144897 – [JSC] Make the NegZero backward propagated flags of ArithMod stricter

Benjamin Poulain

Reported 2015-05-11 21:48:02 PDT

[JSC] Make the NegZero backward propagated flags of ArithMod stricter

Benjamin Poulain

Comment 1 2015-05-11 21:57:49 PDT

Created attachment 252934 [details] Patch

Benjamin Poulain

Comment 2 2015-05-11 22:00:02 PDT

Conf#1 Conf#2 SunSpider: 3d-cube 7.3096+-0.1934 ? 7.4294+-0.2474 ? might be 1.0164x slower 3d-morph 7.8223+-0.0646 ? 7.8898+-0.0493 ? 3d-raytrace 8.7737+-0.2719 ? 8.9476+-0.2205 ? might be 1.0198x slower access-binary-trees 3.1445+-0.1413 3.1245+-0.1169 access-fannkuch 8.2737+-0.1730 ? 8.2788+-0.1923 ? access-nbody 4.1562+-0.1739 4.0845+-0.2074 might be 1.0175x faster access-nsieve 4.3913+-0.2465 ? 4.4570+-0.1088 ? might be 1.0150x slower bitops-3bit-bits-in-byte 1.9205+-0.1489 ? 1.9504+-0.0978 ? might be 1.0156x slower bitops-bits-in-byte 5.6562+-0.1028 ? 5.6813+-0.2050 ? bitops-bitwise-and 2.7701+-0.2415 2.7650+-0.3168 bitops-nsieve-bits 4.3029+-0.0962 ? 4.3074+-0.1332 ? controlflow-recursive 2.9368+-0.1381 2.8655+-0.0571 might be 1.0249x faster crypto-aes 5.9152+-0.2921 ? 6.1016+-0.2937 ? might be 1.0315x slower crypto-md5 3.7109+-0.2635 ? 3.7811+-0.0340 ? might be 1.0189x slower crypto-sha1 3.2431+-0.2131 ? 3.3232+-0.1786 ? might be 1.0247x slower date-format-tofte 12.2433+-0.5438 12.0262+-0.5669 might be 1.0181x faster date-format-xparb 7.5975+-0.4948 7.5175+-0.4549 might be 1.0106x faster math-cordic 4.3057+-0.1172 ? 4.3842+-0.1195 ? might be 1.0182x slower math-partial-sums 9.2627+-0.1811 9.2128+-0.1184 math-spectral-norm 2.9246+-0.2446 ? 2.9700+-0.1551 ? might be 1.0155x slower regexp-dna 9.5955+-0.7898 ? 9.6619+-0.2420 ? string-base64 6.1765+-0.2599 6.0321+-0.2308 might be 1.0239x faster string-fasta 9.3511+-0.3013 9.1705+-0.1365 might be 1.0197x faster string-tagcloud 12.8707+-0.5449 12.5853+-0.4704 might be 1.0227x faster string-unpack-code 27.6194+-0.6628 26.5349+-0.8257 might be 1.0409x faster string-validate-input 6.6133+-0.2757 ? 6.7636+-0.3193 ? might be 1.0227x slower <arithmetic> 7.0341+-0.0320 6.9941+-0.0477 might be 1.0057x faster Conf#1 Conf#2 LongSpider: 3d-cube 1263.8520+-16.7950 1258.0325+-16.6455 3d-morph 1872.7726+-0.4463 ? 1873.0650+-2.3307 ? 3d-raytrace 1044.6783+-7.6353 ? 1047.7736+-10.7386 ? access-binary-trees 1356.6971+-5.0869 1352.9928+-11.1932 access-fannkuch 454.7128+-21.0587 ? 463.3568+-33.1324 ? might be 1.0190x slower access-nbody 1037.6741+-1.2262 1034.9919+-2.1494 access-nsieve 675.0123+-8.0709 674.8629+-5.3561 bitops-3bit-bits-in-byte 50.5023+-1.1438 50.4409+-1.3348 bitops-bits-in-byte 345.6577+-6.4050 ? 347.4278+-4.9835 ? bitops-nsieve-bits 630.5558+-3.3914 630.3899+-1.5603 controlflow-recursive 709.8051+-1.1608 ^ 703.7598+-0.5389 ^ definitely 1.0086x faster crypto-aes 930.4857+-5.5600 ? 939.6304+-6.7545 ? crypto-md5 675.5173+-7.6251 ? 682.2261+-3.1582 ? crypto-sha1 908.8090+-27.0588 ? 910.1366+-5.9269 ? date-format-tofte 993.2455+-9.9953 976.8732+-18.2429 might be 1.0168x faster date-format-xparb 1020.8379+-19.9226 ? 1047.8403+-24.3783 ? might be 1.0265x slower math-cordic 672.7300+-1.5563 ? 673.2540+-1.3231 ? math-partial-sums 1065.7282+-4.0977 ? 1066.7860+-2.8594 ? math-spectral-norm 1073.0081+-2.6639 1072.7660+-1.5614 string-base64 489.3131+-6.4905 485.0983+-3.6640 string-fasta 609.2489+-2.6731 605.3463+-10.4256 string-tagcloud 276.3430+-3.0015 275.7495+-2.5209 <geometric> 692.2664+-2.8553 ? 