manifest: Update to latest upstream #91
Conversation
jerome-pouiller
force-pushed
the
align-with-upstream
branch
5 times, most recently
from
337be93 to
7fd89db
Compare
.github/workflows/upstream-build.yml
Outdated
| - name: Build Entropy test | ||
| working-directory: zephyr-silabs | ||
| shell: bash | ||
| run: | | ||
| west twister -v --inline-logs -K \ | ||
| -s drivers.entropy \ | ||
| -p siwx917_rb4338a | ||
|
|
||
| - name: Build Flash test | ||
| working-directory: zephyr-silabs | ||
| shell: bash | ||
| run: | | ||
| west twister -v --inline-logs \ | ||
| -s drivers.flash.common.default \ | ||
| -p siwx917_rb4338a \ | ||
| -x DTC_OVERLAY_FILE=$(pwd)/tests/drivers/flash/common/boards/siwx917_rb4338a.overlay | ||
|
|
||
| - name: Build DMA test | ||
| working-directory: zephyr-silabs | ||
| shell: bash | ||
| run: | | ||
| west twister -v --inline-logs \ | ||
| -s drivers.dma.chan_blen_transfer \ | ||
| -p siwx917_rb4338a \ | ||
| -x DTC_OVERLAY_FILE=$(pwd)/tests/drivers/dma/chan_blen_transfer/boards/siwx917_rb4338a.overlay \ | ||
| -x EXTRA_CONF_FILE=$(pwd)/tests/drivers/dma/chan_blen_transfer/boards/siwx917_rb4338a.conf | ||
|
|
||
| - name: Build Bluetooth samples | ||
| working-directory: zephyr-silabs | ||
| shell: bash | ||
| run: | | ||
| west twister -v --inline-logs -K \ | ||
| -s sample.bluetooth.peripheral_hr \ | ||
| -s sample.bluetooth.observer \ | ||
| -p xg27_dk2602a \ | ||
| -p siwx917_rb4338a |
There was a problem hiding this comment.
Wouldn't we still want to keep some amount of 917 testing here, just like we do for EFR32?
There was a problem hiding this comment.
Probably yes.
zephyr/module.yml
Outdated
| # Additional roots for boards and DTS files. Zephyr will use the | ||
| # `<board_root>/boards` for additional boards. The `.` is the root of this | ||
| # repository. | ||
| board_root: . | ||
| # Zephyr will use the `<dts_root>/dts` for additional dts files and | ||
| # `<dts_root>/dts/bindings` for additional dts binding files. The `.` is | ||
| # the root of this repository. | ||
| dts_root: . | ||
| # Zephyr will use the `<soc_root>/soc` for additional socs. The `.` is the | ||
| # root of this repository. |
There was a problem hiding this comment.
I'd consider keeping these, so that it remains possible to have boards in this tree in the future. Downstream (to us) users might also want to populate a fork of this tree with their own board definitions possibly (though, a better solution might be to have a separate manifest tree in that case that just points at this one)
west.yml
Outdated
| revision: a36c91b4cc49b9b54030121004a1565d6fc69475 | ||
| revision: pull/85144/head |
There was a problem hiding this comment.
Why not just update this to the latest upstream main branch, since the PR was already merged?
There was a problem hiding this comment.
I think you should likely also start pointing at the upstream hal_silabs tree, i.e. remove the reference to the downstream one (which will eventually be removed completely)
There was a problem hiding this comment.
Why not just update this to the latest upstream
mainbranch, since the PR was already merged?
I see you created this before the upstream PR was merged, which explains the reference. But now it can be updated to point at a real commit.
There was a problem hiding this comment.
I think you should likely also start pointing at the upstream hal_silabs tree
Ideally yes, but for now:
zephyr/modules/hal/silabs$ git diff --stat upstream/main silabs/silabs
[...]
113 files changed, 49889 insertions(+), 5172 deletions(-)
(the differences mainly come from mbedtls/psa).
There was a problem hiding this comment.
Oh, I didn't realize we had that in our downstream tree. That needs to be sorted out first then, i.e. in the short term rebase our hal_silabs on upstream and in the long term upstream any changes, so that we can proceed with the "remove public downstream hal_silabs" plan.
jerome-pouiller
force-pushed
the
align-with-upstream
branch
7 times, most recently
from
7585d6a to
d247962
Compare
dts/Kconfig
Outdated
There was a problem hiding this comment.
Why add a Kconfig file in dts? If you just want to keep the directory around, .gitignore would make more sense, since Kconfig will never be needed in the dts directory.
dts/silabs/Kconfig
Outdated
There was a problem hiding this comment.
This one was a mistake.
jerome-pouiller
force-pushed
the
align-with-upstream
branch
from
d247962 to
0d4eeb1
Compare
|
Revert to the draft state since hal_silabs miss some commits to allow compilation of 917. |
jerome-pouiller
force-pushed
the
align-with-upstream
branch
3 times, most recently
from
831e19f to
49b9a5b
Compare
|
v2:
|
jerome-pouiller
force-pushed
the
align-with-upstream
branch
from
49b9a5b to
d429e72
Compare
west.yml
Outdated
| path: modules/hal/silabs | ||
| - name: zephyr-mbedtls | ||
| remote: silabs | ||
| revision: 6024c415bf050c09e3990bddb6d53d3b4990a2b1 | ||
| path: modules/crypto/mbedtls | ||
| - name: zephyr | ||
| remote: zephyrproject-rtos | ||
| revision: a36c91b4cc49b9b54030121004a1565d6fc69475 | ||
| revision: f63c6113dedb6e31d968aa0bd239c009fa8b8a8e |
There was a problem hiding this comment.
$ git log --pretty=oneline f63c6113dedb6e31d968aa0bd239c009fa8b8a8e..|wc -l
188Would be nice to update to a slightly newer upstream.
Upstream now include support for 917. So, we can drop it from the downstream. We can also drop the tests related to the 917. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Since commit f8a5ebfd2, check_compliance.py run "git grep CONFIG_" in boards/ and soc/ directories. This command will fail if no match is found. This failure prevent check_compliance.py to run properly. The workaround is to add the word "CONFIG_BOARD" (any existing config option will work) in a random file in boards/ and soc/ Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Several things are easier now. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
jerome-pouiller
force-pushed
the
align-with-upstream
branch
from
d429e72 to
09767fa
Compare
Support for 917 is now a part of the upstream.