Reintroduce the InlineHint function attribute.

This time it's for real! I am going to hook this up in the frontends as well.

The inliner has some experimental heuristics for dealing with the inline hint.
When given a -respect-inlinehint option, functions marked with the inline
keyword are given a threshold just above the default for -O3.

We need some experiments to determine if that is the right thing to do.

llvm-svn: 95466

1 files changed, 1 insertions, 0 deletions

diff --git a/llvm/lib/AsmParser/LLParser.cpp b/llvm/lib/AsmParser/LLParser.cpp
index ed743016245..5dd65691a3d 100644
--- a/llvm/lib/AsmParser/LLParser.cpp
+++ b/llvm/lib/AsmParser/LLParser.cpp
@@ -947,6 +947,7 @@ bool LLParser::ParseOptionalAttrs(unsigned &Attrs, unsigned AttrKind) {
     case lltok::kw_noinline:        Attrs |= Attribute::NoInline; break;
     case lltok::kw_readnone:        Attrs |= Attribute::ReadNone; break;
     case lltok::kw_readonly:        Attrs |= Attribute::ReadOnly; break;
+    case lltok::kw_inlinehint:      Attrs |= Attribute::InlineHint; break;
     case lltok::kw_alwaysinline:    Attrs |= Attribute::AlwaysInline; break;
     case lltok::kw_optsize:         Attrs |= Attribute::OptimizeForSize; break;
     case lltok::kw_ssp:             Attrs |= Attribute::StackProtect; break;