On 2026/09/03 02:40, Pali Rohár wrote: > Well, this is not only about mingw32, but also about cygwin. > > And if the gcc still support mingw32 (or at least present that it > support it) then it is not a good idea to explicitly break it just by > an argument "I doubt that somebody is using it". I can understand that > gcc people do not want to support it anymore and will remove it. That is > fine, I have no reason to oppose it, but it should be official and > officially removed. I do not like the fact that some "random mingw-w64" > change for gcc will break something non-mingw-w64 (even if it has small > or zero usage). And still need to think about cygwin.
I agree it would be better to split non-mingw-w64 part into a sperated PR. I'll update my patch later to exclude changes to mingw32.h. Regarding cygwin, either one should doesn't matter, as cygwin doesn't provide all these related libraries (libcoldname, libmoldname, libmingwthrd). BTW, a sad news is that it looks like GCC people have little interest of Windows support (all Cygwin, MinGW, and MinGW-w64). Most of my patches to GCC regarding mingw-w64 just got a '*-*-mingw' tag, then nothing happens, even if I intermittently pinged for about every 2 weeks. _______________________________________________ Mingw-w64-public mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/mingw-w64-public
