IA32: Small assembler tweak for divsd->mulsd sequence

Review URL: https://chromiumcodereview.appspot.com/13447003

Patch from Zheng Liu <zheng.z.liu@intel.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14225 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

src/ia32/lithium-codegen-ia32.cc

@@ -2018,6 +2018,9 @@ void LCodeGen::DoArithmeticD(LArithmeticD* instr) {
       break;
     case Token::DIV:
       __ divsd(left, right);
+      // Don't delete this mov. It may improve performance on some CPUs,
+      // when there is a mulsd depending on the result
+      __ movaps(left, left);
       break;
     case Token::MOD: {
       // Pass two doubles as arguments on the stack.

src/x64/lithium-codegen-x64.cc

@@ -1793,6 +1793,8 @@ void LCodeGen::DoArithmeticD(LArithmeticD* instr) {
       break;
     case Token::DIV:
       __ divsd(left, right);
+      // Don't delete this mov. It may improve performance on some CPUs,
+      // when there is a mulsd depending on the result
       __ movaps(left, left);
       break;
     case Token::MOD: