internal/bytealg: move short string Index implementations into bytealg

Also move the arm64 CountByte implementation while we're here.

Fixes #19792

Change-Id: I1e0fdf1e03e3135af84150a2703b58dad1b0d57e
Reviewed-on: https://go-review.googlesource.com/98518
Run-TryBot: Keith Randall <khr@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

4 files changed, 79 insertions, 214 deletions

diff --git a/src/strings/strings.go b/src/strings/strings.go
index 7d3ed37edd..b0a53fdefd 100644
--- a/src/strings/strings.go
+++ b/src/strings/strings.go
@@ -932,6 +932,85 @@ func EqualFold(s, t string) bool {
 	return s == t
 }
 
+// Index returns the index of the first instance of substr in s, or -1 if substr is not present in s.
+func Index(s, substr string) int {
+	n := len(substr)
+	switch {
+	case n == 0:
+		return 0
+	case n == 1:
+		return IndexByte(s, substr[0])
+	case n == len(s):
+		if substr == s {
+			return 0
+		}
+		return -1
+	case n > len(s):
+		return -1
+	case n <= bytealg.MaxLen:
+		// Use brute force when s and substr both are small
+		if len(s) <= bytealg.MaxBruteForce {
+			return bytealg.IndexString(s, substr)
+		}
+		c := substr[0]
+		i := 0
+		t := s[:len(s)-n+1]
+		fails := 0
+		for i < len(t) {
+			if t[i] != c {
+				// IndexByte is faster than bytealg.IndexString, so use it as long as
+				// we're not getting lots of false positives.
+				o := IndexByte(t[i:], c)
+				if o < 0 {
+					return -1
+				}
+				i += o
+			}
+			if s[i:i+n] == substr {
+				return i
+			}
+			fails++
+			i++
+			// Switch to bytealg.IndexString when IndexByte produces too many false positives.
+			if fails > bytealg.Cutover(i) {
+				r := bytealg.IndexString(s[i:], substr)
+				if r >= 0 {
+					return r + i
+				}
+				return -1
+			}
+		}
+		return -1
+	}
+	c := substr[0]
+	i := 0
+	t := s[:len(s)-n+1]
+	fails := 0
+	for i < len(t) {
+		if t[i] != c {
+			o := IndexByte(t[i:], c)
+			if o < 0 {
+				return -1
+			}
+			i += o
+		}
+		if s[i:i+n] == substr {
+			return i
+		}
+		i++
+		fails++
+		if fails >= 4+i>>4 && i < len(t) {
+			// See comment in ../bytes/bytes_generic.go.
+			j := indexRabinKarp(s[i:], substr)
+			if j < 0 {
+				return -1
+			}
+			return i + j
+		}
+	}
+	return -1
+}
+
 func indexRabinKarp(s, substr string) int {
 	// Rabin-Karp search
 	hashss, pow := hashStr(substr)
diff --git a/src/strings/strings_amd64.go b/src/strings/strings_amd64.go
deleted file mode 100644
index 75e7d0c139..0000000000
--- a/src/strings/strings_amd64.go
+++ /dev/null
@@ -1,79 +0,0 @@
-// Copyright 2015 The Go Authors. All rights reserved.
-// Use of this source code is governed by a BSD-style
-// license that can be found in the LICENSE file.
-
-package strings
-
-import "internal/cpu"
-
-//go:noescape
-
-// indexShortStr returns the index of the first instance of c in s, or -1 if c is not present in s.
-// indexShortStr requires 2 <= len(c) <= shortStringLen
-func indexShortStr(s, c string) int  // ../runtime/asm_amd64.s
-func countByte(s string, c byte) int // ../runtime/asm_amd64.s
-
-var shortStringLen int
-
-func init() {
-	if cpu.X86.HasAVX2 {
-		shortStringLen = 63
-	} else {
-		shortStringLen = 31
-	}
-}
-
-// Index returns the index of the first instance of substr in s, or -1 if substr is not present in s.
-func Index(s, substr string) int {
-	n := len(substr)
-	switch {
-	case n == 0:
-		return 0
-	case n == 1:
-		return IndexByte(s, substr[0])
-	case n == len(s):
-		if substr == s {
-			return 0
-		}
-		return -1
-	case n > len(s):
-		return -1
-	case n <= shortStringLen:
-		// Use brute force when s and substr both are small
-		if len(s) <= 64 {
-			return indexShortStr(s, substr)
-		}
-		c := substr[0]
-		i := 0
-		t := s[:len(s)-n+1]
-		fails := 0
-		for i < len(t) {
-			if t[i] != c {
-				// IndexByte skips 16/32 bytes per iteration,
-				// so it's faster than indexShortStr.
-				o := IndexByte(t[i:], c)
-				if o < 0 {
-					return -1
