_DEBUG_USB_IO_ was not actually used anywhere.
Its last user was the old ft2232 driver removed in Nov 2016
with commit cc2d4f015f72d7c30d613b50572eb9f31fac515a
("Remove since long deprecated ft2232 driver").
Change-Id: I1a98db7c7b03a89cc9347c0a66ec2106d2168c3f Signed-off-by: R. Diez <rdiez-2006@rd10.de>
Reviewed-on: https://review.openocd.org/c/openocd/+/9002 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Marc Schink [Sun, 13 Jul 2025 07:05:40 +0000 (09:05 +0200)]
helper/log: Rework 'debug_level' command
The patch changes the following:
- Use correct return value ERROR_COMMAND_ARGUMENT_INVALID is case an
invalid debug level is provided.
- Do not echo the selected debug level.
- Remove the 'debug_level: ' prefix when the debug level is shown.
This makes processing via Tcl easier.
- Use command_print() in order to provide the error message to the
caller.
Change-Id: Ida84a58c61060497fc36a1926eec7dd30c66cd72 Signed-off-by: Marc Schink <dev@zapb.de>
Reviewed-on: https://review.openocd.org/c/openocd/+/8996 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Add a board configuration file for the MikroElektronika Clicker 4 for
STM32F745VG (MIKROE-6331), which contains their CMSIS-DAP-compatible
on-board CODEGRIP programmer.
Place this into its own subdirectory, since MikroElektronika has quite a
few boards in their portfolio.
Marc Schink [Mon, 28 Jul 2025 07:34:39 +0000 (07:34 +0000)]
target/armv8: Use 'bool' data type for cache validity flag
The variable is already used as boolean value but has the wrong data
type.
Change-Id: Ia54cfbcdad00dc15e1181c05fb97fcbaa435bb21 Signed-off-by: Marc Schink <dev@zapb.de>
Reviewed-on: https://review.openocd.org/c/openocd/+/9059 Tested-by: jenkins Reviewed-by: Richard Allen <rsaxvc@gmail.com> Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Marc Schink [Sun, 20 Jul 2025 11:33:38 +0000 (11:33 +0000)]
target: Make use of str_enabled_disabled()
The data type changes introduced in [1,2] lead to implicit casts from a
boolean to an integer value in the string selection between "enabled" and
"disabled".
Use str_enabled_disabled() to get rid of this implicit cast.
Antonio Borneo [Fri, 4 Jul 2025 14:50:46 +0000 (16:50 +0200)]
tcl: move STM32 MPU files in vendor folder
Move the existing files for STM32MP13x and STM32MP15x in the
folder "st".
Rename the board files using the correct names.
While there, add the missing URL to one of the boards.
Liam Fletcher [Mon, 26 May 2025 09:42:56 +0000 (10:42 +0100)]
tcl: add microchip's pic64gx curiosity config
Microchip's PIC64GX Curiosity Board has a RISC-V core complex with 4
application processors and one monitor processor. The Curiosity kit also
has an on-board debug interface based around an FTDI 4232H device.
This patch adds basic target, interface and board support for PIC64GX
Curiosity Kit.
Liam Fletcher [Mon, 26 May 2025 09:30:07 +0000 (10:30 +0100)]
target: add microchip polarfire soc config
Microchip's PolarFire SoC has a RISC-V core complex with four
application processors and one monitor processor. This basic
configuration can be used to attach to all proccessor's or a single
processor, specified by the run-time argument $COREID
It can be used with most FTDI based debug interfaces and has been tested
with interface/ftdi/olimex-arm-usb-tiny-h.cfg.
Change-Id: I75dd965f1ce550807706d00fe17de887d36f0b02 Signed-off-by: Liam Fletcher <liam.fletcher@microchip.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8877 Reviewed-by: Tomas Vanek <vanekt@fbl.cz> Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Nicolas Derumigny [Tue, 5 Nov 2024 15:49:04 +0000 (15:49 +0000)]
jtag: drivers: xlnx-axi-xvc: Add support for Xilinx XVC over direct bus interface (AXI)
This change allow to use direct mapping of the JTAG interface using
Xilinx Virtual Cable (XVC) over AXI. This merges the existing XVC PCIe
code and the patch proposed by Jeremy Garff
(https://review.openocd.org/c/openocd/+/6594).