692.9399+-2.7121 ? might be 1.0010x slower Conf#1 Conf#2 V8Spider: crypto 72.3904+-1.7558 ? 72.9924+-1.3246 ? deltablue 92.0757+-4.5451 89.1080+-1.4567 might be 1.0333x faster earley-boyer 63.2979+-0.6869 ? 63.7155+-1.8153 ? raytrace 40.5826+-0.9132 39.5854+-0.8786 might be 1.0252x faster regexp 110.5294+-0.4430 110.4039+-0.8382 richards 99.5128+-1.3822 97.9828+-5.5051 might be 1.0156x faster splay 51.7552+-5.7511 51.2334+-2.1943 might be 1.0102x faster <geometric> 71.6816+-0.7101 70.9862+-0.2728 might be 1.0098x faster Conf#1 Conf#2 Octane: encrypt 0.30782+-0.00096 ? 0.30784+-0.00037 ? decrypt 5.64357+-0.00563 ? 5.65390+-0.00746 ? deltablue x2 0.25494+-0.00270 ? 0.26163+-0.01148 ? might be 1.0263x slower earley 0.55129+-0.01116 0.55128+-0.00549 boyer 8.09976+-0.08684 ? 8.11254+-0.06392 ? navier-stokes x2 6.38303+-0.00400 ? 6.38445+-0.01375 ? raytrace x2 1.76462+-0.07431 ? 1.79508+-0.20268 ? might be 1.0173x slower richards x2 0.16467+-0.00166 ? 0.16481+-0.00479 ? splay x2 0.54832+-0.00965 0.54819+-0.00338 regexp x2 42.36274+-0.47627 42.27607+-0.38783 pdfjs x2 61.11648+-0.54304 ? 61.31600+-0.24800 ? mandreel x2 69.57252+-0.49530 ? 70.78021+-1.31388 ? might be 1.0174x slower gbemu x2 59.39880+-0.36790 ? 59.54133+-2.16785 ? closure 0.75914+-0.00074 0.75883+-0.00199 jquery 9.45365+-0.02050 ^ 9.38507+-0.02534 ^ definitely 1.0073x faster box2d x2 17.23735+-0.16354 ? 17.28385+-0.26463 ? zlib x2 548.20516+-14.80934 542.06260+-5.99452 might be 1.0113x faster typescript x2 1087.14520+-12.50388 1067.83325+-33.79316 might be 1.0181x faster <geometric> 9.01903+-0.01935 ? 9.03954+-0.07854 ? might be 1.0023x slower Conf#1 Conf#2 Kraken: ai-astar 484.160+-1.340 477.647+-13.454 might be 1.0136x faster audio-beat-detection 148.758+-1.874 148.038+-0.991 audio-dft 144.341+-2.582 142.427+-2.456 might be 1.0134x faster audio-fft 111.150+-0.947 ? 111.227+-0.954 ? audio-oscillator 99.810+-1.565 ^ 96.983+-0.107 ^ definitely 1.0292x faster imaging-darkroom 139.431+-0.164 ? 140.519+-3.197 ? imaging-desaturate 94.679+-0.801 94.378+-0.360 imaging-gaussian-blur 150.356+-11.917 147.428+-0.788 might be 1.0199x faster json-parse-financial 66.180+-1.782 66.049+-1.211 json-stringify-tinderbox 43.857+-0.565 43.548+-0.118 stanford-crypto-aes 86.467+-1.827 ? 87.066+-0.930 ? stanford-crypto-ccm 79.530+-3.361 ? 80.448+-5.336 ? might be 1.0115x slower stanford-crypto-pbkdf2 179.545+-2.243 178.703+-0.884 stanford-crypto-sha256-iterative 67.883+-1.086 66.730+-0.587 might be 1.0173x faster <arithmetic> 135.439+-1.324 134.371+-1.134 might be 1.0079x faster Conf#1 Conf#2 JSRegress: abs-boolean 3.7945+-0.1032 3.7609+-0.1207 adapt-to-double-divide 17.8882+-0.8635 ? 18.2075+-0.7726 ? might be 1.0178x slower aliased-arguments-getbyval 1.6333+-0.2417 1.5437+-0.2311 might be 1.0580x faster allocate-big-object 3.7910+-0.7192 ? 3.9800+-0.6881 ? might be 1.0499x slower arguments-named-and-reflective 14.1133+-0.2356 13.8932+-0.3077 might be 1.0158x faster arguments-out-of-bounds 17.9368+-0.5696 ? 18.0964+-0.1997 ? arguments-strict-mode 12.2035+-0.9827 ? 12.3861+-0.8032 ? might be 1.0150x slower arguments 10.6924+-0.2066 ? 10.7538+-0.3913 ? arity-mismatch-inlining 1.1474+-0.1422 1.1028+-0.0813 might be 1.0405x faster array-access-polymorphic-structure 9.4302+-0.4946 ? 9.4593+-0.2707 ? array-nonarray-polymorhpic-access 47.3547+-3.6726 ? 48.9702+-0.9082 ? might be 1.0341x slower array-prototype-every 116.2148+-4.3806 ? 