Currently we always allocate string buffers in heap.
For example, in the following code we allocate a temp string
just for comparison:
if string(byteSlice) == "abc" { ... }
This change extends escape analysis to cover []byte->string
conversions and string concatenation. If the result of operations
does not escape, compiler allocates a small buffer
on stack and passes it to slicebytetostring and concatstrings.
Then runtime uses the buffer if the result fits into it.
Size of the buffer is 32 bytes. There is no fundamental theory
behind this number. Just an observation that on std lib
tests/benchmarks frequency of string allocation is inversely
proportional to string length; and there is significant number
of allocations up to length 32.
benchmark old allocs new allocs delta
BenchmarkFprintfBytes 2 1 -50.00%
BenchmarkDecodeComplex128Slice 318 316 -0.63%
BenchmarkDecodeFloat64Slice 318 316 -0.63%
BenchmarkDecodeInt32Slice 318 316 -0.63%
BenchmarkDecodeStringSlice 2318 2316 -0.09%
BenchmarkStripTags 11 5 -54.55%
BenchmarkDecodeGray 111 102 -8.11%
BenchmarkDecodeNRGBAGradient 200 188 -6.00%
BenchmarkDecodeNRGBAOpaque 165 152 -7.88%
BenchmarkDecodePaletted 319 309 -3.13%
BenchmarkDecodeRGB 166 157 -5.42%
BenchmarkDecodeInterlacing 279 268 -3.94%
BenchmarkGoLookupIP 153 135 -11.76%
BenchmarkGoLookupIPNoSuchHost 508 466 -8.27%
BenchmarkGoLookupIPWithBrokenNameServer 245 226 -7.76%
BenchmarkClientServerParallel4 62 61 -1.61%
BenchmarkClientServerParallel64 62 61 -1.61%
BenchmarkClientServerParallelTLS4 79 78 -1.27%
BenchmarkClientServerParallelTLS64 112 111 -0.89%
benchmark old ns/op new ns/op delta
BenchmarkFprintfBytes 381 311 -18.37%
BenchmarkStripTags 2615 2351 -10.10%
BenchmarkDecodeNRGBAGradient
3715887 3635096 -2.17%
BenchmarkDecodeNRGBAOpaque
3047645 2928644 -3.90%
BenchmarkGoLookupIP 153 135 -11.76%
BenchmarkGoLookupIPNoSuchHost 508 466 -8.27%
Change-Id: I9ec01da816945c3329d7be3c7794b520418c3f99
Reviewed-on: https://go-review.googlesource.com/3120
Reviewed-by: Keith Randall <khr@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>
"func @\"\".printsp ()\n"
"func @\"\".printlock ()\n"
"func @\"\".printunlock ()\n"
- "func @\"\".concatstring2 (? string, ? string) (? string)\n"
- "func @\"\".concatstring3 (? string, ? string, ? string) (? string)\n"
- "func @\"\".concatstring4 (? string, ? string, ? string, ? string) (? string)\n"
- "func @\"\".concatstring5 (? string, ? string, ? string, ? string, ? string) (? string)\n"
- "func @\"\".concatstrings (? []string) (? string)\n"
+ "func @\"\".concatstring2 (? *[32]byte, ? string, ? string) (? string)\n"
+ "func @\"\".concatstring3 (? *[32]byte, ? string, ? string, ? string) (? string)\n"
+ "func @\"\".concatstring4 (? *[32]byte, ? string, ? string, ? string, ? string) (? string)\n"
+ "func @\"\".concatstring5 (? *[32]byte, ? string, ? string, ? string, ? string, ? string) (? string)\n"
+ "func @\"\".concatstrings (? *[32]byte, ? []string) (? string)\n"
"func @\"\".cmpstring (? string, ? string) (? int)\n"
"func @\"\".eqstring (? string, ? string) (? bool)\n"
"func @\"\".intstring (? int64) (? string)\n"
- "func @\"\".slicebytetostring (? []byte) (? string)\n"
+ "func @\"\".slicebytetostring (? *[32]byte, ? []byte) (? string)\n"
"func @\"\".slicebytetostringtmp (? []byte) (? string)\n"
"func @\"\".slicerunetostring (? []rune) (? string)\n"
"func @\"\".stringtoslicebyte (? string) (? []byte)\n"
e->noesc = list(e->noesc, n);
break;
+ case OARRAYBYTESTR:
+ n->escloopdepth = e->loopdepth;
+ n->esc = EscNone; // until proven otherwise
+ e->noesc = list(e->noesc, n);
+ break;
+
+ case OADDSTR:
+ n->escloopdepth = e->loopdepth;
+ n->esc = EscNone; // until proven otherwise
+ e->noesc = list(e->noesc, n);
+ // Arguments of OADDSTR do not escape.
