmp, changed to strcmp for posix compatibility
- ST STM32x cortex support added
- corrected stm32x_handle_options_write_command, incorrect options printed
- motorola s19 file loader added
- fix build issues when not using autoconf/automake to build
- flash autoerase <on|off> cmd added, default is off - flash banks are calculated and erased prior to write (flash write_image only)
- cleaned up str7, str9 and stm32 flash drivers
- add verify_image command
- corrected str91x bank1 programming issues
- reset_run now works as expected on cortex-m3
- added support for Hitex STM32 Performance Stick
- added patch by Ãyvind Harboe to Intel Hex Start Segment Address Record (Type 3)
- removed flash write_image - binary compare function has been moved to verify_image command
- fixed problem when write_image is called multiple times
- minimum autoconf 2.59 is now required and verified - due to issues with AS_HELP_STRING
- add support for extended gdb packet R (restart), requires gdb to be started with target extended-remote
- rename flash_init and flash_erase to flash_init_drivers and flash_erase_address_range - stops conflicts with redboot. Thanks Ãyvind Harboe
- added faster gdb packet handling (thanks to oyvind harboe for the patch)
- _DEBUG_GDB_IO_ was left on by mistake
- fix gdb packet retransmission omission
- debug handler file not closed - (thanks to oyvind harboe for the patch)
- add missing files from previous commit (tms470 flash driver)
- add support for cortex_m3 target_request debugmsgs
- updated docs for cfi command, added missing ft2232_layout names
fixed incorrect elf segment size - Thanks Ãyvind Harboe
Added patch to stop SEGFAULT with missing jtag config lines. Thanks Ãyvind Harboe
- fixes issue with reset and arm926ejs core. Thanks Ãyvind Harboe
- added synchronous reset patch, Thanks Ãyvind Harboe
- added synchronous wait/resume patch. Thanks Ãyvind Harboe
- add autoprobe support to the stm32 flash driver
- stm32 flash driver now checks for correct target
- telnet prompt behaves correctly for new synchronous halt/resume/reset commands
- add autoprobe support to flash info command
- added check for revA silicon in stm32 flash driver
- fix programming issue with lpc2101/2
- fix xscale memory leak and warnings. Thanks Ãyvind Harboe
- fix issue with telnet prompt while gdb running
- cleanup and remove time() warning
- solve lots of problems with stuck GDB connections, making it impossible to connect to GDB, handle timeout of acknowledgement, etc.
- combine similar functions in telnet_server.c
- Added a "User:" debug level. These are messages that are intended for the user and are part of normal operation.
- add new non-cfi SST flash device. Thanks Ãyvind Harboe
- removed a raft of unecessary exit() calls. Issuing a reset will solve these ails.
- fix broken JTAG error handling
- arg list now correctly released on error. Thanks Ãyvind Harboe
- reapply fix with telnet prompt while running/halting gdb
- Fixed bug in pathmove for XScale
- add search paths via new arg -s (-search). Thanks Ted Roth
- fix bug when using full paths to config files. Thanks Ted Roth
- remove build warnings
- code reformat (Thanks Pavel Chromy)
- fix issue when target is already halted
- correct line endings from previous commit
- fix line-endings
- added svn props
- svn file props set to LF endings
- update openocd.texi to fdl 1.2
- fix bug with emulated cortex_m3 dcc channel
- added svn prop eol-style native
- updated svn:eol-style prop to native
- contrib directory added
- added svn prop svn:eol-style native
- added svn prop svn:eol-style native
- correct spelling typo in stm32x flash driver
- The elf loader incorrectly assumed that the program header always follows the ELF header. (Thanks Michael Bruck)
- added more TARGET_HALTED checks for the read/write_memory handlers
- add stm32 target script
- 16 and 32 bit unaligned accesses supported
- armv7m control register now set as dirty when switching context
- changed jtag_add_reset errors to warnings
- updated docs with new commands/features
- gdb server was incorrectly sending null terminator on qXfer:features:read: packet
- rename log functions to stop conflicts under win32 (wingdi)
- added note to docs about gdb 6.8 and memory map support
- add missing svn props
- due to previous patch adding AC_C_VARARRAYS, autoconf v2.61 min required
- single core context used, removed debug context as thought unnecessary.
