Antonio Borneo [Fri, 22 May 2020 16:55:34 +0000 (18:55 +0200)]
target/cortex_a: fix memory leak of register cache
There is no method to free the register cache, allocated in
armv4_5, so we get a memory leak.
Issue identified by valgrind.
Implement the method arm_free_reg_cache() and call it in cortex_a
deinit and to exit for error during arm_dpm_setup().
Tested on dual cortex-A stm32mp15x.
This change is inspired from similar fix in commit b01b5fe13a67
("armv7m: Fix memory leak in register caching.").
The same allocation is also used by target types "arm7tdmi",
"arm9tdmi", "arm11" and "xscale" but they all lack the deinit
method and I do not have relevant HW to test the fix. For such
reasons they are not addressed in this patch.
Change-Id: I4da1e1f12e36ec245d1f3b11a4eafcbd9a1d2e25 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/5693 Tested-by: jenkins
Antonio Borneo [Mon, 25 May 2020 19:37:12 +0000 (21:37 +0200)]
libusb_helper: fix memory leak when no adapter is found
When jtag_libusb_open() fails to find the adapter, it returns
error but left libusb initialized causing memory leak of libusb
internal data.
Issue found with valgrind when no adapter or board is connected to
the host, e.g.
valgrind openocd -f board/st_nucleo_f4.cfg
Close the libusb operations if jtag_libusb_open() has to return
error.
Change-Id: Ieb2f110be15705dafe80c099e7d83c07056c2a41 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/5701 Tested-by: jenkins
Antonio Borneo [Tue, 26 May 2020 22:03:45 +0000 (00:03 +0200)]
target/cortex-m: enable C_DEBUGEN during examine
Current code for Cortex M does not set C_DEBUGEN as soon as
possible, (which means during target examine), but later-on either:
1) at command 'halt' (e.g. for 'gdb-attach' event);
2) at command 'soft_reset_halt';
3) at commands 'reset', 'reset halt' or 'reset init';
4) during polling, but only if the target:
= enter in 'double fault', or
= exit from a reset, or
= halts (not possible if C_DEBUGEN is not set)
Plus, if commands in 1) or 2) are executed before the very first
poll of the target, the value of 'cortex_m->dcb_dhcsr' is used not
initialized while writing it back in DCB_DHCSR.
Another side effect of this situation is that it's possible to set
a HW breakpoint with the target running, while C_DEBUGEN is not
set. Accordingly to [1], C1.3.1 "Debug authentication":
When DGBEN is LOW and DHCSR.S_HALT == 0:
...
FPB breakpoints do not generate an entry to Debug state and,
if no DebugMonitor exception is generated, will escalate to
HardFault, Lockup, or be ignored.
On STM32MP15x I get HW breakpoint ignored, while on STM32F411 I
get HardFault.
E.g. following these steps:
- power-on a pre-flashed board that starts running the firmware;
- connect openocd, without halting or resetting the board;
- set a HW breakpoint to some address often executed;
- wait, but the board doesn't halt ...;
- type the command 'halt';
- if the Cortex-M has HardFault it would be visible and the fault
is at the breakpoint address;
- if no HardFault then type the command 'resume';
- wait and the board will finally halt at the HW breakpoint.
A similar issue has been detected on Cortex-A code and fixed by
commit bff87a7f28fb ("target/cortex_a: enable DSCR_HALT_DBG_MODE
during examine").
Follow the same approach and set C_DEBUGEN during examine.
Also, initialize 'cortex_m->dcb_dhcsr' during examine.
[1] ARM DDI 0403E "ARM v7-M Architecture Reference Manual"
Change-Id: I5b0b23403634f7dfce38f104bba9f59c33eb3e99 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/5702 Tested-by: jenkins Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com> Reviewed-by: Tarek BOCHKATI <tarek.bouchkati@gmail.com> Reviewed-by: Moritz Fischer <moritzf@google.com>
Antonio Borneo [Fri, 29 May 2020 10:00:35 +0000 (12:00 +0200)]
log: fix kept_alive() and report expired timeout
The kept_alive() function is called to inform the keep-alive code
that a keep-alive as been just kicked through some other path.
