diff options
| author | 2015-03-27 13:44:44 -0700 | |
|---|---|---|
| committer | 2015-03-27 14:56:51 -0700 | |
| commit | 3e90a96f403cbc353731e6687fe12a088f996cee (patch) | |
| tree | 544800c12984b975ae45658093a04f5662aebfd4 /compiler/optimizing/intrinsics_arm.cc | |
| parent | d68cef75a2b57b5c5ab33065d567446b4510b143 (diff) | |
[optimizing] Do not inline intrinsics
The intrinsics generally have specialized code and the code for them
may be faster than what can be achieved with inlining. Thus inliner
should skip intrinsics.
At the same time, easy methods are not worth intrinsifying: ie String
length and isEmpty. Those can be handled by inliner with no problem
and can actually lead to better code since call is not kept around
through all of the optimizations.
Change-Id: Iab38e6c33f79efa54d845d4871cf26fa9b235ab0
Signed-off-by: Razvan A Lupusoru <razvan.a.lupusoru@intel.com>
Diffstat (limited to 'compiler/optimizing/intrinsics_arm.cc')
| -rw-r--r-- | compiler/optimizing/intrinsics_arm.cc | 2 |
1 files changed, 0 insertions, 2 deletions
diff --git a/compiler/optimizing/intrinsics_arm.cc b/compiler/optimizing/intrinsics_arm.cc index 3fb6a7d2d0..31b4e64e1b 100644 --- a/compiler/optimizing/intrinsics_arm.cc +++ b/compiler/optimizing/intrinsics_arm.cc @@ -903,8 +903,6 @@ UNIMPLEMENTED_INTRINSIC(MathRoundDouble) // Could be done by changing rounding UNIMPLEMENTED_INTRINSIC(MathRoundFloat) // Could be done by changing rounding mode, maybe? UNIMPLEMENTED_INTRINSIC(UnsafeCASLong) // High register pressure. UNIMPLEMENTED_INTRINSIC(SystemArrayCopyChar) -UNIMPLEMENTED_INTRINSIC(StringIsEmpty) // Might not want to do these two anyways, inlining should -UNIMPLEMENTED_INTRINSIC(StringLength) // be good enough here. UNIMPLEMENTED_INTRINSIC(StringIndexOf) UNIMPLEMENTED_INTRINSIC(StringIndexOfAfter) UNIMPLEMENTED_INTRINSIC(ReferenceGetReferent) |