148045 – [JSC] Avoid going to slow path when comparing undefined and null

Benjamin Poulain

Reported 2015-08-14 16:19:11 PDT

[JSC] Avoid going to slow path when comparing undefined and null

Benjamin Poulain

Comment 1 2015-08-14 16:27:11 PDT

Created attachment 259056 [details] Patch

Benjamin Poulain

Comment 2 2015-08-14 16:27:47 PDT

Conf#1 Conf#2 SunSpider: 3d-cube 8.0270+-0.2314 ? 8.3705+-0.4974 ? might be 1.0428x slower 3d-morph 8.1423+-0.2629 8.0778+-0.3080 3d-raytrace 8.6281+-0.2705 ? 8.8495+-0.4114 ? might be 1.0257x slower access-binary-trees 3.1108+-0.1881 3.1095+-0.1234 access-fannkuch 8.6398+-0.1608 8.4928+-0.1323 might be 1.0173x faster access-nbody 4.1830+-0.2632 4.1400+-0.2182 might be 1.0104x faster access-nsieve 4.5781+-0.0761 ? 4.6166+-0.1229 ? bitops-3bit-bits-in-byte 1.9863+-0.0201 ? 2.0385+-0.1665 ? might be 1.0262x slower bitops-bits-in-byte 5.6634+-0.0469 5.5883+-0.0686 might be 1.0134x faster bitops-bitwise-and 2.8320+-0.2365 2.7089+-0.2151 might be 1.0454x faster bitops-nsieve-bits 4.3690+-0.1124 4.2479+-0.0561 might be 1.0285x faster controlflow-recursive 3.0052+-0.0639 ? 3.0807+-0.1469 ? might be 1.0251x slower crypto-aes 6.0765+-0.2871 6.0192+-0.3982 crypto-md5 3.8787+-0.1117 ? 3.8879+-0.1360 ? crypto-sha1 3.3662+-0.1525 3.2835+-0.1581 might be 1.0252x faster date-format-tofte 12.5698+-0.5042 12.1857+-0.5735 might be 1.0315x faster date-format-xparb 7.3695+-0.1444 7.2939+-0.4023 might be 1.0104x faster math-cordic 4.3356+-0.0694 ? 4.3890+-0.0932 ? might be 1.0123x slower math-partial-sums 10.1257+-0.3032 ? 10.1627+-0.3318 ? math-spectral-norm 2.9561+-0.1776 ? 2.9691+-0.1027 ? regexp-dna 10.1000+-0.1767 9.9821+-0.4862 might be 1.0118x faster string-base64 6.4539+-0.4130 ? 6.6208+-0.2183 ? might be 1.0259x slower string-fasta 9.2523+-0.1551 ? 9.3478+-0.0672 ? might be 1.0103x slower string-tagcloud 13.2127+-0.9639 ? 13.2138+-0.3353 ? string-unpack-code 29.0781+-0.9197 29.0722+-1.5443 string-validate-input 6.8618+-0.3863 6.8356+-0.3159 <arithmetic> 7.2616+-0.0529 7.2532+-0.0805 might be 1.0012x faster Conf#1 Conf#2 LongSpider: 3d-cube 1159.4390+-18.3872 ? 1159.9818+-5.5884 ? 3d-morph 1872.3270+-4.0075 1871.2806+-1.6537 3d-raytrace 996.2850+-15.7166 987.9307+-23.3216 access-binary-trees 1321.8011+-3.7252 ? 1322.8964+-5.6430 ? access-fannkuch 449.0647+-21.0879 ? 464.4824+-39.7047 ? might be 1.0343x slower access-nbody 1008.7484+-0.7014 ? 1010.6885+-1.5951 ? access-nsieve 666.4697+-11.4402 666.3622+-8.4761 bitops-3bit-bits-in-byte 51.4730+-0.5074 ? 52.2062+-1.1943 ? might be 1.0142x slower bitops-bits-in-byte 310.0107+-6.2068 ? 314.7267+-7.0319 ? might be 1.0152x slower bitops-nsieve-bits 620.4373+-3.7827 619.4960+-3.6790 controlflow-recursive 686.0615+-1.2504 ^ 676.2185+-1.2180 ^ definitely 1.0146x faster crypto-aes 940.2765+-7.6746 932.1885+-7.4716 crypto-md5 695.4270+-3.4939 ? 700.1296+-10.2146 ? crypto-sha1 964.9257+-2.1927 ? 965.6763+-5.0394 ? date-format-tofte 956.9150+-14.7842 947.0907+-9.3238 might be 1.0104x faster date-format-xparb 1063.1870+-105.9450 1057.6287+-83.9667 hash-map 238.4415+-1.5434 237.1068+-4.6344 math-cordic 654.6338+-1.1860 ? 656.4597+-2.7010 ? math-partial-sums 1126.2252+-4.1284 ? 1130.0621+-3.8194 ? math-spectral-norm 1074.9101+-1.9610 ? 1074.9171+-2.7628 ? string-base64 530.9650+-3.4890 ? 532.9538+-5.7707 ? string-fasta 606.8112+-5.2560 ? 609.0435+-5.2807 ? string-tagcloud 286.3067+-4.6076 282.1616+-2.5118 might be 1.0147x faster <geometric> 657.7898+-3.5459 ? 658.2909+-3.1522 ? might be 1.0008x slower Conf#1 Conf#2 V8Spider: crypto 73.2687+-1.4024 72.1216+-0.9278 might be 1.0159x faster deltablue 92.2487+-2.4600 ? 96.0862+-3.6794 ? might be 1.0416x slower earley-boyer 63.3910+-1.0347 ? 63.5065+-1.2097 ? raytrace 41.3782+-2.1482 ? 41.5938+-1.7031 ? regexp 102.9228+-0.5188 102.2650+-0.8783 richards 93.9991+-2.2165 92.9843+-1.8538 might be 1.0109x faster splay 55.1071+-1.8320 ? 55.6590+-2.1056 ? might be 1.0100x slower <geometric> 71.3838+-0.7175 ? 71.6381+-1.1915 ? might be 1.0036x slower Conf#1 Conf#2 Octane: encrypt 0.27555+-0.01586 ? 0.27879+-0.00453 ? might be 1.0118x slower decrypt 4.83295+-0.00507 ? 4.85875+-0.02968 ? deltablue x2 0.26448+-0.01150 0.26235+-0.01431 earley 0.50115+-0.00201 0.50041+-0.00412 boyer 8.04595+-0.08925 7.93259+-0.10713 might be 1.0143x faster navier-stokes x2 6.37906+-0.03004 6.36829+-0.00458 raytrace x2 1.71163+-0.07924 ? 1.78228+-0.11828 ? might be 1.0413x slower richards x2 0.16984+-0.00142 ? 0.17222+-0.00315 ? might be 1.0140x slower splay x2 0.53225+-0.00674 ? 0.53932+-0.00713 ? might be 1.0133x slower regexp x2 38.34281+-0.58286 37.95919+-0.26565 might be 1.0101x faster pdfjs x2 59.83973+-0.23388 ? 