Fixing one bug can easily uncover another .... in this case,
making sure that we properly invalidate some cached NOR state when
resuming arbitrary target code turned up an issue when the code
wasn't quite arbitrary (and we couldn't know that, but some parts
of OpenOCD assumed the cache would not be invalidated.
Specifically: some flash drivers (like CFI) update that state in loops
with downloaded algorithms, thus invalidating the state as it's probed.
+ Add a new target state flag, to record whether the target is
running downloaded algorithm code.
+ Use that flag to add a special case: "trust" downloaded algorithms
not to corrupt that cached state, bypassing cache invalidation.
Also update some of the documentation to stipulate that this flavor of
trustworthiness is now *required* ... not just a fortuitous acident.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Allow targets to run checks post reset. Used to check
that e.g. DCC downloads have been enabled.
Signed-off-by: Øyvind Harboe <oyvind.harboe@zylin.com>
Don't include <helper/jim.h> from target.h ... not everything
which touches targets needs to be able to talk to Jim. Plus,
most files include this header by another path.
Also, switch the affected files to use the classic sequence
for #included files: all <framework/headers.h> first, then
the "local_headers.h". This helps prevent growth of problematic
layering, by minimizing entanglement.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Most of these happened to be in the target.h file.
Some of those are associated with symbols that could be
removed at some point ... e.g. NVP_ASSERT/true and its
sibling NVP_DEASSERT/false.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Provide and use debug_reason_name() instead of expecting targets
to call Jim_Nvp_value2name_simple(). Less dependency on Jim, and
the code becomes more clear too.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Changes from the flat namespace to heirarchical one. Instead of writing:
#include "types.h"
the following form should be used.
#include <helper/types.h>
The exception is from .c files in the same directory.
Changes from the flat namespace to heirarchical one. Instead of writing:
#include "jim.h"
the following form should be used.
#include <helper/jim.h>
The exception is from .c files in the same directory.
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>
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>
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>
The container_of macro is useful as a general solution. It belongs
in types.h, rather than target.h where it was introduced. Requires
the offsetof macro, which comes from <stddef.h> (moved as well).
Lots of files still include it, often through needless
duplicate inclusion of "log.h"; sigh.
This cleans up the inclusion graph a bunch, so there are
fewer inclusion paths, but it doesn't change much otherwise.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Most files in the tree seem to have ended up including this,
and *quite* needlessly ... only code implementing or using
downloadable algorithms actually needs these declarations.
So take it out of the header files which included it, and put
it in files which use it ... reduce needless interdependencies.
Also: "algorithm.h" doesn't need to include "types.h" again;
it already comes from a different header.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Most files in the tree seem to have ended up including this,
and *quite* needlessly ... only code implementing or using
breakpoints actually needs these declarations.
So take it out of the header files which included it, and put
it in files which use it ... reduce needless interdependencies.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Previously this flag was stored in "target_type", so that for example
if there were two ARM7TDMI targets in a scan chain, both would claim
to have been examined although only the first one actually had its
examine() method called.
Move this state to where it should have been in the first place, and
hide a method that didn't need exposure ... the flag is write-once.
Provide some doxygen. The examine() method is confusing, since it
isn't separating one-time setup from the after-each-reset stuff. And
the ARM7/ARM9 version is, somewhat undesirably, not leaving the debug
state alone after reset ... probably more of an issue for trace setup
than for watchpoints and breakpoints.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Provide a cleaner way to handle single inheritance of targets
in C, using the same model Linux does: structs containing other
structs, un-nested via calls to a "container_of()" macro that
are packaged in typesafe inline functions.
Targets already use this containment idiom, but make it much
more complicated because they un-nest using embedded "void *"
pointers ... in chains of up to five per target, which is all
pure needless complication. (Example: arm92x core, arm9tdmi,
arm7_9, armv4_5 ... on top of the base "target" class.)
Applying this scheme consistently simplifies things, and gets
rid of many error-prone untyped pointers. It won't change any
part of the type model though -- it just simplifies things.
(And facilitates more cleanup later on.)
Rule of thumb: where there's an X->arch_info void* pointer,
access to that pointer can and should be removed. It may be
convenient to set up pointers to some of the embedded structs;
and shrink their current "*_common" names (annoyingly long).
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Fixed bug: if virtual address for working memory was not specified
and MMU was enabled, then address 0 would be used.
Require working address to be specified for both MMU enabled
and disabled case.
