ci failures with x64 new backend · wasmtime

Stream: wasmtime

Topic: ci failures with x64 new backend

Pat Hickey (Aug 05 2020 at 16:28):

on my wiggle PR, there seems to be a CI failure due to some unimplemented instructions in the new backend - just want to confirm we're merging PRs even if they fail this check https://github.com/bytecodealliance/wasmtime/pull/2096/checks?check_run_id=947428091

Wiggle: Fix #2053 by pchickey · Pull Request #2096 · bytecodealliance/wasmtime

This PR implements GuestType for GuestPtr<[T]>, and makes it possible to use witx arrays as members of structs or variants of unions. The impl GuestType for GuestPtr<[T]> where T: Guest...

Chris Fallin (Aug 05 2020 at 16:36):

I think it's actually an intermittent (the panic output is spurious, from #[ignore_panic] tests)

Chris Fallin (Aug 05 2020 at 16:37):

At least, that's what @Benjamin Bouvier (PTO) was saying last week after he turned the new backend CI on

Chris Fallin (Aug 05 2020 at 16:37):

I think we should probably actually remove this CI test until he's back and can take a look

Chris Fallin (Aug 05 2020 at 16:38):

(I'm planning to spend a little time looking at it, but in the meantime I mean)

Chris Fallin (Aug 05 2020 at 16:43):

https://github.com/bytecodealliance/wasmtime/pull/2100

Temporarily disable new x64 backend CI. by cfallin · Pull Request #2100 · bytecodealliance/wasmtime

This CI test has been intermittently failing, which is causing issues with other PRs. We should turn it back off until we can work out why the intermittent failures are occuring.

Pat Hickey (Aug 05 2020 at 16:52):

thanks. yeah, it looks intermittent, not every ci run gets hit

Last updated: Nov 22 2024 at 16:03 UTC