[llvm-commits] [llvm] r41747 - in /llvm/trunk: include/llvm/ lib/Analysis/ lib/AsmParser/ lib/Bitcode/Reader/ lib/Bitcode/Writer/ lib/CodeGen/ lib/CodeGen/SelectionDAG/ lib/ExecutionEngine/ lib/ExecutionEngine/JIT/ lib/Target/CBackend/ lib/Target/MSIL/ lib/Target/X86/ lib/Transforms/IPO/ lib/Transforms/Scalar/ lib/VMCore/ tools/llvm-upgrade/ tools/llvm2cpp/

[Dale Johannesen]

>
>> One other thing.  Right now there is this code in the X86  
>> backend:  X86DAGToDAGISel::EmitSpecialCodeForMain
>>
>> Both of them should just be zapped.  It was a poor band-aid to  
>> support 64-bit doubles with fewer precision problems, but  
>> obviously won't work when we support long double!  In any case,  
>> they should just be nuked now.

This appears to be the only overload of EmitFunctionEntryCode, would  
you like to get rid of that too?