For some completely inexplicable reason this fixes the regression
in svn 2646 for flash write in arm926ejs target. The logs showed
that MMU was disabled in the case below:
https://lists.berlios.de/pipermail/openocd-development/2009-November/011882.html
Signed-off-by: Øyvind Harboe <oyvind.harboe@zylin.com>
- Move definition of 'struct target_type_s' into new 'target_type.h' file.
- Forward delclaration remains in target.h, with comment pointing to new file.
- Replaces #define with #include in source files.
git-svn-id: svn://svn.berlios.de/openocd/trunk@1971 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- Add DEFINE_TARGET_TYPE_S symbol in files that need it defined.
- Forward declare 'struct target_type_s' only, unless that symbol is defined.
git-svn-id: svn://svn.berlios.de/openocd/trunk@1969 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- replaces all calls to target->type->{add,remove}_{break,watch}point.
git-svn-id: svn://svn.berlios.de/openocd/trunk@1967 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- replaces all accesses to target->type->name.
- add documentation in target_s to warn not to access field directly.
git-svn-id: svn://svn.berlios.de/openocd/trunk@1966 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- replaces all calls to target->type->get_gdb_reg_list.
- add documentation in target_s to warn not to invoke callback directly.
git-svn-id: svn://svn.berlios.de/openocd/trunk@1964 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- replaces all calls to target->type->bulk_write_memory.
- add documentation in target_s to warn not to invoke callback directly.
git-svn-id: svn://svn.berlios.de/openocd/trunk@1963 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- replace all checks of target->type->examined with target_was_examined().
- replace all setting of target->type->examined with target_set_examined().
- replace clearing of target->type->examined with target_reset_examined().
- add documentation in target_s to warn not to access field directly.
git-svn-id: svn://svn.berlios.de/openocd/trunk@1962 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- replaces all calls to target->type->run_algorithm.
- add documentation in target_s to warn not to invoke callback directly.
git-svn-id: svn://svn.berlios.de/openocd/trunk@1961 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- replaces all calls to target->type->write_memory.
- add documentation in target_s to warn not to invoke callback directly.
git-svn-id: svn://svn.berlios.de/openocd/trunk@1960 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- replaces all calls to target->type->read_memory.
- add documentation in target_s to warn not to invoke callback directly.
git-svn-id: svn://svn.berlios.de/openocd/trunk@1959 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- cleaned up headers to match rest of code
- added missing svn props for previously added files
git-svn-id: svn://svn.berlios.de/openocd/trunk@987 b42882b7-edfa-0310-969c-e2dbd0fdcd60
converts a number of 'simple string lookup tables' into NVP tables.
These NVP tables will be used by various commands coming in the next patch.
git-svn-id: svn://svn.berlios.de/openocd/trunk@962 b42882b7-edfa-0310-969c-e2dbd0fdcd60
into the target implementation.
Also target_process_reset() is now simpler and has error handling,
e.g. if assert reset fails, then target_process_reset() will propagate
that error.
cmd_ctx was passed in to examine(), which is wrong - removed that.
git-svn-id: svn://svn.berlios.de/openocd/trunk@887 b42882b7-edfa-0310-969c-e2dbd0fdcd60
from pre/post_reset event scripts. Adding the second parameter was
a mistake seen in retrospect. this gives precise control in post_reset
for *when* the post reset speed is set. The pre_reset event was
added *after* the second parameter to jtag_khz/speed
- the target implementations no longer gets involved in the reset mode
scheme. Either they reset a target into a halted mode or not.
target_process_reset()
detects if the reset halt failed or not.
- tcl target event names are now target_N_name. Mainly internal
at this early stage, but best to get the naming right now.
- added hardcoded reset modes from gdb_server.c. I don't know precisely what
these defaults should be or if it should be made configurable. Perhaps some
hardcoded defaults will do for now and it can be made configurable later.
- bugfix in cortex_m3.c for reset_run_and_xxx?
- issue syntax error upon obsolete argument in target command instead of
printing message that will surely drown in the log
git-svn-id: svn://svn.berlios.de/openocd/trunk@849 b42882b7-edfa-0310-969c-e2dbd0fdcd60
- Log output handlers now get a "const char *line"
- Added "const" to parameter.
src/helper/command.c
src/helper/command.h
- New function: 'command_output_text()'
- Log output handlers now get a "const char *line"
src/helper/options.c
- Log output handlers now get a "const char *line"
src/server/telnet_server.c
- DO not transmit NULL bytes via TELNET.
