[libcxx-commits] [PATCH] D122479: [libcxx] Reuse TEXT macro from WIN32 for mapping strings to UNICODE

[Gustavo Henrique Nihei via Phabricator via libcxx-commits]

gustavonihei added a comment.

In D122479#3409111 <https://reviews.llvm.org/D122479#3409111>, @mstorsjo wrote:

> I think the reasoning/argument here is backwards. I wouldn't want to bring in the `TEXT()` macro here; what we do here is unrelated to the unicode switch for the Windows APIs.
>
> So the issue is that the `PS()` macro we use right now can clash with some header on some OS? If that's the case, I don't mind changing it to a slightly longer macro, e.g. `PATHSTR()` or something like that, but please keep it untangled from the Win32 `TEXT()`/`UNICODE` macros.

Yes, it is clashing with another one defined for Xtensa CPUs, commonly included in several OSes:
https://android.googlesource.com/kernel/common/+/fd43fe19b830d6cd0eba08a6c6a5f71a6bd9c1b0/include/asm-xtensa/xtensa/config-linux_be/specreg.h#71
https://github.com/espressif/esp-idf/blob/master/components/xtensa/include/xtensa/specreg.h#L96

I'll update the patch according to your suggestion.
Thanks for the feedback!


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D122479/new/

https://reviews.llvm.org/D122479