Specializing x86 range argument copying
The ARM implementation of range argument copying was specialized in some cases.
For all other architectures, it would fall back to generating memcpy. This patch
updates the x86 implementation so it does not call memcpy and instead generates
loads and stores, favoring movement of 128-bit chunks.
Change-Id: Ic891e5609a4b0e81a47c29cc5a9b301bd10a1933
Signed-off-by: Razvan A Lupusoru <razvan.a.lupusoru@intel.com>
diff --git a/compiler/dex/quick/arm/utility_arm.cc b/compiler/dex/quick/arm/utility_arm.cc
index 07fc6c7..9d3968b 100644
--- a/compiler/dex/quick/arm/utility_arm.cc
+++ b/compiler/dex/quick/arm/utility_arm.cc
@@ -367,6 +367,16 @@
return OpRegRegShift(op, r_dest_src1, r_src2, 0);
}
+LIR* ArmMir2Lir::OpMovRegMem(int r_dest, int r_base, int offset, MoveType move_type) {
+ UNIMPLEMENTED(FATAL);
+ return nullptr;
+}
+
+LIR* ArmMir2Lir::OpMovMemReg(int r_base, int offset, int r_src, MoveType move_type) {
+ UNIMPLEMENTED(FATAL);
+ return nullptr;
+}
+
LIR* ArmMir2Lir::OpCondRegReg(OpKind op, ConditionCode cc, int r_dest, int r_src) {
LOG(FATAL) << "Unexpected use of OpCondRegReg for Arm";
return NULL;