Antonio Borneo [Thu, 7 Feb 2019 08:56:10 +0000 (09:56 +0100)]
command: fix the mode for command "jtag configure"
Before commit 877cec20dca6e78f9f029f0f173879cda101a6c2 ("command:
check command mode for native jim commands") all the jim commands
were erroneously treated as they had mode COMMAND_ANY.
The commit above fixes the code in order to check the mode and
permit running the commands only if the mode is respected.
Those jim commands that have incorrect mode were not detected nor
fixes because the wrong mode was masked by the missing mode check.
After the commit above, the wrong mode triggers error in several
existing configuration scripts.
A complete list of commands that now does not run anymore as
CONFIG_ANY is reported in ticket 225, but most of them have the
mode set correctly.
At least two instances of command "jtag configure" have the wrong
mode.
Fix the mode to CONFIG_ANY for command "jtag configure" in files
src/jtag/aice/aice_transport.c and src/jtag/tcl.c
Change-Id: I3f96c5fd24d7d463712cbaf1295284fe0dc56b23
Ticket: https://sourceforge.net/p/openocd/tickets/225/ Reported-by: Bill Paul <wpaul@users.sourceforge.net> Fixes: 877cec20dca6 ("command: check command mode for native jim commands") Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4886 Reviewed-by: Tomas Vanek <vanekt@fbl.cz> Tested-by: jenkins
Kevin Vermilion [Wed, 30 Jan 2019 19:01:48 +0000 (11:01 -0800)]
at91samd: Add flash programming support for SAMC2?N* parts
Added id, name, flash size and RAM size for following parts to samc20_parts[]:
SAMC20N18A
SAMC20N17A
And the following to samc21_parts[]:
SAMC21N18A
SAMC21N17A
Change-Id: Ie8cf1c531a60bfaed6e814d436d232afb89dae3f Signed-off-by: Kevin Vermilion <kevin.vermilion@gmail.com>
Reviewed-on: http://openocd.zylin.com/4880 Tested-by: jenkins Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
target start_algorithm: Don't copy the IN mem_params fix uninitialised value.
Fix the write only out params TODO on armv7m.c
Fix conditional move depends on uninitialised value.
It was detected while programming a LPC8Nxx with a FTDI adapter.
valgrind --leak-check=full --show-leak-kinds=all --track-origins=yes
[...]
==8696== Conditional jump or move depends on uninitialised value(s)
==8696== at 0x16E4D3: buf_set_u32 (binarybuffer.h:52)
==8696== by 0x16E4D3: ftdi_swd_queue_cmd (ftdi.c:1206)
==8696== by 0x18D76D: swd_queue_ap_write (adi_v5_swd.c:271)
==8696== by 0x18E33B: dap_queue_ap_write (arm_adi_v5.h:382)
==8696== by 0x18E33B: mem_ap_write (arm_adi_v5.c:420)
==8696== by 0x197CD9: target_write_buffer_default (target.c:2176)
==8696== by 0x2464B3: armv7m_start_algorithm (armv7m.c:383)
==8696== by 0x246AEB: armv7m_run_algorithm (armv7m.c:330)
==8696== by 0x19D846: target_run_algorithm (target.c:814)
==8696== by 0x1DF3A6: lpc2000_iap_call.isra.3 (lpc2000.c:818)
==8696== by 0x1E0CF6: lpc2000_erase (lpc2000.c:992)
==8696== by 0x185BDF: flash_driver_erase (core.c:44)
==8696== by 0x18650D: flash_iterate_address_range_inner (core.c:541)
==8696== by 0x18650D: flash_iterate_address_range (core.c:567)
==8696== by 0x18732F: flash_erase_address_range (core.c:584)
==8696== by 0x18732F: flash_write_unlock (core.c:928)
==8696== Uninitialised value was created by a heap allocation
==8696== at 0x4C2BBAF: malloc (vg_replace_malloc.c:299)
==8696== by 0x220EF9: init_mem_param (algorithm.c:30)
==8696== by 0x1DF247: lpc2000_iap_call.isra.3 (lpc2000.c:777)
==8696== by 0x1E0CF6: lpc2000_erase (lpc2000.c:992)
==8696== by 0x185BDF: flash_driver_erase (core.c:44)
==8696== by 0x18650D: flash_iterate_address_range_inner (core.c:541)
==8696== by 0x18650D: flash_iterate_address_range (core.c:567)
==8696== by 0x18732F: flash_erase_address_range (core.c:584)
==8696== by 0x18732F: flash_write_unlock (core.c:928)
==8696== by 0x18ACDF: handle_flash_write_image_command (tcl.c:457)
==8696== by 0x1B7D99: run_command (command.c:623)
==8696== by 0x1B7D99: script_command_run (command.c:208)
==8696== by 0x1B7FD9: command_unknown (command.c:1033)
==8696== by 0x2E2D37: JimInvokeCommand (jim.c:10364)
==8696== by 0x2E3865: Jim_EvalObj (jim.c:10814)
==8696==
==8696== Conditional jump or move depends on uninitialised value(s)
==8696== at 0x16E506: buf_set_u32 (binarybuffer.h:52)
==8696== by 0x16E506: ftdi_swd_queue_cmd (ftdi.c:1207)
==8696== by 0x18D76D: swd_queue_ap_write (adi_v5_swd.c:271)
==8696== by 0x18E33B: dap_queue_ap_write (arm_adi_v5.h:382)
==8696== by 0x18E33B: mem_ap_write (arm_adi_v5.c:420)
==8696== by 0x197CD9: target_write_buffer_default (target.c:2176)
==8696== by 0x2464B3: armv7m_start_algorithm (armv7m.c:383)
==8696== by 0x246AEB: armv7m_run_algorithm (armv7m.c:330)
==8696== by 0x19D846: target_run_algorithm (target.c:814)
