JavaScript Performance Through the Spyglass
Vyacheslav Egorov
Excelsior JET
V8
Dart VM
LuaJIT
Excelsior JET
V8
Dart VM
LuaJIT
no choice
[ask on StackOverflow]
Crankshaft ⇒ IRHydra
[http://mrale.ph/irhydra/2]
# Get the source as described at:
# https://chromium.googlesource.com/v8/v8.git
$ fetch v8
...
$ cd v8
$ make x64.debug
...
# Get the source as described at:
# https://chromium.googlesource.com/v8/v8.git
$ fetch v8
...
$ cd v8
$ make x64.debug
...
$ out/x64.debug/d8 --print-ast test.js
. . FOR at -1
. . . YIELD COUNT 0
. . . BODY at -1
. . . . BLOCK at -1
. . . . . BLOCK NOCOMPLETIONS at -1
. . . . . . EXPRESSION STATEMENT at -1
. . . . . . . INIT at -1
. . . . . . . . VAR PROXY local[0] (mode = LET) "i"
. . . . . . . . VAR PROXY local[2] (mode = TEMPORARY) ".for"
. . . . . . IF at -1
. . . . . . . CONDITION at -1
. . . . . . . . EQ at -1
. . . . . . . . . VAR PROXY local[3] (mode = TEMPORARY) ".for"
. . . . . . . . . LITERAL 1
. . . . . . . THEN at -1
. . . . . . . . EXPRESSION STATEMENT at -1
. . . . . . . . . ASSIGN at -1
. . . . . . . . . . VAR PROXY local[3] (mode = TEMPORARY) ".for"
. . . . . . . . . . LITERAL 0
. . . . . . . ELSE at 200
. . . . . . . . EXPRESSION STATEMENT at 200
. . . . . . . . . POST INC at 200
. . . . . . . . . . VAR PROXY local[0] (mode = LET) "i"
. . . . . . EXPRESSION STATEMENT at -1
. . . . . . . ASSIGN at -1
. . . . . . . . VAR PROXY local[4] (mode = TEMPORARY) ".for"
. . . . . . . . LITERAL 1
. . . . . . IF at 186
. . . . . . . CONDITION at 186
. . . . . . . . LT at 186
. . . . . . . . . VAR PROXY local[0] (mode = LET) "i"
. . . . . . . . . LITERAL 1e+08
. . . . . . . THEN at -1
. . . . . . . . EMPTY at -1
. . . . . . . ELSE at -1
. . . . . . . . BREAK at -1
. . . . . FOR at 168
. . . . . . YIELD COUNT 0
. . . . . . COND at -1
. . . . . . . EQ at -1
. . . . . . . . VAR PROXY local[4] (mode = TEMPORARY) ".for"
VAR PROXY local[0] (mode = LET) "i"
VAR PROXY local[2] (mode = TEMPORARY) ".for"
local[0] (mode = LET) "i"
local[2] (mode = TEMPORARY) ".for"
$ git grep '"\.for"'
src/ast/ast-value-factory.h:
F(dot_for, ".for") \
$ git grep dot_for
src/parsing/parser.cc: const AstRawString* temp_name = ast_value_factory()->dot_for_string();
src/parsing/parser.cc: scope_->NewTemporary(ast_value_factory()->dot_for_string());
src/parsing/parser.cc: scope_->NewTemporary(ast_value_factory()->dot_for_string());
// We are given a for statement
// of the form
//
// labels: for (let/const x = i;
// cond;
// next) body
//
// and rewrite it as follows.
// We are given a for statement of the form
//
// labels: for (let/const x = i; cond; next) body
//
// and rewrite it as follows. Here we write {{ ... }} for init-blocks, ie.,
// blocks whose ignore_completion_value_ flag is set.
//
// {
// let/const x = i;
// temp_x = x;
// first = 1;
// undefined;
// outer: for (;;) {
// let/const x = temp_x;
// {{ if (first == 1) {
// first = 0;
// } else {
// next;
// }
// flag = 1;
// if (!cond) break;
// }}
// labels: for (; flag == 1; flag = 0, temp_x = x) {
// body
// }
// {{ if (flag == 1) // Body used break.
