-
Notifications
You must be signed in to change notification settings - Fork 2k
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
tools/edbg: update to latest upstream version #13233
Conversation
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.
Looks good, I'll test on saml11-xpro and arduino-zero tomorrow.
@@ -0,0 +1,142 @@ | |||
# Atmel SAM C/D/L/R series: |
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.
this looks inefficient, all findstrings will always be checked.
Maybe link them with else?
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.
IMHO, the board
should still pick this. Typically a board knows its CPU_MODEL
, right?
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.
IMHO, the board should still pick this.
No, that's what I was trying to get rid of.
The board already sets CPU_MODEL
- that should be enough.
There is no need to further duplicate that information if it can be avoided.
this looks inefficient, all findstrings will always be checked. Maybe link them with else?
I can sprinkle some else
in to speed up make flash
😉
But I doubt that it makes much of a difference.
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.
But I doubt that it makes much of a difference.
I couldn't measure a difference, so probably it doesn't matter. But it looks inefficient, especially most of the device types are not even supported. Maybe we just uncomment those?
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.
Maybe we just uncomment those?
... and add an error case (`ifeq (,
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.
SAML10, SAML11, SAML21 are happy with the new edbg version. Work as expected.
I don't have opinion about dropping EDBG_DEVICE_TYPE from the board side. Does it impact that much the build system ?
It works like a charm on arduino-zero, samr30-xpro and samr34-xpro. |
Hm, the CI raspi use a precompiled edbg version. We have a chicken-and-egg problem here. |
Nice, the old options are still accepted. I've updated edbg on the CI workers, let's see. |
unrelated build errors. CI wise, this is fine now. |
SAME54-XPRO is also happy with the new version. |
yes, let's go. please squash! |
This version has changed the target names, so add a edbg-devices.inc.mk to automatically select the right target.
EDBG_DEVICE_TYPE is a property of the CPU, is should not be set by every board individually.
Contribution description
Upstream
edbg
changed the target names (they already generated a deprecation warning before).While updating this I moved the cpu -> target name mapping to
edbg-devices.inc.mk
and removed theEDBG_DEVICE_TYPE
declaration from the board Makefiles.I chose to generate a mapping even for MCUs that are not supported by RIOT yet, this allowed for automated generation of the file and no further work should we gain support for these processors in the future.
Testing procedure
You should still be able to flash the boards as before.
I tested this on
same54-xpro
andsamr21-xpro
.Issues/PRs references
none