ARM: Add instruction VFPCompareAndSetFlags to macro assembler

src/arm/code-stubs-arm.cc

 // Copyright 2010 the V8 project authors. All rights reserved.
 // Redistribution and use in source and binary forms, with or without
 // modification, are permitted provided that the following conditions are
 // met:
 //
 //     * Redistributions of source code must retain the above copyright
 //       notice, this list of conditions and the following disclaimer.
 //     * Redistributions in binary form must reproduce the above
 //       copyright notice, this list of conditions and the following
 //       disclaimer in the documentation and/or other materials provided
@@ skipping 848 matching lines @@
              Operand(scratch1, LSL, kPointerSizeLog2 + 1));
 
       Register probe = mask;
       __ ldr(probe,
              FieldMemOperand(scratch1, FixedArray::kHeaderSize));
       __ BranchOnSmi(probe, not_found);
       __ sub(scratch2, object, Operand(kHeapObjectTag));
       __ vldr(d0, scratch2, HeapNumber::kValueOffset);
       __ sub(probe, probe, Operand(kHeapObjectTag));
       __ vldr(d1, probe, HeapNumber::kValueOffset);
-      __ vcmp(d0, d1);
-      __ vmrs(pc);
+      __ VFPCompareAndSetFlags(d0, d1);
       __ b(ne, not_found);  // The cache did not contain this value.
       __ b(&load_result_from_cache);
     } else {
       __ b(not_found);
     }
   }
 
   __ bind(&is_smi);
   Register scratch = scratch1;
   __ and_(scratch, mask, Operand(object, ASR, 1));
@@ skipping 87 matching lines @@
   EmitSmiNonsmiComparison(masm, lhs_, rhs_, &lhs_not_nan, &slow, strict_);
 
   __ bind(&both_loaded_as_doubles);
   // The arguments have been converted to doubles and stored in d6 and d7, if
   // VFP3 is supported, or in r0, r1, r2, and r3.
   if (CpuFeatures::IsSupported(VFP3)) {
     __ bind(&lhs_not_nan);
     CpuFeatures::Scope scope(VFP3);
     Label no_nan;
     // ARMv7 VFP3 instructions to implement double precision comparison.
-    __ vcmp(d7, d6);
-    __ vmrs(pc);  // Move vector status bits to normal status bits.
+    __ VFPCompareAndSetFlags(d7, d6, r0);

[Rodolph Perfetta, 2011/01/10 14:59:36]

No need to clear the flags, if the result is unordered the overflow flag will be
set (the vs condition).

[Søren Thygesen Gjesse, 2011/01/11 08:56:17]

OK, then I am a bit in the dark regarding the clearing of the IOC bit just
before calling vcmp in http://codereview.chromium.org/6142004/ - could you
please explain when to clear this bit before vcmp and when not to. Having
re-read the ARM ARM I actually don't see any need to clear it when using the
vcmpe instruction (which is what we do).

[Søren Thygesen Gjesse, 2011/01/11 12:41:52]

Done.

     Label nan;
     __ b(vs, &nan);
     __ mov(r0, Operand(EQUAL), LeaveCC, eq);
     __ mov(r0, Operand(LESS), LeaveCC, lt);
     __ mov(r0, Operand(GREATER), LeaveCC, gt);
     __ Ret();
 
     __ bind(&nan);
     // If one of the sides was a NaN then the v flag is set.  Load r0 with
     // whatever it takes to make the comparison fail, since comparisons with NaN
@@ skipping 99 matching lines @@
   __ b(eq, &false_result);
 
   // HeapNumber => false iff +0, -0, or NaN.
   __ ldr(scratch, FieldMemOperand(tos_, HeapObject::kMapOffset));
   __ LoadRoot(ip, Heap::kHeapNumberMapRootIndex);
   __ cmp(scratch, ip);
   __ b(&not_heap_number, ne);
 
   __ sub(ip, tos_, Operand(kHeapObjectTag));
   __ vldr(d1, ip, HeapNumber::kValueOffset);
-  __ vcmp(d1, 0.0);
-  __ vmrs(pc);
+  __ VFPCompareAndSetFlags(d1, 0.0, scratch);

[Rodolph Perfetta, 2011/01/10 14:59:36]

Ditto.

[Søren Thygesen Gjesse, 2011/01/11 12:41:52]

Done.

   // "tos_" is a register, and contains a non zero value by default.
   // Hence we only need to overwrite "tos_" with zero to return false for
   // FP_ZERO or FP_NAN cases. Otherwise, by default it returns true.
   __ mov(tos_, Operand(0, RelocInfo::NONE), LeaveCC, eq);  // for FP_ZERO
   __ mov(tos_, Operand(0, RelocInfo::NONE), LeaveCC, vs);  // for FP_NAN
   __ Ret();
 
   __ bind(&not_heap_number);
 
   // Check if the value is 'null'.
@@ skipping 3797 matching lines @@
   if (CpuFeatures::IsSupported(VFP3)) {
     CpuFeatures::Scope scope(VFP3);
 
     // Load left and right operand
     __ sub(r2, r1, Operand(kHeapObjectTag));
     __ vldr(d0, r2, HeapNumber::kValueOffset);
     __ sub(r2, r0, Operand(kHeapObjectTag));
     __ vldr(d1, r2, HeapNumber::kValueOffset);
 
     // Compare operands
-    __ vcmp(d0, d1);
-    __ vmrs(pc);  // Move vector status bits to normal status bits.
+    __ VFPCompareAndSetFlags(d0, d1, r2);

[Rodolph Perfetta, 2011/01/10 14:59:36]

Ditto.

[Søren Thygesen Gjesse, 2011/01/11 12:41:52]

Done.

 
     // Don't base result on status bits when a NaN is involved.
     __ b(vs, &unordered);
 
     // Return a result of -1, 0, or 1, based on status bits.
     __ mov(r0, Operand(EQUAL), LeaveCC, eq);
     __ mov(r0, Operand(LESS), LeaveCC, lt);
     __ mov(r0, Operand(GREATER), LeaveCC, gt);
     __ Ret();
 
@@ skipping 50 matching lines @@
   __ pop(r1);
   __ Jump(r2);
 }
 
 
 #undef __
 
 } }  // namespace v8::internal
 
 #endif  // V8_TARGET_ARCH_ARM