[llvm-commits] [llvm] r150060 - in /llvm/trunk: include/llvm/IntrinsicsX86.td lib/Target/X86/X86ISelLowering.cpp lib/Target/X86/X86InstrSSE.td

[David A. Greene]

Duncan Sands <baldrick at free.fr> writes:

>> Not everthing is Clang.  Lots of people use GCC builtins and expect that
>> code to be honored.  We can't do that without intrinsics.
>
> I think you are swimming against the tide here.  

It's not my desire.  It's a customer requirement.

Really, what's wrong with keeping intrinsics around if people find them
necessary and useful?  It doesn't mean clang has to use them.

We need this ability.  I'm telling you, we do.

> Remember when C programmers used to complain because the compiler
> didn't just directly translate their C into the obvious assembly code
> any more...

None of this historical essay matters.  We have customers who demand
that intrinsics (intrinsics, not general C code), remain, well,
intrinsics.

>> If anything, I would like to see *more* x86 intrinsics.  In fact we have
>> a bunch here that I was planning to send up.
>
> If they can't be expressed simply as generic IR, then an intrinsic is
> appropriate so sure go ahead!

I'm talking about cases where it CAN be expressed with generic IR.

Apple may not have these needs, but others do.

                               -Dave