- added svn props for newly added parport.cfg
- split str71x driver into banks
- added new and missing target scripts - stm32stick and str730
- correct stm32stick config script
- flash.c change stricmp to strcmp
- update docs for flash write_image command
- set erase flag on sector
- set lpc288x[ch] svn file props
- added stm32 stick and luminary eval boards interface configs
- fix typo in openocd.texi
- added svn props for newly added files
- added svn props for newly added files
- added svn props for newly added files
- change str9 flash controller to read array after a level 1 protect
- incorrect ram size for lm3s811 target script
- luminary flash now loader polls when finished
- added ARMV7_GDB_HACKS define to armv7m.h, enabling all gdb hacks to be enabled/disabled for testing
- add cortex_m3 variant luminary to fix reset issue with asserting SRST
- updated cortex_m3 docs regarding luminary reset behaviour
- stm32x default to 128k if flash size failed.
- added new device to luminary flash driver
- add missing svn props
- add -lusb to link for using libftdi under win32
- added info about libftdi support under win32
- updated docs for ft2232_vid_pid command
- added svn props for newly added files
- restored tabs and formatting
- added jlink support, based on Jürgen Stuber patch
- remove error message on shutdown
- add support for newer high density stm32 parts
- jlink jlink_execute_queue returns result
- removed compiler warnings
- update sectors on stm32x mass erase
- added stellaris flash mass_erase command
- updated docs for new stellaris mass_erase command
- corrected error with stm32 page calculation
- fix issue if jlink sends result in single usb in packet
- fixed typo in wp command
flash_write will attempt to pad image sections
- moved flash erase_check target code to target.c
- stellaris flash driver changed warning to info
- added checksum_memory and blank_check_memory for xscale
- str9 driver remove the duplicate target_free_working_area and reformat spaces to tabs.
- luminary, stm32 and str7 flash driver error cleanup
- fix jlink accessing incorrect buffer element
- erase bank using bank erase rather than each sector
- stm32 erase will use mass_erase if all banks selected
- error message cleanup in jlink driver
- probe incorrect for high density stm32 flash
- removed duplicate cortex_m3_register_commands
- added missing install entries for eir-sam7se512 target
- fix compile errors when _DEBUG_INSTRUCTION_EXECUTION_ is defined
- fix reset_halt issue with certain arm cores - address not set in embedded ice reg
- added svn props for newly added files
- update docs for new target_script events
- allocate target_type_t structure for each target, fixes issue when multiple targets of the same type are located on the jtag chain
- fixed line endings with commands.tcl
- added jtag_khz to target scripts
- added event scripts for str73x and str75x targets
- added svn props for newly added files
- added search for libdl
- fixed build issues with win32
- fix typo in memory.tcl
- fix duplicate log entry
- command_run_line will only search once for the command, and execute if found
- apply correct formatting to openocd.c
- fix pre_resume script not being called
- removed target_process_events as only used in handle_resume_command and events will be called anyway by poll
- target_resume is now synchronous with 5sec timeout
- only check normal resume, not debug resume
- added luminary libftdi interface config
- adding missing install entry for luminary-libftdi.cfg
- added svn props for newly added files
- configure check added for tclsh
- remove requirement for file2c.tcl
- fix issue when building in separate build dir
- added svn props for newly added files
- remove unused objcopy from configure.in
- stm32 flash driver correctly handles early silicon
- reverted resume_target to old behaviour
- fix incorrectly registered function openocd_array2mem
- remove build warning from keep_alive
- updated texi for removed daemon_startup command
- added svn props for newly added files
- add missing quotes from CFLAGS_FOR_BUILD
- fix bug with stm32 high density write protection
- more fixes to high density stm32x flash driver
- fix typo
- fix typo in jtag_add_end_state
- comment about dummy ack '+' char from gdb
- merged mips target into svn trunk
- added run_and_halt_time to deprecated/removed functions section
- fix target_examine declaration
- added svn props for previously added file
- remove Jim_InitEmbedded warning
- correct BUILD_ECOSBOARD definition is server.c
- added svn props for newly added files
- fix build issues under win32 (mingw)
- added svn props for newly added files
- added missing parport configs to texi
- fix win32 build issues from previous jim patch
- add support for new gdb QStartNoAckMode, which disables sending of ack's between remote packets
- clear any existing breakpoints/watchpoints when restarting in gdb extended remote mode
- added myself to copyright on files i remember adding large contributions for over the years
- fix typo's
- renamed x7926 driver to aduc702x to match other flash drivers
- update docs for aduc702x flash driver
- convert spaces to tabs in at91sam7.[ch]
- at91sam7.c remove build warnings
- remove build warnings
- remove build warnings from mips_m4k.c and arm11.c
- remove build warnings
- fix tcl_port typo in docs
- make docs more readable
- update docs deprecated section to include links to new commands (if any)
- remove texi warnings
- add link in texi docs
- fix native mingw build if gettimeofday not defined.
