boards/nucleo-X: enable cpy2remed programmer

[krzysztof-cabaj]

Final Contribution description

Finally this PR enables cpy2remed programmer (see #17550) to all STM Nucleo boards.

Initial Contribution description

This PR enables cpy2remed programmer (see #17550) to four STM Nucleo boards: l031k8, l073rz, f303k8 and l432kc .
All boards have initially older ST-LINK firmware - the oldest was:

Version: 0221
Build:   Nov 19 2015 15:23:07

All are updated (without any problems) to minimal firmware which works with cpy2remed programmer - 2.37.36 (V2.J37.M26 - shown in the ST-LinkUpgrade program).

Testing procedure

Go to RIOT/examples/blinky and run command

make BOARD=nucleo-l031k8 PROGRAMMER=cpy2remed flash
make BOARD=nucleo-l073rz PROGRAMMER=cpy2remed flash
make BOARD=nucleo-f303k8 PROGRAMMER=cpy2remed flash
make BOARD=nucleo-l432kc  PROGRAMMER=cpy2remed flash

After a while, green LED (LED0) should start blinking.

Changes in documentaion can be observed using commands:

make doc
xdg-open doc/doxygen/html/group__boards__nucleo-l031k8.html
xdg-open doc/doxygen/html/group__boards__nucleo-l073rz.html
xdg-open doc/doxygen/html/group__boards__nucleo-f303k8.html
xdg-open doc/doxygen/html/group__boards__nucleo-l432kc .html

Issues/PRs references

#17550

[maribu]

See inline. Maybe your editor has forcefully broken the line mid-word at 80 chars?

Note that e.g. links often tend to be longer than 80 chars, so just ignore the CI complaining about overlong lines if the line contains a link. If the link is less than 80 chars, you could also have the link on its own line and break the line early.

[krzysztof-cabaj]

@maribu thanks for spotting and fixing this issue. I commit all changes.

[aabadie]

Why not move all this copy2remed code in a common place and use some Make magic ?

For example, I'm pretty sure the following would work for all nucleo boards:

index eeed8f95d8..661a81aa8d 100644
--- a/boards/common/nucleo/Makefile.include
+++ b/boards/common/nucleo/Makefile.include
@@ -6,3 +6,9 @@ INCLUDES += -I$(RIOTBOARD)/common/stm32/include
 
 # Setup of programmer and serial is shared between STM32 based boards
 include $(RIOTMAKE)/boards/stm32.inc.mk
+
+# variable needed by cpy2remed PROGRAMMER
+# it contains name of ST-Link removable media
+DIR_NAME_AT_REMED = "NODE$(call uppercase,$(subst -,_,$(subst nucleo,,$(BOARD))))"
+
+PROGRAMMERS_SUPPORTED += cpy2remed

[krzysztof-cabaj]

Why not move all this copy2remed code in a common place and use some Make magic ?

In my initial PR (#17550) I first add this at global level but was decided that should be moved to particular boards.

[aabadie]

In my initial PR (#17550) I first add this at global level but was decided that should be moved to particular boards.

It seems that in #17550, you were only adding the cpy2remed to the supported programmers at a (too) central level (stm32.inc.mk) and were only setting the directory for nucleo-l552ze-q.
My proposal above both adapts the supported programmers and defines the directory automatically for all nucleo boards (assuming the directory name can be computed automatically from the board name).

stale

[krzysztof-cabaj]

Ok. I check if suggested solution works for tested Nucleo boards and introduce necessary changes.

[krzysztof-cabaj]

I checked names of nucleo boards and code provided by @aabadie (once again thanks!). It should works for all nucleo boards - two boards have different name and some manual overwrite of DIR_NAME_AT_REMED is needed. I introduce necessary changes for nucleo-l552ze-q and p-nucleo-wb55. I tested it on few boards to which I have access and everything is working. Moreover, I clean previously introduced changes in the Makefile.include files.

I have one question, due to fact that this PR adds new programmer not only for mentioned 4 boards but for all nucleo boards should I open a new PR? Or only change this PR title and description to reflect final changes?

[krzysztof-cabaj]

@benpicco can move this PR forward ?

[benpicco]

Of course, please squash!

[krzysztof-cabaj]

White space error fixed and all commits squashed. @benpicco @maribu thank for rapid response.