// break;
// }}
// }
// }
internals are important
internals are irrelevant
function find(val){
function index(value) {
return [1, 2, 3].indexOf(value);
}
return index(val);
}
function find(val){
function index(value) {
return [1, 2, 3].indexOf(value);
}
return index(val);
}
// => 5464 op/ms
var arr = [1, 2, 3];
function find(val){
function index(value) {
return arr.indexOf(value);
}
return index(val);
}
var arr = [1, 2, 3];
function find(val){
function index(value) {
return arr.indexOf(value);
}
return index(val);
}
// => 14084 op/ms
«array allocation
is expensive!»
var makeArr = () => [1, 2, 3];
function find(val){
function index(value) {
return makeArr().indexOf(value);
}
return index(val);
}
var makeArr = () => [1, 2, 3];
function find(val){
function index(value) {
return makeArr().indexOf(value);
}
return index(val);
}
// => 12048 op/ms
var makeArr = () => [1, 2, 3];
function find(val){
function index(value) {
return makeArr().indexOf(value);
}
return index(val);
}
// => 12048 op/ms
$ perf record d8 --perf-basic-prof test.js
$ perf report
14.80% v8::internal::Factory::NewFunction
9.98% v8::internal::Factory::NewFunctionFromSharedFunctionInfo
8.74% *InnerArrayIndexOf native array.js:1019
7.24% v8::internal::Factory::NewFunctionFromSharedFunctionInfo
5.79% v8::internal::Runtime_NewClosure
4.84% Builtin:ArgumentsAdaptorTrampoline
4.44% v8::internal::Heap::AllocateRaw
4.30% v8::internal::Factory::New
3.74% v8::internal::SharedFunctionInfo::SearchOptimizedCodeMap
3.20% ~index test.js:2
14.80% v8::internal::Factory::NewFunction
9.98% v8::internal::Factory::NewFunctionFromSharedFunctionInfo
8.74% *InnerArrayIndexOf native array.js:1019
7.24% v8::internal::Factory::NewFunctionFromSharedFunctionInfo
5.79% v8::internal::Runtime_NewClosure
4.84% Builtin:ArgumentsAdaptorTrampoline
4.44% v8::internal::Heap::AllocateRaw
4.30% v8::internal::Factory::New<v8::internal::JSFunction>
3.74% v8::internal::SharedFunctionInfo::SearchOptimizedCodeMap
3.20% ~index test.js:2
14.49% *InnerArrayIndexOf native array.js:1019
10.47% LoadIC:A load IC from the snapshot
8.45% ~index b2.js:12
8.05% Stub:FastNewClosureStub
5.77% Builtin:ArgumentsAdaptorTrampoline
5.23% *find2 b2.js:11
14.49% *InnerArrayIndexOf native array.js:1019
10.47% LoadIC:A load IC from the snapshot
8.45% ~index b2.js:12
8.05% Stub:FastNewClosureStub
5.77% Builtin:ArgumentsAdaptorTrampoline
5.23% *find2 b2.js:11
FastNewClosureStub
did not support allocation of functions with literals inside
[now it actually can, example is from February 2016]
function getBlock(buffer) {
var x = new Uint32Array(16);
for (var i = 16; i--;) x[i] = input[i];
for (var i = 20; i > 0; i -= 2) {
quarterRound(x, 0, 4, 8,12);
quarterRound(x, 1, 5, 9,13);
quarterRound(x, 2, 6,10,14);
quarterRound(x, 3, 7,11,15);
quarterRound(x, 0, 5,10,15);
quarterRound(x, 1, 6,11,12);
quarterRound(x, 2, 7, 8,13);
quarterRound(x, 3, 4, 9,14);
}
for (i = 16; i--;) x[i] += input[i];
for (i = 16; i--;) U32TO8_LE(buffer, 4 * i, x[i]);
input[12]++;
return buffer;
}
function getBlock(buffer) {
var x = new Uint32Array(16);
for (var i = 16; i--;) x[i] = input[i];
for (var i = 20; i > 0; i -= 2) {
quarterRound(x, 0, 4, 8,12);
quarterRound(x, 1, 5, 9,13);
quarterRound(x, 2, 6,10,14);
quarterRound(x, 3, 7,11,15);
quarterRound(x, 0, 5,10,15);
quarterRound(x, 1, 6,11,12);
quarterRound(x, 2, 7, 8,13);
quarterRound(x, 3, 4, 9,14);
}
for (i = 16; i--;) x[i] += input[i];
for (i = 16; i--;) U32TO8_LE(buffer, 4 * i, x[i]);
input[12]++;
return buffer;
}
function getBlock(buffer) {
var x = new Uint32Array(16);
for (var i = 16; i--;) x[i] = input[i];
for (var i = 20; i > 0; i -= 2) {
// quarterRound(x, 0, 4, 8,12);
x[ 0] += x[ 4]; x[12] = ((x[12] ^ x[ 0]) << 16) | ((x[12] ^ x[ 0]) >>> 16);
x[ 8] += x[12]; x[ 4] = ((x[ 4] ^ x[ 8]) << 12) | ((x[ 4] ^ x[ 8]) >>> 20);
x[ 0] += x[ 4]; x[12] = ((x[12] ^ x[ 0]) << 8) | ((x[12] ^ x[ 0]) >>> 24);
x[ 8] += x[12]; x[ 4] = ((x[ 4] ^ x[ 8]) << 7) | ((x[ 4] ^ x[ 8]) >>> 25);
// ... so on ...
