cpu, boards: add new STM32C0 and NUCLEO-C071RB [new PR]

[crasbe]

Contribution description

Unfortunately the PR #20939 that added several new STM32C0 based boards went stale, so I took the liberty to take it over.

The fixup commit changes the copyright notice to the new SPDX format and the headerguards to #pragma once as well as fixing a typo in the KConfig file. The flashing information now points to the guide merged in #21767 (following the scheme from #21337).

For the time being I did not include the STM32C0116-DK and STM32C0316-DK because I have neither of them for testing available.

Testing procedure

I have not tested this yet with real hardware, but that will happen soon.

Issues/PRs references

This is continued from #20939.
Tracking Issue #19264.

[riot-ci]

Murdock results

✔️ PASSED

e0f9f11 cpu/stm32: update STM32C0 CMSIS version

Success	Failures	Total	Runtime
10515	0	10516	33m:13s

Artifacts

• Documentation preview

[crasbe]

Well it's never easy... The latest OpenOCD release (0.12.0) does not support the STM32C071 yet. The latest sourcecode does support it, but the standard programming method hla_swd is deprecated (or rather: it throws an error and says that the adapter does not support it).

cbuec@W11nMate:~/RIOTstuff/riot-nrfvendor/RIOT$ sudo BOARD=nucleo-c071rb OPENOCD=/usr/local/bin/openocd make -C tests/sys/shell flash-only
make: Entering directory '/home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/tests/sys/shell'
/home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/dist/tools/openocd/openocd.sh flash /home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/tests/sys/shell/bin/nucleo-c071rb/tests_shell.elf
### Flashing Target ###
Open On-Chip Debugger 0.12.0+dev-02228-ge5888bda3-dirty (2025-10-09-21:04)
Licensed under GNU GPL v2
For bug reports, read
        http://openocd.org/doc/doxygen/bugs.html
Debug adapter doesn't support 'hla_swd' transport
make: *** [/home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/tests/sys/shell/../../../Makefile.include:866: flash-only] Error 1
make: Leaving directory '/home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/tests/sys/shell'

I found this comment that daplink_swd should be used instead (https://review.openocd.org/c/openocd/+/8525/2..7/tcl/board/st_nucleo_c0.cfg#b5) but apparently that is also deprecated again and swd should be used.

cbuec@W11nMate:~/RIOTstuff/riot-nrfvendor/RIOT$ sudo BOARD=nucleo-c071rb OPENOCD=/usr/local/bin/openocd make -C tests/sys/shell flash-only
make: Entering directory '/home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/tests/sys/shell'
/home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/dist/tools/openocd/openocd.sh flash /home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/tests/sys/shell/bin/nucleo-c071rb/tests_shell.elf
### Flashing Target ###
Open On-Chip Debugger 0.12.0+dev-02228-ge5888bda3-dirty (2025-10-09-21:04)
Licensed under GNU GPL v2
For bug reports, read
        http://openocd.org/doc/doxygen/bugs.html
Warn : DEPRECATED! use 'transport select swd', not 'transport select dapdirect_swd' <============================
srst_only separate srst_nogate srst_open_drain connect_assert_srst
DEPRECATED! use 'tcl port' not 'tcl_port'
DEPRECATED! use 'telnet port', not 'telnet_port'
DEPRECATED! use 'gdb port', not 'gdb_port'
Info : STLINK V2J45M31 (API v2) VID:PID 0483:374B
Info : Target voltage: 3.260170
Info : Unable to match requested speed 2000 kHz, using 1800 kHz
Info : Unable to match requested speed 2000 kHz, using 1800 kHz
Info : clock speed 1800 kHz
Info : SWD DPIDR 0x0bc11477
Info : [stm32c0x.cpu] Cortex-M0+ r0p1 processor detected
Info : [stm32c0x.cpu] target has 4 breakpoints, 2 watchpoints
Info : [stm32c0x.cpu] Examination succeed
Info : [stm32c0x.cpu] starting gdb server on 0
Info : Listening on port 46869 for gdb connections
    TargetName         Type       Endian TapName            State
--  ------------------ ---------- ------ ------------------ ------------
 0* stm32c0x.cpu       cortex_m   little stm32c0x.cpu       unknown
[stm32c0x.cpu] halted due to debug-request, current mode: Thread
xPSR: 0xf1000000 pc: 0x080008a4 msp: 0x20000200
Info : device idcode = 0x10016493 (STM32C071xx - Rev Z : 0x1001)
Info : RDP level 0 (0xAA)
Info : flash size = 128 KiB
Info : flash mode : single-bank
Info : Padding image section 1 at 0x08003fdc with 4 bytes (bank write end alignment)
Warn : Adding extra erase range, 0x08003fe0 .. 0x08003fff
auto erase enabled
wrote 16352 bytes from file /home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/tests/sys/shell/bin/nucleo-c071rb/tests_shell.elf in 0.581683s (27.453 KiB/s)
verified 16348 bytes in 0.205263s (77.778 KiB/s)
shutdown command invoked
Done flashing
make: Leaving directory '/home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/tests/sys/shell'

