Remove dead code from sys::windows::c

[pitdicker]

I am not entirely sure I have got everything right, but if it compiles it is ok probably...
I tested it with msvc x86_64 and gnu.

Somehow a lot of EXCEPTION-* constants are dead code when running test, no idea why.
I have put #![cfg_attr(test, allow(dead_code))] at the top for this.

[rust-highfive]

r? @alexcrichton

(rust_highfive has picked a reviewer for you, use r? to override)

[alexcrichton]

@bors: r+ 0a6f35b

Quite the cleanup!

[bors]

⌛ Testing commit 0a6f35b with merge a2ee00f...

[bors]

💔 Test failed - auto-win-gnu-32-nopt-t

[pitdicker]

I can't really get cross-compiling set up, but I think this fixes it for 32-bit.
Should I somehow first test myself, or is it ok to try?

[alexcrichton]

@bors: r+ 94db864

If this bounces I've got a simple script to just mock compile the repo, which may be useful for ensuring that other platforms just compile

[pitdicker]

Thanks!

I have no idea what that script does...

[alexcrichton]

Ah if you tweak the first two lines to uncomment the right stage/triple and just run it inside of a checkout of the compiler it will "bootstrap" that target into a directory called out

[bors]

⌛ Testing commit 94db864 with merge b8e2d98...

[bors]

💔 Test failed - auto-win-gnu-32-nopt-t

[pitdicker]

mingw really does not like me... I have reinstalled msys and 64- and 32-bit mingw and made a fresh checkout and rebuild of rust. But cross-compiling seems out of reach for me. And I tried your script.
Is there any guide for setting this it up?

This pr should work now, but untested for 32-bit.

[alexcrichton]

I tested this out locally and got (for the MSVC targets)

src/libstd/sys/windows/c.rs:1090:5: 1093:58 warning: foreign function is never used: `RaiseException`, #[warn(dead_code)] on by default
src/libstd/sys/windows/c.rs:1090     pub fn RaiseException(dwExceptionCode: DWORD,
src/libstd/sys/windows/c.rs:1091                           dwExceptionFlags: DWORD,
src/libstd/sys/windows/c.rs:1092                           nNumberOfArguments: DWORD,
src/libstd/sys/windows/c.rs:1093                           lpArguments: *const ULONG_PTR);

Both the MinGW targets compiled with no warnings, however.

[pitdicker]

Thanks so much for testing this! A little cleanup is not worth all the effort this takes...

I don't understand why RaiseException is not used on msvc. Both msvc targets use seh.rs.
And seh.rs uses it, except for stage0.
But I thought dead code in stage0 only gives warnings, not errors?

[retep998]

@pitdicker Which is why the message says warning: foreign function is never used: RaiseException, #[warn(dead_code)] on by default, which is a warning and not an error.

[pitdicker]

@retep998 The problem is the later stages deny warnings. So I am not sure if compiling fails with this warning, and if so, why?

[alexcrichton]

@pitdicker oh good point! If that ends up being the only error then this is indeed probably good to go, I only compiled in stage0

@bors: r+ 98fa5ac

[pitdicker]

They say in Dutch when you don't succeed the first two times: third time is shipping rights :). Let's hope

[bors]

⌛ Testing commit 98fa5ac with merge 321a60b...

[alexcrichton]

@bors: retry force

Unfortunately this won't land until #31855 lands

[bors]

⌛ Testing commit 98fa5ac with merge df91cb9...

[bors]

☀️ Test successful - auto-linux-32-nopt-t, auto-linux-32-opt, auto-linux-64-debug-opt, auto-linux-64-nopt-t, auto-linux-64-opt, auto-linux-64-x-android-t, auto-linux-cross-opt, auto-linux-musl-64-opt, auto-mac-32-opt, auto-mac-64-nopt-t, auto-mac-64-opt, auto-mac-ios-opt, auto-win-gnu-32-nopt-t, auto-win-gnu-32-opt, auto-win-gnu-64-nopt-t, auto-win-gnu-64-opt, auto-win-msvc-32-opt, auto-win-msvc-64-opt