mstorsjo wrote:
> Hm, this seems to affect some feature detection (and constraints) later on,
> though I don't see why/how removing the specification of `--unwindlib=none`
> to the driver (leaving it only to the linker) would cause that:
>
> https://github.com/llvm/llvm-project/blob/1c046ca3f3
h-vetinari wrote:
I think the riddle may be explained by @mstorsjo's excellent commit message
from
https://github.com/llvm/llvm-project/commit/4169b5251f58cc0eb2ac6c3b4b9990a51728aac6
(which I just found):
> If the CMake requirement is bumped to 3.14, we could use
`CMAKE_REQUIRED_LINK_OPTIONS`
h-vetinari wrote:
Found some more CMake cruft... Unfortunately, it doesn't improve the situation
with libunwind, but at least that means I should be able to break this out into
a separate PR.
https://github.com/llvm/llvm-project/pull/93429
___
cfe-co
https://github.com/h-vetinari updated
https://github.com/llvm/llvm-project/pull/93429
>From c8a0b127ade4330151951144f6d4841beaf66d23 Mon Sep 17 00:00:00 2001
From: "H. Vetinari"
Date: Sun, 26 May 2024 13:21:46 +1100
Subject: [PATCH 1/3] [runtimes] remove workaround for old CMake when setting
`
https://github.com/h-vetinari updated
https://github.com/llvm/llvm-project/pull/93429
>From c8a0b127ade4330151951144f6d4841beaf66d23 Mon Sep 17 00:00:00 2001
From: "H. Vetinari"
Date: Sun, 26 May 2024 13:21:46 +1100
Subject: [PATCH 1/3] [runtimes] remove workaround for old CMake when setting
`