But kept_alive() erroneously resets the timeout counter without
checking if it has already expired, thus masking a potential
timeout.
Check if timeout counter has expired during kept_alive().
While there, put the timeout values in macros and explicit the
units in the timeout messages.
Antonio Borneo [Fri, 29 May 2020 08:42:43 +0000 (10:42 +0200)]
nor/kinetis: add keep-alive during flash write
Flashing Kinetis MCU MK22FX512VLH12 takes longer than one second,
thus gdb connection can timeout. Before commit 7f260f5009a7
("helper/command: Handle Tcl return values consistently") the
openocd timeout warning was masked by a bug in gdb_put_packet()
that resets the timeout counter if any message is logged out.
The commit above removes one log message and the timeout warning
is now triggered.
While the bug in gdb_put_packet() is still to be fixed, the reason
of the timeout is the lack of keep_alive() calls in kinetis flash
driver.
Add keep_alive() call at every iteration in function
kinetis_write_sections(), that is reported as core write function
in chunks of 1024 bytes in the log from the reporter.
Add also a keep_alive() call at every iteration in function
kinetis_write_inner(). This part is not present in the log but, by
code analysis, it represents another critical loop.
jtag/drivers/bcm2835gpio: bcm2835gpio_init has been updated
For jtag/drivers/bcm2835gpio dev_mem_fd has been updated within
bcm2835gpio_init with the add on of gpio to mem. This permits the
access to memory of GPIO without the need for root access.
For failed attempt, a fallback to original memory follows.
It should be noted that any printed error is relative to original
memory ("dev/mem").
Tested on EFM32GG12B390F board
Update Autotools Mythbuster link to avoid multiple redirects.
While I'm hoping to maintain the chain of redirects for the foreseeable
future, it makes sense to point at the final URL now that Autotools
Mythuster has had its domain for a long while.
Change-Id: I6d912e3be0f81b25bec14746ebfee58f29f144b6 Signed-off-by: Diego Elio Pettenò <flameeyes@flameeyes.com>
Reviewed-on: http://openocd.zylin.com/5709 Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com> Tested-by: jenkins Reviewed-by: Tarek BOCHKATI <tarek.bouchkati@gmail.com> Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Antonio Borneo [Sun, 20 Oct 2019 17:40:28 +0000 (19:40 +0200)]
helper/command: strip replicated command name in log
After commit 0840414f0e57 ("helper/command: do not replace new
commands with ocd_ prefix"), the command name is present in
argv[0], so there is no need to pass it directly. The current code
causes the command name to be logged twice, once explicitly and
then from the content of argv[0]:
openocd -c 'debug_level 3; echo hello; shutdown'
<snip>
Debug: 13 3 command.c:142 script_debug(): command - echo echo hello
<snip>
Debug: 16 4 command.c:142 script_debug(): command - shutdown shutdown
Remove the command name from the arguments of the function
script_debug().
Change-Id: I57860774f450ff717ee71ef9dc07590549a84319 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/5660 Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com> Tested-by: jenkins
Antonio Borneo [Thu, 21 May 2020 11:24:32 +0000 (13:24 +0200)]
stlink: fix open AP for v2j37 and v3j7
The new stlink firmware requires opening the AP before issuing
any operation.
In the current code we have a 'questionable' check about the core
model to set the TAR autoincrement, that is issued without opening
the AP, thus causing a STLINK_BAD_AP_ERROR.
Modify the AP open API to handle this case and open AP#0 before
the memory access to check the core model.
Change-Id: I576955b5094bd41d63ff1fbad7b4fd9433253321 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com> Reported-by: Andreas Bolsch <hyphen0break@gmail.com>
Reviewed-on: http://openocd.zylin.com/5691 Tested-by: jenkins Reviewed-by: Andreas Bolsch <hyphen0break@gmail.com>
Antonio Borneo [Mon, 23 Mar 2020 14:58:01 +0000 (15:58 +0100)]
stlink: default dapdirect to SWD instead of JTAG
When the transport is not specified, OpenOCD uses the first listed
by the selected adapter driver.
The old HLA driver for stlink lists SWD as first. The new driver
dapdirect instead lists JTAG, making more difficult a fallback to
HLA when dapdirect is not available due to old stlink FW.