THAT however clashes with older OpenOCD versions (0.11.0 that is shipped with my WSL Ubuntu) which results in an error:
Error: Debug adapter doesn't support 'swd' transport 🫠

cbuec@W11nMate:~/RIOTstuff/riot-nrfvendor/RIOT$ sudo BOARD=nucleo-c071rb OPENOCD=/usr/bin/openocd make -C tests/sys/shell flash-only
make: Entering directory '/home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/tests/sys/shell'
/home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/dist/tools/openocd/openocd.sh flash /home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/tests/sys/shell/bin/nucleo-c071rb/tests_shell.elf
### Flashing Target ###
Open On-Chip Debugger 0.11.0
Licensed under GNU GPL v2
For bug reports, read
        http://openocd.org/doc/doxygen/bugs.html
Error: Debug adapter doesn't support 'swd' transport

make: *** [/home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/tests/sys/shell/../../../Makefile.include:866: flash-only] Error 1
make: Leaving directory '/home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/tests/sys/shell'

---

Nevertheless here is a successful flashing process and the shell:

cbuec@W11nMate:~/RIOTstuff/riot-nrfvendor/RIOT$ sudo BOARD=nucleo-c071rb make -C tests/sys/shell flash term
make: Entering directory '/home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/tests/sys/shell'
Building application "tests_shell" for "nucleo-c071rb" with CPU "stm32".

