-
Notifications
You must be signed in to change notification settings - Fork 2.6k
Conversation
I think we should probably disable the simd stuff? So that nothing depends on this by accident? |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, this is correct. Leaving it as Request changes until it's done.
Just curious, is there a roadmap for enabling those various WASM features in wasmtime? |
I don't think there is. But it's good to stay conservative unless we can show that enabling them will actually be beneficial in some way. If there's no benefit then there's no point in turning them on just for the sake of turning them on. |
39d636c
to
adabea6
Compare
Just noticed SIMD was already disabled, so there was no need to change anything. |
Also, @nazar-pc FYI we tend to not force-push in our PRs to clean up the history; we just squash everything on merge. |
Will keep that in mind going forward 👍 |
Indeed; I can confirm that we currently disable One extra thing we should add is to also disable |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I see no major performance regressions in benchmarks after this PR, so looks good to me.
Also disable memory64 support in wasmtime
Also disable memory64 support in wasmtime
Also disable memory64 support in wasmtime
Also disable memory64 support in wasmtime
From release notes besides various fixes, there is SIMD support enabled by default.
The primary reason to upgrade was to make Substrate compile on latest nightly, see bytecodealliance/rustix#142 for details.