Plus, in case of JTAG wiring, SWD is still possible, while the
opposite is not.
Reorder the list of transports to grant the default to SWD.
Ake Rehnman [Sun, 29 Mar 2020 19:57:53 +0000 (21:57 +0200)]
stm8 target: make adapter speed settings work
Previously the adapter speed settings were hard-coded to
connect with low speed then switch over to high speed
regardless what was mentioned in the cfg files. Now the
stm8 target intercept adapter speed settings and configure
the stm8 control registers accordingly.
Antonio Borneo [Thu, 26 Mar 2020 09:49:04 +0000 (10:49 +0100)]
tcl/board: add board ST nucleo-8l152r8
The transport SWIM is selected twice, in board and in target, thus
a warning is generated at run-time. It should be fixed by remove a
line in the target file, but does not harm so let's keep it there
for the time being.
Change-Id: I479004dc16005a330d552c8dbd5def61690f9b9f Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/5543 Tested-by: jenkins
Antonio Borneo [Tue, 28 Jan 2020 14:53:06 +0000 (15:53 +0100)]
stlink: simplify handling of SWIM
Now that SWIM is not accessed through HLA anymore, decouple the
SWIM code and remove the conditional execution.
Fix the inconsistency of the return type for stlink_usb_state() in
case of SWIM (returns ERROR_OK while type is enum target_state)
introduced by commit 3de6b5f6e52f ("jtag/drivers/stlink_usb :
implemented and repaired SWIM support").
The code added by commit above in stlink_usb_state() is an hack to
reuse existing HLA API to perform a reconnect.
Move the SWIM specific code from stlink_usb_state() to a dedicated
stlink_swim_op_reconnect() that provides consistent data type.
Change-Id: I3fe175fef00b0735bea6139b057f217a080c9d38 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com> Fixes: 3de6b5f6e52f ("jtag/drivers/stlink_usb : implemented and repaired SWIM support")
Reviewed-on: http://openocd.zylin.com/5532 Tested-by: jenkins
Antonio Borneo [Tue, 4 Feb 2020 11:53:54 +0000 (12:53 +0100)]
stlink: simplify mem R/W with SWIM
Thanks to API separation between SWIM and the other transports, we
can easily split the memory read/write for SWIM from the rest of
the code.
While there, use the macro STLINK_DATA_SIZE as size of data chunks
that can be read/write in SWIM. This was not implemented before.
Change-Id: I7d913c92539007e4d914480bacc0126a8f0e9705 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/5531 Tested-by: jenkins
Antonio Borneo [Mon, 27 Jan 2020 17:45:10 +0000 (18:45 +0100)]
swim: add new transport
Add SWIM and STM8 to documentation and update TODO file.
Introduce transport "swim" and command "swim newtap".
Switch in swim.c from HLA API to the new SWIM API.
Implement in stlink driver the SWIM APIs as wrappers of existing
HLA functions.
Remove any SWIM related reference from HLA files.
Update stm8 config files and stlink-dap interface config file.
Change-Id: I2bb9f58d52900f6eb4df05f979f7ef11fd439c24 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/5530 Tested-by: jenkins
Antonio Borneo [Tue, 4 Feb 2020 10:07:01 +0000 (11:07 +0100)]
swim: fix adapter speed handling
SWIM transport only supports two adapter speeds:
- "low speed" equal to 363 kHz (8 MHz / 22)
- "high speed" equal to 800 kHz (8 MHz / 10)
Replace the previous convention that use "0" or "1" for "low" or
"high" speed with the effective speed in kHz.
Rework the implementation of stlink_speed_swim().
Set low speed in the stm8 config files, because only low speed is
permitted at debug connection; the previous code ignores the
initial value.
Change-Id: I2484c9419a2c554c59eb6b9216339393ab0b54f3 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/5529 Tested-by: jenkins
Antonio Borneo [Sun, 26 Jan 2020 16:00:55 +0000 (17:00 +0100)]
swim: abstract the transport in stm8 target
SWIM is implemented by (ab)using the HLA API. This was acceptable
when OpenOCD code did not provided a clear separation between
transports and related APIs. Still today SWIM in OpenOCD is only
supported by STLink, so the decision to re-use the HLA API was the
simpler way to implement it.