}
for (i = 16; i--;) x[i] += input[i];
for (i = 16; i--;) U32TO8_LE(buffer, 4 * i, x[i]);
input[12]++;
return buffer;
}
«I HEARD INLINE
MUST IMPROVE PERFORMANCE!»
function U32TO8_LE(x, i, u) {
x[i] = u; u >>>= 8;
x[i+1] = u; u >>>= 8;
x[i+2] = u; u >>>= 8;
x[i+3] = u;
}
function U32TO8_LE(x, i, u) {
// ____ ___ ____ __________ _ ____ ___ ___
// `MM' `M' 6MMMMb\ `MMMMMMMMM dM. 6MMMMb\ `MMb dMM'
// MM M 6M' ` MM \ ,MMb 6M' ` MMM. ,PMM
// MM M MM MM d'YM. MM M`Mb d'MM
// MM M YM. MM , ,P `Mb YM. M YM. ,P MM
// MM M YMMMMb MMMMMMM d' YM. YMMMMb M `Mb d' MM
// MM M `Mb MM ` ,P `Mb `Mb M YM.P MM
// MM M MM MM d' YM. MM M `Mb' MM
// YM M MM MM ,MMMMMMMMb MM M YP MM
// 8b d8 L ,M9 MM / d' YM. L ,M9 M `' MM
// YMMMMM9 MYMMMM9 _MMMMMMMMM _dM_ _dMM_MYMMMM9 _M_ _MM_
//
x[i] = u; u >>>= 8;
x[i+1] = u; u >>>= 8;
x[i+2] = u; u >>>= 8;
x[i+3] = u;
}
function U32TO8_LE(x, i, u) {
// ____ ___ ____ __________ _ ____ ___ ___
// `MM' `M' 6MMMMb\ `MMMMMMMMM dM. 6MMMMb\ `MMb dMM'
// MM M 6M' ` MM \ ,MMb 6M' ` MMM. ,PMM
// MM M MM MM d'YM. MM M`Mb d'MM
// MM M YM. MM , ,P `Mb YM. M YM. ,P MM
// MM M YMMMMb MMMMMMM d' YM. YMMMMb M `Mb d' MM
// MM M `Mb MM ` ,P `Mb `Mb M YM.P MM
// MM M MM MM d' YM. MM M `Mb' MM
// YM M MM MM ,MMMMMMMMb MM M YP MM
// 8b d8 L ,M9 MM / d' YM. L ,M9 M `' MM
// YMMMMM9 MYMMMM9 _MMMMMMMMM _dM_ _dMM_MYMMMM9 _M_ _MM_
// ^^^^^^ Don't remove this comment it makes the code 10x faster. ^^^^^^^^^^^
x[i] = u; u >>>= 8;
x[i+1] = u; u >>>= 8;
x[i+2] = u; u >>>= 8;
x[i+3] = u;
}
we say performance
we mean jsperf.com
we say performance
we mean jsperf.com
[or at least we used to]
// split
str.split("e").length - 1;
// while
var count = 0;
while (str.length) {
if (str.charAt(0) === "e") {
count += 1;
}
str = str.slice(1);
}
// split (IT'S OVER 9000K OP/S ON FF!)
str.split("e").length - 1;
// while
var count = 0;
while (str.length) {
if (str.charAt(0) === "e") {
count += 1;
}
str = str.slice(1);
}
function benchmark() {
var start = Date.now();
/* OP */
return (Date.now() - start);
}
what if OP faster than clock?