- remove build warning
- fix issue with jlink/libusb timeout under linux
- remove build warnings
- slight mips32 cleanup/reformat
- fix crash when variant for mips32 target not given
- correct the register hi/lo read - wrong way round
- stops multiple calls to examine from allocating the breakpoint arrays
- add support for hardware breakpoints to mips32 target
- fox small typo in handle_irscan_command
- preserve cortex_m3 C_MASKINTS during resume/step
- fix issue with gdb_target_to_reg missing second byte
- add new cortex_m3 maskisr cmd
- correct corrupt target configs
- fixes segfault using the targets cmd if multiple targets defined
- stm32x flash driver: add support for low density devices
- fix issue with luminary flash driver and tail bytes
- change mips registers to std names
- fix incorrect svn props from last commit
- cleanup target config scripts
- incorrect str9 irmask used in config files
- str9xpec driver now working with new jtag tap config
- update texi to describe str9xpec driver flow
- fix illegal memory access in unregister_command function
- typo with flash bank help command
- added Axiom AXM-0432 to texi
- incorrect url in last commit for usbjtag
- check cortex_m3 FPB is enabled when setting hardware breakpoint
- remove target specific variant and use target->variant member
- add ability for openocd to communicate to gdb using pipes (stdin/stdout).
- add missing svn props from r1243 commit
- fix rlink build under native win32 (mingw)
- add missing svn props from r1270 commit
- fix missing/incorrect svn file props
- add gdb pipe support to native win32 (--pipe option)
- add missing svn props from r1299 commit
- minor doc updates regarding tap creation and gdb connection using pipes
- a few more docs tweaks
Updates and fixes for the manpage from Uwe Hermann
- added mips software breakpoint support
- fix mips issues with newer versions of gdb
- fix LDFLAGS typo in configure.in
- add missing svn props and fix incorrect line endings from last commit
- add missing svn props from 1323 commit
- add missing bitq and rlink files to distribution
- add guess-rev.sh and contrib/libdcc to dist
- make guess-rev.sh work with msys
- revert guess-rev.sh to using bash
- remove unused includes, fixes build issues under FreeBSD
- add --enable-release to docs
- fix issue with configure when building srcdir != builddir
- AC_CHECK_FILE not supported when cross-compiling
- fix native win32 build issues
- issue warning when flash image exceeds configured flash size.
- guess-rev.sh now works as expected when build_dir is not the same as src_dir
- stm32x flash driver - add support for stm32105/107 (connectivity line)
- add missing svn props from previous commit
- add svn props from previous commit
- fix small typo in texi (section HostOS)
- add support for standalone rlink
- fix issue with cortex_m3 reset run. Thanks Perry Hung
- fix incorrect stm32stick.cfg
- remove build warnings
- fix incorrect str9comstick cfg
- revert texi changes from rev 1418 as it breaks build
- docs patch from Thomas Kindler
- add missing svn props from previous commit
- fix another typo in str9comstick.cfg
- update str9 and stm32 comstick configs
- fix typo's in last commit
- add missing usbprog config. patch from Uwe Hermann
- add svn props from previous commit
- add openocd coding style to texi
- fix at91rm9200 warning. Thanks Zach Welch <zw...@su...>
- remove environ warning under win32 build
- fix signed/unsigned build errors under win32. Thanks Zach Welch <zw...@su...>
- vsllink update from SimonQian [sim...@Si...]
