]> git.sur5r.net Git - openocd/log
openocd
14 years agoadd command private data setter/accessor
Zachary T Welch [Sat, 28 Nov 2009 00:30:28 +0000 (16:30 -0800)]
add command private data setter/accessor

Presently, commands registration taks a static handler data pointer.
This patch adds support for commands that require a dynamic pointer,
such as those registered in a dynamic context (e.g. subcommands for a
user-created 'foo.cpu' command).  The command_set_handler_data will
update a command (group) to use a new context pointer, while the
CMD_DATA macro allows command handlers to access the value.
Jim handlers should find this value in interp->cmdPrivData.

14 years agoadd 'command mode' introspective handler
Zachary T Welch [Sat, 28 Nov 2009 20:29:42 +0000 (12:29 -0800)]
add 'command mode' introspective handler

Allows scripts to behave different depending on the current mode.
Also allows introspection of the mode required for commands.

14 years agoremove unknown handler
Zachary T Welch [Sat, 28 Nov 2009 05:47:54 +0000 (21:47 -0800)]
remove unknown handler

Updates command registration to provide top-level handlers for all
commands, rather than falling back onto the 'unknown' command. Instead,
that same handler is registered for placeholders, providing the same
functionality under the root verb command name instead.  This permits
users to implement their own 'unknown' function, and it resolves some
mind-bending breakage related to function object lookup while recursing.

Changes 'ocd_bounce' to call 'ocd_command' and 'ocd_help' from the
wrapper directly, rather than bouncing through their wrappers. This
prevents endless recursion caused by the above changes, whereby the
'command' wrapper's type check would blow the stack to hell and gone.

14 years agoimprove command handler wrapper script
Zachary T Welch [Sat, 28 Nov 2009 03:14:30 +0000 (19:14 -0800)]
improve command handler wrapper script

Adds 'ocd_bouncer' in startup.tcl that is called as a helper for
all command handlers, shrinking the embedded C wrapper to a mere stub.

Jim handlers are called directly, simple handlers get called with the
wrapper to capture and discard their output on error, and placeholders
call help directly (though the unknown handler still does this too).
It attempts to improve the quality of the error messages as well.

14 years agoadd 'command type' introspective handler
Zachary T Welch [Sat, 28 Nov 2009 02:59:14 +0000 (18:59 -0800)]
add 'command type' introspective handler

Adds the 'command' group handler, with the 'type' command producing
a string that tells whether the given command is 'native' (for Jim-based
command handlers), 'simple' (for simple built-in commands), 'group'
for command group placeholders, and 'unknown' if not found in the
command registration tables (e.g. core built-ins functions).

14 years agoARM11: fix warning on amd64
David Brownell [Sat, 28 Nov 2009 19:55:00 +0000 (11:55 -0800)]
ARM11: fix warning on amd64

Previous version of JTAG_DEBUG() macro hid this bug.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
14 years agoARM11: fix dbgtap JTAG_DEBUG
David Brownell [Sat, 28 Nov 2009 19:28:17 +0000 (11:28 -0800)]
ARM11: fix dbgtap JTAG_DEBUG

There is no DEBUG() macro; don't call one!  Always at
least *parse* debug code, to help prevent such errors.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
14 years agotarget: uplevel add_{break,watch}point() error checks
David Brownell [Sat, 28 Nov 2009 18:40:26 +0000 (10:40 -0800)]
target: uplevel add_{break,watch}point() error checks

In target_type.h it's documented that the target must be
halted for add_breakpoint() ... and with slight ambiguity,
also for its add_watchpoint() sibling.  So rather than
verifying that constraint in the CPU drivers, do it in the
target_add_{break,watch}point() routines.

Add minor paranoia on the remove_*point() paths too:  save
the return value, and print it out in in the LOG_DEBUG message
in case it's nonzero.

Note that with some current cores, like all ARMv7 ones I've
looked at, there's no technical issue preventing watchpoint or
breakpoint add/remove operations on active cores.  This model
seems deeply wired into OpenOCD though.

ALSO:  the ARM targets were fairly "good" about enforcing that
constraint themselves.  The MIPS ones were relied on other code
to catch such stuff, but it's not clear such code existed ...
keep an eye out for new issues on MIPS.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
14 years agotarget: remove unused TARGET_EVENT_OLD_* symbols
David Brownell [Sat, 28 Nov 2009 18:36:32 +0000 (10:36 -0800)]
target: remove unused TARGET_EVENT_OLD_* symbols

Just two *_OLD_* symbols left...

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
14 years agoNEWS: mention new reset-assert event
David Brownell [Sat, 28 Nov 2009 02:53:43 +0000 (18:53 -0800)]
NEWS: mention new reset-assert event

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
14 years agoomap3530.cfg: use new "reset-assert" event
David Brownell [Sat, 28 Nov 2009 02:50:31 +0000 (18:50 -0800)]
omap3530.cfg: use new "reset-assert" event

Replaces previous "reset-assert-pre" workaround.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
14 years agoCortex-A8: support "reset-assert" event
David Brownell [Sat, 28 Nov 2009 02:50:26 +0000 (18:50 -0800)]
Cortex-A8: support "reset-assert" event

Use the new "reset-assert" event; else SRST; else fail.
Tested on an OMAP3, using the event.

NOTE:  still doesn't handle "reset halt".  For some reason
neither VCR nor PRCR seemed effective; they held the value
that was written, but VCR didn't trigger debug entry when
the reset vector fired (maybe the vector needs configuring?)
and PRCR refused to hold the chip in reset until deassert()
could force the core into debug state.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
14 years agotarget: groundwork for "reset-assert" event
David Brownell [Sat, 28 Nov 2009 02:50:20 +0000 (18:50 -0800)]
target: groundwork for "reset-assert" event

This defines a "reset-assert" event and a supporting utility
routine, and documents both how targets should implement it
and how config scripts should use it.  Core-specific updates
are needed to make this work.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
14 years agoARM11: write_memory() avoids increment check
David Brownell [Sat, 28 Nov 2009 02:40:37 +0000 (18:40 -0800)]
ARM11: write_memory() avoids increment check

When writing to a chip's "reset yourself" register, the ARM11 code
was reporting a spurious failure.  Just don't bother checking for
correctly incremented pointers given single-unit writes ... it's
a bit faster that way too.  (Reads should likely do the same thing.
For that matter, such checks are usually just a waste...)

Shrink an overlong parameter name, and associated lines'o'code.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
14 years agofix regression causing duplicated output
Zachary T Welch [Fri, 27 Nov 2009 21:51:17 +0000 (13:51 -0800)]
fix regression causing duplicated output

The command refactoring caused subcommand handlers to produce duplicate
output when run.  The problem was introduced by failing to ensure all
such invocations went through a top-level "catcher" script, prefixing
the command name with the 'ocd_' prefix and consuming its results.

The fix is to ensure such a top-level "catcher" script gets created
for each top-level command, regardless of whether it has a handler.
Indeed, this patch removes all command registrations for sub-commands,
which would not have worked in the new registration scheme anyway.

For now, dispatch of subcommands continues to be handled by the new
'unknown' command handler, which gets fixed here to strip the 'ocd_'
prefix if searching for the top-level command name fails initially.
Some Jim commands may be registered with this prefix, and that situation
seems to require the current fallback approach.  Otherwise, that prefix
could be stripped unconditionally and the logic made a little simpler.
The same problem must be handled by the 'help' command handler too,
so its lookup process works as intended.

Overall, the command dispatching remains more complicated than desired,
but this patch fixes the immediate regressions.

14 years agozy1000: keep up with new command registration stuff
Øyvind Harboe [Fri, 27 Nov 2009 19:41:06 +0000 (20:41 +0100)]
zy1000: keep up with new command registration stuff

jim and classic style commands are both supported.