function benchmark(N) {
var start = Date.now();
for (var i = 0; i < N; i++) {
/* OP */
}
return (Date.now() - start) / N;
}
\[\frac{C \times \cancel{N}}{\cancel{N}}\]
while (str.length) {
if (str.charAt(0) === "e") {
count += 1;
}
str = str.slice(1);
}
function benchmark() {
var str = "...";
var start = Date.now();
for (var i = 0; i < N; i++) {
var count = 0;
while (str.length) {
if (str.charAt(0) === "e") {
count += 1;
}
str = str.slice(1);
}
}
return (Date.now() - start) / N;
}
function benchmark() {
var str = "...";
var start = Date.now();
for (var i = 0; i < N; i++) {
var count = 0;
while (str.length) {
if (str.charAt(0) === "e") {
count += 1;
}
str = str.slice(1);
}
} // here str === ""
return (Date.now() - start) / N;
}
repetitions start with str === ""
\[C \times 1 + 0 \times (N - 1)\]
\[\frac{C \times 1 + 0 \times (N - 1)}{N}\]
\[\frac{C}{N} \approx 0\]
benchmark.js misuse
OP should take the same time
when repeatedly executed
JITs optimize
as program runs
\[C_u N_u + C_o N_o\]
[unoptimized + optimized version]
\[{\sum C_{i} N_i}\]
[multiple unoptimized and
optimized versions]
\[{\sum C_{i} N_i} + \color{#dc1e1a}{\frac{1}{\sqrt{2\pi}}\int C(\xi) e ^ {-\frac{\xi^2}{2}} d\xi}\]
[math gets too complicated to approach analytically]
JITs are clever
to measure you need to outsmart
Disclaimer: On slides I will show optimizations as source-to-source transformations. In reality compilers operate on more sophisticated representations.
function benchmark() {
var i = '1561565', j;
var start = Date.now();
for (var n = 0; n < N; n++) {
j = ~~i;
}
return (Date.now() - start) / N;
}
function benchmark() {
var j;
var start = Date.now();
for (var n = 0; n < N; n++) {
j = ~~'1561565';
}
return (Date.now() - start) / N;
}
function benchmark() {
var j;
var start = Date.now();
for (var n = 0; n < N; n++) {
j = ~-1561566;
}
return (Date.now() - start) / N;
}
function benchmark() {
var j;
var start = Date.now();
for (var n = 0; n < N; n++) {
j = 1561565;
}
return (Date.now() - start) / N;
}
«hey, I can trick the compiler!»
function benchmark() {
var i = Date.now().toString(), j;
// ^ not a constant any more
var start = Date.now();
for (var n = 0; n < N; n++) {
j = ~~i;
}
return (Date.now() - start) / N;
}
function benchmark() {
var i = Date.now().toString(), j;
var start = Date.now();
for (var n = 0; n < N; n++) {
j = ~~i;
// ^^^ loop invariant
}
return (Date.now() - start) / N;
}
function benchmark() {
var i = Date.now().toString(), j;
var start = Date.now();
var j0 = ~~i;
// ^^^^^^^^ hoisted from the loop
for (var n = 0; n < N; n++) {
j = j0;
}
return (Date.now() - start) / N;
}
loop invariant code motion
function benchmark() {
var i = Date.now().toString(), j;
var start = Date.now();
var j0 = ~~i;
for (var n = 0; n < N; n++) {
j = j0;
}
return (Date.now() - start) / N;
}
function benchmark() {
var i = Date.now().toString(), j;
var start = Date.now();
var j0 = ~~i;
for (var n = 0; n < N; n++) {
j = j0;
}
return (Date.now() - start) / N;
}
function benchmark() {
var i = Date.now().toString(), j;
var start = Date.now();
var j0 = ~~i;
for (var n = 0; n < N; n++) {
j = j0;
}
return (Date.now() - start) / N;
}
function benchmark() {
var i = Date.now().toString(), j;
var start = Date.now();
var j0 = ~~i;
for (var n = 0; n < N; n++) {
j = j0;
}
return (Date.now() - start) / N;
}
function benchmark() {
var start = Date.now();
for (var n = 0; n < N; n++) {
/* sound of silence */
}
return (Date.now() - start) / N;
}
function benchmark() {
var start = Date.now();
/*
* sound of silence
*/
return (Date.now() - start) / N;
}
// Remember this one?
