8sa1-gcc/libgo/go/runtime/string.go
Ian Lance Taylor 108fdcc56e compiler,runtime: pass only ptr and len to some runtime calls
This ports https://golang.org/cl/227163 to the Go frontend.
This is a step toward moving up to the go1.15rc1 release.

Original CL description:

    cmd/compile,runtime: pass only ptr and len to some runtime calls

    Some runtime calls accept a slice, but only use ptr and len.
    This change modifies most such routines to accept only ptr and len.

    After this change, the only runtime calls that accept an unnecessary
    cap arg are concatstrings and slicerunetostring.
    Neither is particularly common, and both are complicated to modify.

    Negligible compiler performance impact. Shrinks binaries a little.
    There are only a few regressions; the one I investigated was
    due to register allocation fluctuation.

    Passes 'go test -race std cmd', modulo golang/go#38265 and golang/go#38266.
    Wow, does that take a long time to run.

    file      before    after     Δ       %
    compile   19655024  19655152  +128    +0.001%
    cover     5244840   5236648   -8192   -0.156%
    dist      3662376   3658280   -4096   -0.112%
    link      6680056   6675960   -4096   -0.061%
    pprof     14789844  14777556  -12288  -0.083%
    test2json 2824744   2820648   -4096   -0.145%
    trace     11647876  11639684  -8192   -0.070%
    vet       8260472   8256376   -4096   -0.050%
    total     115163736 115118808 -44928  -0.039%

For golang/go#36890

Reviewed-on: https://go-review.googlesource.com/c/gofrontend/+/245099
2020-07-27 17:05:17 -07:00