- fix win32 build after svn 1511 changes
- fix win32 build
- add missing svn props from svn 1565 commit
- fix build issue when HAVE_ELF_H is not defined
- add missing svn props from svn 1768 commit
- add missing svn props from svn 1798 commit
- add support for cortino jtag interface
- hack added to fix a issue with v5/6 jlink
- change signature for adi_jtag_dp_scan and adi_jtag_dp_scan_u32 to use swjdp_common_t *swjdp instead of arm_jtag_t *jtag_info
- add support for different TAR autotincrement sizes as per ARM ADI spec.
- fix texi/pdf issue created in svn r2039
- update openocd online doc url's
- fix bug introduced during r1962
- fix break caused by r2208 when using --pipe option
- correct configure ftd2xx highspeed test
- fix issue when using gdb_port cmd
- remove cygwin build warnings in at91sam3.c
- correct stm32 connectivity line programming as per latest manual
- add stm32 connectivity line tapid to stm32.cfg
- fix build when using a cross compiler - do not try and run any host tools
- change sleep to usleep - fixes issue under win32 build
Gary Carlson <gca...@ca...>
oharboe (1059):
- added time command
cosmetic fixes to debug output + phasing out printf() in favour of logging system. From Pavel Chromy <ch...@as...>
added svn:ignore for files generated by bootstrap + build process.
sharpend JTAG docs w.r.t. hw fifo implementations and retired jtag_cancel_queue() which is inheritely incompatible with a hw fifo concept.
debug_level 3 now prints seconds since start of openocd
disable cfi_write_words(). Broken for spansion, fallback
- fixed target->type->poll() return value
- added -c option that will execute an openocd command
move options handling to separate file to better support embedded implementations of OpenOCD
- added "xscale fast_memory_access" which speeds up memory
Michael Bruck spotted an omission in svn 322
- "flash write_binary" is now "flash write_bank" to clarify the focus of the
- using ERROR_COMMAND_SYNTAX_ERROR to print syntax in a couple of places
dos2unix fix.
committed bugfix from Michael Bruck
From Michael Bruck
- Fix target library path problem w/Windows
Fix from Pavel Chromy.
Pavel Chromy style fixes.
formatting.
Pavel Chromy
- jtag minidriver work in progress
Next step in the JTAG minidriver. This should be fairly close to the final thing, but
jtag_add_reset() minidriver gaffe fixed.
oopss.. another jtag minidriver reset gaffe.
fix naming conflict under Windows.
Nicolas Pitre listed some more devices.
Uwe Hermann fixed some warnings.
Uwe Hermann Add @dircategory and @direntry to the info page -- they're used to add the openocd info page to the global index of info pages in the system (usually /usr/share/info/dir).
- Added TARGET_REQ_DEBUGCHAR target_request debugmsg. This
eCos flash driver.
lingering change for eCos flash driver
jtag_add_shift() fn - minidriver work in progress.
fix copyright.
added some code that will be used in upcoming minidriver work.
code to be used in upcoming minidriver work.
fix from Pavel Chromy
Pavel Chromy cleaned up checks for halted, error messages, etc.
fix gaffe.
Pavel Chromy - cleanup error messages
small line ending cleanup.
numerous fixes from Uwe Hermann
Uwe Hermann: Simplify the manpage by removing lots of redundant information which is
Pavel Chromy spotted a leak
Pavel Chromy spotted duplicated newlines gaffe from yours sincerely.
fixed newline gaffe in OUTPUT()
changed to eCos license.
Michael Bruck:
Michael Bruck:
Summary: passing of variable argument list reduced, strings sent to logging are now formatted just once - more efficient.