// split (IT'S OVER 9000K OP/S ON FF!)
str.split("e").length - 1;
// it's full of dead code.
optimizer eats µbenchmarks
for breakfast
- verify results
- no constants
- no loop invariants
- no dead code
function benchmark() {
var i = Date.now().toString(),
i1 = Date.now().toString(), t;
var j;
var start = Date.now();
for (var n = 0; n < N; n++) {
t = i; i = i1; i1 = t;
j = ~~i;
}
if (i != j || i1 != j) throw "whoa?";
return (Date.now() - start) / N;
}
(potentially)
still not enough
for (var n = 0; n < N; n++) {
t = i; i = i1; i1 = t;
j = ~~i;
}
for (var n = 0; n < (N / 2); n++) {
t = i; i = i1; i1 = t;
j = ~~i;
t = i; i = i1; i1 = t;
j = ~~i;
}
if ((N % 2) === 1) {
t = i; i = i1; i1 = t;
j = ~~i;
}
for (var n = 0; n < (N / 2); n++) {
j = ~~i1;
t = i1;
j = ~~i;
}
for (var n = 0; n < (N / 2); n++) {
j = ~~i1; /* dead */
t = i1; /* dead */
j = ~~i; /* invariant */
}
V8 doesn't do it
but I want to induce paranoia ☺
presumption of performance
reasonable code
reasonably fast
«prototype chains are slow»
var obj =
Object.create(
Object.create(
Object.create(
Object.create(
Object.create({prop: 10})))));
var obj =
Object.create(
Object.create(
Object.create(
Object.create(
Object.create({prop: 10})))));
// LISP tribute ^^^^^
function doManyLookups() {
var counter = 0;
for(var i = 0; i < 1000; i++)
for(var j = 0; j < 1000; j++)
for(var k = 0; k < 1000; k++)
counter += obj.prop;
print('In total: ' + counter);
}
function lookupAndCache() {
var counter = 0;
var value = obj.prop;
for(var i = 0; i < 1000; i++)
for(var j = 0; j < 1000; j++)
for(var k = 0; k < 1000; k++)
counter += value;
print('In total: ' + counter);
}
// State of art benchmark driver.
function measure(f) {
var start = Date.now();
f();
var end = Date.now();
print(f.name + ' took ' +
(end - start) + ' ms.');
}
measure(doManyLookups);
measure(lookupAndCache);
$ node prototype.js
In total: 10000000000
doManyLookups took 8243 ms.
In total: 10000000000
lookupAndCache took 1058 ms.
CONFIRMED
- Object.create({ prop: 10 })))));
+ Object.create({ get prop () { return 10 } })))));
$ node prototype.js
In total: 10000000000
doManyLookups took 1082 ms.
In total: 10000000000
lookupAndCache took 1061 ms.
«what kind of voodoo is this?»
very old V8 did not inline loads from data properties defined on prototypes
$ d8 prototype.js
In total: 10000000000
doManyLookups took 1294 ms.
In total: 10000000000
lookupAndCache took 1189 ms.
$ d8 prototype.js
In total: 10000000000
doManyLookups took 1294 ms.
In total: 10000000000
lookupAndCache took 1189 ms.
prototype chain traversal got LICMed
... and now for something completely different
what if we run benchmark
twice?
measure(doManyLookups);
measure(doManyLookups);
measure(lookupAndCache);
measure(lookupAndCache);
$ d8 prototype.js | grep took
doManyLookups took 1301 ms.
doManyLookups took 3408 ms.
lookupAndCache took 1204 ms.
lookupAndCache took 3406 ms.
'In total: ' + counter type-feedback leaks upwards into the loop and causes excessive boxing of the counter variable
workaround: hide + from representation inference
- print('In total: ' + counter);
+ print('In total: ' + counter.toString());
$ d8 prototype.js | grep took
doManyLookups took 1298 ms.
doManyLookups took 1119 ms.
lookupAndCache took 1188 ms.
lookupAndCache took 982 ms.