==8696== by 0x1DF3A6: lpc2000_iap_call.isra.3 (lpc2000.c:818)
==8696== by 0x1E0CF6: lpc2000_erase (lpc2000.c:992)
==8696== by 0x185BDF: flash_driver_erase (core.c:44)
==8696== by 0x18650D: flash_iterate_address_range_inner (core.c:541)
==8696== by 0x18650D: flash_iterate_address_range (core.c:567)
==8696== by 0x18732F: flash_erase_address_range (core.c:584)
==8696== by 0x18732F: flash_write_unlock (core.c:928)
==8696== Uninitialised value was created by a heap allocation
==8696== at 0x4C2BBAF: malloc (vg_replace_malloc.c:299)
==8696== by 0x220EF9: init_mem_param (algorithm.c:30)
==8696== by 0x1DF247: lpc2000_iap_call.isra.3 (lpc2000.c:777)
==8696== by 0x1E0CF6: lpc2000_erase (lpc2000.c:992)
==8696== by 0x185BDF: flash_driver_erase (core.c:44)
==8696== by 0x18650D: flash_iterate_address_range_inner (core.c:541)
==8696== by 0x18650D: flash_iterate_address_range (core.c:567)
==8696== by 0x18732F: flash_erase_address_range (core.c:584)
==8696== by 0x18732F: flash_write_unlock (core.c:928)
==8696== by 0x18ACDF: handle_flash_write_image_command (tcl.c:457)
==8696== by 0x1B7D99: run_command (command.c:623)
==8696== by 0x1B7D99: script_command_run (command.c:208)
==8696== by 0x1B7FD9: command_unknown (command.c:1033)
==8696== by 0x2E2D37: JimInvokeCommand (jim.c:10364)
==8696== by 0x2E3865: Jim_EvalObj (jim.c:10814)
Change-Id: I50f9a8c4516b686cf62ac3c76f47c53465e949da Signed-off-by: Jean-Christian de Rivaz <jcamdr70@gmail.com>
Reviewed-on: http://openocd.zylin.com/4811 Tested-by: jenkins Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Jan Vojtech [Tue, 4 Sep 2018 09:50:09 +0000 (11:50 +0200)]
flash/nor/stm32f1x: Ability to change user option bytes.
Adds ability to change the user data in STM32F1x/STM32F3x MCU's option byte.
Since OpenOCD prints the content of user data in option byte registers, it
is seems logical to also provide a way how to change this data.
Change-Id: Ie6cb756b4f11b5c6dabd34bc89434a358eb758ff Signed-off-by: Jan Vojtech <honza.vojtech@gmail.com> Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: http://openocd.zylin.com/4663 Tested-by: jenkins Reviewed-by: Tarek BOCHKATI <tarek.bouchkati@gmail.com>
Tomas Vanek [Tue, 18 Dec 2018 16:17:33 +0000 (17:17 +0100)]
target algo: do not write reg_param if direction is PARAM_IN
Without this change xxx_start_algorithm() writes all register
parameters no matter of their direction. It usually results
in writing of uninitialized reg_params[].value - possibly
reported by valgrind.
While on it fix the wrong parameter direction in
kinetis_disable_wdog_algo(). This bug did not have any
impact because of unconditional write of reg_params.
Change-Id: Ia9c6a7b37f77d5eb6e5f5463012dddd50471742b Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: http://openocd.zylin.com/4813 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com> Reviewed-by: Matthias Welwarsky <matthias@welwarsky.de>
Oleksij Rempel [Wed, 27 Jun 2018 12:54:21 +0000 (14:54 +0200)]
jtag: drivers: provide initial support for usb path filtering
With this patch drivers will be able to use usb path filtering.
The path format is identical to the format provided by linux kernel:
bus-port.port....
With this format it should be easier just to copy and paste
path found in dmesg.
Tomas Vanek [Fri, 27 Oct 2017 19:43:25 +0000 (21:43 +0200)]
flash/nor: flash driver and cfg for SAM E54, E53, E51 and D51
The new Microchip (former Atmel) series powered by Cortex-M4 looks
very similar to older M0+ powered SAM D2x at the first sight.
Unfortunately the new series differs a lot in important details.
NVMCTRL has different register addresses, moved important bits
and even changed binary command set. An universal driver for all SAM D/E
would be very complicated. That's why a new driver was derived.
Tested on Microchip SAM E54 Xplained Pro kit (board cfg included).
Adjusted for the restructured dap support.
Checked by valgrind and clang static analyzer.
Change-Id: I26c67047a552076f4b207b9b89285a53d69b4ca4 Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: http://openocd.zylin.com/4272 Tested-by: jenkins Reviewed-by: Andres Vahter <andres.vahter@gmail.com>
Tomas Vanek [Sat, 19 Jan 2019 13:58:49 +0000 (14:58 +0100)]
flash/nor/stm32f1x: fix minor error messages
Change-Id: I1e9e62979c4629c8ba1d5ae89ca7392259969eb6 Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: http://openocd.zylin.com/4858 Tested-by: jenkins Reviewed-by: Christopher Head <chead@zaber.com>
Tomas Vanek [Sat, 19 Jan 2019 09:26:38 +0000 (10:26 +0100)]
cortex_m: fix stepping on FPB rev 1
Stepping in the maskisr auto mode sets breakpoint to step over interrupt
service tasks. If the device has FPB rev 1, setting hard breakpoint
is impossible on address over 0x1fffffff.