118.4680+-2.6570 ? might be 1.0194x slower array-prototype-forEach 116.3635+-3.8296 114.7291+-0.8026 might be 1.0142x faster array-prototype-map 129.2472+-3.1354 128.4584+-3.0453 array-prototype-some 116.4114+-3.0720 ? 116.5173+-1.4461 ? array-splice-contiguous 57.5156+-0.1264 ? 57.5730+-0.7608 ? array-with-double-add 5.6365+-0.1071 5.5610+-0.1842 might be 1.0136x faster array-with-double-increment 4.1852+-0.2282 ? 4.2421+-0.3225 ? might be 1.0136x slower array-with-double-mul-add 7.9681+-0.2136 ^ 7.0914+-0.1705 ^ definitely 1.1236x faster array-with-double-sum 4.3299+-0.1982 ? 4.4008+-0.1490 ? might be 1.0164x slower array-with-int32-add-sub 9.6363+-0.2365 ? 9.6687+-0.1619 ? array-with-int32-or-double-sum 4.4898+-0.1394 4.4832+-0.1030 ArrayBuffer-DataView-alloc-large-long-lived 46.3905+-0.8681 46.2858+-1.3664 ArrayBuffer-DataView-alloc-long-lived 18.3890+-0.3027 ? 18.5048+-0.1385 ? ArrayBuffer-Int32Array-byteOffset 5.3640+-0.1285 5.2371+-0.1392 might be 1.0242x faster ArrayBuffer-Int8Array-alloc-large-long-lived 49.6100+-2.2123 48.4280+-1.0889 might be 1.0244x faster ArrayBuffer-Int8Array-alloc-long-lived-buffer 30.9250+-0.4660 30.7578+-0.6599 ArrayBuffer-Int8Array-alloc-long-lived 17.8150+-0.2555 17.6857+-0.1300 ArrayBuffer-Int8Array-alloc 14.8312+-0.1639 ? 16.8157+-3.7096 ? might be 1.1338x slower asmjs_bool_bug 8.4252+-0.1348 ? 8.5947+-0.1752 ? might be 1.0201x slower assign-custom-setter-polymorphic 4.2513+-0.0812 ? 4.3137+-0.0796 ? might be 1.0147x slower assign-custom-setter 5.8245+-0.0516 5.7598+-0.1950 might be 1.0112x faster basic-set 11.2040+-0.7951 ? 11.3455+-1.2735 ? might be 1.0126x slower big-int-mul 5.9077+-0.1321 5.8718+-0.0889 boolean-test 4.2803+-0.1536 ? 4.3239+-0.1908 ? might be 1.0102x slower branch-fold 4.7489+-0.0434 ? 4.7905+-0.1771 ? branch-on-string-as-boolean 22.7600+-1.0812 22.4290+-0.7247 might be 1.0148x faster by-val-generic 10.6552+-0.3140 ? 10.7367+-0.1239 ? call-spread-apply 39.6992+-1.6918 37.8203+-0.3794 might be 1.0497x faster call-spread-call 32.2085+-1.0318 ? 32.5010+-0.3889 ? captured-assignments 0.6558+-0.0795 0.5726+-0.0996 might be 1.1454x faster cast-int-to-double 8.2469+-0.0735 8.1087+-0.1924 might be 1.0170x faster cell-argument 9.8926+-0.2401 ? 10.0624+-0.5697 ? might be 1.0172x slower cfg-simplify 3.6884+-0.1047 ? 3.7254+-0.1975 ? might be 1.0100x slower chain-getter-access 11.5616+-0.6279 11.4583+-0.6201 cmpeq-obj-to-obj-other 12.2501+-0.2240 ? 12.4130+-0.4124 ? might be 1.0133x slower constant-test 7.9368+-0.2297 7.8431+-0.4481 might be 1.0120x faster create-lots-of-functions 16.4077+-0.4684 ? 16.8812+-0.9182 ? might be 1.0289x slower DataView-custom-properties 52.2923+-1.8012 ? 53.0933+-2.0597 ? might be 1.0153x slower deconstructing-parameters-overridden-by-function 0.7234+-0.1840 ? 0.7692+-0.1228 ? might be 1.0633x slower delay-tear-off-arguments-strictmode 18.4985+-0.2792 ? 18.7222+-0.2169 ? might be 1.0121x slower deltablue-varargs 272.4995+-1.7214 271.1607+-2.4495 destructuring-arguments 23.3422+-0.8125 22.6669+-0.7112 might be 1.0298x faster destructuring-swap 7.8783+-0.1406 7.7079+-0.1061 might be 1.0221x faster direct-arguments-getbyval 1.6253+-0.1951 ? 1.6573+-0.2450 ? might be 1.0197x slower div-boolean-double 5.5299+-0.1419 5.5225+-0.0937 div-boolean 9.7222+-0.2681 ? 9.7355+-0.2732 ? double-get-by-val-out-of-bounds 5.8448+-0.3663 ? 