After commit efd1d642220a ("adapter: switch from struct
jtag_interface to adapter_driver") the transports API are better
split and SWIM can be implemented as a separate set of API. This
would open the possibility to extend OpenOCD for other adapters
that provide SWIM, e.g. versaloon, or through SPI emulation [1].
Introduce a new set of files swim.[ch] to handle the SWIM API.
Beside the API that almost match the transport low-level data
communication (system_reset, read_mem, write_mem), add a further
API reconnect. Today, inside HLA STLink code, the reconnect is
implemented by hacking the HLA API state(). Please notice that
due to this hack the return type is incorrect; stlink_usb_state()
returns ERROR_OK in SWIM mode, while its return type is enum
target_state. Ignore the type mismatch and still call the HLA API
state in the new SWIM API reconnect. Further commit will fix it.
[1] http://kuku.eu.org/?projects/stm8spi/stm8spi
Change-Id: I52018e1e2200cbd41af8e5031f7b35dc761b61d6 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/5528 Tested-by: jenkins
Antonio Borneo [Sun, 26 Jan 2020 15:30:14 +0000 (16:30 +0100)]
adapter: expose HLA interface in struct adapter_driver
The transition of STM8/SWIM out of HLA will require a new struct
swim_ops in struct adapter_driver.
To simplify the development, make the HLA interface temporarily
accessible through the struct adapter_driver.
This commit will be reverted after the swim rework.
Change-Id: I1e4f370cf64641164d7bcaa22f75ac58c9240052 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/5527 Tested-by: jenkins
Antonio Borneo [Wed, 18 Mar 2020 21:35:26 +0000 (22:35 +0100)]
stlink: reduce use of hla specific enum hl_transports
In the driver's code it's widely used the enum hl_transports for
any check concerning the current transport, even for the non-hla
transport dapdirect.
The driver already provides a stlink specific enum stlink_mode
that can be used in place of the hla one.
Replace the hla enum with the stlink one allover the code.
Introduce a hla specific wrapper stlink_usb_hl_open() to cope with
the only API that requires the hla specific enum.
The overall behaviour is not changed, with exception for a debug
message in stlink_usb_open() than now prints the numeric value of
stlink enum in place of the numeric value of the corresponding hla
one.
This change prepares the road for moving SWIM transport out of hla
by removing any reference to the macro HL_TRANSPORT_SWIM from the
stlink driver.
Change-Id: Ieeea34f312245a94bfc1333087afdb8eb9f77139 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/5526 Tested-by: jenkins
Expression like (0xffff << 16) evaluate to type int, which is not able
to hold that value, producing a warning when compiling with
-fsanitize=undefined. This patch makes most of the cortex_m constants
unsigned using the BIT() macro or appending "ul" when possible to fix
the undefined behavior warning.
Signed-off-by: iosabi <iosabi@protonmail.com>
Change-Id: I7af194305ef612d7a32e74eaf9f11dd85fa87f32
Reviewed-on: http://openocd.zylin.com/5583 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com> Reviewed-by: Christopher Head <chead@zaber.com> Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Antonio Borneo [Mon, 6 May 2019 10:18:30 +0000 (12:18 +0200)]
coding style: add missing space when split strings
Long strings are split across few lines; usually split occurs at
the white space between two words.
Check that the space between the two words is still present.
While there, adjust the amount of space between words.
Issue identified by checkpatch script from Linux kernel v5.1 using
the command
Antonio Borneo [Sun, 5 May 2019 23:03:52 +0000 (01:03 +0200)]
coding style: avoid unnecessary line continuations
Line continuation, adding a backslash as last char of the line, is
requested in multi-line macro definition, but is not necessary in
the rest of C code.
Remove it where present.
Identified by checkpatch script from Linux kernel v5.1 using the
command
Antonio Borneo [Sun, 5 May 2019 22:35:52 +0000 (00:35 +0200)]
coding style: add parenthesis around the argument of sizeof
The script checkpatch available in new Linux kernel offers an
experimental feature for automatically fix the code in place.
While still experimental, the feature works quite well for simple
fixes, like parenthesis.