This is useful when using on a Zynq/ZynqMP/uBlaze host with direct
access to the debug bridge over AXI. You can then use the debug bridge
Xilinx IP (AXIXVC) to debug a remote device.
Signed-off-by: Nicolas Derumigny <nicolas.derumigny@inria.fr>
Change-Id: I934591b489e30b400b87772b1437e6030440904c
Reviewed-on: https://review.openocd.org/c/openocd/+/8595 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Tomas Vanek [Thu, 3 Jul 2025 04:52:29 +0000 (06:52 +0200)]
flash/nor/stm32l4x: fix permanent write protection on STM32U5
Unlike other devices supported by this driver STM32U5 devices
have a new UNLOCK bit in FLASH_WRP1AR, WRP1BR, WRP2AR, WRP2BR
registers. Writing zero to this bit makes the write protection
block permanent with no way to unprotect.
Commit 6554d176e926 ("flash/stm32l4x: support STM32U59/U5Ax devices")
and later commits with additional U5 devices lack support for
the UNLOCK bit and therefore makes write protection permanent
without warning.
Introduce the new bit flag F_WRP_HAS_LOCK and mark U5 devices by it.
Set UNLOCK bit in stm32l4_write_one_wrpxy() if F_WRP_HAS_LOCK is set.
Change-Id: I26b97d855e094a21540e3377f367520683af2eac Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: https://review.openocd.org/c/openocd/+/8981 Tested-by: jenkins
Antonio Borneo [Sun, 22 Jun 2025 09:03:41 +0000 (11:03 +0200)]
jtag/drivers: dmem: fix build on Linux 32 bits
On 32 bits machine both 'uintptr_t' and pointers are 32 bit.
The cast
(volatile uint32_t *)((uintptr_t)dmem_emu_virt_base_addr + addr)
fails with error
error: cast to pointer from integer of different size
[-Werror=int-to-pointer-cast]
in lines 100 and 109 because:
- 'addr' is a 'uint64_t';
- adding 'uintptr_t' and 'uint64_t' returns a 64 bit value;
- cast the 64 bit to 'uint32_t *' is an error.
In the code the value passed to 'addr' is always 32 bit wide, so
there is no need to pass it as 'uint64_t'.
Change the type of 'addr' to 'uint32_t'.
Fix also some format string to fit both 32 and 64 bits machines.
Change-Id: I90ff7cd3731cb24a0fc91fe7b69c532b5c698ba0 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8974 Reviewed-by: Nishanth Menon <nm@ti.com> Tested-by: jenkins Reviewed-by: R. Diez <rdiez-2006@rd10.de>
Tomas Vanek [Wed, 18 Jun 2025 10:01:08 +0000 (12:01 +0200)]
flash/nor/rp2xxx: save ACCESSCTRL over ROM API calls
Especially after the flash probe (used in gdb-attach event)
we need to completely restore the original security state to allow
'resume' or gdb 'continue' without injecting strange errors
to application code.
Save all ACCESSCTRL registers potentially changed by triggering CFGRESET.
Restore them at cleanup.
Tomas Vanek [Wed, 18 Jun 2025 07:44:03 +0000 (09:44 +0200)]
flash/nor/rp2xxx: save security state over target algo
RP2040 and RP2350 flash driver runs a ROM API target algorithm
in probe to setup QSPI command interface. The Cortex-M33 core
of RP2350 has to be in secure mode with SAU and MPU switched off
to ensure ROM API call working properly.
Especially after the flash probe (used in gdb-attach event)
we need to completely restore the original security state to allow
'resume' or gdb 'continue' without injecting strange errors
to application code.
Use cortex_m support to set secure mode and to restore it back.
