makefiles/edbg.inc.mk: use FLASHFILE#11172
Conversation
|
@cladmi I never used riotboot yet so could you confirm it works with a second slot flash beyond the first one ? |
|
@dylad Yeah I tested it sorry to not explain it, I will print the result here as reference. |
|
Maybe it is too early for SAML1X :) i guess it is unrelated to this PR. |
|
Testing And the differents steps will do:
And the matching |
|
@dylad yes the board should have |
Good idea, riotboot doesn't work at all on SAML1X right now. This is unrelated to this PR so we will skip these MCUs for now. I'll try to fix this later. Results for SAML21-XPRO: |
|
All my samd21 based board use BOSSAC instead of EDBG so I cannot test on samd21 family. |
|
@dylad making |
but SAML1X uses the same cpu_init as SAM0 so I was hoping this would work easily. |
|
@aabadie do you have any boards you would like to test with this one |
|
@dylad it is not directly related as it was mainly the "flash still works" as I just added the variable name. |
|
I added the missing test procedure that |
|
@cladmi could you rebased please ? I'll re-test on saml10/saml11 right away. |
Update to use FLASHFILE as file to be flashed on the board.
It is now the case for `edbg`. Openocd support still needs special handling.
6a50e3b to
d4775cc
Compare
|
Rebased. |
|
Thanks @cladmi ! |
|
For me, if you agree with the "Test without board" that shows that flashing command is the same, it should be good with this. |
|
Thank you for the review, almost no hack anymore for |
|
Oh I just noticed this changed the rule for flashing @emmanuelsearch @bergzand @kYc0o should I provide a fix for this one quickly or do you think #11254 is ready/tested enough to get through ? |
|
Hmm it is still working but only by luck because of the hack |
Contribution description
Update to use FLASHFILE as file to be flashed on the board.
This also now removes the compatibility hack in
riotboot.Testing procedure
We need to test that boards using edbg would still work with this.
Boards using edbg:
Test flashing normal examples with the board
Test without board
I replaced FLASHER by 'true' to only show the FFLAGS and get the same output with master and this pull request:
FLASHFILE can be updated
When setting FLASHFILE to another value from environment, like $(ELFFILE) it is correctly taken into account.
wdiff output_master output_pr
riotboottestTest running
tests/riotboottest forsamr21-xproandsaml21-xproIssues/PRs references
Split out of #8838