5.9610+-0.1857 ? might be 1.0199x slower double-pollution-getbyval 10.1550+-0.7941 9.6420+-0.2787 might be 1.0532x faster double-pollution-putbyoffset 5.8510+-0.3793 5.6558+-0.1200 might be 1.0345x faster double-to-int32-typed-array-no-inline 2.9271+-0.1443 2.8912+-0.0467 might be 1.0124x faster double-to-int32-typed-array 2.5756+-0.1010 2.4990+-0.1382 might be 1.0307x faster double-to-uint32-typed-array-no-inline 2.9664+-0.1345 ? 2.9725+-0.1348 ? double-to-uint32-typed-array 2.7798+-0.1491 2.7161+-0.0931 might be 1.0234x faster elidable-new-object-dag 55.1592+-1.8840 ? 56.4025+-1.9418 ? might be 1.0225x slower elidable-new-object-roflcopter 60.9514+-0.6353 ? 61.7096+-1.7594 ? might be 1.0124x slower elidable-new-object-then-call 52.1484+-1.3537 ? 52.9410+-4.6067 ? might be 1.0152x slower elidable-new-object-tree 63.5796+-1.6777 63.2422+-0.4494 empty-string-plus-int 7.5394+-0.2742 7.5372+-0.3050 emscripten-cube2hash 44.9832+-2.3722 44.5742+-0.6937 exit-length-on-plain-object 18.9510+-0.2680 18.9167+-0.4894 external-arguments-getbyval 1.5833+-0.1593 ? 1.7605+-0.4121 ? might be 1.1120x slower external-arguments-putbyval 3.2680+-0.2670 ? 3.4031+-0.1595 ? might be 1.0413x slower fixed-typed-array-storage-var-index 1.5685+-0.1420 1.5183+-0.0334 might be 1.0331x faster fixed-typed-array-storage 1.2510+-0.1791 1.2443+-0.0716 Float32Array-matrix-mult 5.8938+-0.2856 ? 6.0864+-0.2528 ? might be 1.0327x slower Float32Array-to-Float64Array-set 71.6873+-3.4572 ? 78.7935+-5.1995 ? might be 1.0991x slower Float64Array-alloc-long-lived 92.0365+-0.8079 91.3535+-0.5279 Float64Array-to-Int16Array-set 91.7327+-2.0636 ? 93.6941+-2.2655 ? might be 1.0214x slower fold-double-to-int 19.0328+-0.2925 18.9493+-0.3670 fold-get-by-id-to-multi-get-by-offset-rare-int 10.9141+-0.3555 10.8231+-0.4204 fold-get-by-id-to-multi-get-by-offset 9.1384+-0.6471 ? 9.2424+-0.3899 ? might be 1.0114x slower fold-multi-get-by-offset-to-get-by-offset 8.9278+-0.6263 7.9127+-0.8924 might be 1.1283x faster fold-multi-get-by-offset-to-poly-get-by-offset 8.3753+-0.4172 ? 8.3837+-0.5913 ? fold-multi-put-by-offset-to-poly-put-by-offset 7.8901+-0.1466 7.8257+-0.1426 fold-multi-put-by-offset-to-put-by-offset 5.9980+-0.6805 ? 6.1899+-0.8342 ? might be 1.0320x slower fold-multi-put-by-offset-to-replace-or-transition-put-by-offset 12.7966+-0.4903 ? 12.9080+-0.4897 ? fold-put-by-id-to-multi-put-by-offset 8.9373+-0.8319 ? 9.0664+-0.6103 ? might be 1.0144x slower fold-put-structure 6.3090+-0.7819 ? 6.4485+-0.7905 ? might be 1.0221x slower for-of-iterate-array-entries 6.4421+-0.2332 ? 6.5075+-0.0881 ? might be 1.0101x slower for-of-iterate-array-keys 4.9989+-0.1544 4.8893+-0.2570 might be 1.0224x faster for-of-iterate-array-values 4.8762+-0.3638 ? 4.9468+-0.2178 ? might be 1.0145x slower fround 23.5386+-1.0792 22.9845+-0.2049 might be 1.0241x faster ftl-library-inlining-dataview 100.9219+-2.4402 ^ 89.1573+-0.6838 ^ definitely 1.1320x faster ftl-library-inlining 87.6165+-10.2108 84.8242+-0.5256 might be 1.0329x faster function-dot-apply 2.7295+-0.3572 2.6504+-0.2114 might be 1.0298x faster function-test 4.3148+-0.2235 ? 4.3710+-0.0714 ? might be 1.0130x slower function-with-eval 135.0314+-2.0755 ? 135.1512+-2.9296 ? gcse-poly-get-less-obvious 24.7178+-0.4004 ? 24.8440+-0.4938 ? gcse-poly-get 24.6845+-0.5207 ? 24.7808+-0.6377 ? gcse 6.6791+-0.3603 6.5291+-0.1416 might be 1.0230x faster get-by-id-bimorphic-check-structure-elimination-simple 3.3361+-0.1340 3.2906+-0.1361 might be 1.0138x faster get-by-id-bimorphic-check-structure-elimination 8.3328+-0.3800 ? 8.4139+-0.1148 ? get-by-id-chain-from-try-block 11.4398+-0.0707 ? 