MIPS: Add float instructions and test coverage, part one

src/mips/constants-mips.h

 // Copyright 2012 the V8 project authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 #ifndef  V8_MIPS_CONSTANTS_H_
 #define  V8_MIPS_CONSTANTS_H_
 #include "src/globals.h"
 // UNIMPLEMENTED_ macro for MIPS.
 #ifdef DEBUG
 #define UNIMPLEMENTED_MIPS()                                                  \
@@ skipping 459 matching lines @@
   CVT_L_S = ((4 << 3) + 5),
   CVT_PS_S = ((4 << 3) + 6),
   // COP1 Encoding of Function Field When rs=D.
   ADD_D = ((0 << 3) + 0),
   SUB_D = ((0 << 3) + 1),
   MUL_D = ((0 << 3) + 2),
   DIV_D = ((0 << 3) + 3),
   SQRT_D = ((0 << 3) + 4),
   ABS_D = ((0 << 3) + 5),
   MOV_D = ((0 << 3) + 6),
   NEG_D = ((0 << 3) + 7),

[paul.l..., 2015/05/09 01:07:47]

We have inconsistency in naming with many of these FP ops, which apparently were
named when we only intended to support doubles. These are defined as MOV.fmt &
NEG.fmt in the manual, and can be used for single or double, by specifying the
format field.

We do not have to clean up the entire world in this CL, but we should note
inconsistency, and make a decision if these should be cleaned in a future CL.

[Djordje.Pesic, 2015/05/14 13:22:30]

Done.

   ROUND_L_D = ((1 << 3) + 0),
   TRUNC_L_D = ((1 << 3) + 1),
   CEIL_L_D = ((1 << 3) + 2),
   FLOOR_L_D = ((1 << 3) + 3),
   ROUND_W_D = ((1 << 3) + 4),
   TRUNC_W_D = ((1 << 3) + 5),
   CEIL_W_D = ((1 << 3) + 6),
   FLOOR_W_D = ((1 << 3) + 7),
+  RECIP = ((2 << 3) + 5),
+  RSQRT_D = ((2 << 3) + 6),

[paul.l..., 2015/05/09 01:07:47]

These two should be named consistently. Since they are both *.fmt, you should
remove the _D from RSQRT.

Should we even implement instructions for these? The manual says both are
approximations: "The numeric accuracy of this operation is implementation
dependent; it does not meet the accuracy specified by the IEEE 754 Floating
Point standard. The computed result differs from both the exact result and the
IEEE-mandated representation of the exact result by no more than two units in
the least-significant place (ULP)."

If we have a need now, put them in. If we do not have a need that can tolerate
the inaccuracy, we might want to skip these. If we do not skip, we should add
comments to the assembler files where they are defined, so someone does not use
them in a way that requires accuracy.

[dusmil.imgtec, 2015/05/11 13:13:24]

For the sake of simulator completeness, let's leave them in code.
Their names already mean approximation, so if programmers used them they used to
know their behavior.

[Djordje.Pesic, 2015/05/14 13:22:30]

Done.

[Djordje.Pesic, 2015/05/14 13:22:30]

Done.

   MIN = ((3 << 3) + 4),
   MINA = ((3 << 3) + 5),
   MAX = ((3 << 3) + 6),
   MAXA = ((3 << 3) + 7),
   CVT_S_D = ((4 << 3) + 0),
   CVT_W_D = ((4 << 3) + 4),
   CVT_L_D = ((4 << 3) + 5),
   C_F_D = ((6 << 3) + 0),
   C_UN_D = ((6 << 3) + 1),
   C_EQ_D = ((6 << 3) + 2),
@@ skipping 38 matching lines @@
   CMP_SAT = ((3 << 3) + 0),  // Reserved, not implemented.
   CMP_SOR = ((3 << 3) + 1),
   CMP_SUNE = ((3 << 3) + 2),
   CMP_SNE = ((3 << 3) + 3),
   CMP_SUGE = ((3 << 3) + 4),  // Reserved, not implemented.
   CMP_SOGE = ((3 << 3) + 5),  // Reserved, not implemented.
   CMP_SUGT = ((3 << 3) + 6),  // Reserved, not implemented.
   CMP_SOGT = ((3 << 3) + 7),  // Reserved, not implemented.
 
   SEL = ((2 << 3) + 0),
+  MOVZ_C = ((2 << 3) + 2),
+  MOVN_C = ((2 << 3) + 3),

[paul.l..., 2015/05/09 01:07:47]

Other FP instructions that have _C suffix (to differentiate from integer
version), have an explicit comment:
 // COP1 on FPR registers. 
Please add that here.

[Djordje.Pesic, 2015/05/14 13:22:30]

Done.

   SELEQZ_C = ((2 << 3) + 4),  // COP1 on FPR registers.
+  MOVF = ((2 << 3) + 1),      // Function field for MOVT.fmt and MOVF.fmt
   SELNEZ_C = ((2 << 3) + 7),  // COP1 on FPR registers.
   // COP1 Encoding of Function Field When rs=PS.
   // COP1X Encoding of Function Field.
   MADD_D = ((4 << 3) + 1),
 
   NULLSF = 0
 };
 
 
 // ----- Emulated conditions.
@@ skipping 408 matching lines @@
 // JS argument slots size.
 const int kJSArgsSlotsSize = 0 * Instruction::kInstrSize;
 // Assembly builtins argument slots size.
 const int kBArgsSlotsSize = 0 * Instruction::kInstrSize;
 
 const int kBranchReturnOffset = 2 * Instruction::kInstrSize;
 
 } }   // namespace v8::internal
 
 #endif    // #ifndef V8_MIPS_CONSTANTS_H_