Signed-off-by: Øyvind Harboe <oyvind.harboe@zylin.com>
14 years agoupdate minidummy interface driver command handling
Zachary T Welch [Fri, 27 Nov 2009 19:02:54 +0000 (11:02 -0800)]
update minidummy interface driver command handling

Changes the interface definition field reference from register_commands
to commands, which allows the module to compile.

14 years agofix 'nand info' command
Zachary T Welch [Fri, 27 Nov 2009 17:57:25 +0000 (09:57 -0800)]
fix 'nand info' command

Move device argument parsing after check for number of arguments;
otherwise, calling this command without any arguments would access
argv[0] before checking whether it even existed.

14 years agozy1000: keep up with changes to log_init()
Øyvind Harboe [Fri, 27 Nov 2009 18:41:22 +0000 (19:41 +0100)]
zy1000: keep up with changes to log_init()

fn's and return value for log_init() changed to void.

Signed-off-by: Øyvind Harboe <oyvind.harboe@zylin.com>
14 years agominidummy: fix compilation error
Øyvind Harboe [Fri, 27 Nov 2009 08:10:37 +0000 (09:10 +0100)]
minidummy: fix compilation error

during refactoring a search and replace error crept in

Signed-off-by: Øyvind Harboe <oyvind.harboe@zylin.com>
14 years agoARM NAND I/O header documentation update.
Dean Glazeski [Mon, 16 Nov 2009 18:47:45 +0000 (12:47 -0600)]
ARM NAND I/O header documentation update.

Fixed the header file to properly specify the doxygen documentation for the
items defined in it.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
14 years agoARM NAND I/O documentation update.
Dean Glazeski [Mon, 16 Nov 2009 18:43:39 +0000 (12:43 -0600)]
ARM NAND I/O documentation update.

This updates the functions in the file to all have doxygen comments
describing what they do.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
14 years agoARM NAND I/O refactor code copying.
Dean Glazeski [Mon, 16 Nov 2009 18:26:21 +0000 (12:26 -0600)]
ARM NAND I/O refactor code copying.

Created a function for copying code to the working area on
a target.  The NAND write and read functions are updated to
include use of this function.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
14 years agoARM NAND I/O read function.
Dean Glazeski [Sun, 15 Nov 2009 08:32:38 +0000 (02:32 -0600)]
ARM NAND I/O read function.

Implementation of the NAND read function for ARM NAND I/O that
includes running a local algorithm on a device to increase the
performance of block reads.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
14 years agoNAND Flash documentation update.
Dean Glazeski [Tue, 17 Nov 2009 20:16:23 +0000 (14:16 -0600)]
NAND Flash documentation update.

Updated doxygen comments for different interface structures for
the NAND interface.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
15 years agoXScale: add stub {read,write}_phys routines
David Brownell [Thu, 26 Nov 2009 20:00:56 +0000 (12:00 -0800)]
XScale: add stub {read,write}_phys routines

Just make these fail, instead of letting them write over
potentially random memory.  Users should be able to work
around the lack of real implementations by disbling the
MMU by hand ... until someone provides a Real Fix.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
15 years agoupdate bug reporting information
Uwe Hermann [Tue, 24 Nov 2009 19:42:42 +0000 (20:42 +0100)]
update bug reporting information

The Berlios bug-tracker is disabled, bug reports go to the list.

Signed-off-by: Zachary T Welch <zw@superlucidity.net>
15 years agofix typos in documentation
Uwe Hermann [Tue, 24 Nov 2009 19:28:21 +0000 (20:28 +0100)]
fix typos in documentation

Fixes some minor typos in the top-level documentation files.

Signed-off-by: Zachary T Welch <zw@superlucidity.net>
15 years agofix typos in source files
Uwe Hermann [Tue, 24 Nov 2009 19:01:15 +0000 (20:01 +0100)]
fix typos in source files

Correct some spelling errors in source comments and printed output.