11.4900+-0.0942 ? get-by-id-check-structure-elimination 7.6108+-0.3019 7.4026+-0.2078 might be 1.0281x faster get-by-id-proto-or-self 25.3595+-4.9000 23.2469+-1.9983 might be 1.0909x faster get-by-id-quadmorphic-check-structure-elimination-simple 4.0696+-0.0822 4.0540+-0.1658 get-by-id-self-or-proto 25.6520+-2.7374 25.3458+-2.7601 might be 1.0121x faster get-by-val-out-of-bounds 5.8560+-0.1009 ? 5.8619+-0.2035 ? get_callee_monomorphic 3.8079+-0.2955 3.7728+-0.2220 get_callee_polymorphic 4.6575+-0.1588 ? 4.7205+-0.3895 ? might be 1.0135x slower getter-no-activation 5.7689+-0.1171 5.7370+-0.0675 getter-richards 131.9443+-1.1034 130.6263+-3.5667 might be 1.0101x faster getter 7.2563+-0.1091 7.1871+-0.2076 global-var-const-infer-fire-from-opt 1.2357+-0.2769 ? 1.2534+-0.2881 ? might be 1.0144x slower global-var-const-infer 1.2004+-0.1161 1.1989+-0.1982 HashMap-put-get-iterate-keys 33.4813+-0.8561 33.0273+-0.7685 might be 1.0137x faster HashMap-put-get-iterate 32.9943+-0.4902 32.9052+-0.4417 HashMap-string-put-get-iterate 35.4672+-0.9759 35.4174+-1.4848 hoist-make-rope 14.0502+-1.3038 ? 14.4888+-1.0001 ? might be 1.0312x slower hoist-poly-check-structure-effectful-loop 6.9377+-0.1705 ? 7.0413+-0.1439 ? might be 1.0149x slower hoist-poly-check-structure 4.7935+-0.1787 ? 4.8484+-0.0641 ? might be 1.0115x slower imul-double-only 10.1744+-0.3366 9.7892+-2.1420 might be 1.0393x faster imul-int-only 12.0106+-1.4385 11.9690+-1.3316 imul-mixed 9.6106+-0.2018 ? 9.8112+-0.7798 ? might be 1.0209x slower in-four-cases 25.3843+-0.1192 ? 25.6031+-0.3216 ? in-one-case-false 13.1120+-0.2849 12.7961+-0.2232 might be 1.0247x faster in-one-case-true 13.1743+-0.2551 12.8223+-0.2953 might be 1.0275x faster in-two-cases 13.5348+-0.2292 13.3450+-0.2822 might be 1.0142x faster indexed-properties-in-objects 3.8690+-0.0740 ^ 3.7035+-0.0556 ^ definitely 1.0447x faster infer-closure-const-then-mov-no-inline 4.6185+-0.0301 ? 4.6240+-0.0853 ? infer-closure-const-then-mov 25.0302+-1.2156 24.9347+-0.8634 infer-closure-const-then-put-to-scope-no-inline 14.4721+-0.3668 14.3403+-0.1539 infer-closure-const-then-put-to-scope 27.8299+-0.1599 ? 27.9474+-0.2842 ? infer-closure-const-then-reenter-no-inline 61.3676+-0.7282 ? 61.5118+-0.8874 ? infer-closure-const-then-reenter 29.2018+-0.6622 ? 29.7175+-0.2820 ? might be 1.0177x slower infer-constant-global-property 4.7179+-0.1646 ? 4.8483+-0.5238 ? might be 1.0276x slower infer-constant-property 3.2775+-0.0889 ? 3.3710+-0.1020 ? might be 1.0285x slower infer-one-time-closure-ten-vars 14.6194+-0.2598 ? 14.7477+-0.5086 ? infer-one-time-closure-two-vars 14.0472+-0.3765 14.0467+-0.4701 infer-one-time-closure 14.1493+-0.2449 ? 14.1992+-0.9306 ? infer-one-time-deep-closure 24.6380+-0.6716 24.2572+-0.1191 might be 1.0157x faster inline-arguments-access 6.0715+-0.0752 5.9248+-0.2334 might be 1.0248x faster inline-arguments-aliased-access 6.1038+-0.2681 5.8441+-0.2390 might be 1.0444x faster inline-arguments-local-escape 6.0081+-0.1769 ? 6.0330+-0.2716 ? inline-get-scoped-var 5.6774+-0.0211 ? 5.7180+-0.0206 ? inlined-put-by-id-transition 16.5682+-0.4644 16.5279+-0.7493 int-or-other-abs-then-get-by-val 6.7702+-0.2313 6.7225+-0.3356 int-or-other-abs-zero-then-get-by-val 28.1090+-0.5016 ^ 26.6490+-0.3681 ^ definitely 1.0548x faster int-or-other-add-then-get-by-val 5.5082+-0.1212 5.3670+-0.1969 might be 1.0263x faster int-or-other-add 7.8910+-0.1046 ? 7.9225+-0.3286 ? int-or-other-div-then-get-by-val 5.1843+-0.0814 ? 