On Wed, Nov 12, 2008 at 5:07 PM, Daniel Dunbar <[EMAIL PROTECTED]> wrote: > Out of curiosity, was this motivated by a test case? Should we care > about monitoring performance of this?
It's just a hunch, really; I don't think we need to monitor performance of just this bit. I'd like to monitor performance of overload resolution as a whole at some point, however. - Doug > On Wed, Nov 12, 2008 at 1:52 PM, Douglas Gregor <[EMAIL PROTECTED]> wrote: >> Author: dgregor >> Date: Wed Nov 12 15:52:20 2008 >> New Revision: 59173 >> >> URL: http://llvm.org/viewvc/llvm-project?rev=59173&view=rev >> Log: >> Slightly improve type hash >> >> Modified: >> cfe/trunk/include/clang/AST/TypeOrdering.h >> >> Modified: cfe/trunk/include/clang/AST/TypeOrdering.h >> URL: >> http://llvm.org/viewvc/llvm-project/cfe/trunk/include/clang/AST/TypeOrdering.h?rev=59173&r1=59172&r2=59173&view=diff >> >> ============================================================================== >> --- cfe/trunk/include/clang/AST/TypeOrdering.h (original) >> +++ cfe/trunk/include/clang/AST/TypeOrdering.h Wed Nov 12 15:52:20 2008 >> @@ -42,7 +42,8 @@ >> } >> >> static unsigned getHashValue(clang::QualType Val) { >> - return (unsigned)Val.getAsOpaquePtr(); >> + return (unsigned)Val.getAsOpaquePtr() ^ >> + ((unsigned)Val.getAsOpaquePtr() >> 9); >> } >> >> static bool isEqual(clang::QualType LHS, clang::QualType RHS) { >> >> >> _______________________________________________ >> cfe-commits mailing list >> [email protected] >> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits >> > _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