Use soft type breakpoint for adresses over 0x1fffffff if FPB is rev 1.
This may eventually fail if the code memory is not writeable, but there
is nothing to do in such case.
Change-Id: Ibdeeb506903a35d550b64f82c24c37a668de62b3 Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: http://openocd.zylin.com/4857 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
samr34/r35 combine SAML21 and SX1276 (lora transceiver). This one was found on
xplaned pro evaluation kit. Ids for other r34/r35 chips are apparently not yet
documented.
Antonio Borneo [Wed, 29 Aug 2018 09:57:42 +0000 (11:57 +0200)]
arm_adi_v5: rewrite dap_to_jtag and dap_to_swd
The functions dap_to_jtag() and dap_to_swd() have been introduced by 3ef9beb52cd0 ("ADIv5 DAP ops switching to JTAG or SWD modes") in
arm_adi_v5.c by using the JTAG queue only.
Later, in 6f8b8593d63b ("ADIv5 transport support moves to separate
files") the functions has been moved in adi_v5_swd.c and adi_v5_jtag.c
but keeping the dependency from JTAG queue.
The functions does not work if the current transport is not JTAG.
Move back the functions in arm_adi_v5.c, replace the input parameter
"target" with "dap", use the transport to detect if the JTAG queue is
present, in case of SWD transport use the proper method, for other
transports report error.
Reuse the ADI v5 sequences already present in jtag/swd.h.
Also, OpenOCD does not support switching to another transport after
the initial selection, so do not change DAP's ops vector.
Antonio Borneo [Sat, 19 Jan 2019 23:52:07 +0000 (00:52 +0100)]
command: initialize the command mode for every command
All the commands in OpenOCD have been inspected and have the
command mode initialize, apart for two of them.
This is not critical, because the uninitialized value (0) is
equivalent to the enum COMMAND_EXEC, that is also the correct
value for the two mentioned commands.
To keep the code consistent, initialize the command mode to
COMMAND_EXEC.
Change-Id: Iaf043364cbd1005418d787ed045a3ec653612382 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4861 Tested-by: jenkins Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Peter Lawrence [Thu, 17 Jan 2019 00:00:34 +0000 (18:00 -0600)]
tcl: Support for Analog Devices ADSP-SC58x / ADSP-SC584-EZBRD
The original script was broken by changes to the Cortex-A code. The
recent introduction of the mem_ap target provided a new mechanism to
allow the script to be fixed. This also adds an example board script
for the ADSP-SC584-EZBRD.
Change-Id: I36bc1ac6b6c036539f4175f1e65223ba10a35355 Signed-off-by: Peter Lawrence <majbthrd@gmail.com>
Reviewed-on: http://openocd.zylin.com/4855 Tested-by: jenkins Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Correct ZynqMP configuration to be appropriately named
The xilinx_ultrascale.cfg target is actually the configuration for a
ZynqMP, which is a combination of an UltraScale+ FPGA core and a quad
core A53. Update the filename/comments to reflect this, and include
the tap IDs for all known FPGA cores for this part.
Antonio Borneo [Tue, 15 Jan 2019 14:51:13 +0000 (15:51 +0100)]
arm_opcode: fix encoding of ARMv5 breakpoint instruction
The encoding of BKPT instruction is 0xE12###7#, where the four '#'
characters should be replaced by the 16 bits immediate value.
The macro uses an incorrect shift amount, thus the immediate value
is not properly coded and bits 20~23 of the opcode could get
corrupted.
Antonio Borneo [Wed, 29 Aug 2018 08:55:48 +0000 (10:55 +0200)]
arm_adi_v5: fix and update sequences to spec IHI 0031E
Fix the SWD line reset sequence accordingly to Arm specification IHI
0031E that requires at least 2 idle clocks after the 50 clocks with
SWDIO high.
Fix the value of the activation code in the (currently unused)
sequence dormant-to-SWD.
Make each sequence's length multiple of 8, so it is compatible with
adapters that have such limitation (e.g. buspirate) and try to split
and comment each part of the sequence (when possible keep each part
byte aligned, inspired from commit 3ef9beb52cd0). This slightly
increases the sequence length but does not impact run-time
performance because these are rarely used sequences.
Add the missing sequence dormant-to-JTAG and JTAG-to-dormant, not
used yet.
On devices that implements the dormant state, IHI 0031E deprecates
the direct switching between SWD and JTAG, and recommends using a
transition through dormant. This is not implemented.
Tarek BOCHKATI [Fri, 21 Dec 2018 18:34:58 +0000 (19:34 +0100)]
cortex_m: fix bug in poll() machine state (external resume awareness)
This patch covers the fact that cortex_m could be resumed externally by
Cross Trigger Interface or by direct write to DHSCR ...
To reproduce:
- halt the target
- then run the core through DHCSR (mww 0xe000edf0 0xa05f0001)
=> this resumes the core, but target state in OpenOCD remains HALTED.
Tarek BOCHKATI [Thu, 3 Jan 2019 17:16:45 +0000 (18:16 +0100)]
cortex_a_poll: minor code factorization to enhance readability
cortex_a_debug_entry and update_halt_gdb are called in two consecutive
conditions which are complementary, so externalizing the common code
makes the conditions' body lighter
With the removal of LOG_DEBUG(" ") since it does not look too informative
Antonio Borneo [Tue, 8 Jan 2019 08:54:30 +0000 (09:54 +0100)]
helper/command: check for malloc failure in __command_name
If malloc fails in __command_name, the following strcpy will
segfault, thus preventing __command_name to return.