550 lines
12 KiB
Go

// Copyright 2014 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 runtime
import (
"internal/bytealg"
"unsafe"
)
// For gccgo, use go:linkname to export compiler-called functions.
//
//go:linkname concatstrings
//go:linkname slicebytetostring
//go:linkname slicebytetostringtmp
//go:linkname stringtoslicebyte
//go:linkname stringtoslicerune
//go:linkname slicerunetostring
//go:linkname intstring
// Temporary for C code to call:
//go:linkname gostringnocopy
//go:linkname findnull
// The constant is known to the compiler.
// There is no fundamental theory behind this number.
const tmpStringBufSize = 32
type tmpBuf [tmpStringBufSize]byte
// concatstrings implements a Go string concatenation x+y+z+...
// The operands are passed in the slice a.
// If buf != nil, the compiler has determined that the result does not
// escape the calling function, so the string data can be stored in buf
// if small enough.
func concatstrings(buf *tmpBuf, p *string, n int) string {
var a []string
*(*slice)(unsafe.Pointer(&a)) = slice{unsafe.Pointer(p), n, n}
// idx := 0
l := 0
count := 0
for _, x := range a {
n := len(x)
if n == 0 {
continue
}
if l+n < l {
throw("string concatenation too long")
}
l += n
count++
// idx = i
}
if count == 0 {
return ""
}
// If there is just one string and either it is not on the stack
// or our result does not escape the calling frame (buf != nil),
// then we can return that string directly.
// Commented out for gccgo--no implementation of stringDataOnStack.
// if count == 1 && (buf != nil || !stringDataOnStack(a[idx])) {
// return a[idx]
// }
s, b := rawstringtmp(buf, l)
for _, x := range a {
copy(b, x)
b = b[len(x):]
}
return s
}
// slicebytetostring converts a byte slice to a string.
// It is inserted by the compiler into generated code.
// ptr is a pointer to the first element of the slice;
// n is the length of the slice.
// Buf is a fixed-size buffer for the result,
// it is not nil if the result does not escape.
func slicebytetostring(buf *tmpBuf, ptr *byte, n int) (str string) {
if n == 0 {
// Turns out to be a relatively common case.
// Consider that you want to parse out data between parens in "foo()bar",
// you find the indices and convert the subslice to string.
return ""
}
if raceenabled {
racereadrangepc(unsafe.Pointer(ptr),
uintptr(n),
getcallerpc(),
funcPC(slicebytetostring))
}
if msanenabled {
msanread(unsafe.Pointer(ptr), uintptr(n))
}
if n == 1 {
stringStructOf(&str).str = unsafe.Pointer(&staticbytes[*ptr])
stringStructOf(&str).len = 1
return
}
var p unsafe.Pointer
if buf != nil && n <= len(buf) {
p = unsafe.Pointer(buf)
} else {
p = mallocgc(uintptr(n), nil, false)
}
stringStructOf(&str).str = p
stringStructOf(&str).len = n
memmove(p, unsafe.Pointer(ptr), uintptr(n))
return
}
func rawstringtmp(buf *tmpBuf, l int) (s string, b []byte) {
if buf != nil && l <= len(buf) {
b = buf[:l]
s = slicebytetostringtmp(&b[0], len(b))
} else {
s, b = rawstring(l)
}
return
}
// slicebytetostringtmp returns a "string" referring to the actual []byte bytes.
//
// Callers need to ensure that the returned string will not be used after
// the calling goroutine modifies the original slice or synchronizes with
// another goroutine.
//
// The function is only called when instrumenting
// and otherwise intrinsified by the compiler.
//
// Some internal compiler optimizations use this function.
// - Used for m[T1{... Tn{..., string(k), ...} ...}] and m[string(k)]
// where k is []byte, T1 to Tn is a nesting of struct and array literals.
// - Used for "<"+string(b)+">" concatenation where b is []byte.
// - Used for string(b)=="foo" comparison where b is []byte.
func slicebytetostringtmp(ptr *byte, n int) (str string) {
if raceenabled && n > 0 {
racereadrangepc(unsafe.Pointer(ptr),
uintptr(n),
getcallerpc(),
funcPC(slicebytetostringtmp))
}
if msanenabled && n > 0 {
msanread(unsafe.Pointer(ptr), uintptr(n))
}
stringStructOf(&str).str = unsafe.Pointer(ptr)
