cmd/compile: optimize bvec routines

The recent improvements to the prove pass
make it possible to provide bounds
hints to the compiler in some bvec routines.

This speeds up the compilation of the code in

name  old time/op       new time/op       delta
Pkg         7.93s ± 4%        7.69s ± 3%  -2.98%  (p=0.000 n=29+26)

While we're here, clean up some C-isms.

Updates #13554
Updates #20393

Change-Id: I47a0ec68543a9fc95c5359c3f37813fb529cb4f0
Reviewed-on: https://go-review.googlesource.com/110560
Run-TryBot: Josh Bleecher Snyder <josharian@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>

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

@@ -118,8 +118,8 @@ func (bv bvec) Next(i int32) int32 {
 }
 
 func (bv bvec) IsEmpty() bool {
-	for i := int32(0); i < bv.n; i += wordBits {
-		if bv.b[i>>wordShift] != 0 {
+	for _, x := range bv.b {
+		if x != 0 {
 			return false
 		}
 	}
@@ -127,15 +127,18 @@ func (bv bvec) IsEmpty() bool {
 }
 
 func (bv bvec) Not() {
-	i := int32(0)
-	w := int32(0)
-	for ; i < bv.n; i, w = i+wordBits, w+1 {
-		bv.b[w] = ^bv.b[w]
+	for i, x := range bv.b {
+		bv.b[i] = ^x
 	}
 }
 
 // union
 func (dst bvec) Or(src1, src2 bvec) {
+	if len(src1.b) == 0 {
+		return
+	}
+	_, _ = dst.b[len(src1.b)-1], src2.b[len(src1.b)-1] // hoist bounds checks out of the loop
+
 	for i, x := range src1.b {
 		dst.b[i] = x | src2.b[i]
 	}
@@ -143,6 +146,11 @@ func (dst bvec) Or(src1, src2 bvec) {
 
 // intersection
 func (dst bvec) And(src1, src2 bvec) {
+	if len(src1.b) == 0 {
+		return
+	}
+	_, _ = dst.b[len(src1.b)-1], src2.b[len(src1.b)-1] // hoist bounds checks out of the loop
+
 	for i, x := range src1.b {
 		dst.b[i] = x & src2.b[i]
 	}
@@ -150,6 +158,11 @@ func (dst bvec) And(src1, src2 bvec) {
 
 // difference
 func (dst bvec) AndNot(src1, src2 bvec) {
+	if len(src1.b) == 0 {
+		return
+	}
+	_, _ = dst.b[len(src1.b)-1], src2.b[len(src1.b)-1] // hoist bounds checks out of the loop
+
 	for i, x := range src1.b {
 		dst.b[i] = x &^ src2.b[i]
 	}