Benchmark.prototype.setup = function () {
var obj = { }
var _a = 0;
Object.defineProperty(obj, 'a', {
get: function () { return _a; },
set: function (val) { _a = val; }
});
};
suite.add('Accessor', function () {
obj.a = 2;
obj.a;
});
$ v5.3/d8 concat.js
| Accessor |
x 3,840,922 ops/sec ±1.54% |
$ v5.3/d8 concat.js
| Accessor |
x 3,840,922 ops/sec ±1.54% |
$ v3.31/d8 concat.js
| Accessor |
x 874,731,387 ops/sec ±1.09% |
$ v5.3/d8 concat.js
| Accessor |
x 3,840,922 ops/sec ±1.54% |
$ v3.31/d8 concat.js
| Accessor |
x 874,731,387 ops/sec ±1.09% |
function f() {
var obj = { }
var _a = 0;
Object.defineProperty(obj, 'a', {
get: function () { return _a; },
set: function(val) { _a = val; }
});
for (var i = 0; i < 1e6; i++) {
obj.a = 2;
obj.a;
}
}
f(); // => 2ms
f(); // => 287ms
f(); // => 2ms
f(); // => 287ms
Object.defineProperty(obj, 'a', {
/* ... */
});
print('obj has fast properties: ' +
%HasFastProperties(obj));
$ d8 --allow-natives-syntax
obj has fast properties: true
f() took 2ms
obj has fast properties: false
f() took 287ms
hidden class transition clash
var obj1 = { }
Object.defineProperty(obj1, 'a', {
get: function () { /* ... */ },
set: function(val) { /* ... */ }
});
var obj2 = { }
Object.defineProperty(obj2, 'a', {
get: function () { /* ... */ },
set: function(val) { /* ... */ }
});
// Initial hudden class for obj1/obj2 is the same,
// but after defineProperty they have different ones.
Object.defineProperty(obj, 'a', {
/* ... */
});
Object.create(obj);
f(); // => 4ms
f(); // => 3ms
f(); // => 4ms
f(); // => 3ms
f(); // => 3ms
f(); // => 9ms
f(); // => 28ms
function g() {
var obj = { }
var _a = 0;
obj.a = {
get: function () { return _a; },
set: function(val) { _a = val; }
};
for (var i = 0; i < 1e6; i++) {
obj.a.set(2);
obj.a.get();
}
}
g(); // => 23ms
g(); // => 15ms
g(); // => 16ms
g(); // => 22ms
g(); // => 16ms
function Box(_a) {
this._a = _a;
}
Box.prototype.get = function () {
return this._a;
};
Box.prototype.set = function (val) {
this._a = val;
};
function h() {
var obj = { }
obj.a = new Box(0);
for (var i = 0; i < 1e6; i++) {
obj.a.set(2);
obj.a.get();
}
}
h(); // => 2ms
h(); // => 1ms
h(); // => 0ms
h(); // => 1ms
h(); // => 1ms
we can look into other VMs!
$ jsc test.js
f(); // => 3ms
f(); // => 3ms
f(); // => 16ms
f(); // => 17ms
f(); // => 11ms
$ jsc --useConcurrentJIT=false \
--showDisassembly=true \
test.js >& code.asm
$ jsc --useConcurrentJIT=false \
--showDisassembly=true \
test.js >& code.asm
# there'll be three DFG versions of function f
82:<!7:loc14> GetLocal(Untyped:@178, JS|MustGen|UseAsOther, Final, loc6(O<Final>/FlushedCell), machine:loc5, R:Stack(-7), bc#102) predicting Final
0x22d0b7400516: mov -0x30(%rbp), %rax
84:<!0:-> CheckStructure(Cell:@82, MustGen, [%CS:Object], R:JSCell_structureID, Exits, bc#102)
0x22d0b740051a: cmp $0xff, (%rax)
0x22d0b7400520: jnz 0x22d0b7400871
85:< 2:loc13> GetGetterSetterByOffset(KnownCell:@82, KnownCell:@82, JS|UseAsOther, Othercell, id5{a}, 0, inferredType = Top, R:NamedProperties(5), Exits, bc#102)
0x22d0b7400526: mov 0x10(%rax), %rsi
86:< 1:loc9> GetSetter(KnownCell:@85, JS|UseAsOther, Function, R:GetterSetter_setter, Exits, bc#102)
0x22d0b740052a: mov 0x18(%rsi), %rdx
87:<!0:-> MovHint(Untyped:@82, MustGen, loc22, W:SideState, ClobbersExit, bc#102)
89:<!0:-> MovHint(Untyped:@81, MustGen, loc21, W:SideState, ClobbersExit, bc#102, ExitInvalid)
91:<!0:-> ExitOK(MustGen, W:SideState, bc#102)
