LIU Hao
@oltolm Someone expressed negative thoughts in mingw-w64 so I am afraid there is currently no plan on it. A proper fix for it requires hacking the CRT to invoke thread-local...
> gcc14 win32 thread model seems to support c++11 thread. gcc14 will be released soon, so maybe consider use win32 thread model instead. I don't think it will solve the...
> How does Clang solve this problem? It has nothing to do with Clang; the magic lies in UCRT. We have https://github.com/mingw-w64/mingw-w64/blob/dbfdf802580c0d6b90b91995dab019f2a7787a8e/mingw-w64-crt/crt/tls_atexit.c#L108-L119 which causes UCRT to invoke destructors of thread-local...
> So the advantage of MCF over MSVCRT is correctness and performance and the advantage of MCF over UCRT is performance? Is that correct? Yes. > Why not create a...
> I know about it, it's great, but as I said I can't use it without pacman. pacman is too convenient. How about this? https://github.com/lhmouse/MINGW-packages One reason that I do...
@oltolm pacman packages can be found here: https://files.lhmouse.com/pkg-snapshots/ . Source scripts are in https://github.com/lhmouse/MINGW-packages . Installing these packages locally with `pacman -U` will produce environments for building everything else.
Suggestion: 1. The installer can create the `c:\msys64` directory before installing anything, then copy the ACL from `c:\program files`, so most installed files will be protected from non-priviledged processes. 2....
And Intel syntax shall be the default. That's the only standard and documented syntax.
> Closing, as setting the format seems to work (already fixed by #63 and related changes in the backend & docker projects). The default may be a good idea, too,...
Is there any progress on this issue? This has always been disrupting and annoying for me, as it always duplicates opening parentheses unlike any other editor.