60.00888+-0.40818 ? mandreel x2 69.93256+-0.61686 69.62872+-0.44826 gbemu x2 59.82779+-1.28386 ? 60.29241+-0.65040 ? closure 0.89632+-0.00175 0.89594+-0.00590 jquery 11.69179+-0.06452 11.67285+-0.10563 box2d x2 17.24493+-0.25505 ? 17.40000+-0.10062 ? zlib x2 574.63719+-10.86984 572.46226+-4.67871 typescript x2 1096.98853+-32.30140 ? 1119.18933+-16.95256 ? might be 1.0202x slower <geometric> 8.99476+-0.03262 ? 9.04214+-0.05265 ? might be 1.0053x slower Conf#1 Conf#2 Kraken: ai-astar 366.954+-7.110 358.306+-22.822 might be 1.0241x faster audio-beat-detection 83.350+-0.155 82.378+-1.091 might be 1.0118x faster audio-dft 130.904+-4.564 127.778+-2.067 might be 1.0245x faster audio-fft 59.235+-0.745 59.018+-0.487 audio-oscillator 101.311+-0.431 100.828+-0.458 imaging-darkroom 94.925+-0.306 94.838+-0.415 imaging-desaturate 90.572+-0.955 90.379+-0.763 imaging-gaussian-blur 151.483+-1.093 150.710+-2.667 json-parse-financial 69.007+-1.263 ? 69.180+-0.636 ? json-stringify-tinderbox 41.754+-0.661 41.543+-0.833 stanford-crypto-aes 63.410+-2.167 ? 63.569+-0.416 ? stanford-crypto-ccm 59.182+-2.108 57.932+-2.425 might be 1.0216x faster stanford-crypto-pbkdf2 142.639+-2.715 141.572+-1.771 stanford-crypto-sha256-iterative 58.184+-1.739 57.436+-2.095 might be 1.0130x faster <arithmetic> 108.065+-0.846 106.819+-1.662 might be 1.0117x faster Conf#1 Conf#2 JSRegress: abc-forward-loop-equal 58.2887+-1.0064 ? 58.3814+-0.5516 ? abc-postfix-backward-loop 56.7443+-1.1515 ? 57.2455+-2.1390 ? abc-simple-backward-loop 56.7975+-1.5868 56.4058+-0.5100 abc-simple-forward-loop 56.9490+-0.4733 56.7919+-0.6236 abc-skippy-loop 38.8353+-0.2932 ? 39.2732+-0.7482 ? might be 1.0113x slower abs-boolean 3.8264+-0.1971 3.7173+-0.1145 might be 1.0293x faster adapt-to-double-divide 19.0785+-0.6318 18.9803+-0.3377 aliased-arguments-getbyval 1.7449+-0.1748 1.6933+-0.3307 might be 1.0305x faster allocate-big-object 3.8170+-0.2369 3.4373+-0.3056 might be 1.1105x faster arguments-named-and-reflective 15.1081+-0.5961 15.0075+-0.1418 arguments-out-of-bounds 18.0140+-0.2171 ? 18.4821+-0.3800 ? might be 1.0260x slower arguments-strict-mode 14.1133+-0.4487 14.0585+-0.6840 arguments 12.7502+-0.1603 12.5034+-0.7468 might be 1.0197x faster arity-mismatch-inlining 1.2207+-0.1538 1.2190+-0.1256 array-access-polymorphic-structure 9.2256+-0.0982 ? 9.3348+-0.2856 ? might be 1.0118x slower array-nonarray-polymorhpic-access 43.8837+-1.6009 43.5362+-2.3322 array-prototype-every 117.4352+-0.3903 117.4315+-0.8044 array-prototype-forEach 116.4421+-3.7636 ? 117.4316+-3.8972 ? array-prototype-map 129.6927+-3.6762 128.7234+-1.3312 array-prototype-reduce 111.0988+-2.5063 110.4459+-2.9779 array-prototype-reduceRight 110.7897+-1.7052 ? 110.9873+-2.8425 ? array-prototype-some 119.0731+-2.0873 118.7554+-2.3622 array-splice-contiguous 33.2384+-0.6302 33.0129+-0.7028 array-with-double-add 5.6268+-0.1444 ? 5.7300+-0.1647 ? might be 1.0184x slower array-with-double-increment 4.1437+-0.1400 ? 4.1591+-0.2038 ? array-with-double-mul-add 7.2116+-0.1274 ? 7.2646+-0.1430 ? array-with-double-sum 4.4636+-0.1600 4.3229+-0.2244 might be 1.0325x faster array-with-int32-add-sub 9.7039+-0.2661 ? 9.8176+-0.5627 ? might be 1.0117x slower array-with-int32-or-double-sum 4.4205+-0.2741 ? 4.4442+-0.1325 ? ArrayBuffer-DataView-alloc-large-long-lived 45.4847+-1.9772 ? 47.4412+-3.7053 ? might be 1.0430x slower ArrayBuffer-DataView-alloc-long-lived 17.9303+-0.7373 17.9271+-0.6407 ArrayBuffer-Int32Array-byteOffset 5.7529+-0.1323 5.7103+-0.3444 ArrayBuffer-Int8Array-alloc-large-long-lived 46.7865+-1.8971 ? 47.7119+-1.7964 ? might be 1.0198x slower ArrayBuffer-Int8Array-alloc-long-lived-buffer 29.9709+-1.3758 ? 30.4168+-0.9992 ? might be 1.0149x slower ArrayBuffer-Int8Array-alloc-long-lived 16.9316+-0.1045 ? 17.7052+-1.0998 ? might be 1.0457x slower ArrayBuffer-Int8Array-alloc 14.4175+-0.3142 14.3056+-0.5656 asmjs_bool_bug 9.7740+-0.2169 9.7323+-0.1355 assign-custom-setter-polymorphic 4.1656+-0.1176 ? 4.2510+-0.1250 ? might be 1.0205x slower assign-custom-setter 5.7785+-0.1751 5.7062+-0.2531 might be 1.0127x faster basic-set 11.9575+-0.7414 11.7069+-0.6779 might be 1.0214x faster big-int-mul 5.8705+-0.1417 ? 5.9474+-0.2081 ? might be 1.0131x slower boolean-test 4.4860+-0.0408 4.3840+-0.1352 might be 1.0232x faster branch-fold 4.7819+-0.0496 ? 4.8207+-0.1278 ? branch-on-string-as-boolean 26.6865+-0.3951 26.3751+-1.4106 might be 1.0118x faster by-val-generic 9.0914+-0.4552 ? 9.3495+-0.3739 ? might be 1.0284x slower call-spread-apply 38.4805+-0.4369 ? 39.7807+-1.9166 ? might be 1.0338x slower call-spread-call 33.1773+-1.7863 ? 