Antonio Borneo [Sat, 21 Jun 2025 10:11:24 +0000 (12:11 +0200)]
helper: types: fix proper return type in example of ARRAY_SIZE()
The example in the comment above the declaration of the macro
ARRAY_SIZE() assigns the value to a variable of type 'unsigned'
that is not allowed by the coding style (should be 'unsigned int')
and is not correct since the macro uses 'sizeof()' and the type
returned is 'size_t'.
Henrik Brix Andersen [Wed, 11 Dec 2024 08:34:57 +0000 (09:34 +0100)]
jtag: drivers: xlnx-pcie-xvc: use correct TMS polarity during pathmove
The xlnx_pcie_xvc_execute_pathmove() function checks whether TMS
should be high or low for transitioning from the current state to the
next state, but then calls xlnx_pcie_xvc_transact() with the opposite
level, leading to invalid state transitions.
Fix the polarity of TMS in the calls to xlnx_pcie_xvc_transact() to
match the required TMS level.
Change-Id: I2383e41fb70063e26aa69fabcf728df597607934 Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8613 Reviewed-by: Moritz Fischer <moritzf@google.com> Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com> Reviewed-by: Nicolas Derumigny <nicolas.derumigny@inria.fr>
Jan Matyas [Tue, 17 Jun 2025 11:17:23 +0000 (13:17 +0200)]
openocd.c: 'init' should fail if GDB service cannot be created
If it is not possible to create a GDB service for a certain target
(for example the given TCP port is already occupied), the "init"
command should fail, but it currently does not.
Fix this by checking the return code of gdb_target_add_all().
Steps to reproduce:
1) Make the port 3333/tcp occupied. For example by:
nc -l 3333
2) In another terminal, launch OpenOCD. Use the gdb_port
3333 (which is the default). For example:
Antonio Borneo [Sat, 14 Jun 2025 12:39:37 +0000 (14:39 +0200)]
jtag: interfaces: sort the drivers by alphabetic order
Add comments to require the list of drivers to be kept sorted.
While there:
- align the check on BUILD_PRESTO and BUILD_USB_BLASTER;
- fix indentation of the closing parenthesis.
Change-Id: Ic78281b1cdfb5bf72ea41427233e76516001b429 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8950 Tested-by: jenkins
kryvosheiaivan [Wed, 28 May 2025 13:52:42 +0000 (16:52 +0300)]
cmsis-dap: Fix freeing pending transfers on close
Freeing pending transfers on shutdown is done in openOCD
and on libusb side. This created concurrency in freeing
memory and segmentation faults:
https://github.com/libusb/libusb/issues/1627
Bug is reproduced better if many targets are laucnhed.
Bug was reproduced with CMSIS-DAP on targets:
cyw20829, psoc4, stm32l5 if launching multiple times.
Proposed working fix: if some transfers pending/in-flight
on 'shutdown' then apply libusb_handle_events_timeout_completed()
to make transfer complete. In all cases transfer completed
due to tests.
Change-Id: I44621ac6096791714910220d04614d0a19ce47bd Signed-off-by: kryvosheiaivan <Ivan.Kryvosheia@infineon.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8876 Reviewed-by: Tomas Vanek <vanekt@fbl.cz> Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Marc Schink [Wed, 9 Oct 2024 09:14:21 +0000 (11:14 +0200)]
adapter: Deprecate Gateworks GW16012 driver
The adapter is not available for years now. There is also no information
about this device from Gateworks. The poor hardware availability and the
lack of users prevents testing, maintenance and adaptations to future
changes.
Mark the adapter as deprecated as a first step to give potential users
the opportunity to upgrade the hardware until the next OpenOCD release.
Change-Id: I037325a6b018b26608733a36bef30db2785858f8 Signed-off-by: Marc Schink <dev@zapb.de>
Reviewed-on: https://review.openocd.org/c/openocd/+/8651 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
The adapter is not available for years now and Amontec is not even a
company anymore. The poor hardware availability and the lack of users
prevents testing, maintenance and adaptations to future changes.
Mark the adapter as deprecated as a first step to give potential users
the opportunity to upgrade the hardware until the next OpenOCD release.
