Steven Johnson
Steven Johnson
...sigh. Is there any path or plan for them to implement them?
Since WebGPU API issues have come up again, I see these are still unimplemented; is there any likely path to them getting there, or should we assume this is just...
(From offline discussions: while the spec is getting close to done, it's not clear whether any browsers have plans to implement it yet, nor whether Emscripten et al have plans...
> Here's a failure case that segfaults instead of throwing a bounds error: ...does this mean the current state of `main` is dodgy? (I had just started a GitHub ->...
> They'll still show up in the IR, but they'll be compiled to no-ops. I ran into trouble when outright deleting them to do with the way the simplifier strips...
(briefly emerges from the depths...) IMHO option 2 is the better answer, upgrading Dawn isn't hard. (submerges once again, bloop)
> I'm not sure how disturbed I am by that. It seems less bad than being able to construct meaningless targets. It may be unreasonably hard to make it impossible...
> I don't add tests to simd_op_check_x86 Understood, but is there any other reasonable way to add checks for this? This sort of stuff is often pretty fragile.
Is this PR still an active one? It seems pretty stale.
What's the story, needs more work?