Thank you!

LGTM

It's not obvious to me how anything could depend on cxx-headers target (not the component) since that target did not exist until r334468 which has landed only recently (just before r334477). The motivation behind this change was just consistency since all other targets in libc++ use underscores rather than dashes, so reverting that change is fine with me if it's breaking you.

LGTM

LGTM

In D45604#1141891, @Hahnfeld wrote:

Should ENABLE_PER_TARGET_RUNTIME_DIR be declared at LLVM's top level CMakeLists.txt? If I read the patch correctly, it's in clang so llvm/runtimes/ relies on the default value being OFF (hard to tell because the patch doesn't follow the current layout of the SVN repository but rather a git monorepo?)

Ping?

Personally, I'd prefer to spell out all the headers explicitly rather than using globing which is more error-prone to future changes. I also agree with @kubamracek's comment about using crt (which may be also mistaken for crt*.o) to compiler_rt.

In D47169#1141824, @Hahnfeld wrote:

The messages seem to be on llvm-commits (http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20180528/557038.html and http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20180611/thread.html#559742), but in any case you should have received a direct notification from Phabricator. Maybe you might want to check your filters?

In D47169#1134764, @Hahnfeld wrote:

Due to the lack of replies, I've reverted this change in rL334903 to unbreak my build. Feel free to submit a patch that doesn't regress the build system, thanks.