On Sun, 2007-04-22 at 13:45 -0700, Chris Lattner wrote: > > Index: llvm/include/llvm/AbstractTypeUser.h > > diff -u llvm/include/llvm/AbstractTypeUser.h:1.27 llvm/include/llvm/ > > AbstractTypeUser.h:1.28 > > --- llvm/include/llvm/AbstractTypeUser.h:1.27 Mon Jun 5 11:29:06 2006 > > +++ llvm/include/llvm/AbstractTypeUser.h Sun Apr 22 13:49:32 2007 > > @@ -14,6 +14,10 @@ > > #ifndef LLVM_ABSTRACT_TYPE_USER_H > > #define LLVM_ABSTRACT_TYPE_USER_H > > > > +// IMPORTANT: Do not include this file directly. Include Type.h > > instead. > > +// Some versions of GCC can't handle the inlined method > > PATypeHolder::dropRef() > > +// correctly otherwise. > > + > > Ugh. Is this really true in all cases? Should we add something like: > > #ifndef LLVM_TYPE_H > #error "dont' do this" > #endif > > to the top of this file?
If that's the case then ATU.h should just be folded into Type.h and we don't ever have to talk about this again :) Reid. > > -Chris > _______________________________________________ > llvm-commits mailing list > llvm-commits@cs.uiuc.edu > http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits _______________________________________________ llvm-commits mailing list llvm-commits@cs.uiuc.edu http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