Pavel Chromy: clarify error messages
do not write single chars to log file.
- clean up target output strings a bit
- image.c and fileio.c now uses logging to propagate error strings.
remove warnings.
formatting fixes from Pavel Chromy
lingering patch from Pavel Chromy.
Uwe Hermann: mproves the manpage text a bit and adds the
Pavel Chromy: add missing newlines.
Uwe Hermann:
Uwe Hermann: Updated version which applies cleanly after the recent whitespace fixes.
Pavel Chromy fix: the guess-rev.sh scripts to retrieve SVN revision returns the result including new line
Uwe Hermann: Add a --version switch for openocd.
Tweaked logging output.
fixed stack corruption. Introduced when _check_value was phased out.
added an #error in case anybody tries to compile that broken code.
error handling fix.
wip. fixed gaffe in jtag_add_shift()
log_remove_callback
Now uses jtag_add_shift() via embeddedice_write_reg_inner().
improve DCC bulk write performance by using jtag_add_shift() + tweaked embedded ICE communication.
backed out jtag_add_shift()
fix memory corruption regression introduced in 335
wrote up explaining why tests are done on committed code.
wrote up explaining why tests are done on committed code.
target scripts for test suite.
fixed FSF address. Uwe Hermann.
Uwe Hermann tightned up comments, etc. to follow OpenOCD policy
some comments from Dominic
added at91sam9260.cfg, nslu2.cfg, pxa255.cfg, pxa255_sst.cfg
added fill_malloc test. Not quite sure what to do here yet, but it would be good to have something...
I fixed all the targets even though
Michael Bruck:
Michael Bruck: fix warnings.
Michael Bruck - fix warnings.
Fixed GDB timeout crash - regression introduced back when log_add/remove_callback was added.
Bogus error message in GDB removed
telnet_port can now be invoked multiple times
* fixed malloc corruption in target->debug_reason
Pavel Chromy: memory leak in at91sam7 flash driver, possible incorrect pointer conversion in gpnvm command handling,
added jtag_add_dr_out(). Better for hw fifo, same for software fifo.
- This speeds up dcc arm7_9 bulk write a little bit and exercises the jtag_add_dr_out() codepath
warnings & comments fixes.
make debug code w.r.t. incorrect args for bypass stricter.
Michael Bruck: 64 bit va_list fix for crash
Michael Bruck:
Pavel Chromy: performance tweak of gdb_put_packet_inner() removing malloc and avoiding memcpy of larger blocks of data,
Pavel Chromy: faster alloc_printf()
comment.
tms is never referenced, confusing old code left behind probably.
use jtag_execute_queue() instead of jtag->execute_queue()
retired
Pavel Chromy: va_copy should always be paired with va_end.
Cosmetic fixes from Uwe Hermann
removed excessive debug output. Perhaps a debug_level 4 should be introduced? Hopefully it can be avoided as this was the only case where debug output was going through the roof.
Removed code that inserted prompt after printing asynchronous information. Current implementation was broken beyond repair.
wi-9c target scripts
alloc_printf and alloc_vprintf
Pavel Chromy, the patch fixes an issue with PRESTO & FTD2XX under Linux.
minor corrections for target scripts.
Added some timeout handling to XScale so OpenOCD doesn't get
Edgar Grimberg, fix arm926ejs_examine_debug_reason return value.
Michael Bruck: fixed warnings
Asynchronous output information from e.g. a halt is now displayed again.
- Improves error handling upon GDB connect
typo.
Michael Bruck: fix warning
added stm32.cfg to install list
- LOG_SILENT can be used to silence the log(needed in upcoming patches)
- Fixed various error handling when looking for memory leaks
- the jtag chain is examined and validated after GDB & telnet servers
- fixed a problem with big endian XScale and GDB register packets.
prettier async output
Pavel Chromy: the attached patch refines PRESTO support and makes it work with libftdi.
Pavel Chromy: hopefully perfection for async output and prompt/partially typed command line handling.
more target scripts wip.
more target scripts wip.
backed out changes from 483.