Signed-off-by: Zachary T Welch <zw@superlucidity.net>
15 years agostyleguide: add some embedded style rules.
Øyvind Harboe [Tue, 24 Nov 2009 19:26:40 +0000 (20:26 +0100)]
styleguide: add some embedded style rules.

Embedded and pthreads rely on modest and predictable
stack usage.

Signed-off-by: Øyvind Harboe <oyvind.harboe@zylin.com>
15 years agofix 'flash protect' and 'flash erase_sector'
Eric Wetzel [Thu, 26 Nov 2009 15:54:45 +0000 (07:54 -0800)]
fix 'flash protect' and 'flash erase_sector'

Command upgrading introduced two off-by-one bugs in the flash commands.
This patch fixes the 'flash {protect,erase_sector}' commands to check
that they have been passed the correct number of arguments.

Ammended during commit to fix help text for 'erase_address' too.

15 years agotarget: create and use target_name()
David Brownell [Thu, 26 Nov 2009 00:38:08 +0000 (16:38 -0800)]
target: create and use target_name()

Several of the sites now using target_type_name() really
ought to be using an instance-specific name.  Create a
function called target_name(), accessing the instance's
own (command) name.

Use it in several places that really should be displaying
instance-specific names.  Also in several places which
were already doing so, but which had no wrapper to call.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
15 years agotarget: target_get_name() --> target_type_name()
David Brownell [Thu, 26 Nov 2009 00:38:08 +0000 (16:38 -0800)]
target: target_get_name() --> target_type_name()

There are two names that may matter on a per-target basis.
One is a per-instance name (for example, "at91sam7s.cpu").
The other is the name of its type (for example, "arm7tdmi"),
which is shared among multiple targets.

Currently target_get_name() returns the type name, which is
misleading and is rarely appropriate for target diagnostics.
Rename that as target_type_name().

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
15 years agoomap3530.cfg: yes we have SRAM!
David Brownell [Thu, 26 Nov 2009 00:19:53 +0000 (16:19 -0800)]
omap3530.cfg: yes we have SRAM!

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
15 years agoARM: minor armv4/armv5 cleanup
David Brownell [Thu, 26 Nov 2009 00:18:22 +0000 (16:18 -0800)]
ARM: minor armv4/armv5 cleanup

Lines of 300+ characters are still bad; debug tweaks.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
15 years agoARM: comment tweaks in ADIv5
David Brownell [Thu, 26 Nov 2009 00:14:45 +0000 (16:14 -0800)]
ARM: comment tweaks in ADIv5

"OptimoDE DESS" is ARM's semicustom DSPish stuff.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
15 years agoARM7/9: shrink run_algorithm_inner() lines
David Brownell [Thu, 26 Nov 2009 00:11:26 +0000 (16:11 -0800)]
ARM7/9: shrink run_algorithm_inner() lines

300+ characters is unreasonable.  So is half that.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
15 years agoNEWS updates
David Brownell [Thu, 26 Nov 2009 00:05:35 +0000 (16:05 -0800)]
NEWS updates

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
15 years agoupdate NEWS with recent developments
Zachary T Welch [Fri, 20 Nov 2009 15:06:32 +0000 (07:06 -0800)]
update NEWS with recent developments

Mention changes to flash bank command syntax, 'nand verify' command,
command error handling and reporting, and help/usage command upgrades.

15 years agouse ARRAY_SIZE macro
Zachary T Welch [Wed, 25 Nov 2009 04:32:10 +0000 (20:32 -0800)]
use ARRAY_SIZE macro

Search and destroy lingering cases where the ARRAY_SIZE macro should
be used to convey more intrinsic meaning in the OpenOCD code.

15 years agosupport OPENOCD_DEBUG_LEVEL environment setting
Zachary T Welch [Mon, 23 Nov 2009 20:14:05 +0000 (12:14 -0800)]
support OPENOCD_DEBUG_LEVEL environment setting

Detect the OPENOCD_DEBUG_LEVEL setting in log_init(), allowing the
very early startup phases to be debugged.