35.5188+-2.3682 ? might be 1.0706x slower captured-assignments 0.7251+-0.2070 0.7153+-0.1676 might be 1.0137x faster cast-int-to-double 8.3672+-0.1694 8.2828+-0.1660 might be 1.0102x faster cell-argument 10.2355+-0.7282 10.1572+-0.2452 cfg-simplify 4.0626+-0.1700 3.9446+-0.2435 might be 1.0299x faster chain-getter-access 12.1132+-0.6099 12.0158+-0.5454 cmpeq-obj-to-obj-other 13.6918+-0.6339 13.4770+-1.0689 might be 1.0159x faster constant-test 8.0485+-0.1622 8.0262+-0.2435 create-lots-of-functions 16.2736+-0.2387 ? 16.7440+-0.2835 ? might be 1.0289x slower cse-new-array-buffer 3.1590+-0.1747 ? 3.2490+-0.2491 ? might be 1.0285x slower cse-new-array 3.3322+-0.2206 ? 3.3499+-0.2354 ? DataView-custom-properties 53.9090+-1.8937 ? 54.6787+-2.4591 ? might be 1.0143x slower delay-tear-off-arguments-strictmode 18.9810+-0.5663 ? 19.0701+-0.9397 ? deltablue-varargs 268.6152+-3.7894 ? 271.9018+-6.9076 ? might be 1.0122x slower destructuring-arguments 239.3279+-3.4525 ? 239.4290+-4.4579 ? destructuring-parameters-overridden-by-function 0.7950+-0.1433 0.7490+-0.1560 might be 1.0614x faster destructuring-swap 7.8615+-0.1651 7.7859+-0.2656 direct-arguments-getbyval 1.6850+-0.1650 ? 1.7838+-0.1640 ? might be 1.0586x slower div-boolean-double 5.6132+-0.1508 5.6122+-0.1082 div-boolean 9.9347+-0.3788 ? 10.0549+-0.1559 ? might be 1.0121x slower double-get-by-val-out-of-bounds 6.0786+-0.4350 ? 6.3256+-0.3239 ? might be 1.0406x slower double-pollution-getbyval 9.9087+-0.2003 9.8135+-0.2419 double-pollution-putbyoffset 5.5997+-0.3304 ? 5.6437+-0.1694 ? double-real-use 40.6090+-0.5327 40.5966+-1.1797 double-to-int32-typed-array-no-inline 2.9141+-0.1541 2.7863+-0.1717 might be 1.0458x faster double-to-int32-typed-array 2.6895+-0.0675 2.6578+-0.0791 might be 1.0119x faster double-to-uint32-typed-array-no-inline 2.9337+-0.2020 ? 2.9900+-0.1195 ? might be 1.0192x slower double-to-uint32-typed-array 2.7392+-0.1450 ? 2.8625+-0.1935 ? might be 1.0450x slower elidable-new-object-dag 55.5491+-2.1268 ? 56.4929+-1.9160 ? might be 1.0170x slower elidable-new-object-roflcopter 56.0391+-0.9879 ? 56.3384+-0.6889 ? elidable-new-object-then-call 57.3751+-1.9965 ? 58.2582+-2.2884 ? might be 1.0154x slower elidable-new-object-tree 66.3618+-1.6502 ? 67.8672+-2.5131 ? might be 1.0227x slower empty-string-plus-int 7.5840+-0.2597 7.5654+-0.2521 emscripten-cube2hash 45.5349+-1.0407 ? 45.7130+-1.0894 ? exit-length-on-plain-object 20.0347+-2.3064 19.1815+-0.3366 might be 1.0445x faster external-arguments-getbyval 1.7682+-0.2180 1.7424+-0.0926 might be 1.0149x faster external-arguments-putbyval 3.1785+-0.1449 ? 3.2415+-0.1525 ? might be 1.0198x slower fixed-typed-array-storage-var-index 1.6366+-0.2326 1.6014+-0.1639 might be 1.0220x faster fixed-typed-array-storage 1.2861+-0.0914 1.2835+-0.1077 Float32Array-matrix-mult 6.2012+-0.2162 6.1163+-0.2436 might be 1.0139x faster Float32Array-to-Float64Array-set 75.7306+-2.7376 ? 85.0920+-30.7306 ? might be 1.1236x slower Float64Array-alloc-long-lived 93.1265+-1.2259 92.6470+-0.6640 Float64Array-to-Int16Array-set 92.6533+-0.4591 91.6345+-1.8214 might be 1.0111x faster fold-double-to-int 19.5321+-0.5629 19.2605+-0.7033 might be 1.0141x faster fold-get-by-id-to-multi-get-by-offset-rare-int 12.0860+-0.4289 11.8972+-0.5459 might be 1.0159x faster fold-get-by-id-to-multi-get-by-offset 9.4568+-0.5245 9.3918+-0.3502 fold-multi-get-by-offset-to-get-by-offset 8.7472+-0.6326 ? 8.7896+-0.8539 ? fold-multi-get-by-offset-to-poly-get-by-offset 8.3160+-0.7068 ? 8.6014+-0.4161 ? might be 1.0343x slower fold-multi-put-by-offset-to-poly-put-by-offset 7.5986+-0.6523 ? 7.9963+-0.5620 ? might be 1.0523x slower fold-multi-put-by-offset-to-put-by-offset 6.6580+-0.6721 ? 6.7843+-0.5302 ? might be 1.0190x slower fold-multi-put-by-offset-to-replace-or-transition-put-by-offset 13.2728+-0.2875 12.6271+-0.4430 might be 1.0511x faster fold-put-by-id-to-multi-put-by-offset 9.6065+-0.6182 9.2958+-0.5993 might be 1.0334x faster fold-put-structure 6.8015+-0.5881 6.7257+-0.5777 might be 1.0113x faster for-of-iterate-array-entries 16.7695+-0.2466 16.7354+-0.6625 for-of-iterate-array-keys 5.3076+-0.1228 5.0300+-0.4303 might be 1.0552x faster for-of-iterate-array-values 5.0027+-0.2522 4.8495+-0.2939 might be 1.0316x faster fround 24.4231+-0.7553 ? 25.6017+-1.6071 ? might be 1.0483x slower ftl-library-inlining-dataview 93.6871+-4.0901 ? 95.8419+-6.1818 ? might be 1.0230x slower ftl-library-inlining 90.3807+-0.8303 ? 90.3939+-1.7109 ? function-dot-apply 3.0753+-0.2306 2.9447+-0.1353 might be 1.0444x faster function-test 4.3166+-0.1368 ? 4.3642+-0.1923 ? might be 1.0110x slower function-with-eval 152.2069+-1.7286 151.0617+-0.5950 gcse-poly-get-less-obvious 24.9995+-0.3918 ? 25.1913+-0.8000 ? gcse-poly-get 25.0765+-0.4001 ? 25.0788+-0.5365 ? gcse 6.6767+-0.1871 ? 