This patch has been created automatically with the script under
review for inclusion in OpenOCD, using the command:
Antonio Borneo [Sun, 5 May 2019 22:03:03 +0000 (00:03 +0200)]
coding style: remove useless break after a goto or return
In a switch/case statement, a break placed after a goto or return
is never executed.
The script checkpatch available in Linux kernel v5.1 issues a
warning for such unused break statements.
In the process of reviewing the new checkpatch for its inclusion
in OpenOCD, let's get rid of these warnings.
The script checkpatch is unable to fixup automatically this case.
Thanks to having "break" command using a single code line, this
patch has been generated using the script below:
find src/ -type f -exec ./tools/scripts/checkpatch.pl -q \
--types UNNECESSARY_BREAK -f {} \; \
| sed -n '/^#/{s/^.*FILE: //;s/:$//;s/:/ /;p}' \
| awk 'function P() {print "sed -i '\''"b"'\'' "a};
{
if ($1!=a) {
if (a) {P()};
a=$1;
b=$2"{d}";
} else {
b=b";"$2"{d}"
}
};
END {P()}'
Change-Id: I56ca098faa5fe8d1e3f712dc0a029a3f10559d99 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/5617 Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com> Tested-by: jenkins
Antonio Borneo [Sat, 25 Apr 2020 23:07:27 +0000 (01:07 +0200)]
tcl: remove trailing whitespace
The script checkpatch available in new Linux kernel offers an
experimental feature for automatically fix the code in place.
While still experimental, the feature works quite well for simple
fixes, like spacing.
This patch has been created automatically with the script under
review for inclusion in OpenOCD, using the command:
find tcl/ -type f -exec ./tools/scripts/checkpatch.pl \
-q --types TRAILING_WHITESPACE --fix-inplace -f {} \;
The patch only changes amount and position of whitespace, thus
the following commands show empty diff
git diff -w
git log -w -p
git log -w --stat
Change-Id: Ie7e3a236f4db9c70019e3b3c7e851edbd3a9dd84 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/5616 Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com> Tested-by: jenkins
Antonio Borneo [Sat, 25 Apr 2020 23:25:32 +0000 (01:25 +0200)]
tcl: fix typo and spelling
Identified by checkpatch script from Linux kernel v5.7-rc1 using
the command
find tcl/ -type f -exec ./tools/scripts/checkpatch.pl \
-q --types TYPO_SPELLING --strict -f {} \;
Change-Id: I7b523f0ab5ec047ff167742a44c29984ac672cf4 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/5615 Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com> Tested-by: jenkins
Antonio Borneo [Sat, 25 Apr 2020 23:39:16 +0000 (01:39 +0200)]
doc: fix typo and spelling
Identified by checkpatch script from Linux kernel v5.7-rc1 using
the command
find doc/ -type f -exec ./tools/scripts/checkpatch.pl \
-q --types TYPO_SPELLING --strict -f {} \;
Change-Id: I1269ac966027439e16eb6e63179e43925bec37fa Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/5614 Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com> Tested-by: jenkins
Antonio Borneo [Mon, 27 Apr 2020 14:52:31 +0000 (16:52 +0200)]
doc: fix texinfo warning on @deffn not at the line beginning
Commit 87a4158acf56 ("drivers: xds110: Clean up command syntax and
documentation") does not follow the documentation structure that
lists the adapters in cpt 2 "Debug Adapter Hardware" then lists
the adapter commands in cpt 8.2 "Interface Drivers"; it puts all
in cpt 2.
While doing that, uses an incorrect texinfo syntax that causes the
following warnings at compile time:
doc/openocd.texi:543: warning: @deffn should only appear at the beginning of a line
doc/openocd.texi:547: warning: @deffn should only appear at the beginning of a line
doc/openocd.texi:552: warning: @deffn should only appear at the beginning of a line
Move the documentation of xds110 commands in the proper chapter
and fix the texinfo syntax.