- Log output handlers now get a "const char *line"
src/server/gdb_server.c
- Log output handlers now get a "const char *line"
*** THIS INCLUDES PORTIONS OF A PATCH FROM Oyvind sent
previously to the mailing list for TCL users try
src/target/target.c
*** THIS INCLUDES PORTIONS OF A PATCH FROM Oyvind sent
previously to the mailing list for TCL users try
src/target/target.h
*** THIS INCLUDES PORTIONS OF A PATCH FROM Oyvind sent
previously to the mailing list for TCL users try
src/openocd.c
- **MAJOR** Work: New TCL/Jim function: mem2array
- **MAJOR** Work: Redirect Tcl/Jim stdio output to remote users.
- Previously: TCL output did not go to GDB.
- Previously: TCL output did not go to TELNET
- Previously: TCL output only goes to control console.
- This fixes that problem.
+ Created callbacks:
+openocd_jim_fwrite()
+openocd_jim_fread()
+openocd_jim_vfprintf()
+openocd_jim_fflush()
+openocd_jim_fgets()
src/Makefile.am
- New TCL files.
- Future note: This should be more automated. As the list of
'tcl' files grows maintaning this list will suck.
src/Jim.c
- ** THIS INCLUDES A PREVIOUS PATCH I SENT EARLIER **
- that impliments many [format] specifies JIM did not support.
- Jim_FormatString() - **MAJOR** work.
- Previously only supported "%s" "%d" and "%c"
- And what support existed had bugs.
- NEW: *MANY* formating parameters are now supported.
- TODO: The "precision" specifier is not supported.
** NEW ** This patch.
- Jim_StringToWide() test if no conversion occured.
- Jim_StringToIndex() test if no conversion occured.
- Jim_StringToDouble() test if no conversion occured.
** NEW ** This Patch. Major Work.
- Previously output from JIM did not get sent to GDB
- Ditto: Output to Telnet session.
- Above items are now fixed - By introducing callbacks
new function pointers in the "interp" structure.
- Helpers that call the callbacks.
- New function: Jim_fprintf()
- New function: Jim_vfprintf()
- New function: Jim_fwrite()
- New function: Jim_fread()
- New function: Jim_fflush()
- New function: Jim_fgets()
By default: the output is to STDIO as previous.
The "openocd.c" - redirects the output as needed.
- Jim_Panic() - Send panic to both STDERR and the interps
specified STDERR output as a 2nd choice.
- Now JIM's "stdin/stdout/stderr" paramters are "void *"
and are no longer "FILE *".
src/Jim.h
- **MAJOR**
- New JIM STDIO callbacks in the "interp" structure.
- change: "stdin/stdout/stderr" are now "void *" cookies.
- New JIM stdio helper functions.
git-svn-id: svn://svn.berlios.de/openocd/trunk@755 b42882b7-edfa-0310-969c-e2dbd0fdcd60
to daemon_startup(still supported).
- print warning if srst and trst change state at the same time when srst_and_trst
is seperate
- reset now performs a trst, examines and validates the jtag chain before targets
assert reset
- if startup fails to examine and validate the jtag chain, try a reset before
trying again
git-svn-id: svn://svn.berlios.de/openocd/trunk@552 b42882b7-edfa-0310-969c-e2dbd0fdcd60
Wrote a default flash erase check fn which uses CFI's target algorithm
w/fallback to memory reads.
- "flash info" no longer prints erase status as it is stale.
- "flash erase_check" now prints erase status. erase check can take a
*long* time. Work in progress
- arm7/9 with seperate srst & trst now supports reset init/halt
after a power outage. arm7/9 no longer makes any assumptions
about state of target when reset is asserted.
- fixes for srst & trst capable arm7/9 with reset init/halt
- prepare_reset_halt retired. This code needs to be inside
assert_reset anyway
- haven't been able to get stm32 write algorithm to work. Fallback
flash write does work. Haven't found a version of openocd trunk
where this works.
- added target_free_all_working_areas_restore() which can
let be of restoring backups. This is needed when asserting
reset as the target must be assumed to be an unknown state.
Added some comments to working areas API
- str9 reset script fixes
- some guidelines
- fixed dangling callbacks upon reset timeout
git-svn-id: svn://svn.berlios.de/openocd/trunk@536 b42882b7-edfa-0310-969c-e2dbd0fdcd60
David Brownell
Øyvind Harboe
Zachary T Welch
dbrownell
oharboe
duane
zwelch
kc8apf
ntfreak