cmd/compile: give mparith{2,3}.go files more meaningful names

Also, relocate related const and type definitions from go.go.

Change-Id: Ieb9b672da8dd510ca67022b4f7ae49a778a56579
Reviewed-on: https://go-review.googlesource.com/20080
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Dave Cheney <dave@cheney.net>
Reviewed-by: Keith Randall <khr@golang.org>

src/cmd/compile/internal/gc/go.go

@@ -6,7 +6,6 @@ package gc
 
 import (
 	"bytes"
-	"cmd/compile/internal/big"
 	"cmd/internal/obj"
 )
 
@@ -17,32 +16,6 @@ const (
 	MaxStackVarSize = 10 * 1024 * 1024
 )
 
-const (
-	// Maximum size in bits for Mpints before signalling
-	// overflow and also mantissa precision for Mpflts.
-	Mpprec = 512
-	// Turn on for constant arithmetic debugging output.
-	Mpdebug = false
-)
-
-// Mpint represents an integer constant.
-type Mpint struct {
-	Val  big.Int
-	Ovf  bool // set if Val overflowed compiler limit (sticky)
-	Rune bool // set if syntax indicates default type rune
-}
-
-// Mpflt represents a floating-point constant.
-type Mpflt struct {
-	Val big.Float
-}
-
-// Mpcplx represents a complex constant.
-type Mpcplx struct {
-	Real Mpflt
-	Imag Mpflt
-}
-
 type Val struct {
 	// U contains one of:
 	// bool     bool when n.ValCtype() == CTBOOL

src/cmd/compile/internal/gc/mpfloat.go

@@ -13,6 +13,25 @@ import (
 
 // implements float arithmetic
 
+const (
+	// Maximum size in bits for Mpints before signalling
+	// overflow and also mantissa precision for Mpflts.
+	Mpprec = 512
+	// Turn on for constant arithmetic debugging output.
+	Mpdebug = false
+)
+
+// Mpflt represents a floating-point constant.
+type Mpflt struct {
+	Val big.Float
+}
+
+// Mpcplx represents a complex constant.
+type Mpcplx struct {
+	Real Mpflt
+	Imag Mpflt
+}
+
 func newMpflt() *Mpflt {
 	var a Mpflt
 	a.Val.SetPrec(Mpprec)

src/cmd/compile/internal/gc/mpint.go

@@ -10,7 +10,14 @@ import (
 	"fmt"
 )
 
-/// implements fix arithmetic
+// implements integer arithmetic
+
+// Mpint represents an integer constant.
+type Mpint struct {
+	Val  big.Int
+	Ovf  bool // set if Val overflowed compiler limit (sticky)
+	Rune bool // set if syntax indicates default type rune
+}
 
 func mpsetovf(a *Mpint) {
 	a.Val.SetUint64(1) // avoid spurious div-zero errors