The actual calls to command_name() implement the correct check
for the NULL pointer, but propagate error -ENOMEM, that is not
an error value coherent within OpenOCD. Plus, in one case it
overwrites an already detected error.
Check the pointer returned by malloc and, in case of failure,
issue an error message and return the NULL pointer.
Let the caller of command_name() to keep the already detected
error or to return ERROR_FAIL in case of end of memory.
Change-Id: I151a24569409777dd5bc09a3daf5dba2b8e2829b Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4838 Tested-by: jenkins Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Antonio Borneo [Fri, 4 Jan 2019 17:08:20 +0000 (18:08 +0100)]
drivers/sysfsgpio: fix usage messages
The notation "(tck tms tdi tdo)* " is incorrect, because it means the
quadruple of gpio can be repeated on the command-line.
The correct syntax of the command requires instead to provide either
all the four gpio numbers (in order to set the values) or to pass an
empty command-line (to dump the values previously set).
Change the .usage field to "[tck tms tdi tdo]".
Change similarly the corresponding .usage field for SWD command.
Add the .usage field for the commands that individually set each gpio.
Change-Id: If5b3c618097b71dfe7fcf988fb3c1499ae03a6d5 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4833 Tested-by: jenkins Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
stlink_usb: Submit multiple USB URBs at once to improve performance
Commands to stlink devices are typically comprised of multiple
transactions with each transaction completing before moving to the next.
This change allows for multiple USB transactions to be issued at once
followed by a check that all transactions completed successfully. This
improves performance on some machines where there is a large turn-around
time between USB transfers such as is seen on some virtual machines.
This change is only supported when compiled with libusb1 as libusb1
supports and asynchronous interface.
Multi-transaction queueing introduced in this change paves the way for
improving speed of other transactions in the future such as memory and
register reads where multiple USB transactions in succession are
required to complete a command. Multiple USB transactions can be
submitted at once using jtag_libusb_bulk_transfer_n function.
Change-Id: I924e049217a789ef445b14e00aa1983576970fbf Signed-off-by: Austin Phillips <austin_phillips@hotmail.com>
Reviewed-on: http://openocd.zylin.com/4484 Tested-by: jenkins Reviewed-by: Andreas Bolsch <hyphen0break@gmail.com> Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk> Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Antonio Borneo [Thu, 4 Oct 2018 09:39:50 +0000 (11:39 +0200)]
libusb: add debug message on adapter not found due to wrong serial
When few adapters of the same type are in use, the serial string is
the way to select the right one.
Currently a serial string that does not match any of the connected
adapters will just fail the open, without specific information to
track the issue.
Add a specific message to highlight that the open failure is caused
by a serial mismatch.
Change-Id: I5cb77f1045cc746e532d395b2e5ced40a23ab638 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4701 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Antonio Borneo [Thu, 4 Oct 2018 15:54:19 +0000 (17:54 +0200)]
libusb0: return oocd error values
Commit d1b74376336814266236054f925a9964b87dd8a5 fixes libusb1 to
return OpenOCD error values instead of negative errors in Linux
kernel's style.
The same fix should be applied to libusb0 too.
Fix return value of libusb0 to uniform it to OpenOCD style.
Antonio Borneo [Thu, 1 Nov 2018 14:08:21 +0000 (15:08 +0100)]
target/aarch64: add support for multi-architecture gdb
GDB can be built for multi-architecture through the command
./configure --enable-targets=all && make
Such multi-architecture GDB requires the target's architecture to
be selected either manually by the user through the GDB command
"set architecture" or automatically by the target description sent
by the remote target (i.e. OpenOCD).
Commit e65acd889c61a424c7bd72fdee5d6a3aee1d8504 ("gdb_server: add
support for architecture element") already provides the required
infrastructure to support multi-architecture gdb.
aarch64-linux-gnu-gdb 8.2 uses "aarch64" as default architecture,
but also supports the value "aarch64:ilp32" and all the values
supported by arm-none-eabi-gdb.
These values can be displayed on arm gdb prompt by typing
"set architecture " followed by a TAB for autocompletion.
Set the gdb architecture value for aarch64 target to "aarch64".
Change-Id: I63e9769f47d8e73f048eb84fa73e082dd1c8e52c Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4755 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Antonio Borneo [Thu, 1 Nov 2018 13:50:27 +0000 (14:50 +0100)]
target/arm: add support for multi-architecture gdb
GDB can be built for multi-architecture through the command
./configure --enable-targets=all && make
Such multi-architecture GDB requires the target's architecture to
be selected either manually by the user through the GDB command
"set architecture" or automatically by the target description sent
by the remote target (i.e. OpenOCD).
Commit e65acd889c61a424c7bd72fdee5d6a3aee1d8504 ("gdb_server: add
support for architecture element") already provides the required
infrastructure to support multi-architecture gdb.
arm-none-eabi-gdb 8.2 uses "arm" as default architecture, but also
supports the following values: "arm_any", "armv2", "armv2a",
"armv3", "armv3m", "armv4", "armv4t", "armv5", "armv5t", "armv5te",
"armv5tej", "armv6", "armv6k", "armv6kz", "armv6-m", "armv6s-m",
"armv6t2", "armv7", "armv7e-m", "armv8-a", "armv8-m.base",
"armv8-m.main", "armv8-r", "ep9312", "iwmmxt", "iwmmxt2", "xscale".
These values can be displayed on arm gdb prompt by typing
"set architecture " followed by a TAB for autocompletion.
