Stream: git-wasmtime

Topic: wasmtime / Issue #1231 fix yanix compile with armv7


view this post on Zulip Wasmtime GitHub notifications bot (Mar 05 2020 at 15:47):

alexcrichton commented on Issue #1231:

Could the .into() and the : i64 annotations both be removed? It doesn't seem like there should be any reason to explicitly require i64 when there's a try_into to u64 afterwards which should handle any type conversion.

Additionally, if the goal is to compile yanix on a few platforms, it seems like we should consider adding CI to verify that yanix compiles correctly for other platforms.

view this post on Zulip Wasmtime GitHub notifications bot (Mar 06 2020 at 14:35):

kubkon commented on Issue #1231:

I think @alexcrichton is right that the .into() here is unnecessary to begin with, and simply removal of i64 annotation should fix the issue.

@alexcrichton about verifying builds on additional platforms, I totally agree! We should definitely start putting out some sort of coverage and not only for yanix but for wasi-common and perhaps wasmtime as a whole. Do you have some ideas for this BTW? The only thing that I can think that should work in Actions would be downloading the target and cross-compiling to it. This however, will generate a handful of linker errors, unless you know of a way to stop at the compilation stage?

view this post on Zulip Wasmtime GitHub notifications bot (Mar 06 2020 at 19:18):

alexcrichton commented on Issue #1231:

I think it should be easy enough in CI to do something like:

$ rustup target add armv7-unknown-linux-gnueabihf
$ cargo check -p wasi-common --target armv7-unknown-linux-gnueabihf

and that should be good enough for ensuring things keep compiling.

view this post on Zulip Wasmtime GitHub notifications bot (Mar 08 2020 at 14:11):

stefson commented on Issue #1231:

the E0308 type of error has been fixed in the meantime, so I'm going to close this and ask for some help with the other error.


Last updated: Dec 23 2024 at 12:05 UTC