Minor cleanup of FT2232:
- make Olimex glue warn about Olimex issues instead of JTAGkey issues;
- make some data static+const;
- don't export some internal symbols.
git-svn-id: svn://svn.berlios.de/openocd/trunk@2037 b42882b7-edfa-0310-969c-e2dbd0fdcd60
Remove pernicious whitespace from src/jtag/*c files; mostly
the end-of-line flavor for now, although there's more.
git-svn-id: svn://svn.berlios.de/openocd/trunk@2036 b42882b7-edfa-0310-969c-e2dbd0fdcd60
Convert the Interface/Dongle Config chapter's section
on drivers to use the @deffn syntax, and integrate the
presentation of the driver-specific commands with the
relevant driver. Alphabetize.
Cross-checked against the code ... several adapters were
not listed, and a few commands weren't.
(Maintainers for the versaloon and zy1000 drivers would be
good candidates to add the commands missing from those
sections...)
git-svn-id: svn://svn.berlios.de/openocd/trunk@2035 b42882b7-edfa-0310-969c-e2dbd0fdcd60
Convert the str9xpec driver info to use @deffn; alphabetize;
add the missing part_id command.
Convert the mflash support to use @deffn; alphabetize.
git-svn-id: svn://svn.berlios.de/openocd/trunk@2034 b42882b7-edfa-0310-969c-e2dbd0fdcd60
Update docs for most of the remaining commands in jtag.c:
- switch to @deffn
- these are just the "low level" JTAG commands
- resolve much goofage!
* remove docs for non-existent commands
* add missing docs for some existing commands
* fix incorrect docs for some commands
- just index TAP states overall, not individually
- current name is "RUN/IDLE" not "IDLE"
Cross checked against the source.
This also creates an "Interface Drivers" section, analagous to how
(NOR) Flash and NAND drivers are presented; that's not yet sorted.
git-svn-id: svn://svn.berlios.de/openocd/trunk@2014 b42882b7-edfa-0310-969c-e2dbd0fdcd60
Rework the TAP creation documentation.
- Try to use "TAP" not "tap" everywhere; it's an acronym.
- Update the associated "target config files" section:
* reference the "TAP Creation" chapter for details
* simplify: reference interesting multi-tap config files
* let's not forget CPU configuration (*before* workspace setup)
* streamline it a bit
* move that workspace-vs-mmu issue to a better location
- Clean up TAP creation doc mess
* switch to @deffn
* (re)organize the remaining stuff
* reference the "Config File Guidelines" chapter
- Tweak the "Target Configuration" chapter
* rename as "CPU configuration"; unconfuse vs. target/*.cfg
* bring out that it's not just there for GDB
* move TAP events to the TAP chapter, where they belong (bugfix)
git-svn-id: svn://svn.berlios.de/openocd/trunk@2013 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- Cloned the src/jtag/jtag.c file to src/jtag/interface.c.
- For each for of those files, deleted the contents of the other.
- Add new source file to automake input.
git-svn-id: svn://svn.berlios.de/openocd/trunk@2012 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- Move the jtag_interface structure definition.
- Move the Cable API declarations.
- Add new header file to automake input.
The next patch will move the implementation to interface.c.
git-svn-id: svn://svn.berlios.de/openocd/trunk@2008 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- Wraps all minidriver API functions using API front-ends:
- Outlines jtag_add_dr_out() and jtag_alloc_in_value32().
- Adds interface_ prefix to existing jtag_alloc_invalue_32 routines.
- Re-inline these interface definitions in new header file.
- Re-inline parts of the (mini)driver implementations in minidriver.h.
- Replace INCLUDE_JTAG_MINIDRIVER_H with #include directives.
The next patch will finish removing '#ifdef HAVE_JTAG_MINIDRIVER_H'
from jtag.h.
git-svn-id: svn://svn.berlios.de/openocd/trunk@2006 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- change SWJDP_IR/DR_APACC to DAP_IR/DR_APACC to conform with ARM_ADI docs.
- add swjdp->memaccess_tck field and code for extra tck clocks before accessing memory bus
- Set default memaccess value to 8 for Cortex-M3.
- Add dap memaccess command.
- document all armv7 dap cmds.
- Original patch submitted by Magnus Lundin [lundin@mlu.mine.nu].
git-svn-id: svn://svn.berlios.de/openocd/trunk@2005 b42882b7-edfa-0310-969c-e2dbd0fdcd60
v5/6 jlink seems to have an issue if the first tap move is not divisible by 8, so we send a TLR on first power up
git-svn-id: svn://svn.berlios.de/openocd/trunk@2004 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- Initial support for FT2232H/FT4232H devices from FTDI.
- Add --enable-ftd2xx-highspeed option to configure script.
- Original patch submitted by Joern Kaipf <lists@joernline.de>.
git-svn-id: svn://svn.berlios.de/openocd/trunk@1998 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- Moves references to global jtag interface to default core implementation.
- Missed this reference in the earlier "pointless" patch. Mea culpa.
Important: this has a side-effect. Previously, the error return inside
the interface routine short-circuited the remainder of that function
when 'init' has not been called. With this patch, the command queue
will be cleared in the case that 'init' has been called. Since that
case indicates a buggy script, this does not seem to be a problem.
git-svn-id: svn://svn.berlios.de/openocd/trunk@1995 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- Add jtag_callback_queue_reset() to reset the callback queue.
- Make interface_jtag_execute_queue() use new helper function.
git-svn-id: svn://svn.berlios.de/openocd/trunk@1994 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- Make interface_jtag_execute_queue call new helper function.
- Add default_interface_jtag_execute_queue to wrap jtag interface access.
This patch may look useless on its own, but it helps to isolate the core
JTAG variables from the interface_jtag_* routines, so the later can be
moved into jtag_driver.c in a pending patch.
git-svn-id: svn://svn.berlios.de/openocd/trunk@1993 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- Rename last_command_pointer as next_command_pointer, because this variable
stores the address where jtag_queue_command() will store a command pointer.
- Make that variable static, since it is only used internally in jtag.c.
- Remove superfluous accessor for that now-static variable.
- Deobfuscate use of variables in jtag_command_queue.
- Add jtag_command_queue_reset helper function.
- Use it in interface_jtag_execute_queue.
git-svn-id: svn://svn.berlios.de/openocd/trunk@1992 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- Factor jtag_add_scan_check to call new jtag_add_scan_check_alloc helper.
- Use conditional logic to define two versions of the helper.
- These helpers will be moved to other files in future patches.
git-svn-id: svn://svn.berlios.de/openocd/trunk@1989 b42882b7-edfa-0310-969c-e2dbd0fdcd60
Make the TCL "drscan" and "irscan" commands finish in RUN/IDLE
unless the user specifies otherwise ... usually they'd choose
something like DRPAUSE or IRPAUSE, avoiding RUN/IDLE.
The current "end" state is whatever the preceding commands left
in "cmd_queue_end_state", which to TCL scripts isn't knowable.
This change should forestall various surprises/bugs.
Also check that any "end" state specified is safe in case this
adapter's JTAG clock is free-running. For now, just issue a
warning; eventually a hard failure is probably correct.
git-svn-id: svn://svn.berlios.de/openocd/trunk@1988 b42882b7-edfa-0310-969c-e2dbd0fdcd60
zwelch
oharboe
ntfreak