Change-Id: I0b3f0fe0c687f194bb02e2d81aca86fcd4fdd718 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com> Fixes: 87a4158acf56 ("drivers: xds110: Clean up command syntax and documentation")
Reviewed-on: http://openocd.zylin.com/5613 Tested-by: jenkins Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com> Reviewed-by: Edward Fewell <efewell@ti.com>
Antonio Borneo [Mon, 27 Apr 2020 21:19:14 +0000 (23:19 +0200)]
helper/ioutil: silence gcc-8 on strncpy
Starting from version 8, gcc issues a warning if strncpy could be
able to truncate a string (so without adding the zero-termination
char in destination) by copying exactly "size" char from a source
string not shorter than "size".
Such truncation from strncpy could actually be the desired code
behaviour, but the way to silent gcc only locally (without global
disabling with -Wno-stringop-truncation) through pragma has other
side effects on portability.
In current code, the source string is always "eth0", because has
been checked right above. So this is a false positive from gcc,
being always strlen("eth0") < 16, the sizeof(ifreq.ifr_name).
Silent gcc by decrementing the "size" and remove:
error: ‘strncpy’ specified bound 16 equals destination
size [-Werror=stringop-truncation]
This file is only optionally compiled together with ZY1000 driver
with --enable-zy1000 --enable-ioutil. This combination is not
checked in jenkins, so the error passed unnoticed.
Plus, the configure flags above are both deprecated!
Change-Id: I229e66227cfd3513139feeaffa47a6e1ec00767b Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/5631 Tested-by: jenkins Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Antonio Borneo [Fri, 1 May 2020 16:30:00 +0000 (18:30 +0200)]
helper/command: register all commands through register_commands()
The commands "ocd_find" and "capture" are registered directly
through the jim API, instead of the common way to describe them
in a struct command_registration that is then passed to the helper
register_commands().
This cause the two commands above to not have either "help" nor
"usage" string nor a properly identified "mode".
Since the following line registers the commands listed in struct
command_builtin_handlers, simply add the two commands above in the
same struct.
Change-Id: Id6ee11dac3b18364deeed65ee8e18ad80152750a Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/5644 Tested-by: jenkins Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Tim Newsome [Thu, 17 Oct 2019 22:12:55 +0000 (15:12 -0700)]
Add get_thread_reg() and set_reg() for hwthread
get_thread_reg() allows gdb to request the register value of a specific
"thread."
set_reg() allows register writes without getting a giant list of
registers first.
Signed-off-by: Tim Newsome <tim@sifive.com>
Change-Id: I87faa1c8793916b9ee476dd696f0695a07ca2b41
Reviewed-on: http://openocd.zylin.com/5324 Tested-by: jenkins
Tarek BOCHKATI [Mon, 9 Mar 2020 18:10:56 +0000 (19:10 +0100)]
flash/stm32l4x: always use stm32l4_get_flash_reg
this change is a preparation for STM32L5 support on top of L4 driver
STM32L5 flash is quite similar to L4 flash, mainly register names
and offsets and some bits are changed.
a table with register offset will be introduced, thus correct register
addresses will be obtained using this table and the driver internal
function 'stm32l4_get_flash_reg' will be responsible of this.
Change-Id: I74bf61a83fe53575623640af0328b3253ecc796f Signed-off-by: Tarek BOCHKATI <tarek.bouchkati@gmail.com>
Reviewed-on: http://openocd.zylin.com/5508 Tested-by: jenkins Reviewed-by: Andreas Bolsch <hyphen0break@gmail.com> Reviewed-by: Michael Jung <mijung@gmx.net> Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Tomas Vanek [Sun, 20 Oct 2019 08:08:07 +0000 (10:08 +0200)]
target/armv7m: cosmetic refactorization
Introduce a variable 'size' and reduce the number of dereferencing
*reg_list_size by using the variable.
Change-Id: I3bdf1485a4ed8e34435e8acb3efd0df8d802508c Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: http://openocd.zylin.com/5326 Tested-by: jenkins Reviewed-by: Christopher Head <chead@zaber.com> Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Antonio Borneo [Wed, 22 Apr 2020 20:39:07 +0000 (22:39 +0200)]
jtag: remove unused function adapter_driver_modules_load()
Commit c2cecc74b0ac ("Move JTAG interface list to new files.")
merged in mid 2009 introduces an unused and empty function that
the developer expects to use for loading adapter drivers from
shared libraries. This have never happened and the function is
still empty and unused.