92:<!0:-> CheckCell(Cell:@86, Untyped:@82, MustGen, <0x1035cc970, Function>, set#EoFqmc/<nogen>:[0x1035a1de0], Exits, bc#102)
0x22d0b740052e: mov $0x1035cc970, %r11
0x22d0b7400538: cmp %r11, %rdx
0x22d0b740053b: jnz 0x22d0b7400887
--> set#EoFqmc:<0x103597620, bc#102, SetterCall, known callee: Cell: 0x1035cc970 (%BA:Function), ID: 49, numArgs+this = 2, stackOffset = -28 (loc0 maps to loc28)>
82:<!7:loc14> GetLocal(Untyped:@178, JS|MustGen|UseAsOther, Final, loc6(O<Final>/FlushedCell), machine:loc5, R:Stack(-7), bc#102) predicting Final
0x22d0b7400516: mov -0x30(%rbp), %rax
84:<!0:-> CheckStructure(Cell:@82, MustGen, [%CS:Object], R:JSCell_structureID, Exits, bc#102)
0x22d0b740051a: cmp $0xff, (%rax)
0x22d0b7400520: jnz 0x22d0b7400871
85:< 2:loc13> GetGetterSetterByOffset(KnownCell:@82, KnownCell:@82, JS|UseAsOther, Othercell, id5{a}, 0, inferredType = Top, R:NamedProperties(5), Exits, bc#102)
0x22d0b7400526: mov 0x10(%rax), %rsi
86:< 1:loc9> GetSetter(KnownCell:@85, JS|UseAsOther, Function, R:GetterSetter_setter, Exits, bc#102)
0x22d0b740052a: mov 0x18(%rsi), %rdx
87:<!0:-> MovHint(Untyped:@82, MustGen, loc22, W:SideState, ClobbersExit, bc#102)
89:<!0:-> MovHint(Untyped:@81, MustGen, loc21, W:SideState, ClobbersExit, bc#102, ExitInvalid)
91:<!0:-> ExitOK(MustGen, W:SideState, bc#102)
92:<!0:-> CheckCell(Cell:@86, Untyped:@82, MustGen, <0x1035cc970, Function>, set#EoFqmc/<nogen>:[0x1035a1de0], Exits, bc#102)
0x22d0b740052e: mov $0x1035cc970, %r11
0x22d0b7400538: cmp %r11, %rdx
0x22d0b740053b: jnz 0x22d0b7400887
--> set#EoFqmc:<0x103597620, bc#102, SetterCall, known callee: Cell: 0x1035cc970 (%BA:Function), ID: 49, numArgs+this = 2, stackOffset = -28 (loc0 maps to loc28)>
82 GetLocal(@178)
84 CheckStructure(@82)
85 GetGetterSetterByOffset(@82)
86 GetSetter(@85)
92 CheckCell(Cell:@86, ...)
--> set#EoFqmc:<... known callee ...>
80 CheckStructure(@78)
81 GetGetterSetterByOffset(@78)
82 GetSetter(@81)
88 Call(@82)
80 CheckStructure(@78)
81 GetGetterSetterByOffset(@78)
82 GetSetter(@81)
88 Call(@82)
method call
vs.
function call
function mk(word) {
var len = word.length;
if (len > 255) return undefined;
var i = len >> 2;
return String.fromCharCode(
(word.charCodeAt( 0) & 0x03) << 14 |
(word.charCodeAt( i) & 0x03) << 12 |
(word.charCodeAt( i+i) & 0x03) << 10 |
(word.charCodeAt(i+i+i) & 0x03) << 8 |
len
);
}
Benchmark.prototype.setup = function() {
function mk(word) {
/* ... */
}
var MK = function() { };
MK.prototype.mk = mk;
var mker = new MK;
};
suite
.add('Function', function() {
var key = mk('www.wired.com');
key = mk('www.youtube.com');
key = mk('scorecardresearch.com');
key = mk('www.google-analytics.com');
})
.add('Method', function() {
var key = mker.mk('www.wired.com');
key = mker.mk('www.youtube.com');
key = mker.mk('scorecardresearch.com');
key = mker.mk('www.google-analytics.com');
})
$ d8 method-vs-function.js
| Function |
x 4,149,776 ops/sec ±0.62% |
| Method |
x 682,273,122 ops/sec ±0.72% |
Fastest is Method
keep secret
what I am going to show you now
--- a/method-function.js
+++ b/method-function.js
@@ -2,6 +2,9 @@
load("../benchmark.js");
Benchmark.prototype.setup = function() {
+ "Speed" + "your" + "JS" + "with" +
+ "this" + "one" + "weird" + "trick";
+
function mk(word) {
var len = word.length;
if (len > 255) return undefined;
$ d8 method-vs-function.js
| Function |
x 695,708,197 ops/sec ±0.38% |
| Method |
x 692,496,013 ops/sec ±0.29% |
Fastest is Function,Method
Function was never optimized!