6.6859+-0.0741 ? get-by-id-bimorphic-check-structure-elimination-simple 3.3588+-0.0701 3.2603+-0.0416 might be 1.0302x faster get-by-id-bimorphic-check-structure-elimination 8.4265+-0.3332 8.4072+-0.0614 get-by-id-chain-from-try-block 6.6638+-0.2946 ? 6.7987+-0.1497 ? might be 1.0202x slower get-by-id-check-structure-elimination 7.5994+-0.0663 ? 7.6259+-0.0724 ? get-by-id-proto-or-self 24.9205+-2.8048 23.7673+-1.6365 might be 1.0485x faster get-by-id-quadmorphic-check-structure-elimination-simple 3.9924+-0.1408 ? 4.0090+-0.0424 ? get-by-id-self-or-proto 23.0740+-0.6575 ? 23.4334+-1.5997 ? might be 1.0156x slower get-by-val-out-of-bounds 5.8887+-0.2031 ? 5.8917+-0.4282 ? get-by-val-with-string-bimorphic-check-structure-elimination-simple 3.8891+-0.1209 ! 4.3538+-0.1061 ! definitely 1.1195x slower get-by-val-with-string-bimorphic-check-structure-elimination 9.9590+-0.1676 ? 10.1454+-0.1757 ? might be 1.0187x slower get-by-val-with-string-chain-from-try-block 7.8618+-0.2827 ? 7.9338+-0.2038 ? get-by-val-with-string-check-structure-elimination 9.2508+-0.1385 9.0905+-0.2060 might be 1.0176x faster get-by-val-with-string-proto-or-self 23.7261+-1.4576 23.4355+-0.9578 might be 1.0124x faster get-by-val-with-string-quadmorphic-check-structure-elimination-simple 5.0061+-0.1733 4.9163+-0.0732 might be 1.0183x faster get-by-val-with-string-self-or-proto 23.9263+-1.9701 ? 24.7756+-2.7752 ? might be 1.0355x slower get-by-val-with-symbol-bimorphic-check-structure-elimination-simple 4.7784+-0.1242 ! 5.1998+-0.1189 ! definitely 1.0882x slower get-by-val-with-symbol-bimorphic-check-structure-elimination 19.4578+-0.3987 ? 19.6690+-0.6428 ? might be 1.0109x slower get-by-val-with-symbol-chain-from-try-block 8.0378+-0.1686 7.8813+-0.1372 might be 1.0199x faster get-by-val-with-symbol-check-structure-elimination 18.2907+-0.4349 18.2406+-0.0967 get-by-val-with-symbol-proto-or-self 24.0062+-2.0889 23.8724+-2.0738 get-by-val-with-symbol-quadmorphic-check-structure-elimination-simple 6.1989+-0.0546 6.1830+-0.1724 get-by-val-with-symbol-self-or-proto 23.9543+-2.6164 ? 24.7161+-2.2229 ? might be 1.0318x slower get_callee_monomorphic 3.6133+-0.1993 ? 3.6172+-0.3239 ? get_callee_polymorphic 4.4749+-0.1678 ? 4.5038+-0.2935 ? getter-no-activation 5.9998+-0.6117 5.7992+-0.1417 might be 1.0346x faster getter-prototype 18.1263+-0.3187 18.0215+-0.4027 getter-richards 132.0611+-1.5368 ? 132.3125+-3.5260 ? getter 7.2862+-0.1544 7.2575+-0.1168 global-object-access-with-mutating-structure 8.7153+-0.3302 8.3966+-0.6224 might be 1.0380x faster global-var-const-infer-fire-from-opt 1.4174+-0.1257 1.3477+-0.2471 might be 1.0517x faster global-var-const-infer 1.0780+-0.1889 ? 1.1154+-0.1466 ? might be 1.0347x slower HashMap-put-get-iterate-keys 33.5554+-0.7109 ? 33.5730+-0.3699 ? HashMap-put-get-iterate 32.8280+-0.3127 ? 32.8465+-0.2581 ? HashMap-string-put-get-iterate 37.2803+-1.5704 35.9612+-0.9820 might be 1.0367x faster hoist-make-rope 16.0734+-2.2169 ? 16.9504+-2.1026 ? might be 1.0546x slower hoist-poly-check-structure-effectful-loop 7.0628+-0.0791 7.0062+-0.1961 hoist-poly-check-structure 4.8928+-0.2016 ? 4.9135+-0.1598 ? imul-double-only 10.4675+-0.8363 ? 11.2904+-0.9380 ? might be 1.0786x slower imul-int-only 12.6352+-0.6035 12.6135+-0.7586 imul-mixed 10.6093+-0.2451 ? 10.9760+-0.4745 ? might be 1.0346x slower in-four-cases 27.5613+-0.3486 27.5470+-0.1757 in-one-case-false 14.7556+-0.2779 14.5455+-0.3250 might be 1.0144x faster in-one-case-true 14.5605+-0.2746 ? 14.6133+-0.1683 ? in-two-cases 15.3001+-0.2979 ? 15.4854+-0.1410 ? might be 1.0121x slower indexed-properties-in-objects 3.7530+-0.1958 ? 3.7867+-0.0916 ? infer-closure-const-then-mov-no-inline 4.5552+-0.1658 4.4617+-0.1518 might be 1.0210x faster infer-closure-const-then-mov 25.6282+-0.6965 25.5356+-0.4920 infer-closure-const-then-put-to-scope-no-inline 14.8829+-0.2470 14.7369+-0.1857 infer-closure-const-then-put-to-scope 28.0590+-0.2565 ? 28.1827+-0.8191 ? infer-closure-const-then-reenter-no-inline 71.6585+-0.9315 71.5441+-0.3208 infer-closure-const-then-reenter 28.3509+-0.2362 ? 28.6378+-0.3355 ? might be 1.0101x slower infer-constant-global-property 4.8878+-0.2838 4.7686+-0.1509 might be 1.0250x faster infer-constant-property 3.2859+-0.1009 ? 3.3238+-0.0871 ? might be 1.0115x slower infer-one-time-closure-ten-vars 14.6336+-1.1479 14.5790+-0.8828 infer-one-time-closure-two-vars 14.4915+-0.7028 14.1931+-0.4429 might be 1.0210x faster infer-one-time-closure 14.7592+-0.6940 ? 14.8199+-0.3540 ? infer-one-time-deep-closure 25.3856+-0.1929 25.1282+-0.2910 might be 1.0102x faster inline-arguments-access 5.8500+-0.4612 ? 5.8824+-0.2444 ? inline-arguments-aliased-access 5.9479+-0.1476 ? 