5.2502+-0.0258 ? might be 1.0127x slower int-or-other-max-then-get-by-val 5.1105+-0.1592 5.0105+-0.0270 might be 1.0200x faster int-or-other-min-then-get-by-val 4.9470+-0.1015 ? 5.0676+-0.1000 ? might be 1.0244x slower int-or-other-mod-then-get-by-val 5.0851+-0.0481 5.0602+-0.1634 int-or-other-mul-then-get-by-val 4.9180+-0.1630 ? 5.0287+-0.0731 ? might be 1.0225x slower int-or-other-neg-then-get-by-val 6.0945+-0.1385 6.0811+-0.2230 int-or-other-neg-zero-then-get-by-val 28.2449+-0.4971 ^ 26.5869+-0.1542 ^ definitely 1.0624x faster int-or-other-sub-then-get-by-val 5.9839+-0.2048 5.9072+-0.0922 might be 1.0130x faster int-or-other-sub 4.2643+-0.0920 ? 4.3174+-0.1687 ? might be 1.0125x slower int-overflow-local 5.6223+-0.1745 5.5192+-0.1816 might be 1.0187x faster Int16Array-alloc-long-lived 62.8265+-0.8644 ? 63.2474+-1.3574 ? Int16Array-bubble-sort-with-byteLength 36.9628+-0.2351 ? 37.2084+-1.4977 ? Int16Array-bubble-sort 36.8652+-0.2441 36.5529+-0.9840 Int16Array-load-int-mul 1.9390+-0.1463 ? 1.9760+-0.1439 ? might be 1.0191x slower Int16Array-to-Int32Array-set 72.1407+-0.9760 71.5993+-6.4119 Int32Array-alloc-large 33.4929+-2.1877 32.8245+-0.8520 might be 1.0204x faster Int32Array-alloc-long-lived 70.8380+-1.6034 70.1285+-0.8858 might be 1.0101x faster Int32Array-alloc 4.6058+-0.1210 4.5278+-0.1702 might be 1.0172x faster Int32Array-Int8Array-view-alloc 9.3418+-0.2109 ? 9.5227+-0.3550 ? might be 1.0194x slower int52-spill 7.9166+-0.1507 ? 7.9464+-0.4096 ? Int8Array-alloc-long-lived 56.6159+-2.2806 55.8309+-0.9080 might be 1.0141x faster Int8Array-load-with-byteLength 4.8105+-0.1966 4.7040+-0.1648 might be 1.0226x faster Int8Array-load 4.7913+-0.1458 4.7292+-0.1061 might be 1.0131x faster integer-divide 13.3164+-0.1605 13.2405+-0.2207 integer-modulo 2.7169+-0.0919 2.6797+-0.3391 might be 1.0139x faster is-boolean-fold-tricky 5.7089+-0.0350 5.7061+-0.2629 is-boolean-fold 4.0048+-0.1956 3.9996+-0.0895 is-function-fold-tricky-internal-function 14.7151+-0.6315 ? 15.0859+-1.1559 ? might be 1.0252x slower is-function-fold-tricky 5.8403+-0.2112 ? 5.8754+-0.1681 ? is-function-fold 4.0667+-0.2411 4.0357+-0.0841 is-number-fold-tricky 5.6593+-0.0902 ? 5.7488+-0.1979 ? might be 1.0158x slower is-number-fold 3.9664+-0.1979 3.9446+-0.0896 is-object-or-null-fold-functions 4.0342+-0.0534 ? 4.1187+-0.1309 ? might be 1.0210x slower is-object-or-null-fold-less-tricky 5.8753+-0.1213 5.8466+-0.1712 is-object-or-null-fold-tricky 7.7351+-0.0730 7.7138+-0.1424 is-object-or-null-fold 4.0592+-0.1044 3.9694+-0.2292 might be 1.0226x faster is-object-or-null-trickier-function 5.9594+-0.0972 ? 5.9911+-0.0729 ? is-object-or-null-trickier-internal-function 15.0306+-0.0936 ? 15.0627+-0.2578 ? is-object-or-null-tricky-function 5.8070+-0.1713 5.7728+-0.2028 is-object-or-null-tricky-internal-function 11.2480+-0.2475 ? 11.6612+-0.7947 ? might be 1.0367x slower is-string-fold-tricky 5.7197+-0.0561 ? 5.7622+-0.1922 ? is-string-fold 3.9505+-0.1476 3.9387+-0.0979 is-undefined-fold-tricky 4.5173+-0.1549 ? 4.5582+-0.1011 ? is-undefined-fold 3.9946+-0.1334 3.9243+-0.0609 might be 1.0179x faster large-int-captured 6.8173+-0.2549 6.7831+-0.3603 large-int-neg 20.7515+-0.4438 20.7252+-0.2866 large-int 18.5352+-0.3275 ? 18.8012+-0.8286 ? might be 1.0144x slower logical-not 6.0776+-0.3505 ? 6.1487+-0.2695 ? might be 1.0117x slower lots-of-fields 19.3743+-0.4078 19.0927+-0.6115 might be 1.0147x faster make-indexed-storage 4.2888+-0.4545 ? 4.3552+-0.5310 ? might be 1.0155x slower make-rope-cse 6.6602+-0.2540 ? 