-
Notifications
You must be signed in to change notification settings - Fork 2.1k
boards/samd21-based: model Kconfig #17355
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
boards/samd21-based: model Kconfig #17355
Conversation
For |
There also seems to be an issue with the |
T
This is missing from Kconfig actually |
5c1747d
to
c93fa56
Compare
Thanks, I fixed this now |
Let's see how this works |
BTW: I rebased to resolve conflicts |
ugg, tests/periph_pm still has some issues... |
This one now needs a rebase! |
c93fa56
to
456e146
Compare
Rebased and pushed a fix attempt |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It would be nice to have another maintainers opinion on the use of cpu specific module selection in an application.
# selects ztimer_no_periph_rtt to select the same modules in Kconfig and make | ||
ifneq (,$(filter-out sam3,$(filter sam%,$(CPU)))) | ||
ifneq (,$(filter ztimer%,$(USEMODULE))) | ||
USEMODULE += ztimer_no_periph_rtt |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It seems this works and it is commented well enough. I feel like cpu specific code should not be in the application, but as this is somehow a workaround for the rather complex rtt/ztimer maybe we allow it for now until a better solution is found.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Makefile.board.dep
is there exactly for this reason, sure most of the time its based on $(BOARD)
but IMO this is exactly the kind of usage. So far we have force Kconfig
to match make
weird behaviour, I see no harm in doing it the other way around here. IMO we shouldn't stall this one!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Fair enough. It is documented and everything is passing. Please squash!
maybe ping @maribu or @kaspar030 ? |
456e146
to
f18bd93
Compare
- fxos8700 - hdc1000 - pir - pulse_counter
For sam0 there is a conflict between rtt and rtc, make resolves this based on feature conflicts and the feature to be included depends on dependency resolution. Kconfig can't rely on order of inclusion therefore ztimer_no_periph_rtt is implied to aboid ztimer_msec selecting rtt (its the case for BOARDs using stdio_rtt
f18bd93
to
600acb7
Compare
Rebase to fix conflict |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
ACK, if everything is documented and everything is green it is better to have the boards in then not. This should prevent the whack-a-mole problem with the Kconfig migration.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
ACK! (all but my lately force-pushed changes)
Next will be #17402 |
Contribution description
This models Kconfig for all samd21-based boards. Currently this includes 28d8445 and c9b81f2 from #17299 as a dependency. Some minor changes on drivers are also included.
The list of boards is:
Testing procedure
Issues/PRs references
Depends on #17299
Part of #16875