5.9579+-0.1923 ? inline-arguments-local-escape 5.9890+-0.2159 5.9855+-0.4742 inline-get-scoped-var 5.8372+-0.2177 5.7009+-0.0838 might be 1.0239x faster inlined-put-by-id-transition 17.2598+-0.3707 17.0693+-0.8540 might be 1.0112x faster int-or-other-abs-then-get-by-val 6.9661+-0.0644 6.9592+-0.2572 int-or-other-abs-zero-then-get-by-val 27.6118+-0.4366 27.4520+-0.8042 int-or-other-add-then-get-by-val 5.9312+-0.1685 ? 6.1255+-0.1639 ? might be 1.0328x slower int-or-other-add 8.0950+-0.1849 8.0850+-0.1023 int-or-other-div-then-get-by-val 5.3392+-0.0792 5.3247+-0.1693 int-or-other-max-then-get-by-val 5.3975+-0.0527 5.3712+-0.1339 int-or-other-min-then-get-by-val 5.2209+-0.0749 ? 5.2571+-0.0123 ? int-or-other-mod-then-get-by-val 5.2366+-0.1751 5.1975+-0.0816 int-or-other-mul-then-get-by-val 5.1489+-0.1549 5.1454+-0.0817 int-or-other-neg-then-get-by-val 6.5280+-0.0678 6.4048+-0.1053 might be 1.0192x faster int-or-other-neg-zero-then-get-by-val 27.9598+-1.0573 27.4907+-0.2768 might be 1.0171x faster int-or-other-sub-then-get-by-val 6.0895+-0.1239 6.0663+-0.2082 int-or-other-sub 4.8351+-0.2339 ? 4.8837+-0.1231 ? might be 1.0101x slower int-overflow-local 5.8470+-0.1376 ? 6.0366+-0.6931 ? might be 1.0324x slower Int16Array-alloc-long-lived 65.1121+-0.8853 64.9979+-1.1488 Int16Array-bubble-sort-with-byteLength 38.0549+-1.0541 ? 38.2034+-1.2812 ? Int16Array-bubble-sort 37.6967+-0.4592 37.5710+-0.4605 Int16Array-load-int-mul 2.0907+-0.1680 2.0167+-0.0937 might be 1.0367x faster Int16Array-to-Int32Array-set 72.5335+-1.4316 72.4548+-1.5258 Int32Array-alloc-large 33.5204+-2.1511 33.2973+-0.6012 Int32Array-alloc-long-lived 74.6233+-1.9901 73.9258+-1.3238 Int32Array-alloc 4.7817+-0.5218 4.5452+-0.2219 might be 1.0520x faster Int32Array-Int8Array-view-alloc 9.2092+-0.1216 ? 9.2820+-0.6863 ? int52-spill 6.8708+-0.2395 6.8447+-0.4100 Int8Array-alloc-long-lived 57.6243+-0.9129 57.5596+-0.9197 Int8Array-load-with-byteLength 4.7959+-0.1344 4.7903+-0.1777 Int8Array-load 4.8262+-0.0853 4.7898+-0.1565 integer-divide 13.5342+-0.5166 13.5288+-0.2657 integer-modulo 2.8031+-0.2086 ? 2.8085+-0.2523 ? is-boolean-fold-tricky 5.8223+-0.1483 5.8112+-0.1002 is-boolean-fold 3.9602+-0.0909 ? 4.0553+-0.2242 ? might be 1.0240x slower is-function-fold-tricky-internal-function 14.8935+-1.0499 ? 15.0900+-1.0239 ? might be 1.0132x slower is-function-fold-tricky 5.9641+-0.0830 ? 5.9827+-0.2145 ? is-function-fold 4.0333+-0.1328 ? 4.0934+-0.1103 ? might be 1.0149x slower is-number-fold-tricky 5.8030+-0.1511 5.7949+-0.2084 is-number-fold 3.9904+-0.2336 ? 4.0317+-0.1055 ? might be 1.0104x slower is-object-or-null-fold-functions 4.2217+-0.1751 4.1281+-0.1735 might be 1.0227x faster is-object-or-null-fold-less-tricky 6.0062+-0.0782 5.9191+-0.2302 might be 1.0147x faster is-object-or-null-fold-tricky 7.7667+-0.1416 7.7462+-0.1164 is-object-or-null-fold 4.0212+-0.0839 ? 4.1353+-0.0708 ? might be 1.0284x slower is-object-or-null-trickier-function 6.0673+-0.1468 6.0487+-0.1671 is-object-or-null-trickier-internal-function 15.6751+-2.3187 15.1358+-0.0881 might be 1.0356x faster is-object-or-null-tricky-function 5.9968+-0.1564 ? 6.0485+-0.0739 ? is-object-or-null-tricky-internal-function 11.4572+-0.9460 11.3889+-0.0788 is-string-fold-tricky 5.8289+-0.1140 5.8242+-0.1477 is-string-fold 3.9767+-0.1623 ? 4.0387+-0.0869 ? might be 1.0156x slower is-undefined-fold-tricky 4.5299+-0.1682 ? 4.5640+-0.1095 ? is-undefined-fold 4.0998+-0.0728 4.0361+-0.0185 might be 1.0158x faster JSONP-negative-0 0.4077+-0.0999 ? 0.4131+-0.1058 ? might be 1.0133x slower large-int-captured 6.8360+-0.1693 6.7210+-0.2996 might be 1.0171x faster large-int-neg 21.2775+-0.2361 ? 21.5086+-0.5034 ? might be 1.0109x slower large-int 19.3978+-0.4221 19.2560+-0.7604 load-varargs-elimination 30.5300+-1.0810 ? 32.9486+-7.1416 ? might be 1.0792x slower logical-not-weird-types 4.9949+-0.1023 ? 5.0725+-0.1853 ? might be 1.0155x slower logical-not 6.5330+-0.5789 6.1904+-0.1397 might be 1.0554x faster lots-of-fields 17.8360+-0.4898 ? 18.2853+-0.2093 ? might be 1.0252x slower make-indexed-storage 4.1973+-0.6226 ? 4.3390+-0.2378 ? might be 1.0338x slower make-rope-cse 6.3795+-0.2929 ? 6.5083+-0.2448 ? might be 1.0202x slower marsaglia-larger-ints 53.6818+-0.8781 ? 53.9678+-1.3525 ? marsaglia-osr-entry 28.9672+-0.7918 ? 29.1805+-0.5315 ? math-with-out-of-bounds-array-values 31.9413+-0.1630 ? 32.2396+-0.1519 ? max-boolean 3.3460+-0.0867 ? 3.3618+-0.1283 ? method-on-number 23.4180+-0.4877 ? 23.6349+-1.1880 ? min-boolean 3.3389+-0.1112 ? 3.3865+-0.2082 ? might be 1.0143x slower minus-boolean-double 4.4142+-0.5865 4.2569+-0.1090 might be 1.0370x faster minus-boolean 3.3251+-0.1084 3.2740+-0.0791 might be 1.0156x faster misc-strict-eq 56.0356+-1.1150 ? 