Change-Id: Idd9fb75588246bc39e12ea17a71435ed77f0f50b Signed-off-by: Marc Schink <dev@zapb.de>
Reviewed-on: https://review.openocd.org/c/openocd/+/8349 Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com> Tested-by: jenkins
Antonio Borneo [Tue, 3 Jun 2025 09:52:59 +0000 (11:52 +0200)]
HACKING: describe keeping the 'Change-Id' on new patch versions
We often get on Gerrit a new version of an old patch with a new
'Change-Id' value. This breaks the history of the review, adding
more work to the review process.
Describe in HACKING why the hook 'commit-msg' is required and how
to handle the 'Change-Id' on new patch versions.
Change-Id: I5c060b19f966add7422704912b38e1ab2f788e5f Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8940 Tested-by: jenkins
Electric Worry [Thu, 29 May 2025 10:31:59 +0000 (11:31 +0100)]
tcl/board/orange_pi_zero_3: Add Orange Pi Zero 3 board
The Orange Pi Zero 3 is an SBC that uses an Allwinner H618
SoC. As such, JTAG support is fully available, however the
SoC multiplexes JTAG function with UART1 and microSD.
Unfortunately Xunlong has used UART1 for the Wifi-BT
chip, leaving JTAG accessible only via the microSD using
a microSD breakout board (for example).
Change-Id: I0dc078cd2f3176815271917eb5e948cc8ef94525 Signed-off-by: Electric Worry <me@electricworry.net>
Reviewed-on: https://review.openocd.org/c/openocd/+/8938 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com> Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Antonio Borneo [Tue, 3 Jun 2025 08:34:29 +0000 (10:34 +0200)]
configure: hide build issue of amt_jtagaccel driver by disabling it
With commit d8a2f6dbcf5f ("configure.ac: show the Amontec
JTAG-Accelerator driver in the config summary") the driver
amt_jtagaccel is now build by default on Linux.
This highlights the dependency of some include files, dependency
that is not properly managed and that can cause build failure.
The driver is queued to be dropped soon, so there is no real
interest to fix the dependencies.
Change the default so the driver is not built if the user does not
require it at configure time.
Change-Id: Ifb74e2c802abda290efbf59ca4ce02048c94e6f8 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8939 Reviewed-by: R. Diez <rdiez-2006@rd10.de> Tested-by: jenkins
Antonio Borneo [Wed, 28 May 2025 14:58:44 +0000 (16:58 +0200)]
configure: fix for missing include files on Linux
By merging [1] and [2], the drivers 'xlnx-pcie-xvc' and 'linuxspidev'
are now build by default on Linux.
This highlights the dependency of some include files under subfolder
'linux' that are not installed by default in all Linux boxes.
Add the check in 'configure' for the presence of the include file
and conditionally enable the build of the driver.
Change-Id: Ie88645c3455ab07622f069a0cc7bf09d1a5a2c75 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Link: [1] 7214c8be46f7 ("configure: show adapter Xilinx XVC/PCIe in the configuration summary")
Link: [2] 83e0293f7ba3 ("Add Linux SPI device SWD adapter support")
Reviewed-on: https://review.openocd.org/c/openocd/+/8935 Tested-by: jenkins
Kevin Yang [Mon, 12 Oct 2020 20:22:47 +0000 (13:22 -0700)]
target/armv8: Handle modeswitch for aarch32 secure EL3
For aarch32 secure EL3
- Change target_el to 3 for SVC/ABT/IRQ/FIQ/UND/SYS for aarch32 secure
EL3
- Do not update SPSR for SYS, behavior is UNPREDICTABLE (ARMv8-A F5.1.121)
- Do not execute DRPS for SYS, behavior is UNPREDICTABLE (ARMv8-A
F5.1.51)
Change-Id: Ic1484665cd53afcccb5c20b152993a3f0407f8a2 Signed-off-by: Kevin Yang <kangyang@google.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/5854 Tested-by: jenkins Reviewed-by: Matthias Welwarsky <matthias@welwarsky.de> Reviewed-by: Plamena Marinova <pmarinova@hilscher.com> Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Jonathan Bell [Wed, 21 May 2025 14:37:25 +0000 (15:37 +0100)]
tcl: fix broken Raspberry Pi website links
raspberrypi.com is the home for technical information, raspberrypi.org
is the Foundation's site (though there are intelligent redirects).
