Stream: git-wasmtime

Topic: wasmtime / Issue #1535 Support MinGW as a wasmtime target


view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 16:53):

alexcrichton opened Issue #1535:

It looks like Go at least only supports MinGW as a toolchain to link with, but wasmtiem does not currently have MinGW support. If we'd like to support Windows support for Wasmtime in Go I think we'll need to add a MinGW target as well as CI/releases for it.

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 16:53):

alexcrichton labeled Issue #1535:

It looks like Go at least only supports MinGW as a toolchain to link with, but wasmtiem does not currently have MinGW support. If we'd like to support Windows support for Wasmtime in Go I think we'll need to add a MinGW target as well as CI/releases for it.

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 17:31):

tschneidereit commented on Issue #1535:

Do you know how complicated that would be initially—and how much of a maintenance burden it'd be on an ongoing basis? If neither is a concern, this makes sense to me. Otherwise, I think we should let the be driven by use cases.

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 17:36):

tschneidereit commented on Issue #1535:

Ah, I now see the wasmtime-go issue. I guess the question is how many people will want to run Go programs on Windows, which I don't have a good handle on.

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 17:46):

alexcrichton commented on Issue #1535:

Oh sorry I should have clarified, but I definitely agree that we should let use cases and difficulty of implementation guide this in terms of priorities. I don't think it's clear how motivated this is right now (I don't know how often it'd be wanted on Go for Windows). I suspect this would be relatively difficult because of stack unwinding, but I might give this a go later today and see if it "trivially works" or "trivially doesn't work". Basically I think it's probably worth a little bit of effort to see how much effort it would take for us to support this.

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 17:55):

tschneidereit commented on Issue #1535:

That makes a lot of sense to me, yes. Thank you for the clarification!

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 20:31):

alexcrichton commented on Issue #1535:

Turns out that with https://github.com/bytecodealliance/wasmtime/pull/1490 all tests pass. I think there's probably some difference in how big the stack is when a segfault comes in which causes issues today, but once https://github.com/bytecodealliance/wasmtime/pull/1490 lands I can look into adding this to CI. Don't think it'll be too too hard!

view this post on Zulip Wasmtime GitHub notifications bot (Apr 24 2020 at 22:45):

fitzgen closed Issue #1535:

It looks like Go at least only supports MinGW as a toolchain to link with, but wasmtiem does not currently have MinGW support. If we'd like to support Windows support for Wasmtime in Go I think we'll need to add a MinGW target as well as CI/releases for it.


Last updated: Oct 23 2024 at 20:03 UTC