15 years agolog: improve initialization
Zachary T Welch [Sat, 21 Nov 2009 00:36:32 +0000 (16:36 -0800)]
log: improve initialization

Removes redundant assignment of start_ms from log_register_commands().
Eliminates command_context parameter and return value.
Adds Doxygen comment block for this API call.

15 years agoupdate command handler documentation
Zachary T Welch [Wed, 25 Nov 2009 02:47:35 +0000 (18:47 -0800)]
update command handler documentation

Adds sections on command registration and chaining, giving an overview
to developers that want to use these features.

15 years agoadd 'testee' target type
Zachary T Welch [Sun, 22 Nov 2009 06:09:53 +0000 (22:09 -0800)]
add 'testee' target type

Alliteration aside, this should provide the final piece of the puzzle
for developers that want to get started writing a new target type.
In this way, it also seeks to complement the 'dummy' interface driver
and 'faux' NOR flash driver.

15 years agoimprove command handling examples
Zachary T Welch [Sun, 22 Nov 2009 03:27:20 +0000 (19:27 -0800)]
improve command handling examples

Removes hello and foo commands from top-level registration.  Instead,
the dummy interface driver and faux flash driver have been augmented
to register these commands as sub-commands.

15 years agoadd script_command_run helper
Zachary T Welch [Wed, 25 Nov 2009 03:51:49 +0000 (19:51 -0800)]
add script_command_run helper

Eliminates duplicated code in script_command and handle_unknown_command.
Fixes bug with duplicated help output generated by placeholder commands.

15 years agoencapsulate and re-use log capture, retval setup
Zachary T Welch [Wed, 25 Nov 2009 03:13:03 +0000 (19:13 -0800)]
encapsulate and re-use log capture, retval setup

Factors log capture while running script commands, eliminating
duplicated code between script_command and jim_capture.  Factors
setting a command's Jim "retval" into a new helper as well.

Using these new helpers in the new unknown command handler's
fixes possible regressions caused by these bits being missing.

15 years agocombine help and usage command handlers
Zachary T Welch [Wed, 25 Nov 2009 06:18:28 +0000 (22:18 -0800)]
combine help and usage command handlers

Remove duplicated handler code by checking the running command name.

15 years agoimprove usage and help command output
Zachary T Welch [Mon, 23 Nov 2009 20:00:31 +0000 (12:00 -0800)]
improve usage and help command output

Rewrite formatting code in C, removing last remenants of TCL help code.
Sinificantly improves the readability by using smarter indent and wrap.

15 years agoallow scripts to update usage information
Zachary T Welch [Mon, 23 Nov 2009 20:16:27 +0000 (12:16 -0800)]
allow scripts to update usage information

The add_usage_text command uses the same C handler, which was updated
to support its new polymorphic role.  This patch updates the two script
commands that needed this support: 'find' and 'script'.

15 years agohttpd: use register_commands()
Zachary T Welch [Tue, 24 Nov 2009 18:58:32 +0000 (10:58 -0800)]
httpd: use register_commands()

Updates httpd_start() to use register_commands() for 'readform' and
'writeform' commands.  Adds server/httpd.h to export the new signatures
for this function (and httpd_stop), which allows removing the obsoleted
declarations inside openocd.c.

15 years agoadd jim_handler to command_registration
Zachary T Welch [Mon, 23 Nov 2009 23:03:04 +0000 (15:03 -0800)]
add jim_handler to command_registration

Adding jim_handler field to command_registration allows removing the
register_jim helper.  All command registrations now go through the
register_command{,s}() functions.

15 years agorefactor command_new to use command_registration
Zachary T Welch [Mon, 23 Nov 2009 23:01:12 +0000 (15:01 -0800)]
refactor command_new to use command_registration

Save stack space: use a struct.  Makes it easier to add new parameters.

15 years agoremove register_commands from etm_capture_driver
Zachary T Welch [Mon, 23 Nov 2009 16:24:02 +0000 (08:24 -0800)]
remove register_commands from etm_capture_driver

Converts callback to an array of command_registration records.
Moves oocd_trace driver definition to end of file to eliminate
useless forward declaration.