56.7390+-2.5648 ? might be 1.0126x slower mod-boolean-double 11.8713+-0.1148 11.6815+-0.2303 might be 1.0162x faster mod-boolean 8.7926+-0.1691 ? 8.8211+-0.0791 ? mul-boolean-double 4.9789+-0.1572 4.8437+-0.1000 might be 1.0279x faster mul-boolean 3.5997+-0.1455 3.5095+-0.1243 might be 1.0257x faster neg-boolean 4.3529+-0.1378 ? 4.4579+-0.1597 ? might be 1.0241x slower negative-zero-divide 0.5853+-0.1294 0.5838+-0.1161 negative-zero-modulo 0.5977+-0.1097 0.5197+-0.0967 might be 1.1500x faster negative-zero-negate 0.5341+-0.1058 0.5035+-0.1102 might be 1.0609x faster nested-function-parsing 71.2968+-1.2420 71.2443+-1.2230 new-array-buffer-dead 146.5917+-2.1576 ? 148.3995+-2.7854 ? might be 1.0123x slower new-array-buffer-push 9.9073+-0.1162 9.7909+-0.3816 might be 1.0119x faster new-array-dead 25.8542+-1.4003 25.7987+-4.4359 new-array-push 5.3424+-0.4147 ? 5.5535+-0.2305 ? might be 1.0395x slower no-inline-constructor 50.8151+-1.2009 ? 51.1027+-1.0621 ? number-test 4.2703+-0.1234 ? 4.3605+-0.0821 ? might be 1.0211x slower object-closure-call 7.9652+-0.3756 7.8865+-0.2630 object-get-own-property-symbols-on-large-array 5.0026+-0.3170 4.9579+-0.6391 object-test 4.1442+-0.2170 ? 4.3013+-0.1065 ? might be 1.0379x slower obvious-sink-pathology-taken 176.7546+-4.8419 174.1758+-1.1313 might be 1.0148x faster obvious-sink-pathology 159.4089+-2.0055 ? 161.3255+-3.1990 ? might be 1.0120x slower obviously-elidable-new-object 51.1613+-1.2211 ? 51.2842+-1.1859 ? plus-boolean-arith 3.3473+-0.0711 3.3409+-0.0574 plus-boolean-double 4.2972+-0.1395 4.2307+-0.1639 might be 1.0157x faster plus-boolean 3.3223+-0.2221 3.2604+-0.1024 might be 1.0190x faster poly-chain-access-different-prototypes-simple 4.8985+-0.1158 4.8448+-0.0621 might be 1.0111x faster poly-chain-access-different-prototypes 4.4503+-0.1148 ? 4.4659+-0.1686 ? poly-chain-access-simpler 4.9628+-0.0838 4.9313+-0.0923 poly-chain-access 5.3068+-0.1142 5.2956+-0.0592 poly-stricteq 76.8267+-1.1512 76.0848+-0.4150 polymorphic-array-call 1.7773+-0.2478 ? 2.0270+-0.1608 ? might be 1.1405x slower polymorphic-get-by-id 4.1860+-0.1253 ? 4.2405+-0.0842 ? might be 1.0130x slower polymorphic-put-by-id 41.3494+-4.0694 41.0753+-3.3648 polymorphic-structure 23.4633+-0.2322 ? 23.4886+-0.3526 ? polyvariant-monomorphic-get-by-id 11.6137+-0.2547 11.5886+-0.2014 proto-getter-access 11.9475+-0.8872 ? 12.0590+-0.1658 ? prototype-access-with-mutating-prototype 8.3276+-0.4275 8.1007+-0.1082 might be 1.0280x faster put-by-id-replace-and-transition 12.7149+-0.5136 12.3842+-0.3897 might be 1.0267x faster put-by-id-slightly-polymorphic 3.6617+-0.2835 3.5425+-0.1349 might be 1.0336x faster put-by-id 18.2053+-0.4786 18.0837+-0.5460 put-by-val-direct 0.5677+-0.0934 ? 0.5967+-0.1105 ? might be 1.0511x slower put-by-val-large-index-blank-indexing-type 8.6953+-0.7273 ? 8.9435+-0.8237 ? might be 1.0285x slower put-by-val-machine-int 3.4595+-0.2861 ? 3.4799+-0.2068 ? rare-osr-exit-on-local 18.5200+-0.0614 18.1951+-0.6399 might be 1.0179x faster register-pressure-from-osr 26.6147+-0.7066 ? 26.6753+-0.4447 ? repeat-multi-get-by-offset 29.4469+-0.2531 ? 29.4535+-0.6374 ? setter-prototype 13.1982+-0.7287 ? 13.5646+-1.2188 ? might be 1.0278x slower setter 7.5005+-0.5115 7.3199+-0.1835 might be 1.0247x faster simple-activation-demo 34.6340+-1.0875 34.3439+-0.6341 simple-getter-access 15.7452+-0.5904 15.5635+-0.4802 might be 1.0117x faster simple-poly-call-nested 9.0623+-0.0959 ? 9.1170+-0.1691 ? simple-poly-call 1.6585+-0.0576 1.6577+-0.0966 sin-boolean 27.8470+-1.6245 ? 29.3789+-1.0582 ? might be 1.0550x slower singleton-scope 80.3478+-0.4796 ? 81.0616+-1.0841 ? sink-function 15.9830+-0.7611 ? 16.0695+-0.9293 ? sink-huge-activation 22.7417+-0.6455 ? 22.9886+-0.8724 ? might be 1.0109x slower sinkable-new-object-dag 93.6950+-2.8021 92.2120+-2.8376 might be 1.0161x faster sinkable-new-object-taken 70.5927+-4.0143 ? 73.9823+-1.9807 ? might be 1.0480x slower sinkable-new-object 54.0338+-1.0998 ? 54.3998+-1.0238 ? slow-array-profile-convergence 4.1617+-0.2578 3.9125+-0.1634 might be 1.0637x faster slow-convergence 3.7563+-0.1554 ? 3.8419+-0.1855 ? might be 1.0228x slower slow-ternaries 31.3099+-0.5119 30.5288+-0.5804 might be 1.0256x faster sorting-benchmark 25.6982+-0.7340 ? 26.0173+-0.4139 ? might be 1.0124x slower sparse-conditional 1.7617+-0.0591 1.7247+-0.0674 might be 1.0215x faster splice-to-remove 19.3740+-0.4099 ? 19.4282+-0.3897 ? string-char-code-at 20.2340+-0.3546 ? 20.4597+-0.4211 ? might be 1.0112x slower string-concat-object 3.0804+-0.1520 ? 3.1559+-0.2780 ? might be 1.0245x slower string-concat-pair-object 3.1973+-0.3670 ? 3.2776+-0.5712 ? might be 1.0251x slower string-concat-pair-simple 16.4532+-0.2753 ? 16.8273+-0.4268 ? might be 1.0227x slower string-concat-simple 17.1302+-0.2733 17.0754+-0.2276 string-cons-repeat 10.6773+-0.3595 ? 