+ break;
+
case OADDR:
n->esc = EscNone; // until proven otherwise
e->noesc = list(e->noesc, n);
case OMAKECHAN:
case OMAKEMAP:
case OMAKESLICE:
+ case OARRAYBYTESTR:
+ case OADDSTR:
case ONEW:
case OCLOSURE:
case OCALLPART:
case OSLICE3:
case OSLICEARR:
case OSLICE3ARR:
+ case OSLICESTR:
// Conversions, field access, slice all preserve the input value.
escassign(e, dst, src->left);
break;
case OMAKECHAN:
case OMAKEMAP:
case OMAKESLICE:
+ case OARRAYBYTESTR:
+ case OADDSTR:
case OMAPLIT:
case ONEW:
case OCLOSURE:
case OSLICEARR:
case OSLICE3:
case OSLICE3ARR:
+ case OSLICESTR:
escwalk(e, level, dst, src->left);
break;
func printlock()
func printunlock()
-func concatstring2(string, string) string
-func concatstring3(string, string, string) string
-func concatstring4(string, string, string, string) string
-func concatstring5(string, string, string, string, string) string
-func concatstrings([]string) string
+func concatstring2(*[32]byte, string, string) string
+func concatstring3(*[32]byte, string, string, string) string
+func concatstring4(*[32]byte, string, string, string, string) string
+func concatstring5(*[32]byte, string, string, string, string, string) string
+func concatstrings(*[32]byte, []string) string
func cmpstring(string, string) int
func eqstring(string, string) bool
func intstring(int64) string
-func slicebytetostring([]byte) string
+func slicebytetostring(*[32]byte, []byte) string
func slicebytetostringtmp([]byte) string
func slicerunetostring([]rune) string
func stringtoslicebyte(string) []byte
static Mpint mpzero;
static void walkprintfunc(Node**, NodeList**);
+// The constant is known to runtime.
+enum
+{
+ tmpstringbufsize = 32,
+};
+
void
walk(Node *fn)
{
goto ret;
case OARRAYBYTESTR:
- // slicebytetostring([]byte) string;
- n = mkcall("slicebytetostring", n->type, init, n->left);
+ a = nodnil();
+ if(n->esc == EscNone) {
+ // Create temporary buffer for string on stack.
+ t = aindex(nodintconst(tmpstringbufsize), types[TUINT8]);
+ a = nod(OADDR, temp(t), N);
+ }
+ // slicebytetostring(*[32]byte, []byte) string;
+ n = mkcall("slicebytetostring", n->type, init, a, n->left);
goto ret;
case OARRAYBYTESTRTMP:
static Node*
addstr(Node *n, NodeList **init)
{
- Node *r, *cat, *slice;
+ Node *r, *cat, *slice, *buf;
NodeList *args, *l;
int c;
+ vlong sz;
Type *t;
// orderexpr rewrote OADDSTR to have a list of strings.
if(c < 2)
yyerror("addstr count %d too small", c);
+ buf = nodnil();
+ if(n->esc == EscNone) {
+ sz = 0;
+ for(l=n->list; l != nil; l=l->next) {
+ if(n->op == OLITERAL)
+ sz += n->val.u.sval->len;
+ }
+ // Don't allocate the buffer if the result won't fit.
+ if(sz < tmpstringbufsize) {
+ // Create temporary buffer for result string on stack.
