zephyr: update to today's "main" #9455
Conversation
lyakh
requested review from
lgirdwood,
plbossart,
mmaka1,
lbetlej,
dbaluta and
kv2019i
as code owners
west.yml
Outdated
| @@ -43,7 +43,7 @@ manifest: | |||
|
|
|||
| - name: zephyr | |||
| repo-path: zephyr | |||
| revision: 689d1edee1d57f052b1d4572d67618c0b0e2b8a4 | |||
| revision: 632d6b9416c1c1813f3304ef29a6eda7954af3c3 | |||
There was a problem hiding this comment.
That's a good question what to write to these commits (and what should be expected). I've been including a short history for what changes this is pulling in from Zephyr, but this has not been a hard requirement. If I look at Zephyr conventions, it seems git commit description culture on west.yml updates is quite varied. Many "update to latest" mixed with commits that have detailed changelogs of pulled in changes.
Oh well, I'll let others chime in, maybe I should stop making the verbose commits as well and just do "update to today's" as well. It's true anyone can check the changes from the source as well, and this can lower the barrier to submit west.yml changes (= you don't need to write a novel to submit one).
There was a problem hiding this comment.
@kv2019i I like your lists of relevant updates, but since you weren't requiring them in the past... ;-) How do you make them? Something like
git log A..B -- adsp_path_1 adsp_path_2 adsp_path_3...
or by pulling all merged PRs with the "Intel ADSP" label or?.. If it can be scripted well enough, we can do that
There was a problem hiding this comment.
I'm fine with "update to Zephyr latest to bring in A, B and C for SOF X, Y and Z features". Keep it simple.
There was a problem hiding this comment.
@lyakh It's mostly manual what I do. I run "git log --pretty=oneline --abbrev-commit 494f88070f6bb909389630264e4f89c71072ae53.. 689d1edee1d57f052b1d4572d67618c0b0e2b8a4 " and then browse through the commit list and pick up commits that seem relevant to highlight. If it's a big update, I just search for keywords (audio/dai/dma/dsp/intel/nxp) and not go through the full list.
But yeah, let's go with a slimmer approach. That seems more aligned with how west.yml is managed in e.g. Zephyr and keeps the barrier-to-entry low.
There was a problem hiding this comment.
Keep in mind these SHA1s can point to literally anything, including unmerged forks. That's just how Github shared storage works.
https://lwn.net/Articles/884105/
Zephyr CI now has an "imposter commit" check:
- github: workflows: Update manifest action to detect impostor commits zephyrproject-rtos/zephyr#75790
Short of such a check, I think every SHA1 in the west.yml should have at least a one-line comment with the name of the branch it comes from. Not just for safety and security but also for convenience.
There was a problem hiding this comment.
maybe we should add such a check
There was a problem hiding this comment.
I suspect adding a comment is several orders of magnitude less effort than automating such a check :-)
|
Wow, LNL alsabat 100% PASS (or SKIP) in https://sof-ci.01.org/sofpr/PR9455/build7827/devicetest/index.html |
|
Interesting LLEXT regression in https://github.com/thesofproject/sof/actions/runs/10791954297/job/29930534306?pr=9455 |
|
Failures with D3 entry on Intel PTL with this one, so the "Internal Intel CI System/merge/build" failure is real. |
lyakh
force-pushed
the
zephyr
branch
2 times, most recently
from
b1878cb to
1ed4444
Compare
Update Zephyr to include:
1dde70637d2a ("Intel: ACE: move hpsram_mask to a data section")
dc5f1bfb3f2e ("west: fix for Python prior to 3.10")
and other fixes and improvements.
Signed-off-by: Guennadi Liakhovetski <[email protected]>
|
@LaurentiuM1234 this west update may fix your linker PR too ? |
|
I don't fully get how this can work without @LaurentiuM1234 's changes to the linker scripts? I.e. I was expecting we'd need the changes from #9372 to upgrade Zephyr...? Or this is a imx93 case not covered in the github CI flows...? |
ah, interesting, I guess the resulting warnings from zephyrproject-rtos/zephyr@3f2790b don't break the CI job, but they're still there. |
|
I can see a column of red on a single MTL DUT that looks like a GFX error cascading to cause all the other tests to fail. Will rerun CI to be sure. |
|
SOFCI TEST |
for the record: https://sof-ci.01.org/sofpr/PR9455/build8170/devicetest/index.html |
|
@lgirdwood quickbuild re-run successful |
Update Zephyr to include f94427deecf0 ("Intel: ADSP: move HPSRAM mask into assembly").