15 years agoremove target_type register_command callback
Zachary T Welch [Mon, 23 Nov 2009 16:17:01 +0000 (08:17 -0800)]
remove target_type register_command callback

Uses chaining of command_registration structures to eliminate all
target_type register_callback routines.  Exports the command_handler
registration arrays for those target types that are used by others.

15 years agoxscale: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:06 +0000 (07:43 -0800)]
xscale: use register_commands()

15 years agotrace: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:06 +0000 (07:43 -0800)]
trace: use register_commands()

15 years agotarget_request: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:06 +0000 (07:43 -0800)]
target_request: use register_commands()

15 years agotarget: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:06 +0000 (07:43 -0800)]
target: use register_commands()

15 years agooocd_trace: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:06 +0000 (07:43 -0800)]
oocd_trace: use register_commands()

15 years agoetm_dummy: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:06 +0000 (07:43 -0800)]
etm_dummy: use register_commands()

15 years agoetm: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:06 +0000 (07:43 -0800)]
etm: use register_commands()

15 years agoetb: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:06 +0000 (07:43 -0800)]
etb: use register_commands()

15 years agocortex_m3: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:06 +0000 (07:43 -0800)]
cortex_m3: use register_commands()

15 years agocortex_a8: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:06 +0000 (07:43 -0800)]
cortex_a8: use register_commands()

15 years agoarmv7m: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:06 +0000 (07:43 -0800)]
armv7m: use register_commands()

15 years agoarmv7a: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:05 +0000 (07:43 -0800)]
armv7a: use register_commands()

15 years agoarmv4_5: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:05 +0000 (07:43 -0800)]
armv4_5: use register_commands()

15 years agoarm9tdmi: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:05 +0000 (07:43 -0800)]
arm9tdmi: use register_commands()

15 years agoarm966e: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:05 +0000 (07:43 -0800)]
arm966e: use register_commands()

15 years agoarm926ejs: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:05 +0000 (07:43 -0800)]
arm926ejs: use register_commands()

15 years agoarm920t: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:05 +0000 (07:43 -0800)]
arm920t: use register_commands()

15 years agoarm7_9_common: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:05 +0000 (07:43 -0800)]
arm7_9_common: use register_commands()

15 years agoarm720t: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:05 +0000 (07:43 -0800)]
arm720t: use register_commands()

15 years agoarm11: use register_commands()
Zachary T Welch [Mon, 23 Nov 2009 15:43:05 +0000 (07:43 -0800)]
arm11: use register_commands()

15 years agoremove nand_controller->register_callbacks
Zachary T Welch [Sun, 22 Nov 2009 14:24:19 +0000 (06:24 -0800)]
remove nand_controller->register_callbacks

Replace flash_driver callback with pointer to command_registration.
Eliminates all related routines and allows drivers to omit commands.

15 years agoremove flash_driver->register_callbacks
Zachary T Welch [Sun, 22 Nov 2009 14:12:04 +0000 (06:12 -0800)]
remove flash_driver->register_callbacks

Replace flash_driver callback with pointer to command_registration.
Eliminates all related routines and allows drivers to omit commands.

15 years agotms470: use register_commands()
Zachary T Welch [Sun, 22 Nov 2009 12:13:56 +0000 (04:13 -0800)]
tms470: use register_commands()

15 years agostr9xpec: use register_commands()
Zachary T Welch [Sun, 22 Nov 2009 12:13:56 +0000 (04:13 -0800)]
str9xpec: use register_commands()

15 years agostr9x: use register_commands()
Zachary T Welch [Sun, 22 Nov 2009 12:13:56 +0000 (04:13 -0800)]
str9x: use register_commands()

15 years agostr7x: use register_commands()
Zachary T Welch [Sun, 22 Nov 2009 12:13:56 +0000 (04:13 -0800)]
str7x: use register_commands()