+ t = aindex(nodintconst(tmpstringbufsize), types[TUINT8]);
+ buf = nod(OADDR, temp(t), N);
+ }
+ }
+
// build list of string arguments
- args = nil;
+ args = list1(buf);
for(l=n->list; l != nil; l=l->next)
args = list(args, conv(l->n, types[TSTRING]));
t->bound = -1;
slice = nod(OCOMPLIT, N, typenod(t));
slice->alloc = n->alloc;
- slice->list = args;
+ slice->list = args->next; // skip buf arg
+ args = list1(buf);
+ args = list(args, slice);
slice->esc = EscNone;
- args = list1(slice);
}
cat = syslook(namebuf, 1);
r = nod(OCALL, cat, N);
}
}
+func BenchmarkFprintfBytes(b *testing.B) {
+ data := []byte(string("0123456789"))
+ var buf bytes.Buffer
+ for i := 0; i < b.N; i++ {
+ buf.Reset()
+ Fprintf(&buf, "%s", data)
+ }
+}
+
func BenchmarkFprintIntNoAlloc(b *testing.B) {
var x interface{} = 123456
var buf bytes.Buffer
"unsafe"
)
-func concatstrings(a []string) string {
+// 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, a []string) string {
idx := 0
l := 0
count := 0
if count == 0 {
return ""
}
- if count == 1 {
+
+ // 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.
+ if count == 1 && (buf != nil || !stringDataOnStack(a[idx])) {
return a[idx]
}
- s, b := rawstring(l)
+ s, b := rawstringtmp(buf, l)
l = 0
for _, x := range a {
copy(b[l:], x)
return s
}
-func concatstring2(a [2]string) string {
- return concatstrings(a[:])
+func concatstring2(buf *tmpBuf, a [2]string) string {
+ return concatstrings(buf, a[:])
}
-func concatstring3(a [3]string) string {
- return concatstrings(a[:])
+func concatstring3(buf *tmpBuf, a [3]string) string {
+ return concatstrings(buf, a[:])
}
-func concatstring4(a [4]string) string {
- return concatstrings(a[:])
+func concatstring4(buf *tmpBuf, a [4]string) string {
+ return concatstrings(buf, a[:])
}
-func concatstring5(a [5]string) string {
- return concatstrings(a[:])
+func concatstring5(buf *tmpBuf, a [5]string) string {
+ return concatstrings(buf, a[:])
}
-func slicebytetostring(b []byte) string {
- if raceenabled && len(b) > 0 {
+// Buf is a fixed-size buffer for the result,
+// it is not nil if the result does not escape.
+func slicebytetostring(buf *tmpBuf, b []byte) string {
+ l := len(b)
+ if l == 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 && l > 0 {
racereadrangepc(unsafe.Pointer(&b[0]),
- uintptr(len(b)),
+ uintptr(l),
getcallerpc(unsafe.Pointer(&b)),
funcPC(slicebytetostring))
}
- s, c := rawstring(len(b))
+ s, c := rawstringtmp(buf, l)
copy(c, b)
return s
}
+// stringDataOnStack reports whether the string's data is
+// stored on the current goroutine's stack.
+func stringDataOnStack(s string) bool {
+ ptr := uintptr((*stringStruct)(unsafe.Pointer(&s)).str)
+ stk := getg().stack
+ return stk.lo <= ptr && ptr < stk.hi
+}
+
+func rawstringtmp(buf *tmpBuf, l int) (s string, b []byte) {
+ if buf != nil && l <= len(buf) {
+ b = buf[:l]
+ s = slicebytetostringtmp(b)
+ } else {
+ s, b = rawstring(l)
+ }
+ return
+}
+
func slicebytetostringtmp(b []byte) string {
// Return a "string" referring to the actual []byte bytes.
// This is only for use by internal compiler optimizations
t.Fatalf("want 0 allocs, got %v", n)
}
}
+
+func TestStringOnStack(t *testing.T) {
+ s := ""
+ for i := 0; i < 3; i++ {
+ s = "a" + s + "b" + s + "c"
+ }
+
+ if want := "aaabcbabccbaabcbabccc"; s != want {
+ t.Fatalf("want: '%v', got '%v'", want, s)
+ }
+}
*sp = (*sp)[1:2] // ERROR "quux ignoring self-assignment to \*sp"
*bp = (*bp)[1:2] // ERROR "quux ignoring self-assignment to \*bp"
}
+
+type StructWithString struct {
+ p *int
+ s string
+}
+
+// This is escape analysis false negative.
+// We assign the pointer to x.p but leak x.s. Escape analysis coarsens flows
+// to just x, and thus &i looks escaping.
+func fieldFlowTracking() {
+ var x StructWithString
+ i := 0 // ERROR "moved to heap: i"
+ x.p = &i // ERROR "&i escapes to heap"
+ sink = x.s
+}
+
+// String operations.