stringStructOf(&str).len = n
return
}
func stringtoslicebyte(buf *tmpBuf, s string) []byte {
var b []byte
if buf != nil && len(s) <= len(buf) {
*buf = tmpBuf{}
b = buf[:len(s)]
} else {
b = rawbyteslice(len(s))
}
copy(b, s)
return b
}
func stringtoslicerune(buf *[tmpStringBufSize]rune, s string) []rune {
// two passes.
// unlike slicerunetostring, no race because strings are immutable.
n := 0
for range s {
n++
}
var a []rune
if buf != nil && n <= len(buf) {
*buf = [tmpStringBufSize]rune{}
a = buf[:n]
} else {
a = rawruneslice(n)
}
n = 0
for _, r := range s {
a[n] = r
n++
}
return a
}
func slicerunetostring(buf *tmpBuf, a []rune) string {
if raceenabled && len(a) > 0 {
racereadrangepc(unsafe.Pointer(&a[0]),
uintptr(len(a))*unsafe.Sizeof(a[0]),
getcallerpc(),
funcPC(slicerunetostring))
}
if msanenabled && len(a) > 0 {
msanread(unsafe.Pointer(&a[0]), uintptr(len(a))*unsafe.Sizeof(a[0]))
}
var dum [4]byte
size1 := 0
for _, r := range a {
size1 += encoderune(dum[:], r)
}
s, b := rawstringtmp(buf, size1+3)
size2 := 0
for _, r := range a {
// check for race
if size2 >= size1 {
break
}
size2 += encoderune(b[size2:], r)
}
return s[:size2]
}
type stringStruct struct {
str unsafe.Pointer
len int
}
// Variant with *byte pointer type for DWARF debugging.
type stringStructDWARF struct {
str *byte
len int
}
func stringStructOf(sp *string) *stringStruct {
return (*stringStruct)(unsafe.Pointer(sp))
}
func intstring(buf *[4]byte, v int64) (s string) {
if v >= 0 && v < runeSelf {
stringStructOf(&s).str = unsafe.Pointer(&staticbytes[v])
stringStructOf(&s).len = 1
return
}
var b []byte
if buf != nil {
b = buf[:]
s = slicebytetostringtmp(&b[0], len(b))
} else {
s, b = rawstring(4)
}
if int64(rune(v)) != v {
v = runeError
}
n := encoderune(b, rune(v))
return s[:n]
}
// rawstring allocates storage for a new string. The returned
// string and byte slice both refer to the same storage.
// The storage is not zeroed. Callers should use
// b to set the string contents and then drop b.
func rawstring(size int) (s string, b []byte) {
p := mallocgc(uintptr(size), nil, false)
stringStructOf(&s).str = p
stringStructOf(&s).len = size
*(*slice)(unsafe.Pointer(&b)) = slice{p, size, size}
return
}
// rawbyteslice allocates a new byte slice. The byte slice is not zeroed.
func rawbyteslice(size int) (b []byte) {
cap := roundupsize(uintptr(size))
p := mallocgc(cap, nil, false)
if cap != uintptr(size) {
memclrNoHeapPointers(add(p, uintptr(size)), cap-uintptr(size))
}
*(*slice)(unsafe.Pointer(&b)) = slice{p, size, int(cap)}
return
}
// rawruneslice allocates a new rune slice. The rune slice is not zeroed.
func rawruneslice(size int) (b []rune) {
if uintptr(size) > maxAlloc/4 {
throw("out of memory")
}
mem := roundupsize(uintptr(size) * 4)
p := mallocgc(mem, nil, false)
if mem != uintptr(size)*4 {
memclrNoHeapPointers(add(p, uintptr(size)*4), mem-uintptr(size)*4)
}
*(*slice)(unsafe.Pointer(&b)) = slice{p, size, int(mem / 4)}
return
}
// used by cmd/cgo
func gobytes(p *byte, n int) (b []byte) {
if n == 0 {
return make([]byte, 0)
}
if n < 0 || uintptr(n) > maxAlloc {
panic(errorString("gobytes: length out of range"))
}
bp := mallocgc(uintptr(n), nil, false)
memmove(bp, unsafe.Pointer(p), uintptr(n))
*(*slice)(unsafe.Pointer(&b)) = slice{bp, n, n}
return
}
// This is exported via linkname to assembly in syscall (for Plan9).
//go:linkname gostring
func gostring(p *byte) string {
l := findnull(p)
if l == 0 {
return ""
}
s, b := rawstring(l)
memmove(unsafe.Pointer(&b[0]), unsafe.Pointer(p), uintptr(l))
return s
}
func gostringn(p *byte, l int) string {
if l == 0 {
return ""
}
s, b := rawstring(l)
memmove(unsafe.Pointer(&b[0]), unsafe.Pointer(p), uintptr(l))
return s
}
func index(s, t string) int {
if len(t) == 0 {
return 0
}
for i := 0; i < len(s); i++ {
if s[i] == t[0] && hasPrefix(s[i:], t) {
return i
}
}
return -1
}
func contains(s, t string) bool {
return index(s, t) >= 0
}
func hasPrefix(s, prefix string) bool {