Set the gdb architecture value for all arm targets to "arm".
Change-Id: I176cb89878606e1febd546ce26543b3e7849500a Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4754 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Antonio Borneo [Thu, 1 Nov 2018 11:07:03 +0000 (12:07 +0100)]
target/stm8: add support for multi-architecture gdb
GDB can be built for multi-architecture through the command
./configure --enable-targets=all && make
Such multi-architecture GDB requires the target's architecture to
be selected either manually by the user through the GDB command
"set architecture" or automatically by the target description sent
by the remote target (i.e. OpenOCD).
Commit e65acd889c61a424c7bd72fdee5d6a3aee1d8504 ("gdb_server: add
support for architecture element") already provides the required
infrastructure to support multi-architecture gdb.
The gdb patches for stm8 are still not merged in the official
repository and are temporarily hosted in
https://stm8-binutils-gdb.sourceforge.io/
The latest patch set
stm8-binutils-gdb-sources-2018-03-04.tar.gz
define only one possible value ("stm8") for this architecture; it
can be displayed typing "set architecture " followed by a TAB for
autocompletion in gdb for stm8.
Set the gdb architecture value for stm8 to "stm8".
Change-Id: I643ceba662de46cecf061d1dc672b9178a077f1b Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4753 Tested-by: jenkins Reviewed-by: Ake Rehnman <ake.rehnman@gmail.com> Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Cody P Schafer [Mon, 27 Aug 2018 14:50:46 +0000 (10:50 -0400)]
armv7m: always set xPSR.T=1 when starting an algorithm
xPSR.T sets the processor to Thumb mode when set to 1. ARMv7-M only
supports execution of Thumb instructions, so it must always be set to 1.
If xPSR.T is set to 0 on armv7m, a usage fault is generated when a
instruction execution is attempted.
On armv7m, issuing a reset causes the vector table to be examined. PC
and xPSR.T are loaded from the vector table at byte offset 4. xPSR.T is
taken from the least significant bit this value, PC from the remaining
bits. This occurs even with `reset halt`, as the reset itself causes
this load to occur without the execution of any instructions.
As a result of this, following a reset with a "bad" value programmed in
the vector table, openocd would be unable to run algorithms on the
target, as running them would immediately result in a usage fault due to
xPSR.T being unset (0).
Allow algorithms to run regardless of the content of the vector table by
explicitly setting xPSR so that xPSR.T=1 prior to executing an
algorithm. One can think of this as openocd more closely emulating a
reset or branch instruction in executing it's algorithms.
Ticket: https://sourceforge.net/p/openocd/tickets/203/ Signed-off-by: Cody P Schafer <openocd@codyps.com>
Change-Id: I4dc3427ab195d06c3fd780ea768027fefccc4c28
Reviewed-on: http://openocd.zylin.com/4658 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Edward Fewell [Wed, 5 Dec 2018 23:54:42 +0000 (17:54 -0600)]
icepick.cfg: add cancel reset bit to TAP register writes
The Agama family of devices (CC26x2/CC13x2) required an
additional bit to be set when adding the core's TAP into
the scan chain. The cancel reset bit 0x10000 tells the
ICEPick to take the bus out of reset so that the other
bits will take effect. This bit is a NOP on other devices
and ICEPicks, so the change shouldn't adversely affect
other devices.
Change-Id: I9245eef0936ea7eea28ae84ab5e8ce05fa63af40 Signed-off-by: Edward Fewell <efewell@ti.com>
Reviewed-on: http://openocd.zylin.com/4789 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Edward Fewell [Thu, 6 Dec 2018 20:47:29 +0000 (14:47 -0600)]
drivers: xds110: Add support for XDS110 stand-alone probe
The XDS110 stand-alone version has the ability to supply
voltage to the target board via it's AUX FUNCTIONS port.
Added command to enable setting the voltage on the XDS110
stand-alone.
Change-Id: I2f21c4a3d15ed99e649f3a83973c5e724c4bfeb6 Signed-off-by: Edward Fewell <efewell@ti.com>
Reviewed-on: http://openocd.zylin.com/4793 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Jerome Forissier [Fri, 14 Dec 2018 15:31:17 +0000 (16:31 +0100)]
HACKING: add note about refs/for/master
The fact that one needs to always push contributions to a single
remote reference (refs/for/master) might seem odd to people unfamiliar
with Gerrit. GitHub, for instance, hosts personal repositories where
developers typically create topic branches for each contribution and
use a proprietary mecanism to request a review (the "pull request").
More generally, one normally does not expect to be able to push
non-fast-forwarding stuff to a remote branch.
This commit adds a clarifying note to the patch guidelines.
Change-Id: Ia750b815b82b18e92b6109c07f451000dcbecf9b Signed-off-by: Jerome Forissier <jerome.forissier@linaro.org>
Reviewed-on: http://openocd.zylin.com/4806 Tested-by: jenkins Reviewed-by: Christopher Head <chead@zaber.com> Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Rod Boyce [Sun, 16 Dec 2018 17:41:39 +0000 (17:41 +0000)]
NOR: lpc2000 Add support for LPC84x devices
These devices differ from LPC8xx devices in that they have a different
IAP entry point, but everything else is the same. Using Tcl to pass
different IAP entry point.
no new Clang analyser warnings and no new build sanitizers issues.