+
+func slicebytetostring0() {
+ b := make([]byte, 20) // ERROR "does not escape"
+ s := string(b) // ERROR "string\(b\) does not escape"
+ _ = s
+}
+
+func slicebytetostring1() {
+ b := make([]byte, 20) // ERROR "does not escape"
+ s := string(b) // ERROR "string\(b\) does not escape"
+ s1 := s[0:1]
+ _ = s1
+}
+
+func slicebytetostring2() {
+ b := make([]byte, 20) // ERROR "does not escape"
+ s := string(b) // ERROR "string\(b\) escapes to heap"
+ s1 := s[0:1] // ERROR "moved to heap: s1"
+ sink = &s1 // ERROR "&s1 escapes to heap"
+}
+
+func slicebytetostring3() {
+ b := make([]byte, 20) // ERROR "does not escape"
+ s := string(b) // ERROR "string\(b\) escapes to heap"
+ s1 := s[0:1]
+ sink = s1
+}
+
+func addstr0() {
+ s0 := "a"
+ s1 := "b"
+ s := s0 + s1 // ERROR "s0 \+ s1 does not escape"
+ _ = s
+}
+
+func addstr1() {
+ s0 := "a"
+ s1 := "b"
+ s := "c"
+ s += s0 + s1 // ERROR "s0 \+ s1 does not escape"
+ _ = s
+}
+
+func addstr2() {
+ b := make([]byte, 20) // ERROR "does not escape"
+ s0 := "a"
+ s := string(b) + s0 // ERROR "string\(b\) does not escape" "string\(b\) \+ s0 does not escape"
+ _ = s
+}
+
+func addstr3() {
+ s0 := "a"
+ s1 := "b"
+ s := s0 + s1 // ERROR "s0 \+ s1 escapes to heap"
+ s2 := s[0:1]
+ sink = s2
+}
*sp = (*sp)[1:2] // ERROR "quux ignoring self-assignment to \*sp"
*bp = (*bp)[1:2] // ERROR "quux ignoring self-assignment to \*bp"
}
+
+type StructWithString struct {
+ p *int
+ s string
+}
+
+// This is escape analysis false negative.
+// We assign the pointer to x.p but leak x.s. Escape analysis coarsens flows
+// to just x, and thus &i looks escaping.
+func fieldFlowTracking() {
+ var x StructWithString
+ i := 0 // ERROR "moved to heap: i"
+ x.p = &i // ERROR "&i escapes to heap"
+ sink = x.s
+}
+
+// String operations.
+
+func slicebytetostring0() {
+ b := make([]byte, 20) // ERROR "does not escape"
+ s := string(b) // ERROR "string\(b\) does not escape"
+ _ = s
+}
+
+func slicebytetostring1() {
+ b := make([]byte, 20) // ERROR "does not escape"
+ s := string(b) // ERROR "string\(b\) does not escape"
+ s1 := s[0:1]
+ _ = s1
+}
+
+func slicebytetostring2() {
+ b := make([]byte, 20) // ERROR "does not escape"
+ s := string(b) // ERROR "string\(b\) escapes to heap"
+ s1 := s[0:1] // ERROR "moved to heap: s1"
+ sink = &s1 // ERROR "&s1 escapes to heap"
+}
+
+func slicebytetostring3() {
+ b := make([]byte, 20) // ERROR "does not escape"
+ s := string(b) // ERROR "string\(b\) escapes to heap"
+ s1 := s[0:1]
+ sink = s1
+}
+
+func addstr0() {
+ s0 := "a"
+ s1 := "b"
+ s := s0 + s1 // ERROR "s0 \+ s1 does not escape"
+ _ = s
+}
+
+func addstr1() {
+ s0 := "a"
+ s1 := "b"
+ s := "c"
+ s += s0 + s1 // ERROR "s0 \+ s1 does not escape"
+ _ = s
+}
+
+func addstr2() {
+ b := make([]byte, 20) // ERROR "does not escape"
+ s0 := "a"
+ s := string(b) + s0 // ERROR "string\(b\) does not escape" "string\(b\) \+ s0 does not escape"
+ _ = s
+}
+
+func addstr3() {
+ s0 := "a"
+ s1 := "b"
+ s := s0 + s1 // ERROR "s0 \+ s1 escapes to heap"
+ s2 := s[0:1]
+ sink = s2
+}