6.7415+-0.1084 ? might be 1.0122x slower marsaglia-larger-ints 54.8768+-0.9281 ? 55.2280+-0.5981 ? marsaglia-osr-entry 28.1051+-0.6934 27.7675+-0.8214 might be 1.0122x faster max-boolean 3.3352+-0.0683 ? 3.4354+-0.1859 ? might be 1.0300x slower method-on-number 23.5604+-0.8280 22.9657+-0.2893 might be 1.0259x faster min-boolean 3.2449+-0.1198 ? 3.3211+-0.1225 ? might be 1.0235x slower minus-boolean-double 4.1608+-0.1510 ? 4.1917+-0.0342 ? minus-boolean 3.2672+-0.0593 3.2636+-0.1782 misc-strict-eq 57.3501+-2.2081 56.6557+-1.7333 might be 1.0123x faster mod-boolean-double 11.6487+-0.3847 11.6268+-0.4287 mod-boolean 8.7117+-0.2160 8.6805+-0.1101 mul-boolean-double 4.8589+-0.0672 4.8303+-0.1400 mul-boolean 3.4401+-0.0973 ? 3.5140+-0.0557 ? might be 1.0215x slower neg-boolean 4.2925+-0.1087 ? 4.3388+-0.1195 ? might be 1.0108x slower negative-zero-divide 0.4872+-0.0816 ? 0.5070+-0.1021 ? might be 1.0407x slower negative-zero-modulo 0.4832+-0.1082 ? 0.4930+-0.0881 ? might be 1.0203x slower negative-zero-negate 0.4709+-0.1207 0.4270+-0.0058 might be 1.1028x faster nested-function-parsing 57.5099+-0.3040 ? 58.0176+-0.9158 ? new-array-buffer-dead 140.5359+-1.0499 139.3040+-0.4759 new-array-buffer-push 9.6796+-0.2998 ? 9.7462+-0.1699 ? new-array-dead 23.4716+-2.1226 ? 23.8286+-2.4989 ? might be 1.0152x slower new-array-push 5.5055+-0.2401 5.3200+-0.3137 might be 1.0349x faster no-inline-constructor 50.5279+-1.0798 ? 51.1728+-0.6761 ? might be 1.0128x slower number-test 4.2537+-0.1823 4.2252+-0.1826 object-closure-call 7.2874+-0.3855 7.2421+-0.2154 object-test 4.1723+-0.0339 4.1319+-0.1694 obvious-sink-pathology-taken 175.2603+-2.0277 ? 175.9230+-4.0101 ? obvious-sink-pathology 158.5150+-1.4957 ? 158.7575+-3.4426 ? obviously-elidable-new-object 48.6227+-4.7699 48.5007+-4.2135 plus-boolean-arith 3.3040+-0.1473 ? 3.3662+-0.0968 ? might be 1.0188x slower plus-boolean-double 4.2151+-0.1384 4.2063+-0.1331 plus-boolean 3.2280+-0.1462 3.2060+-0.1412 poly-chain-access-different-prototypes-simple 3.8087+-0.0899 ? 3.8745+-0.0931 ? might be 1.0173x slower poly-chain-access-different-prototypes 3.1277+-0.0464 ? 3.1635+-0.1390 ? might be 1.0114x slower poly-chain-access-simpler 3.8171+-0.1214 ? 3.8849+-0.1965 ? might be 1.0178x slower poly-chain-access 3.1659+-0.1579 3.0557+-0.0797 might be 1.0361x faster poly-stricteq 70.5832+-0.3731 70.0180+-0.4769 polymorphic-array-call 1.8113+-0.0916 ? 1.8762+-0.2050 ? might be 1.0359x slower polymorphic-get-by-id 4.1880+-0.2150 ? 4.2225+-0.1165 ? polymorphic-put-by-id 40.6406+-4.2658 40.2591+-5.4946 polymorphic-structure 17.8063+-0.2945 17.7448+-0.2439 polyvariant-monomorphic-get-by-id 11.4618+-0.2949 ? 11.5225+-0.1027 ? proto-getter-access 11.1813+-0.6074 ? 11.4205+-0.2494 ? might be 1.0214x slower put-by-id-replace-and-transition 12.5341+-0.2077 12.3593+-0.2537 might be 1.0141x faster put-by-id-slightly-polymorphic 3.5233+-0.1444 ? 3.5973+-0.0966 ? might be 1.0210x slower put-by-id 18.8417+-0.2920 18.4876+-0.5025 might be 1.0192x faster put-by-val-direct 0.5215+-0.1281 0.4846+-0.1155 might be 1.0762x faster put-by-val-large-index-blank-indexing-type 8.1812+-0.5271 ? 8.3521+-0.5195 ? might be 1.0209x slower put-by-val-machine-int 3.5709+-0.3911 3.5501+-0.2827 rare-osr-exit-on-local 18.1245+-0.6528 ? 18.5234+-0.1240 ? might be 1.0220x slower register-pressure-from-osr 25.8982+-0.5275 25.8140+-0.3145 setter 6.5530+-0.0859 6.5151+-0.1493 simple-activation-demo 31.7686+-1.2001 30.8278+-0.3875 might be 1.0305x faster simple-getter-access 14.8934+-0.6572 14.8862+-0.2539 simple-poly-call-nested 10.0604+-0.0965 ? 