"make" -C /home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/pkg/cmsis/
...
"make" -C /home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/sys/ztimer
   text    data     bss     dec     hex filename
  16172     176    2512   18860    49ac /home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/tests/sys/shell/bin/nucleo-c071rb/tests_shell.elf
/home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/dist/tools/openocd/openocd.sh flash /home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/tests/sys/shell/bin/nucleo-c071rb/tests_shell.elf
### Flashing Target ###
Open On-Chip Debugger 0.12.0+dev-02228-ge5888bda3-dirty (2025-10-09-21:04)
Licensed under GNU GPL v2
For bug reports, read
        http://openocd.org/doc/doxygen/bugs.html
Warn : DEPRECATED! use 'transport select swd', not 'transport select dapdirect_swd'
srst_only separate srst_nogate srst_open_drain connect_assert_srst
DEPRECATED! use 'tcl port' not 'tcl_port'
DEPRECATED! use 'telnet port', not 'telnet_port'
DEPRECATED! use 'gdb port', not 'gdb_port'
Info : STLINK V2J45M31 (API v2) VID:PID 0483:374B
Info : Target voltage: 3.260170
Info : Unable to match requested speed 2000 kHz, using 1800 kHz
Info : Unable to match requested speed 2000 kHz, using 1800 kHz
Info : clock speed 1800 kHz
Info : SWD DPIDR 0x0bc11477
Info : [stm32c0x.cpu] Cortex-M0+ r0p1 processor detected
Info : [stm32c0x.cpu] target has 4 breakpoints, 2 watchpoints
Info : [stm32c0x.cpu] Examination succeed
Info : [stm32c0x.cpu] starting gdb server on 0
Info : Listening on port 40007 for gdb connections
    TargetName         Type       Endian TapName            State
--  ------------------ ---------- ------ ------------------ ------------
 0* stm32c0x.cpu       cortex_m   little stm32c0x.cpu       unknown
[stm32c0x.cpu] halted due to debug-request, current mode: Thread
xPSR: 0xf1000000 pc: 0x080008a4 msp: 0x20000200
Info : device idcode = 0x10016493 (STM32C071xx - Rev Z : 0x1001)
Info : RDP level 0 (0xAA)
Info : flash size = 128 KiB
Info : flash mode : single-bank
Info : Padding image section 1 at 0x08003fdc with 4 bytes (bank write end alignment)
Warn : Adding extra erase range, 0x08003fe0 .. 0x08003fff
auto erase enabled
wrote 16352 bytes from file /home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/tests/sys/shell/bin/nucleo-c071rb/tests_shell.elf in 0.588064s (27.155 KiB/s)
verified 16348 bytes in 0.216457s (73.755 KiB/s)
shutdown command invoked
Done flashing
/home/cbuec/RIOTstuff/riot-nrfvendor/RIOT/dist/tools/pyterm/pyterm -p "/dev/ttyACM0" -b "115200" -ln "/tmp/pyterm-root" -rn "2025-10-09_23.09.35-tests_shell-nucleo-c071rb"
2025-10-09 23:09:35,526 # Connect to serial port /dev/ttyACM0
Welcome to pyterm!
Type '/exit' to exit.
help
2025-10-09 23:10:03,240 # help
2025-10-09 23:10:03,243 # Command              Description
2025-10-09 23:10:03,246 # ---------------------------------------
2025-10-09 23:10:03,251 # bufsize              Get the shell's buffer size
2025-10-09 23:10:03,254 # start_test           starts a test
2025-10-09 23:10:03,257 # end_test             ends a test
2025-10-09 23:10:03,261 # echo                 prints the input command
2025-10-09 23:10:03,265 # empty                print nothing on command
2025-10-09 23:10:03,269 # periodic             periodically print command
2025-10-09 23:10:03,273 # xfa_test1            xfa test command 1
2025-10-09 23:10:03,276 # xfa_test2            xfa test command 2
2025-10-09 23:10:03,281 # app_metadata         Returns application metadata
2025-10-09 23:10:03,286 # pm                   interact with layered PM subsystem
2025-10-09 23:10:03,291 # ps                   Prints information about running threads.
2025-10-09 23:10:03,296 # version              Prints current RIOT_VERSION
2025-10-09 23:10:03,299 # reboot               Reboot the node
> version
2025-10-09 23:10:06,674 # version
2025-10-09 23:10:06,678 # 2025.10-devel-430-g73149-pr/stm32c0_new
>

[AnnsAnns]

It's really hard to find any code to complain about in a crasbe PR 😔 I can confirm that both the board and docs build though (Can not flash, I thought I had one at home but its a different nucleo)

[crasbe]

It's really hard to find any code to complain about in a crasbe PR 😔

Oh stop it you 😳 👉 👈

But credit where credit is due, that code was not written by me (and I also can't and won't take credit for the mistakes 🤣 ).

[AnnsAnns]

That's true, thank you jasonparkerbissell / jparker324 (Are these two people? 😅) for the OG PR.

Considering that this is a revamp of the old PR, I'd count your successful flash test as enough for this but I also wouldn't mind if somebody else could verify it. Either way, thank you for your final touches to get this merge-able 🐸 👍

[AnnsAnns]

Did Murdock fall asleep?

[crasbe]

No, the merge just takes too long and times out. We currently don't have enough CI capacity without `mobi6`.

[AnnsAnns]

Oh :(