Implement fp to bits methods as intrinsics.

Rationale:
Better optimization, better performance.

Results on libcore benchmark:

Most gain is from moving the invariant call out of the loop
after we detect everything is a side-effect free intrinsic.
But generated code in general case is much cleaner too.

Before:
timeFloatToIntBits() in 181 ms.
timeFloatToRawIntBits() in 35 ms.
timeDoubleToLongBits() in 208 ms.
timeDoubleToRawLongBits() in 35 ms.

After:
timeFloatToIntBits() in 36 ms.
timeFloatToRawIntBits() in 35 ms.
timeDoubleToLongBits() in 35 ms.
timeDoubleToRawLongBits() in 34 ms.

bug=11548336

Change-Id: I6e001bd3708e800bd75a82b8950fb3a0fc01766e

1 files changed, 4 insertions, 0 deletions

diff --git a/compiler/optimizing/intrinsics.cc b/compiler/optimizing/intrinsics.cc
index 316e86b4c9..3ed0278871 100644
--- a/compiler/optimizing/intrinsics.cc
+++ b/compiler/optimizing/intrinsics.cc
@@ -130,6 +130,10 @@ static Intrinsics GetIntrinsic(InlineMethod method) {
     case kIntrinsicFloatCvt:
       return ((method.d.data & kIntrinsicFlagToFloatingPoint) == 0) ?
           Intrinsics::kFloatFloatToRawIntBits : Intrinsics::kFloatIntBitsToFloat;
+    case kIntrinsicFloat2Int:
+      return Intrinsics::kFloatFloatToIntBits;
+    case kIntrinsicDouble2Long:
+      return Intrinsics::kDoubleDoubleToLongBits;
 
     // Floating-point tests.
     case kIntrinsicFloatIsInfinite: