Schrijvers Luc

@brad0 are we infected by any of those regressions mentioned at: https://discourse.llvm.org/t/18-1-8-released/79725#release-notes-2 ? EDIT: related commits upstream https://github.com/llvm/llvm-project/commit/3b5b5c1ec4a3095ab096dd780e84d7ab81f3d7ff https://github.com/llvm/llvm-project/commit/72c9425a79fda8e9001fcde091e8703f9fb2a43a

Luckaly no changes needed for an update, build completed on 64bit just now. If things are fine on 32bit maybe we coculd update? pokes @korli @davidkaroly :)

+1 from me, imho only README/COPYRIGHT, manpages, infopages and the likes should be provided in the base packages.

Thought ... how about moving the documentations for packages to a seperate _doc package instead? At repology there are other examples on how other OS's do this, shouldn't be too...

OK, will start it from there. :)

Doesn't seem to change much? ``` Found ninja-1.13.0 at /bin/ninja + /bin/ninja -j16 [124/1481] Compiling Fortran object scipy/sparse/linalg/_propack/liblib__spropack.a.p/PROPACK_single_printstat.F.o f951: Warning: command-line option '-Wno-multichar' is valid for C/C++/ObjC/ObjC++ but not for...

> + %{!,f95:%{!,f77:%{!Wmultichar: -Wno-multichar}}} %(cc1_cpu) %{profile:-p}" Still no go (checking with scipy): ``` [171/1481] Compiling Fortran object scipy/sparse/linalg/_propack/liblib__zpropack.a.p/PROPACK_complex16_dbsvd.F.o f951: Warning: command-line option '-Wno-multichar' is valid for C/C++/ObjC/ObjC++ but not for...

> Theorically this would mean to apply for c c++ c-* c++-* objective-c* objective-c++*: > > ``` > + %{,c|,c++|,c-*|,-c++-*|,objective-c*|,objective-c++*:%{!Wmultichar: -Wno-multichar}} %(cc1_cpu) %{profile:-p}" > ``` This probably isn't how it...

Checked @pulkomandy 's approach, still lot's of warnings. diff should be good: ```diff diff --git a/gcc/config/haiku.h b/gcc/config/haiku.h index ff0b29a..6fd7daf 100644 --- a/gcc/config/haiku.h +++ b/gcc/config/haiku.h @@ -42,16 +42,15 @@ Boston, MA...

I guess most of us are tackling other issues (and maybe forgot here), for myself, I've given up on rebuilding gcc for the x'th amount of times, doesn't mean it's...