Martin Kinkelin

Yep, this seems like a Wine-specific (?) dub issue; this definitely works on native Windows (with `ldc2.exe` being the default driver, not `ldmd2.exe`). **Edit**: I.e. it's somehow using the `DMDCompiler`...

@thewilsonator: Do you want to land any of the open dcompute PRs for v1.35 final? I'd like to release this weekend.

2 remaining failures: * `std.internal.math.gammafunction` unittests, with enabled optimizations only * lit-test `driver/config_diag.d` These work for Cirrus CI, on macOS 12 (not 14, and surely a different Xcode version too).

The gammafunction module consistently fails on the new M1 GHA runners for the vanilla-LLVM jobs too, using vanilla LLVM 16 & 17. The `config_diag.d` lit-test works there though (different LLVM,...

@JohanEngelen: So wrt. gammafunction, I'd expect you to see it too, with the regular Phobos unittest runner. - Wrt. the thrown exception for the extra .conf, I'm wondering how the...

[Draft because of random `Pure virtual function called!` errors (at compiler runtime) in first experiments in #4604, very roughly 0-5 per CI run.] The previous issues are resolved by now.

I've hit exactly the same `-fsanitize=leak` problem at work - on Debian 12 only; Debian 11 and Ubuntu 20/22 all worked. With exactly the same binary LDC package (FWIW, built...

Nope, there are still a few test errors to iron out.

Hmm, I don't know. I just hate that our CMake build doesn't work out of the box with recent Xcode, requiring that dead-ugly `-DLDC_LINK_MANUALLY=OFF -DD_COMPILER_FLAGS=-gcc=/usr/bin/c++` workaround.

Thx for the nice test case. This ultimately boils down to `__ctfe` being a runtime-constant, so the `.idup` call still ends up being codegen'd at the IR level, and then...