Dan Gohman

Ideally, clang should also have an option to emit code which checks for overflow when incrementing the stack. I started some work in this direction in https://reviews.llvm.org/D70515, however I've not...

As @sbc100 mentioned, you can pass `-Wl,--stack-first` to clang. Beyond that, what's needed is stack overflow checking in LLVM's wasm backend. https://reviews.llvm.org/D70515 is a possible starting point, if anyone wants...

It means that overflowing the stack isn't likely to clobber global variables without trapping, because most likely it'll wrap around to 4GiB which most often won't be allocated, so accessing...

We don't need new ad-hoc agreements between runtimes and applications; we just need to improve our tools.

I agree that it would be good to document more things. I've now submitted https://github.com/bytecodealliance/wasmtime/pull/2703 to document `-Wl,--stack-first`.

> Each syscall should ideally have its own entry point. Contrast this with one big syscall entry point where the host implementation must switch on a syscall number. Let's be...

See also https://github.com/WebAssembly/wasi-libc/issues/304 which is a similar issue. `__WASI_RIFLAGS_RECV_PEEK` was anticipating adding sockets support to WASI using the witx IDL. The standards world that the witx IDL was meant to...

Even though `mkstemp` etc. don't hard-code a temporary directory, common usage of them does hard-code a path, often `/tmp`. If we provide this feature, programs using may implicitly depend on...

I'm looking to balance the desires of people using WASI today, with the desire to avoid building up a compatibility burden that may delay or even prevent the really cool...

It is the application's job to clean up after itself, but if applications actually did this job, tmp cleaner daemons / cron jobs wouldn't be a thing, and they are...