15 years agostm32x: use register_commands()
Zachary T Welch [Sun, 22 Nov 2009 12:13:56 +0000 (04:13 -0800)]
stm32x: use register_commands()

15 years agostellaris: use register_commands()
Zachary T Welch [Sun, 22 Nov 2009 12:13:56 +0000 (04:13 -0800)]
stellaris: use register_commands()

15 years agopic32mx: use register_commands()
Zachary T Welch [Sun, 22 Nov 2009 12:13:56 +0000 (04:13 -0800)]
pic32mx: use register_commands()

15 years agonand: use register_commands()
Zachary T Welch [Sun, 22 Nov 2009 12:13:56 +0000 (04:13 -0800)]
nand: use register_commands()

Eliminates 'nand_cmd' global variable.

15 years agomflash: use register_commands()
Zachary T Welch [Sun, 22 Nov 2009 12:13:56 +0000 (04:13 -0800)]
mflash: use register_commands()

15 years agolpc3180_nand_controller: use register_commands()
Zachary T Welch [Sun, 22 Nov 2009 12:13:56 +0000 (04:13 -0800)]
lpc3180_nand_controller: use register_commands()

15 years agolpc2900: use register_commands()
Zachary T Welch [Sun, 22 Nov 2009 12:13:56 +0000 (04:13 -0800)]
lpc2900: use register_commands()

15 years agolpc2000: use register_commands()
Zachary T Welch [Sun, 22 Nov 2009 12:13:56 +0000 (04:13 -0800)]
lpc2000: use register_commands()

15 years agoflash: use register_commands()
Zachary T Welch [Sun, 22 Nov 2009 12:13:56 +0000 (04:13 -0800)]
flash: use register_commands()

Eliminates 'flash_cmd' global variable.

15 years agoecos: use register_commands()
Zachary T Welch [Sun, 22 Nov 2009 12:13:56 +0000 (04:13 -0800)]
ecos: use register_commands()

15 years agocfi: use register_commands()
Zachary T Welch [Sun, 22 Nov 2009 12:13:56 +0000 (04:13 -0800)]
cfi: use register_commands()

15 years agoavrf: use register_commands()
Zachary T Welch [Sun, 22 Nov 2009 12:13:56 +0000 (04:13 -0800)]
avrf: use register_commands()

15 years agoat91sam7: use register_commands()
Zachary T Welch [Sun, 22 Nov 2009 12:13:56 +0000 (04:13 -0800)]
at91sam7: use register_commands()

15 years agoat91sam3: use register_commands()
Zachary T Welch [Sun, 22 Nov 2009 12:13:56 +0000 (04:13 -0800)]
at91sam3: use register_commands()

15 years agoremove register_callbacks from jtag interface
Zachary T Welch [Sat, 21 Nov 2009 06:01:59 +0000 (22:01 -0800)]
remove register_callbacks from jtag interface

Changes the jtag_interface->register_callbacks field to a list of
commands to be registered.  Changes callback to invocation of
register_commands() with that command registration list.  Removes all
JTAG interface driver register_command callback functions, which the
previous commits had converted into identical calls.

15 years agovsllink: use register_commands()
Zachary T Welch [Sat, 21 Nov 2009 05:06:28 +0000 (21:06 -0800)]
vsllink: use register_commands()

Use register_commands() with command registration array.

15 years agojtag: use register_commands()
Zachary T Welch [Sat, 21 Nov 2009 05:02:28 +0000 (21:02 -0800)]
jtag: use register_commands()

Use register commands with command registration array.

15 years agopresto: use register_commands()
Zachary T Welch [Sat, 21 Nov 2009 05:00:23 +0000 (21:00 -0800)]
presto: use register_commands()

Use register_commands() with command registration array.

15 years agoparport: use register_commands()
Zachary T Welch [Sat, 21 Nov 2009 04:17:52 +0000 (20:17 -0800)]
parport: use register_commands()

Use register_commands() with a command registration array.