Several pages have moved around, fix these.
Also tweak a few comments for style and correctness.
Signed-off-by: Jonathan Bell <jonathan@raspberrypi.com>
Change-Id: I7f52bcc362fb213b50987e3a42866fe4a6fec883
Reviewed-on: https://review.openocd.org/c/openocd/+/8885 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com> Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Jonathan Bell [Wed, 21 May 2025 14:26:20 +0000 (15:26 +0100)]
doc: bcm2835gpio: remove broken link and clarify usage
The peripheral address details have been removed from the documentation
part of the website (instead presented in the SoC datasheets).
Pi 5 GPIOs are provided by the RP1 southbridge, which requires the use
of libgpiod. The associated Linux driver for the GPIO interface must be
used when bitbashing on Pi 5, as the timing of pin state changes is not
guaranteed across the PCIe link without special treatment.
Using libgpiod, the typical maximum swclk speed is 1MHz.
Signed-off-by: Jonathan Bell <jonathan@raspberrypi.com>
Change-Id: I8b2c44ec5edd71abaa0a763ba4d4f4603a211348
Reviewed-on: https://review.openocd.org/c/openocd/+/8884 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com> Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Jacek Wuwer [Wed, 24 Apr 2024 08:15:38 +0000 (10:15 +0200)]
jtag/vdebug: implement a polling mechanism
This change adds a polling mechanism to the driver. When not busy
the driver issues a wait, allowing the target to advance time.
The wait period gets adjusted to match the polling setting.
Change-Id: I67f481d05d7c5ce5352b5cb97de78dbaa97d82ae Signed-off-by: Jacek Wuwer <jacekmw8@gmail.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8221 Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com> Tested-by: jenkins
Marc Schink [Mon, 12 May 2025 04:47:22 +0000 (06:47 +0200)]
tcl/target: Move nordic configuration files
Move target configuration files into a dedicated vendor directory as
required by the new guideline for configuration files.
Note that the moved files are still accessible via the old path to ensure
backwards compatibility. This works because of the extended file search in
vendor folders.
Change-Id: If3935985769dc543e8c7d72cda590c9d79303abb Signed-off-by: Marc Schink <dev@zapb.de>
Reviewed-on: https://review.openocd.org/c/openocd/+/8905 Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com> Tested-by: jenkins
Antonio Borneo [Wed, 21 May 2025 13:19:28 +0000 (15:19 +0200)]
startup.tcl: extend the file search in vendor folders
The TCL configuration files are going to be dispatched in vendor
specific folders.
Old user configuration files will fail to find the new files to
include, so a set of fallback files reporting the deprecation
should replace the renamed files.
To prevent such enormous proliferation of fallback files, extend
the search of files in the vendor folders too.
For non-trivial renames, a dedicated table is added in the file
tcl/file_renaming.cfg to track old --> new file names.
The deprecated message is then part of the extended search.
and
- path/to/a/certain/config_file
trigger search of:
- path/to/a/certain/${vendor}/config_file
among a possible vendors list.
This is a temporarily feature that should be removed as soon as
possible to prevent clashing on files with the same name.
The names in tcl/file_renaming.cfg are for demonstration purpose
only and should be dropped when the first real entries are added.
Change-Id: If4793fef27dc570d5df4ff4d77a5e36004f394f6 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8929 Tested-by: jenkins
Lucien Dufour [Thu, 15 May 2025 12:31:25 +0000 (14:31 +0200)]
cortex_a: Use endianness for soft breakpoints
Fix endianness for cortex_r4 and cortex_r5 when inserting software
breakpoints. Because the cortex_a target is used by the cortex_r
architecture and some chips start in BE by default (e.g. TMS570)
projects / users / borneoa / openocd-next.git / log