Change-Id: I2d654dd250f416e74262c0228cad8713a283402f Signed-off-by: Rod Boyce <developer@teamboyce.co.uk>
Reviewed-on: http://openocd.zylin.com/4684 Reviewed-by: Jean-Christian de Rivaz <jcamdr70@gmail.com> Tested-by: jenkins Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Oleksij Rempel [Wed, 26 Dec 2018 11:47:11 +0000 (12:47 +0100)]
flash: nor: ath79: remove base calculation
Currently it is impossible to flash ELF with correct offsets. The reason
is a bogus offset calculation extracted from base.
Since any other spi drivers do not care about base, do the same for
ath79 as well.
Tomas Vanek [Fri, 7 Dec 2018 17:45:06 +0000 (18:45 +0100)]
target: allow moving and resizing working area on running target
target configure -work-area-xxx calls target_free_all_working_areas()
and sets the desired new parameter. Without this change the working area
does not get reallocated if it has been allocated before.
target_free_all_working_areas() results in work area containing one block
marked as free.
Completely free working area in target_free_all_working_areas()
Change-Id: I79c681082f32f2a96a2b40eb3b8751e427549693 Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: http://openocd.zylin.com/4797 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Liming Sun [Mon, 5 Nov 2018 15:26:05 +0000 (10:26 -0500)]
target: armv8: Add TARGET_HALTED check for gdb connect
This commit adds TARGET_HALTED check in armv8_get_core_reg32()
and armv8_set_core_reg32() to void a crash issue when gdb connects
but fails to halt the ARM core. Similar logic can be found in
armv8_get_core_reg() and armv8_set_core_reg().
Below is the call stack information of this case when gdb connects.
(gdb) bt
regnum=regnum@entry=0, dpm=0x990110) at src/target/armv8_dpm.c:657
r=0x9c7240, regnum=0, mode=<optimized out>) at src/target/armv8_dpm.c:974
at src/target/armv8.c:1487
packet=0x8ec8e0 <gdb_packet_buffer.9962> "g",
packet_size=<optimized out>, connection=<optimized out>)
at src/server/gdb_server.c:1200
at src/server/gdb_server.c:3180
command_context=command_context@entry=0x935010)
at src/server/server.c:566
...
Antonio Borneo [Mon, 3 Dec 2018 17:50:43 +0000 (18:50 +0100)]
target/cortex_a: enable DSCR_HALT_DBG_MODE during examine
Arm architecture reference manual DDI0406C reports at page 2024 in
table C3-1 the processor behaviour on debug events depending on
the debug-mode (none, monitor or halt), mode selected through the
bits MDBGen and HDBGen in DSCR register.
The halt request is served independently from the debug-mode. Thus
it's useless to enable the halt debug-mode in cortex_a_halt() by
setting the bit HDBGen (macro DSCR_HALT_DBG_MODE).
On the other side, halting for a breakpoint, a watchpoint or a
vector catch requires being in halt debug-mode.
Today HDBGen is set only in cortex_a_halt(), so we are forced to
halt the core at least once before it can be halted for hitting a
breakpoint/watchpoint/vector-catch. This is annoying since there
is no need to halt the target to set a HW breakpoint.
Move in cortex_a_init_debug_access() the selection of the halt
debug-mode, so the mode is set during examine.
To prevent a misconfigured hardware breakpoint/watchpoint/vector
catch to halt the target when OpenOCD has already quit, return to
debug-mode none at OpenOCD exit.
YanLin Zhu [Fri, 14 Dec 2018 05:17:15 +0000 (13:17 +0800)]
adi_v5_jtag: fix build break when open DEBUG_WAIT macro
DEBUG_WAIT is useful to debug adi_jtag issue,
and the WCR register is replaced by DLCR for DP registers update
in commit 150b7d26f213398d717bf46744811b48834a3744.
Change-Id: I3faa9ea8a6adacd3d5275e40382801da731db32f Signed-off-by: YanLin Zhu <zhuyanlin@pinecone.net>
Reviewed-on: http://openocd.zylin.com/4804 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com> Reviewed-by: Christopher Head <chead@zaber.com>
Antonio Borneo [Mon, 22 Oct 2018 07:06:50 +0000 (09:06 +0200)]
helper/startup.tcl: fix execution stack frame of wrapped commands
The OpenOCD commands that have been wrapped with 'ocd_bouncer' are
executed within two levels of nested proc's:
# see register_command_handler() in src/helper/command.c
proc my_command {args} {eval ocd_bouncer my_command $args}
# see ocd_bouncer in src/helper/startup.tcl
proc ocd_bouncer {name args} {
... [eval ocd_my_command $args] ...
}
This causes the stack frame of 'ocd_my_command' to be the same one
of proc 'ocd_bouncer', thus two levels below the stack frame of the
caller of 'my_command'. This is an issue with commands that receive
a variable by name and have to resolve them to access the value.
E.g. the command
<target> mem2array arrayname bitwidth address count
is wrapped; it receives the name of the array but fails to resolve
it in the current stack frame. Instead, the commands
mem2array arrayname bitwidth address count
ocd_<target> mem2array arrayname bitwidth address count
are not wrapped and can directly access the array because they share
the same stack frame of the caller.
Same situation with the symmetric commands 'array2mem'.
How to test:
within a telnet connection, run the following set of commands,
eventually replacing the address 0x08000000 with a valid readable
address of your <target>,
unset -nocomplain v1 v2 v3
info vars v?
mem2array v1 32 0x08000000 1
<target> mem2array v2 32 0x08000000 1
ocd_<target> mem2array v3 32 0x08000000 1
info vars v?
and notice that only v1 and v3 are now allocated. The array v2 has
been allocated in the temporarily stack frame of proc ocd_bouncer,
together with its local variables, and then lost when proc ended.