11.0948+-0.4536 ? might be 1.0391x slower string-cons-tower 10.2848+-0.1581 ? 10.3610+-0.3745 ? string-equality 21.7126+-0.1496 ? 21.9532+-0.3876 ? might be 1.0111x slower string-get-by-val-big-char 10.0490+-0.1254 10.0023+-0.4743 string-get-by-val-out-of-bounds-insane 5.2590+-0.2229 ? 5.3862+-0.1892 ? might be 1.0242x slower string-get-by-val-out-of-bounds 7.1464+-0.1624 ? 7.2068+-0.2292 ? string-get-by-val 4.7917+-0.0569 ? 4.8239+-0.1096 ? string-hash 2.7605+-0.2240 ? 2.7938+-0.1950 ? might be 1.0121x slower string-long-ident-equality 17.8384+-0.2092 ? 17.8799+-0.2846 ? string-out-of-bounds 18.3049+-0.4735 18.1622+-0.4944 string-repeat-arith 42.9844+-0.6000 ? 43.5773+-0.3102 ? might be 1.0138x slower string-sub 83.1382+-1.7331 ? 83.6303+-1.3284 ? string-test 4.3893+-0.1275 4.3191+-0.1713 might be 1.0163x faster string-var-equality 58.0057+-2.1510 57.7563+-1.3119 structure-hoist-over-transitions 3.7261+-0.1110 3.5363+-0.2063 might be 1.0537x faster substring-concat-weird 57.8778+-0.6721 57.5803+-0.4848 substring-concat 62.9575+-1.2210 62.8000+-0.8354 substring 68.9705+-1.0684 ? 69.3724+-1.1392 ? switch-char-constant 3.4149+-0.1785 3.4055+-0.0449 switch-char 7.9844+-0.1109 7.9273+-0.0954 switch-constant 10.8530+-0.2543 ? 10.9650+-0.4122 ? might be 1.0103x slower switch-string-basic-big-var 32.4445+-1.1210 ? 32.8615+-0.2657 ? might be 1.0129x slower switch-string-basic-big 24.0043+-2.6799 23.9440+-0.7564 switch-string-basic-var 31.1492+-0.7383 ? 31.6539+-0.8241 ? might be 1.0162x slower switch-string-basic 19.4895+-0.2728 ? 19.6692+-0.3980 ? switch-string-big-length-tower-var 27.8477+-0.2533 ? 27.9973+-0.4179 ? switch-string-length-tower-var 21.0109+-0.5697 20.4600+-0.1604 might be 1.0269x faster switch-string-length-tower 15.0373+-0.2367 14.8991+-0.4807 switch-string-short 15.4006+-1.0349 14.9617+-0.3680 might be 1.0293x faster switch 15.2067+-0.4234 ? 15.2610+-0.6598 ? tear-off-arguments-simple 4.3414+-0.2207 ? 4.4305+-0.2040 ? might be 1.0205x slower tear-off-arguments 6.1068+-0.2501 ? 6.2754+-0.5947 ? might be 1.0276x slower temporal-structure 17.2863+-0.1451 ? 17.7263+-0.8057 ? might be 1.0255x slower to-int32-boolean 20.5172+-0.2551 20.3916+-0.0693 try-catch-get-by-val-cloned-arguments 20.5646+-0.5122 ? 21.1485+-0.5862 ? might be 1.0284x slower try-catch-get-by-val-direct-arguments 9.3717+-0.4897 9.2588+-0.1561 might be 1.0122x faster try-catch-get-by-val-scoped-arguments 10.8785+-0.3245 ? 10.9254+-0.3944 ? typed-array-get-set-by-val-profiling 38.7723+-2.6927 36.8466+-4.0883 might be 1.0523x faster undefined-property-access 457.2129+-1.7109 ? 457.6590+-3.8175 ? undefined-test 4.4434+-0.1290 4.4350+-0.1362 unprofiled-licm 28.5782+-1.3192 ? 28.7718+-0.7783 ? varargs-call 18.0337+-0.5452 17.7856+-0.2607 might be 1.0139x faster varargs-construct-inline 37.4577+-0.2933 ? 37.5870+-0.2524 ? varargs-construct 30.5790+-0.6985 ? 30.7068+-0.4164 ? varargs-inline 11.4510+-0.2759 11.2527+-0.1883 might be 1.0176x faster varargs-strict-mode 13.2001+-0.1470 13.1055+-0.3059 varargs 13.3032+-0.2276 ? 13.3486+-0.2960 ? weird-inlining-const-prop 3.0400+-0.2184 ? 3.0494+-0.3200 ? <geometric> 11.5936+-0.0241 ? 11.6044+-0.0436 ? might be 1.0009x slower Conf#1 Conf#2 AsmBench: bigfib.cpp 675.1662+-20.0359 ? 687.8543+-8.8076 ? might be 1.0188x slower cray.c 630.7591+-5.3183 630.0542+-4.3589 dry.c 649.3080+-27.0975 ? 657.5900+-1.6403 ? might be 1.0128x slower FloatMM.c 914.6072+-1.9265 ? 914.8857+-1.5510 ? gcc-loops.cpp 5915.6310+-19.7154 5902.3281+-18.0942 n-body.c 1673.1592+-1.8106 ? 1673.2680+-5.1556 ? Quicksort.c 573.3643+-4.0996 568.9533+-8.8464 stepanov_container.cpp 4906.5717+-70.8814 ? 4907.2338+-36.5477 ? Towers.c 381.1053+-5.3566 ? 385.2407+-3.0688 ? might be 1.0109x slower <geometric> 1115.8817+-6.2704 ? 1119.8371+-1.8141 ? might be 1.0035x slower Conf#1 Conf#2 CompressionBench: huffman 79.1904+-3.0202 78.7522+-2.5757 arithmetic-simple 432.7852+-1.1907 ? 433.1183+-2.0611 ? arithmetic-precise 362.8991+-2.4463 ? 365.2471+-2.1638 ? arithmetic-complex-precise 362.5283+-0.9336 361.9175+-6.2496 arithmetic-precise-order-0 446.4127+-0.9046 445.4664+-1.0449 arithmetic-precise-order-1 424.2969+-1.7938 ? 426.9127+-1.9588 ? arithmetic-precise-order-2 498.8552+-10.4078 492.1649+-7.7838 might be 1.0136x faster arithmetic-simple-order-1 504.3336+-1.1539 ? 504.3896+-1.8974 ? arithmetic-simple-order-2 575.7477+-12.0423 569.9174+-1.6031 might be 1.0102x faster lz-string 418.5870+-7.7481 411.3253+-6.1849 might be 1.0177x faster <geometric> 372.7365+-1.1920 371.3632+-2.0338 might be 1.0037x faster Conf#1 Conf#2 Geomean of preferred means: <scaled-result> 79.5155+-0.1377 79.4912+-0.2766 might be 1.0003x faster