-				}
-				i += o
-			}
-			if s[i:i+n] == substr {
-				return i
-			}
-			fails++
-			i++
-			// Switch to indexShortStr when IndexByte produces too many false positives.
-			// Too many means more that 1 error per 8 characters.
-			// Allow some errors in the beginning.
-			if fails > (i+16)/8 {
-				r := indexShortStr(s[i:], substr)
-				if r >= 0 {
-					return r + i
-				}
-				return -1
-			}
-		}
-		return -1
-	}
-	return indexRabinKarp(s, substr)
-}
diff --git a/src/strings/strings_generic.go b/src/strings/strings_generic.go
deleted file mode 100644
index ac3b8dce85..0000000000
--- a/src/strings/strings_generic.go
+++ /dev/null
@@ -1,55 +0,0 @@
-// Copyright 2015 The Go Authors. All rights reserved.
-// Use of this source code is governed by a BSD-style
-// license that can be found in the LICENSE file.
-
-// +build !amd64,!s390x
-
-package strings
-
-// TODO: implements short string optimization on non amd64 platforms
-// and get rid of strings_amd64.go
-
-// Index returns the index of the first instance of substr in s, or -1 if substr is not present in s.
-func Index(s, substr string) int {
-	n := len(substr)
-	switch {
-	case n == 0:
-		return 0
-	case n == 1:
-		return IndexByte(s, substr[0])
-	case n == len(s):
-		if substr == s {
-			return 0
-		}
-		return -1
-	case n > len(s):
-		return -1
-	}
-	c := substr[0]
-	i := 0
-	t := s[:len(s)-n+1]
-	fails := 0
-	for i < len(t) {
-		if t[i] != c {
-			o := IndexByte(t[i:], c)
-			if o < 0 {
-				return -1
-			}
-			i += o
-		}
-		if s[i:i+n] == substr {
-			return i
-		}
-		i++
-		fails++
-		if fails >= 4+i>>4 && i < len(t) {
-			// See comment in ../bytes/bytes_generic.go.
-			j := indexRabinKarp(s[i:], substr)
-			if j < 0 {
-				return -1
-			}
-			return i + j
-		}
-	}
-	return -1
-}
diff --git a/src/strings/strings_s390x.go b/src/strings/strings_s390x.go
deleted file mode 100644
index b2e459b04e..0000000000
--- a/src/strings/strings_s390x.go
+++ /dev/null
@@ -1,80 +0,0 @@
-// Copyright 2016 The Go Authors. All rights reserved.
-// Use of this source code is governed by a BSD-style
-// license that can be found in the LICENSE file.
-
-package strings
-
-//go:noescape
-
-// indexShortStr returns the index of the first instance of sep in s,
-// or -1 if sep is not present in s.
-// indexShortStr requires 2 <= len(sep) <= shortStringLen
-func indexShortStr(s, sep string) int // ../runtime/asm_$GOARCH.s
-
-// supportsVX reports whether the vector facility is available.
-// indexShortStr must not be called if the vector facility is not
-// available.
-func supportsVX() bool // ../runtime/asm_s390x.s
-
-var shortStringLen = -1
-
-func init() {
-	if supportsVX() {
-		shortStringLen = 64
-	}
-}
-
-// Index returns the index of the first instance of substr in s, or -1 if substr is not present in s.
-func Index(s, substr string) int {
-	n := len(substr)
-	switch {
-	case n == 0:
-		return 0
-	case n == 1:
-		return IndexByte(s, substr[0])
-	case n == len(s):
-		if substr == s {
-			return 0
-		}
-		return -1
-	case n > len(s):
-		return -1
-	case n <= shortStringLen:
-		// Use brute force when s and substr both are small
-		if len(s) <= 64 {
-			return indexShortStr(s, substr)
-		}
-		c := substr[0]
-		i := 0
-		t := s[:len(s)-n+1]
-		fails := 0
-		for i < len(t) {
-			if t[i] != c {
-				// IndexByte skips 16/32 bytes per iteration,
-				// so it's faster than indexShortStr.
-				o := IndexByte(t[i:], c)
-				if o < 0 {
-					return -1
-				}
-				i += o
-			}
-			if s[i:i+n] == substr {
-				return i
-			}
-			fails++
-			i++
-			// Switch to indexShortStr when IndexByte produces too many false positives.
-			// Too many means more that 1 error per 8 characters.
-			// Allow some errors in the beginning.
-			if fails > (i+16)/8 {
-				r := indexShortStr(s[i:], substr)
-				if r >= 0 {
-					return r + i
-				}
-				return -1
-			}
-		}
-		return -1
-	}
-	return indexRabinKarp(s, substr)
-}