rdar://12100355 (part 1)

This revision attempts to recognize following population-count pattern:

 while(a) { c++; ... ; a &= a - 1; ... },
  where <c> and <a>could be used multiple times in the loop body.

 TODO: On X8664 and ARM, __buildin_ctpop() are not expanded to a efficent 
instruction sequence, which need to be improved in the following commits.

Reviewed by Nadav, really appreciate!


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@168931 91177308-0d34-0410-b5e6-96231b3b80d8
diff --git a/lib/Target/X86/X86ISelLowering.cpp b/lib/Target/X86/X86ISelLowering.cpp
index d1552cd..d8e7263 100644
--- a/lib/Target/X86/X86ISelLowering.cpp
+++ b/lib/Target/X86/X86ISelLowering.cpp
@@ -17670,6 +17670,17 @@
   return -1;
 }
 
+ScalarTargetTransformInfo::PopcntHwSupport
+X86ScalarTargetTransformImpl::getPopcntHwSupport(unsigned TyWidth) const {
+  assert(isPowerOf2_32(TyWidth) && "Ty width must be power of 2");
+  const X86Subtarget &ST = TLI->getTargetMachine().getSubtarget<X86Subtarget>();
+
+  // TODO: Currently the __builtin_popcount() implementation using SSE3
+  //   instructions is inefficient. Once the problem is fixed, we should
+  //   call ST.hasSSE3() instead of ST.hasSSE4().
+  return ST.hasSSE41() ? Fast : None;
+}
+
 unsigned
 X86VectorTargetTransformInfo::getArithmeticInstrCost(unsigned Opcode,
                                                      Type *Ty) const {