Remove it.
Change-Id: I7c88dbf8a9747e96e5ca4e6e7038ac0f232604fd Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/5601 Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com> Tested-by: jenkins
Antonio Borneo [Mon, 20 Apr 2020 21:54:38 +0000 (23:54 +0200)]
server: set tcp port and bind address before init
All the servers open the listening socket during openocd "init";
it's not possible to change the tcp port or the bind address
after "init".
In current code, the call order during "init" permits to change
the port and bind address of tcl and telnet server if the related
command is issued after "init" in the same script or on openocd
command line. This is not guaranteed to work if the call order
during "init" changes, so it's incorrect.
Change the commands "bindto" and "*_port" to COMMAND_CONFIG.
Change-Id: Id88f225a67a872b4bcaf3b799466bddedc248015 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com> Reported-by: Christopher Head <chead@zaber.com> Fixes: https://sourceforge.net/p/openocd/tickets/264/
Reviewed-on: http://openocd.zylin.com/5595 Reviewed-by: Christopher Head <chead@zaber.com> Tested-by: jenkins
flash/nor/stm32l4x: cast wrpxxr_mask to uint16_to to print
Fix build error on Mac OS X Catalina (10.15.4) caused by formatting
stm32l4_info->wrpxxr_mask, which is uint32_t, as uint16_t in the debug
log message. Adding casting to uint16_t before substitution because only
lower 16 bits are significant for debug purposes.
Change-Id: Iddb87cd156dfc84ab1f91cd15a1ddee6b646d412 Signed-off-by: Ilya Kharin <akscram@gmail.com>
Reviewed-on: http://openocd.zylin.com/5590 Tested-by: jenkins Reviewed-by: Andreas Bolsch <hyphen0break@gmail.com> Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Aurélien Martin [Mon, 22 Jul 2019 21:13:29 +0000 (23:13 +0200)]
nrf5: Refresh the watchdog while flashing
If watchdog is enabled, there's no way we can disable it while the flashing firmware
is running. (Halt disables it, but software reset doesn't.) So let's have the flashing
firmware refresh the watchdog regularly, in case it has been enabled by previously
running software. Failure to do so could lead to a watchdog reset in the middle of
the chip bieng programmed.
Change-Id: I79d41593948aae0080480e891552e1c2ee3ccbd0 Signed-off-by: Aurélien Martin <martaurel@gmail.com>
Reviewed-on: http://openocd.zylin.com/5266 Tested-by: jenkins Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Florian Fainelli [Mon, 17 Jun 2019 22:46:11 +0000 (15:46 -0700)]
Remove BUILD_TARGET64
BUILD_TARGET64 creates a larger test matrix and mostly gates the
building of the aarch64/armv8 target, make that unconditional, which
would help fixing any issues with 64-bit address types anyway.
Rebased by Antonio Borneo after commit 1fbe8450a9dd ("mips: Add
MIPS64 support")
Liming Sun [Sat, 4 Apr 2020 02:51:13 +0000 (22:51 -0400)]
jtag/drivers/rshim: Disable the driver by default
This is a follow-up change of commit 6d6a69d5 to disable it by
default. The driver was introduced in 6d6a69d5 and enabled by
default in order to run the jenkins build.
Signed-off-by: Liming Sun <lsun@mellanox.com>
Change-Id: I5c5fc6711b971b65dd5846a6163025879044ec40
Reviewed-on: http://openocd.zylin.com/5563 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Liming Sun [Fri, 10 May 2019 15:02:31 +0000 (11:02 -0400)]
jtag/drivers: add debugging support for Mellanox BlueField SoC
This commits adds debugging support for the Mellanox BlueField
SoC via rshim, which is an interface accessible from external USB
or PCIe (for SmartNIC case) via the rshim driver. It implements
the arm dap interfaces based on the existing dapdirect framework.
