On Sunday 18 May 2025 14:57:12 Martin Storsjö wrote: > On Sun, 18 May 2025, Pali Rohár wrote: > > > On Friday 09 May 2025 18:07:22 Pali Rohár wrote: > > > On Tuesday 06 May 2025 19:59:31 Martin Storsjö wrote: > > > > On Thu, 17 Apr 2025, Pali Rohár wrote: > > > > > > > > > Function __main() is function for executing gcc c++ constructors. > > > > > Move it > > > > > into guarded code which prevents duplicate code execution, after > > > > > executing > > > > > msvc c++ constructors (__xc_a-__xc_z callbacks). > > > > > --- > > > > > mingw-w64-crt/crt/crtexe.c | 2 +- > > > > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > > > > > > > diff --git a/mingw-w64-crt/crt/crtexe.c b/mingw-w64-crt/crt/crtexe.c > > > > > index 620f08b9467d..46e61fcecb94 100644 > > > > > --- a/mingw-w64-crt/crt/crtexe.c > > > > > +++ b/mingw-w64-crt/crt/crtexe.c > > > > > @@ -208,6 +208,7 @@ __tmainCRTStartup (void) > > > > > _amsg_exit (8); /* _RT_SPACEARG */ > > > > > > > > > > _initterm (__xc_a, __xc_z); > > > > > + __main (); /* C++ initialization. */ > > > > > > > > > > __native_startup_state = __initialized; > > > > > } > > > > > @@ -229,7 +230,6 @@ __tmainCRTStartup (void) > > > > > _fpreset (); > > > > > > > > > > duplicate_ppstrings (argc, &argv); > > > > > - __main (); /* C++ initialization. */ > > > > > #ifdef _UNICODE > > > > > __winitenv = envp; > > > > > #else > > > > > -- > > > > > > > > This makes the constructors be called before _pei386_runtime_relocator - > > > > this will quite certainly break things in some setup. Yes I know a later > > > > patch fixes the order again, but we shouldn't be moving things in such > > > > an > > > > order. > > > > > > > > // Martin > > > > > > Ou, I did not notice this. This needs to be reworked, so the code is > > > always called in the correct order. > > > > Is this right now the only issue needed resolve? > > I think so (I don't offhand remember if I commented something else - I think > the other discussion threads have been resolved by just discussing the > mattera). > > // Martin
Ok, in this case just move "[PATCH 6/9]" before this one ([PATCH 3/9]). When I run git rebase -i it applied cleanly without any conflict. So I think that it is not needed to resend whole patch series. _______________________________________________ Mingw-w64-public mailing list Mingw-w64-public@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mingw-w64-public