Fixed by executing the wrapped commands with the command 'uplevel'
instead of 'eval'. The amount of levels to skip is checked to avoid
errors in the unusual case 'ocd_bouncer' is called directly without
the first level of wrapper.
Change-Id: Iff90fb8921faf9b5ab04f61062a530578cc20d78 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4731 Tested-by: jenkins Reviewed-by: Christopher Head <chead@zaber.com> Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Bohdan Tymkiv [Wed, 5 Dec 2018 09:57:26 +0000 (11:57 +0200)]
flash/nor/core: Fix chunk size calculation in default_flash_mem_blank_check
Slow version of blank check procedure reads target memory sector-by-sector
using 1 KB chunks. Due to bug in chunk size calculation algorithm the actual
size of the chunk is always 1 KB even if sector size is smaller.
This causes out-of-boundary read of the last sector.
Steps to reproduce:
1) Use target with small sectors (e.g. psoc6 with 512-byte sectors)
2) set WORKAREASIZE_CM0 0
3) flash erase_check 1
Running slow fallback erase check - add working memory
Info : SWD DPIDR 0x6ba02477
Error: Failed to read memory at 0x14008000
unknown error when checking erase state of flash bank #1 at 0x14000000
Bank is erased
Change-Id: I03d0d5fb3a1950ae6aac425f5e24c7fd94b38325 Signed-off-by: Bohdan Tymkiv <bhdt@cypress.com>
Reviewed-on: http://openocd.zylin.com/4785 Reviewed-by: Tomas Vanek <vanekt@fbl.cz> Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Steven Stallion [Thu, 25 Oct 2018 02:29:03 +0000 (21:29 -0500)]
esirisc: support eSi-Trace
This patch adds support for instruction tracing to eSi-RISC targets. The
command interface is borrowed heavily from ETM; eSi-Trace uses a less
sophisticated model for tracing, however the setup and usage is similar.
This patch also cleans up the command interfaces of the other esirisc
command groups and adds additional debugging information to log messages
when dealing with CSRs.
This patch "finalizes" support for 32-bit eSi-RISC targets.
Change-Id: Ia2a9de79a3c7c066240b5212721fb1b7584a9a45 Signed-off-by: Steven Stallion <stallion@squareup.com>
Reviewed-on: http://openocd.zylin.com/4780 Tested-by: jenkins Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Antonio Borneo [Fri, 13 Jul 2018 07:01:47 +0000 (09:01 +0200)]
armv7a_mmu: s/LOG_ERROR/LOG_WARNING/ on address translation failure
When GDB analyses the status of the target it try to guess the
current stack frame and issues few memory read.
E.g. on ARM targets GDB uses R11 value as a potential frame-pointer
and reads at the address pointed by R11.
The address of such memory read is not always valid and can trigger
an address translation failure.
Replace LOG_ERROR with LOG_WARNING in case the virtual address does
not have a hit in TTB; print the virtual address in the warning
message and discriminate the two identical messages with [1]/[2].
Change-Id: I288b8cd26bec2543c4f1c16b7c06dc47d5d843d1 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4602 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
bob [Fri, 23 Nov 2018 13:40:39 +0000 (00:40 +1100)]
flash/nor: Add support for sector erase in stm32l4x.c for L4+ family
Updates support for L4+ device id: 0x470 added by #4310
Extends #4641 to account for L4+ use of multiple DBANK option bits
Enables L4+ 1M and 2M devices to be programmed using sector erase
Change-Id: I42bb379d7d97986f4506423e3da503d07c787c6b Signed-off-by: bob <rea952@gmail.com>
Reviewed-on: http://openocd.zylin.com/4777 Tested-by: jenkins Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Edward Fewell [Wed, 5 Dec 2018 22:05:00 +0000 (16:05 -0600)]
flash/nor: update CC26xx/CC13xx support
Added fixes found in additional code reviews.
Remove inappropriate use of bank_number field and updated
documentation to reflect the change.
Restored functionality to cc2538.cfg file because previous
change removed the cc26xx.cfg file because the flash support
changes made it obsolete. Rolled the previous cc26xx.cfg
file into cc2538.cfg and updated it to work with other
recent changes. Tested using a SmartRF06 Evaluation
board with embedded XDS100v3 and external XDs110.
Change-Id: Ia19d00cf8055c5c0f1acc53aa23fd06a80fd2ebc Signed-off-by: Edward Fewell <efewell@ti.com>
Reviewed-on: http://openocd.zylin.com/4787 Tested-by: jenkins Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Antonio Borneo [Mon, 26 Nov 2018 15:49:05 +0000 (16:49 +0100)]
helper/log: remove dead code in log_forward()
Commit f1be0e6af2e204805a8bc2e8cadf828b9fa46c98 ("HELPER/LOG:
review unused symbols") removed a set of unused functions and,
as consequence, there is no code anymore that sets the variable
"log_forward_count". But, the commit above did not removed the
code (now dead) that depends on "log_forward_count" set.
Remove the code dependant on "log_forward_count" set and the
variable itself.
Change-Id: I6efe93d1dccbe13c409c5bc55ba47a2684c0e3ac Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4779 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Antonio Borneo [Mon, 24 Sep 2018 13:25:46 +0000 (15:25 +0200)]
stlink: add support for STLINK-V3
Extend the driver to include the minimal functionality to support
the HLA model.
Due to the small change in the name (ST-LINK/V2 => STLINK-V3), fix
the existing names in the comments in udev rules.