Pavel Chromy: telnet line buffer size checking, history does not store repeating lines, improved history printing, log callback tweak
- fixed jtag_add_reset(). It no longer causes jtag_execute_queue() to
JTAG_END_STATE is now retired. It is no longer queued.
retire unused code.
- retired unused jtag events. The code was incorrect
tidy up output a bit. No longer show "accepted connection" inside Telnet session.
With the recent changes the TRST needs to happen for every reset.
reduce compare noise. If someone should be crazy enough to try to run OpenOCD under eCos, then they'v got some hooks to point them in the general direction.
tinkered a bit with testing matrix.
Corrections from Dominic
more info about latest working version
fixed regression in XScale introduce in 297
This moves common code into functions so as to make it clear
The bitbang driver leaves the TCK 0 when in idle
xscale now passes w/bitbang in 505
Dominic undid some of my damage. CLK should now be set as intended in all cases.
marked infinite loop in code w/TODO and fixed warning.
XScale excessive waiting fix.
print ms in debug_level 3 logs. Seconds is not enough.
- adds two speeds to jtag_speed. reset and post reset speed. Default
added jtag_khz for use with target library
clarified jtag_khz command.
Michael Bruck: fixed warnings
reset and post reset speed & jtag_khz command documented.
added profile command. It was added to simplify evaluation by testers.
hooks for multithreading. Disable nagle
TAP_SD/SI are now forbidden end states.
Edgar Grimberg: added needed delays
make jtag_add_statemove() internal to the driver.
Removed exit()'s. A reset is usually enough to work around these, reducing cycle times
target_call_timer_callbacks_now() now invokes periodic callbacks immediately
Do not assert trst in srst_only case even if srst_pulls_trst.
- fixes for jtag_khz committed.
optional count argument to mwX
Added timeout(instead of infinite loop) to soft_reset_halt
fix flash info - now reports erased state properly
Edgar Grimberg sharpened the str912 target script.
- Work on fixing erase check. Many implementations are plain broken.
added query of reset speed
- reverted some of the changes that possibly broke arm926ejs. Waiting
- the reset mode parameter is now DEPRECATED. It is implemented
avoid patch trouble by isolating troublesome line...
at91eb40a.elf moved to src/target/ecos
- only if "reset halt" or "reset init" are issued will the reset vector be set up
Spen fixed various issues w.r.t. setting up the sockets for optimal performance.
The endstate now reports the endstate of the queue instead of
Pavel pointed out that jtag_add_tlr() is better than jtag_add_tms().
added a couple of LOG_ERROR() messages to improve logfile.
lm3s6965.cfg contributed by Edwin Olson
Edwin Olson found bug & tested fix for flash write_image for stellaris.
use jtag.c's cur/end_state global variables.
- added "init" command. "init" and "reset" at end of startup script is equivalent
Don Porges fixed c99 issues.
back out Don Porges workaround for some device with manufcaturer # 0x1a2
target lib wip
stop using variable sized arrays. That's something that belongs to C++ and not C.
Improved XScale performance for embedded hosted OpenOCD
added target->type->examine(). Eventually this will allow for bringing up telnet/gdb *before* jtag chain has been validated + it might fix some reset halt problems seen as examine() needs to run after TRST has been asserted.
Wip - split target setup and target examination
Reset wip. Just adding hooks. This is just to reduce the size of the actual change, no change in behaviour.
Reset wip. Just adding hooks. This is just to reduce the size of the actual change, no change in behaviour.
fixed gaffe for default examine implementation
found two more gaffes for reset wip
- Set up ICE registers after TRST
allows launching OpenOCD w/telnet+gdb server w/the
allows launching OpenOCD w/telnet+gdb server w/the
A dummy driver to test codepath w/no contact w/target.
More robust handling of unknown target state for step/continue packet.
TRST is asserted *before* target->type->assert_reset() is invoked. Removed old code.
Added a few tweaks while playing with texi.
moved out stuff that wasn't already moved from openocd.pdf to the target library.
retired fast_memory_access. It's always fast now.
target read/write is no longer attempted for target_xxx() functions when the target has not been examined(fails w/error).
