Bump Wasmtime to v47.0.0#3605
Conversation
|
Looks like I fouled up the regex and this one upgraded some of the wasmtime-related bits but not wasmtime itself. Another spectacular start. ETA: yep it used to be |
|
I fixed the bump manually. There's still a bunch of errors but it's possible they're real breaking changes. |
|
Okay I think I fixed the compile errors but it was extremely "randomly incant at bindgen until it produces the thing that would make my life least painful" rather than any understanding of what I was doing so one of you giant brains had better review the heck out of it! |
Signed-off-by: spinframeworkbot <202838904+spinframeworkbot@users.noreply.github.com>
Signed-off-by: itowlson <ivan.towlson@fermyon.com>
Signed-off-by: itowlson <ivan.towlson@fermyon.com>
cedb743 to
bce7e24
Compare
Signed-off-by: itowlson <ivan.towlson@fermyon.com>
bd9246a to
e892922
Compare
Signed-off-by: itowlson <ivan.towlson@fermyon.com>
e892922 to
d437be9
Compare
|
Okay I had to comment out the dependency review action because:
So we need a better solution for whatever this is meant to achieve, and I don't know what that is. ETA: AARGH I CAN'T COMMENT IT OUT BECAUSE SOMEONE HAS MADE IT A REQUIRED CHECK AARGH |
cfd3ba5 to
d437be9
Compare
|
Was going to start to do the 47 bump but saw this (yay ty!) and changes here look good to me. Both test failures look to be: Do you know if that's spurious? If not I can help try to dig in |
|
@alexcrichton I am not sure. When I test locally, I get the error from #3613, and this is something different. Since, in principle, nothing has changed in this area since #3603, I suspect it's real - but I am not sure... ETA: My "local test gets 3613" was running a test under the Spin CLI, where I could view the results and diagnose behaviour better. In |
|
@alexcrichton I'm now testing in a release CLI and seeing a panic on the |
|
Hm yeah this might all just point back at the same issue... I'm testing locally to see if I can reproduce CI's failure, and if I can I'll see if updating to Wasmtime's |
|
Ok cherry-picking the fix for #3613 doesn't help, and printing the stderr is revealing: Definitely looks like a bug in wit-bindgen or wasmtime, I'll keep digging |
|
Indeed, a bug in Wasmtime. I'll work on getting that landed + backporting the call_hook fix and get a 47.0.2 tomorrow |
|
Ok Wasmtime 47.0.2 is out now, if you want to update to that here I tink it should get past those tests in CI |
|
Looking good locally; thank you, Alex. |
|
@fibonacci1729 What do you think we should do about the "review new dependencies" failure? I think all this stuff is indirect and not readily fixable, and most of it only happened because the cargo update in the PR spammed everything. |
|
@itowlson Would scoping the update work here: |
Signed-off-by: itowlson <ivan.towlson@fermyon.com>
4b9c2d5 to
fe56f0a
Compare
|
Additionally, i think we could make the |
Fie! I looked for that option! Turns out you're right, I missed it's not a named flag:
|
|
Bumps Wasmtime to v47.0.0 prerelease