Change-Id: Ied33e38063a6da81d9bf249ed195444d7cdf4f03 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4717 Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk> Tested-by: Spencer Oliver <spen@spen-soft.co.uk>
Antonio Borneo [Mon, 1 Oct 2018 08:25:35 +0000 (10:25 +0200)]
stlink: fix printed version for new STM8 nucleo board
The ST-LINK/V2.1 embedded in the new nucleo boards for STM8 does
not follow the normal versioning rules, and puts mass-storage and
swim version in the field normally used respectively for jtag and
mass-storage version.
Apply the check suggested by STMicroelectronics to discriminate
the two cases and print the correct version.
Change-Id: I0dd1da11013be3f1e56084489e28cfba98bb07af Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4716 Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk> Tested-by: Spencer Oliver <spen@spen-soft.co.uk>
Antonio Borneo [Mon, 24 Sep 2018 13:26:40 +0000 (15:26 +0200)]
stlink: remove reset pulse when entering in JTAG
Until version J14 the behaviour of ST-Link was to send a reset
pulse to the target when the debug connection is started in JTAG
mode. No reset pulse is sent, instead, in SWD mode.
Version J15 introduces a new parameter to avoid the reset pulse in
JTAG mode, aligning the behaviour with SWD.
This reset from the ST-Link, if propagated to the target, prevents
attaching a running target.
Actually this reset pulse is very short (few microsecond) and can
be easily filtered out by an on-board capacitor, usually present
on the reset wire (mainly to filter the bounces of the reset
button). Moreover, most of the use cases for ST-Link are with SWD
(not with JTAG) and this has probably further masked this JTAG
specific behaviour.
OpenOCD can tolerate it but requires the flag "connect_assert_srst"
to the command "reset_config", but the flag is not present in any
configurations in folder tcl. This enforces the guess it was not
noticed due to on-board capacitors or missing connection of reset
pin or ST-Link only used in SWD; so it's safe applying this patch.
Change the default behaviour to avoid reset in JTAG at connection.
There is no need to manage the ST-Link version here, since every
parameter that is not recognized by older ST-Link is treated as
"connect in JTAG with reset pulse", keeping backward compatibility.
Change-Id: Idc97a1457279e3970fd0839cadbff22d9b0302d4 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4713 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Antonio Borneo [Sun, 23 Sep 2018 15:05:05 +0000 (17:05 +0200)]
stlink: simplify api version handling
The API version is hardcoded in the STLINK device and depends on
the specific firmware version. There is no way to switch between
different API. Thus, it make no sense to keep in two separate
variables the current and the max API supported because they are
always equal to "the only" API supported by the specific STLINK.
Collapse h->version.jtag_api_max and h->jtag_api in a single
variable h->version.jtag_api
Remove an extra print of api version; this is already advertised
while printing the STLINK version.
Change-Id: I0519694a34748e5a0aa330932c1256080a6d5f20 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4712 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Antonio Borneo [Thu, 20 Sep 2018 08:49:43 +0000 (10:49 +0200)]
stlink: simplify maintenance of version and features
The number of stlink firmware version is growing, each carrying new
features. Today's code has several check distributed here and there
and it's already hard to track them and verify the correctness.
The introduction of STLINK-V3 will make the situation much worst,
and the code much less readable.
Add a "flags" bitmask in the struct stlink_usb_version to allow
setting individual bits for each feature available or for specific
quirks and workarounds.
This patch does not implement setting nor testing "flags"; it would
be introduced in following patches, one bit at a time.
Change-Id: I09d78202646a6c8330731f8aa96dc9d295fa5655 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4706 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Antonio Borneo [Thu, 20 Sep 2018 08:32:03 +0000 (10:32 +0200)]
stlink: check for SWD support
The old ST-Link/V1 can either run an obsolete firmware with API_V1 or
a more recent firmware (from V1J11 to last V1J13) with API_V2.
SWD is only provided by the latter API.
Return error is SWD is selected on adapters that does not support it.
Change-Id: Iac4ba54d191ba80fb445a7cd3d8c2621dc7a4846 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4705 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Recent stlink version V2J26 introduces support for 16 bit memory
read/write.
Add the new API and modify the wrappers stlink_usb_{read|write}_mem
For older version of stlink keep the same behavior as in current
code.
While there, fix some minor typo in comments inside the wrappers.
This fixes ticket #204 ("st-link can't flash stm32f2x based flash
without target algorithms").
Change-Id: Id7d404e588f10a4b0f8a93d7ca44cef8e4c49b4f
Author: Laurent LEMELE <laurent.lemele@st.com> Rebased-by: Antonio Borneo <borneo.antonio@gmail.com>
Ticket: https://sourceforge.net/p/openocd/tickets/204/ Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4704 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk> Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
ST-Link uses two separate API for setting the interface speed in SWD
and JTAG.
Add the missing API for JTAG and run the proper API depending on the
selected mode.
Change-Id: I71e955e23c1d82c2ea1c8bfade7612c5a0377a74
Author: Laurent LEMELE <laurent.lemele@st.com> Rebased-by: Antonio Borneo <borneo.antonio@gmail.com> Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4703 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Antonio Borneo [Sun, 23 Sep 2018 07:53:30 +0000 (09:53 +0200)]
stlink: add usb pid for v2.1 without mass storage device
New version of ST-Link/V2.1 without mass storage device.
From debug point of view, it is compatible with existing ST-Link/V2.1
It uses a new USB PID because the USB endpoints and interfaces are
different from usual ST-Link/V2.1
Add the new PID in the driver, in the tcl interface script and in the
udev configuration script.
Change-Id: Id2e1b5a5d0347c5d951a86a9cdb76be52cfd4ea3 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: http://openocd.zylin.com/4702 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>