return len(s) >= len(prefix) && s[:len(prefix)] == prefix
}
func hasSuffix(s, suffix string) bool {
return len(s) >= len(suffix) && s[len(s)-len(suffix):] == suffix
}
const (
maxUint = ^uint(0)
maxInt = int(maxUint >> 1)
)
// atoi parses an int from a string s.
// The bool result reports whether s is a number
// representable by a value of type int.
func atoi(s string) (int, bool) {
if s == "" {
return 0, false
}
neg := false
if s[0] == '-' {
neg = true
s = s[1:]
}
un := uint(0)
for i := 0; i < len(s); i++ {
c := s[i]
if c < '0' || c > '9' {
return 0, false
}
if un > maxUint/10 {
// overflow
return 0, false
}
un *= 10
un1 := un + uint(c) - '0'
if un1 < un {
// overflow
return 0, false
}
un = un1
}
if !neg && un > uint(maxInt) {
return 0, false
}
if neg && un > uint(maxInt)+1 {
return 0, false
}
n := int(un)
if neg {
n = -n
}
return n, true
}
// atoi32 is like atoi but for integers
// that fit into an int32.
func atoi32(s string) (int32, bool) {
if n, ok := atoi(s); n == int(int32(n)) {
return int32(n), ok
}
return 0, false
}
//go:nosplit
func findnull(s *byte) int {
if s == nil {
return 0
}
// Avoid IndexByteString on Plan 9 because it uses SSE instructions
// on x86 machines, and those are classified as floating point instructions,
// which are illegal in a note handler.
if GOOS == "plan9" {
p := (*[maxAlloc/2 - 1]byte)(unsafe.Pointer(s))
l := 0
for p[l] != 0 {
l++
}
return l
}
// pageSize is the unit we scan at a time looking for NULL.
// It must be the minimum page size for any architecture Go
// runs on. It's okay (just a minor performance loss) if the
// actual system page size is larger than this value.
const pageSize = 4096
offset := 0
ptr := unsafe.Pointer(s)
// IndexByteString uses wide reads, so we need to be careful
// with page boundaries. Call IndexByteString on
// [ptr, endOfPage) interval.
safeLen := int(pageSize - uintptr(ptr)%pageSize)
for {
t := *(*string)(unsafe.Pointer(&stringStruct{ptr, safeLen}))
// Check one page at a time.
if i := bytealg.IndexByteString(t, 0); i != -1 {
return offset + i
}
// Move to next page
ptr = unsafe.Pointer(uintptr(ptr) + uintptr(safeLen))
offset += safeLen
safeLen = pageSize
}
}
func findnullw(s *uint16) int {
if s == nil {
return 0
}
p := (*[maxAlloc/2/2 - 1]uint16)(unsafe.Pointer(s))
l := 0
for p[l] != 0 {
l++
}
return l
}
//go:nosplit
func gostringnocopy(str *byte) string {
ss := stringStruct{str: unsafe.Pointer(str), len: findnull(str)}
s := *(*string)(unsafe.Pointer(&ss))
return s
}
func gostringw(strw *uint16) string {
var buf [8]byte
str := (*[maxAlloc/2/2 - 1]uint16)(unsafe.Pointer(strw))
n1 := 0
for i := 0; str[i] != 0; i++ {
n1 += encoderune(buf[:], rune(str[i]))
}
s, b := rawstring(n1 + 4)
n2 := 0
for i := 0; str[i] != 0; i++ {
// check for race
if n2 >= n1 {
break
}
n2 += encoderune(b[n2:], rune(str[i]))
}
b[n2] = 0 // for luck
return s[:n2]
}
// These two functions are called by code generated by cgo -gccgo.
//go:linkname __go_byte_array_to_string __go_byte_array_to_string
func __go_byte_array_to_string(p unsafe.Pointer, l int) string {
if l == 0 {
return ""
}
s, c := rawstringtmp(nil, l)
memmove(unsafe.Pointer(&c[0]), p, uintptr(l))
return s
}
//go:linkname __go_string_to_byte_array __go_string_to_byte_array
func __go_string_to_byte_array(s string) []byte {
return stringtoslicebyte(nil, s)
}
// parseRelease parses a dot-separated version number. It follows the
// semver syntax, but allows the minor and patch versions to be
// elided.
func parseRelease(rel string) (major, minor, patch int, ok bool) {
// Strip anything after a dash or plus.
for i := 0; i < len(rel); i++ {
if rel[i] == '-' || rel[i] == '+' {
rel = rel[:i]
break
}
}
next := func() (int, bool) {
for i := 0; i < len(rel); i++ {
if rel[i] == '.' {
ver, ok := atoi(rel[:i])
rel = rel[i+1:]
return ver, ok
}
}
ver, ok := atoi(rel)
rel = ""
return ver, ok
}
if major, ok = next(); !ok || rel == "" {
return
}
if minor, ok = next(); !ok || rel == "" {
return
}
patch, ok = next()
return
}