Change-Id: I18eb1c54293ec2c581f853e0e55b3f96d7978b56 Signed-off-by: Liming Sun <lsun@mellanox.com>
Reviewed-on: http://openocd.zylin.com/5457 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Antonio Borneo [Wed, 10 Apr 2019 10:46:54 +0000 (12:46 +0200)]
tools/checkpatch.sh: remove flag --no-tree
Commit c5d89883165e02ea4f318e3cb0ba40d1fb6f04d1 ("checkpatch.pl:
check for openocd tree, not for kernel tree") has already fixed
the check for OpenOCD tree, thus we do not need to skip it in the
shell wrapper.
Remove flag --no-tree from the shell wrapper.
Change-Id: I8be497258624d89bde7742fee141a8f56bf9188e Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/5110 Tested-by: jenkins
Antonio Borneo [Fri, 14 Feb 2020 15:49:45 +0000 (16:49 +0100)]
tcl: stm32mp15x: add target and board config files
The stm32mp15x has one or two Cortex-A7 (depending on the P/N) and
one Cortex-M4.
The second core is automatically detected by the target script.
In "engineering boot" all the cores are accessible.
In "production boot" the Cortex-M4 is kept in reset state after
power-on or NRST.
The board DK2 includes a ST-Link/V2, but only SWD is connected.
scan-build-9:
Description: Potential leak of memory pointed to by 'cmd'
File: /home/vanekt/openocd/scanbuild9/../src/jtag/drivers/ulink.c
Line: 1075
Description: Potential leak of memory pointed to by 'cmd'
File: /home/vanekt/openocd/scanbuild9/../src/jtag/drivers/ulink.c
Line: 1275
ulink_append_xxx_cmd() functions allocate memory for cmd
and then call ulink_allocate_payload(), which allocates cmd->payload_out
or cmd->payload_in.
ulink_append_queue() checks the size of queue and if the new payload
does not fit, calls ulink_execute_queued_commands() and then
ulink_post_process_queue(). If any of these two fails, an error is
returned, allocated cmd struct leaks and the queue is left in an undefined
state.
Change ulink_append_queue() flow to proceed to appending cmd to the queue
even in the case of fail in previous ulink_execute_queued_commands()
or ulink_post_process_queue(). In case of fail then clear the queue
including the last appended cmd.
scan-build-9:
Description: Access to field 'payload_in' results in a dereference
of a null pointer (loaded from field 'queue_start')
File: src/jtag/drivers/ulink.c
Line: 2216
Set input/output_signals conditionally if ulink_append_get_signals_cmd()
and ulink_execute_queued_commands() returns no error.
Do not fail driver initialisation as the signals are only printed.
Change-Id: I6c842f0e9a604712abf7444a2fa95ba5810de1ff Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: http://openocd.zylin.com/5520 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com> Reviewed-by: Oleksij Rempel <linux@rempel-privat.de>
core.c:446:
The left operand of '>>' is a garbage value
There are many places where an error code returned from nand->controller
operations are ignored. To keep the change minimal, the error checks are
added only to reading of extended nand info as it was suspected
to be the cause of the warning.
Addition of the error checks did not fix the warning.
scan-build-9 report was inspected and IMHO the warning is bogus:
the term (nand->device->erase_size == 0) cannot give false at line 395
and then evaluate true at line 462. Fixed by zeroing id_buff.
Change-Id: I97ed7ce0fdf1aa23d746d5fb898bacd050e20ae8 Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: http://openocd.zylin.com/5518 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com> Reviewed-by: Oleksij Rempel <linux@rempel-privat.de>
Evgeniy Didin [Tue, 17 Mar 2020 11:06:24 +0000 (14:06 +0300)]
target/arc: remove saving context during reset
In arc_poll() function we handle the cases, when jtag indicates, that
processor is halted, but target->state is not TARGET_HALTED.
In case, when processor was halted and target->state was TARGET_RUNNING,
we should save context. At the same time if target->state was TARGET_RESET
we do not need to save context.
Changes: 16.04:
Fix - Move setting target->state = TARGET_HALT after
"target->state == TARGET_RUNNIG" check, otherwise
this check makes no sense
While on it add a comment explaining the relation of uxTopUsedPriority
and configMAX_PRIORITIES, introduce config_max_priorities and
change types to unsigned.
Change-Id: I4371c8882470d13ee7360ef21b132c56ecb95af8 Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: http://openocd.zylin.com/5577 Tested-by: jenkins
projects / fw / openocd / log