Kick in ETM (and ETB) support for ARM11. Tested on OMAP 2420,
so update that configuration. (That's an ARM1136ejs, ETB,
OpenGL ES1.1, C55x DSP, etc.)
Also update the other ARM11 ETM + ETB targets in the tree
to set up these modules. (Not tested.)
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Various cores with an ETB have its TAP misnamed ... either as a
boundary scan TAP or as the iMX "Secure JTAG Controller" (which
is, among other things, a JRC that could be used to shorten
scan chains).
Use the correct name for these TAPs, which we can recognize since
their IDs were assigned by ARM and these chips all document the
presence of an ETB. The 0x2b900f0f is ETB11; the 0x1b900f0f
is an older module, just called "ETB".
Also shrink the ETB's IR configuration; the default IR-Capture
value is fine, and the mask can specify that all four bits are
safe to check (per ARM documentation).
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
General rule, this is all board-specific and doesn't belong
in target config files. Some of these were just cosmetic.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Here's a patch for the double-reset problem on STM32. I've tested
downloading and debugging with GDB and Eclipse, and everything seems
to work fine.
This effectively sets reset_config to none. trst_only would also
be ok, but that's better left to a board configuration file since
not all boards wire it up.
The NVIC is used to trigger reset, which at least on this chip also
pulses nSRST so the whole system does get rest -- exactly once.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
The semantics of "-work-area-virt 0" (or phys) changed with
the patch to require specifying physical or virtrual work
area addresses. Specifying zero was previously a NOP. Now
it means that address zero is valid.
This patch addresses three related issues:
- MMU-less processors should never specify work-area-virt;
remove those specifications. Such processors include
ARM7TDMI, Cortex-M3, and ARM966.
- MMU-equipped processors *can* specify work-area-virt...
but zero won't be appropriate, except in mischievous
contexts (which hide null pointer exceptions).
Remove those specs from those processors too. If any of
those mappings is valid, someone will need to submit a
patch adding it ... along with a comment saying what OS
provides the mapping, and in which context. Example,
say "works with Linux 2.6.30+, in kernel mode". (Note
that ARM Linux doesn't map kernel memory to zero ...)
- Clarify docs on that "-virt" and other work area stuff.
Seems to me work-area-virt is quite problematic; not every
operating system provides such static mappings; if they do,
they're not in every MMU context...
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Gets rid of the runtime warning "stm32.bs: nonstandard IR mask"
[dbrownell@users.sourceforge.net: line lengths, note issue, section ref]
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
This gets rid of runtime warnings from the use of numbers.
STM32 and LPC2103 were tested. Other LPC updates are the
same, and so are safe. The CFI updates match other tested
changes now in the tree.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Now I can issue "reset halt" and have everything act smoothly;
the vector_catch hardware is obviously not kicking in, but the
rest of the reset sequence acts sanely.
- TAP "setup" event enables the DAP, not omap3_dbginit
(resolving a chicken/egg bug I noted a while back)
- Remove stuff from omap3_dbginit which should never be
used in event handlers
- Cope better with slow clocking during reset
Also, stop hard-wiring the target name: use the input params in
the standard way, and set up $_TARGETNAME as an output param.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Startup now mostly works, except that the initial target state
is "unknown" ... previously, it refused to even start.
Getting that far required fixing the ircapture value (which
can never have been correct!) and the default JTAG clock rate,
then providing custom reset script.
The "reset" command is still iffy. DCSR updates, and loading
the debug handler, report numerous DR/IR capture failures.
But once that's done, "poll" reports that the CPU is halted
(which it shouldn't be, this was "reset run"!), due to the
rather curious reason "target-not-halted".
Summary: you still can't debug these parts, but it's closer.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Without some extra delay after releasing SRST, we seemed to
be trying to talk to the TAP before it was ready to respond.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
The model is that this fires after scanchain verification, when it's
safe to call "jtag tapenable $TAPNAME". So it will fire as part of
non-error paths of "init" and "reset" command processing. However it
will *NOT* trigger during "jtag_reset" processing, which skips all
scan chain verification, or after verification errors.
ALSO:
- switch DaVinci chips to use this new mechanism
- log TAP activation/deactivation, since their IDCODEs aren't verified
- unify "enum jtag_event" scripted event notifications
- remove duplicative JTAG_TAP_EVENT_POST_RESET
git-svn-id: svn://svn.berlios.de/openocd/trunk@2800 b42882b7-edfa-0310-969c-e2dbd0fdcd60
It can be sped up later, once it's known the PLLs are active.
Note that modern tools from TI all use adaptive clocking; and
that if that's done with OpenOCD, "too fast" is also a non-issue.
git-svn-id: svn://svn.berlios.de/openocd/trunk@2740 b42882b7-edfa-0310-969c-e2dbd0fdcd60
Update the board config for the DaVinci DM355 EVM so the reset-init
event handler does the rest of the work it should do:
- minor PLL setup bugfixes
- initialize the DDR2 controller
- probe both NAND banks
- initialize UART0
- enable the icache
git-svn-id: svn://svn.berlios.de/openocd/trunk@2699 b42882b7-edfa-0310-969c-e2dbd0fdcd60
David Brownell
Thomas Kindler
Øyvind Harboe
Freddie Chopin
Wookey
John Rigby
dbrownell
mlu
oharboe