Added checks for target->type->read/write_memory, soft_reset_halt and run_algorithm that the target has been examined.
- sw_bkpts fails if the target is not halted. The side effect is
fix SEGFAULT regression in cortex after TRST fixes
Edgar Grimberg added some missing scripts from the install
Update Embedded ICE registers explicitly during target->type->examine() instead
fix for gaffe in 555 that stopped JTAG chain examine + validate from running.
Edwin Olsen: improves breakpoint handling on cortex-m3 parts. Specifically, this patch allows expressions to be evaluated in GDB that contain function calls.
Nicolas Pitre fixed regression.
Enabled auto erase, gdb memory map and gdb flash programming by default.
retired auto_erase, added optional erase arg to flash write_image.
Karl RobinSod <kar...@gm...> added lpc288x support. Some work remaining, committing for test/collaboration purposes.
deleted obsolete stuff.
wip on reporting bugs.
Andre Renaud support for the am29sl800db CFI flash chip (id: 0x0004, 0x226b).
added fast option.
fixed doc for GDB memory map and flash program defaults to follow code
Tim Hudson: removed setting jtag_speed directly (which should not be done).
regions outside flash are now read/write nocache by default. GDB 6.8 would mark non-flash areas as inacessible by default.
Tim Hudson contributed at91sam9260 target config files + some interface definitions.
Pavel Chromy's on chip flash loader
Edgar's new test cases.
Edgar's new test cases.
Close dangling file handle
make debug_level 3 useful again... the log just drowns. perhaps introduce debug_level 4?
wip instructions for building Cortex toolchain
set communication speeds
Edgar Grimberg added a new rule for target scripts. Flash + verify must succeed.
Edgar's added support for printing jtag_khz
Edgar's naming convention fixes.
lpc2148 fixes from Edgar Grimberg
now compiles again.
Tim Hudson worked on English language.
add missing reset script
The target library is now the authorotative source of config examples
Edgar Grimberg found tiny memory leak
increase packet size to 16kBytes. Very slightly faster.
Edgar Grimberg plugged a leak found w/Valgrind.
This matters for embedded devices, but is probably not observably better for PC hosted OpenOCD.
print available memory in log_level 3
disable code that prints available mem under windows. Does not compile.
Michael Fischer spotted a problem in the reset routines for srst_pulls_trst. It is a bit of a mystery why this was only visible w/LPC2148.
printing available memory fails on win + mac. Disable for now until a robust way of doing this can be found.
added stack trace.
updated guidelines.
Michael Fischer found and Edgar Grimberg fixed generic crash in timers reproduceable with at91fr40008
fix mode output when illegal arm mode is detected. Now prints illegal mode for index -1.
Michael Fischer found this bogus warning. Fixed.
print out an error if srst_pulls_trst is not specified for
Fix crash when mode number fetched from the target is invalid.
reverted change in 658 and simply removed the busted warning for now.
Edgar Grimberg fixes some memory handling issues and
moved srst_pulls_trst check into arm7_9_common.c. Not tested yet, if it is broken it should "only" print bogus warnings or not print a warning when it should have.
comment.
jtag_exexcute_queue() now logs error when it is invoked before 'init' command.
clock 1/0 in idle now configurable.
found out why str912 reset halt failed.
Fredrik Hederstierna: fix leak + clean up return codes
Fredrik Hederstierna: fix leak + clean up return codes
Pavel Chromy: fix logging syntax error + formatting & removing obsolete comments.
Pavel Chromy: TCK returns to zero before state of reset line is changed.
Nicolas Pitre: Feroceon fixes
GDB timeout fix. If a script takes a long time and does not produce any output, ping between every command.
fallback for no implementation of multi word CFI write. Successful codepath not affected.
Spen spotted a bug in warning for missing srst_pulls_trst
Pavel Chromy: TAP state changed to TLR when SRST goes active with srst_pulls_trst, as in all other drivers
Pavel Chromy:
Pavel Chromy:
added pre/post_reset scripts based on Pieter Conradie's ideas.
Fixed...
[truncated message content] |