— heuristics that decide when to optimize are based (among other things) on the amount of initialized inline caches
— function call does not go through an IC, but method call does
— Method had enough initialized ICs to trigger optimizations, but Function didn't!
— Until fake + ICs were added in the setup section
yada yada
measuring almost empty loop again!
can function call be faster than method call?
// for (var item in list) { ... }
var sum = 0;
for (var i = 0, L = arr.length;
i < arr.length;
++i) {
if (arr.length !== L)
H.throwConcurrentModificationError(arr);
var item = list[i];
sum += item;
}
// for (var item in list) { ... }
var sum = 0;
for (var i = 0, L = arr.length;
i < arr.length;
++i) {
// if (arr.length !== L)
// H.throwConcurrentModificationError(arr);
var item = list[i];
sum += item;
}
$ d8 iteration.js
| WithCheck |
x 396,792 ops/sec ±0.37% |
| WithoutCheck |
x 456,653 ops/sec ±0.82% |
Fastest is WithoutCheck (by 18%)
// for (var item in list) { ... }
var sum = 0;
for (var i = 0, L = arr.length;
i < arr.length;
++i) {
if (arr.length !== L)
H.throwConcurrentModificationError(arr);
var item = list[i];
sum += item;
}
// for (var item in list) { ... }
var sum = 0;
for (var i = 0, L = arr.length;
i < arr.length;
++i) {
if (arr.length !== L)
(0,H.throwConcurrentModificationError)(arr);
var item = list[i];
sum += item;
}
$ d8 iteration.js
| WithCheck |
x 396,792 ops/sec ±0.37% |
| WithoutCheck |
x 456,653 ops/sec ±0.82% |
| WithHack |
x 462,698 ops/sec ±0.48% |
Fastest is WithoutCheck,WithHack
18% speedup by replacing
o.f()
with
(0,o.f)()
in the code that never executes
algorithms first
µbenchmarks last
WAIT!
what about the first example?
function getBlock(buffer) {
var x = new Uint32Array(16);
for (var i = 16; i--;) x[i] = input[i];
for (var i = 20; i > 0; i -= 2) {
// quarterRound(x, 0, 4, 8,12);
x[ 0] += x[ 4]; x[12] = ((x[12] ^ x[ 0]) << 16) | ((x[12] ^ x[ 0]) >>> 16);
x[ 8] += x[12]; x[ 4] = ((x[ 4] ^ x[ 8]) << 12) | ((x[ 4] ^ x[ 8]) >>> 20);
x[ 0] += x[ 4]; x[12] = ((x[12] ^ x[ 0]) << 8) | ((x[12] ^ x[ 0]) >>> 24);
x[ 8] += x[12]; x[ 4] = ((x[ 4] ^ x[ 8]) << 7) | ((x[ 4] ^ x[ 8]) >>> 25);
// ... so on ...
}
for (i = 16; i--;) x[i] += input[i];
for (i = 16; i--;) U32TO8_LE(buffer, 4 * i, x[i]);
input[12]++;
return buffer;
}
repetitive deopt is always V8 bug
— inlining U32TO8_LE used to break safe-uint32 analysis
— adding long comment into U32TO8_LE disables inlining
— there are more sane workarounds
function getBlock(buffer) {
var x = new Uint32Array(16);
for (var i = 16; i--;) x[i] = input[i];
for (var i = 20; i > 0; i -= 2) {
// ...
}
for (i = 16; i--;) x[i] += input[i];
for (i = 16; i--;) U32TO8_LE(buffer, 4 * i, x[i]|0);
// immediately truncate to int32 ^^
input[12]++;
return buffer;
}
never assume that a language feature has to be slow
talk to VM people
report bugs