Geoffrey Garen

Comment 3 2015-08-17 11:42:19 PDT

Comment on attachment 259056 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=259056&action=review > Source/JavaScriptCore/ChangeLog:8 > + Comparing null and undefined to themself or eachoter is always true. themselves or each other

Filip Pizlo

Comment 4 2015-08-17 11:54:07 PDT

Comment on attachment 259056 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=259056&action=review I don't completely object to this optimization being done in FixupPhase, but it seems like the kind of thing that belongs in AbstractInterpreter so that it can fully participate in the constant folding fixpoint. > Source/JavaScriptCore/dfg/DFGFixupPhase.cpp:379 > + if (node->child1()->shouldSpeculateOther() && node->child2()->shouldSpeculateOther()) { > + m_insertionSet.insertNode(m_indexInBlock, SpecNone, Check, node->origin, > + Edge(node->child1().node(), OtherUse)); > + m_insertionSet.insertNode(m_indexInBlock, SpecNone, Check, node->origin, > + Edge(node->child2().node(), OtherUse)); > + m_graph.convertToConstant(node, jsBoolean(true)); > + } Why not do this in AbstractInterpreter? > Source/JavaScriptCore/dfg/DFGFixupPhase.cpp:453 > + if (node->child1()->shouldSpeculateOther() && node->child2()->shouldSpeculateOther()) { > + m_insertionSet.insertNode(m_indexInBlock, SpecNone, Check, node->origin, > + Edge(node->child1().node(), OtherUse)); > + m_insertionSet.insertNode(m_indexInBlock, SpecNone, Check, node->origin, > + Edge(node->child2().node(), OtherUse)); > + m_graph.convertToConstant(node, jsBoolean(true)); > + break; > + } Ditto.

Benjamin Poulain

Comment 5 2015-08-18 21:05:53 PDT

*** This bug has been marked as a duplicate of bug 148157 ***

Attachments
Patch (8.43 KB, patch) 2015-08-14 16:27 PDT, Benjamin Poulain	ggaren: review+	Details Formatted Diff Diff
View All Add attachment proposed patch, testcase, etc.