10.1072+-0.2043 ? simple-poly-call 1.6812+-0.0481 1.6375+-0.1275 might be 1.0267x faster sin-boolean 25.9542+-3.9631 ? 27.1741+-1.2065 ? might be 1.0470x slower singleton-scope 80.0787+-0.9497 ? 80.3128+-1.0196 ? sink-function 16.0769+-0.5833 ? 16.1719+-0.2111 ? sinkable-new-object-dag 95.1334+-2.4884 ? 95.3279+-0.3655 ? sinkable-new-object-taken 72.3713+-3.9753 69.7118+-0.9512 might be 1.0381x faster sinkable-new-object 52.7667+-0.7292 ? 54.1821+-1.6146 ? might be 1.0268x slower slow-array-profile-convergence 4.1476+-0.2671 3.9612+-0.2767 might be 1.0470x faster slow-convergence 3.7038+-0.2067 ? 3.8448+-0.3165 ? might be 1.0381x slower sorting-benchmark 27.9146+-0.3085 27.8145+-1.2866 sparse-conditional 1.5826+-0.0829 1.5635+-0.1087 might be 1.0122x faster splice-to-remove 20.7971+-0.3262 20.5209+-0.4813 might be 1.0135x faster string-char-code-at 19.5857+-0.5127 19.3538+-0.1160 might be 1.0120x faster string-concat-object 3.1873+-0.3417 ? 3.3613+-0.4287 ? might be 1.0546x slower string-concat-pair-object 3.2205+-0.4088 3.0599+-0.0914 might be 1.0525x faster string-concat-pair-simple 17.1075+-1.2399 16.9958+-0.8957 string-concat-simple 17.2462+-0.3932 ? 17.3129+-0.6979 ? string-cons-repeat 11.0538+-0.3612 ? 11.4277+-0.1620 ? might be 1.0338x slower string-cons-tower 10.5306+-0.2541 ? 10.8833+-0.1230 ? might be 1.0335x slower string-equality 23.2255+-0.1180 23.1381+-0.3916 string-get-by-val-big-char 9.7953+-0.3864 ? 9.9029+-0.4138 ? might be 1.0110x slower string-get-by-val-out-of-bounds-insane 5.1325+-0.2759 ? 5.1718+-0.0634 ? string-get-by-val-out-of-bounds 6.8584+-0.2233 6.7662+-0.0928 might be 1.0136x faster string-get-by-val 4.7028+-0.1222 ? 4.7576+-0.0936 ? might be 1.0117x slower string-hash 2.7128+-0.1405 ? 2.7485+-0.1139 ? might be 1.0132x slower string-long-ident-equality 18.6025+-0.4546 ? 18.9578+-0.8985 ? might be 1.0191x slower string-out-of-bounds 17.8392+-0.6445 17.6395+-0.4371 might be 1.0113x faster string-repeat-arith 40.0675+-0.6953 39.7419+-1.5478 string-sub 78.1337+-0.5404 ? 78.5480+-1.8337 ? string-test 4.2216+-0.1311 ? 4.2657+-0.0606 ? might be 1.0105x slower string-var-equality 44.3915+-0.3925 44.2338+-0.3619 structure-hoist-over-transitions 3.6465+-0.1920 3.5515+-0.2402 might be 1.0268x faster substring-concat-weird 54.9457+-1.5171 54.5009+-0.7568 substring-concat 58.1820+-1.0266 ? 58.6414+-0.7792 ? substring 63.0558+-1.1730 ? 63.6432+-1.3737 ? switch-char-constant 3.3437+-0.1854 ? 3.3683+-0.0900 ? switch-char 10.2978+-5.7226 7.6990+-0.1366 might be 1.3375x faster switch-constant 11.5632+-0.3825 11.2023+-0.3314 might be 1.0322x faster switch-string-basic-big-var 26.7884+-1.1830 ? 27.3973+-0.3936 ? might be 1.0227x slower switch-string-basic-big 22.1875+-2.1259 22.1460+-7.0286 switch-string-basic-var 27.1140+-0.7521 ? 27.6895+-0.6894 ? might be 1.0212x slower switch-string-basic 16.9702+-1.9015 16.8500+-2.3765 switch-string-big-length-tower-var 25.7384+-0.3289 25.6803+-0.3324 switch-string-length-tower-var 19.7745+-0.3919 19.6990+-0.2505 switch-string-length-tower 15.9852+-1.8733 15.9324+-0.7049 switch-string-short 13.7515+-0.3980 ? 13.8318+-0.2621 ? switch 16.1736+-2.4674 ? 16.1791+-2.3074 ? tear-off-arguments-simple 4.4478+-0.3313 ? 4.4960+-0.1647 ? might be 1.0108x slower tear-off-arguments 6.4111+-0.2702 6.4070+-0.3037 temporal-structure 15.1785+-0.1756 ? 15.3463+-0.2748 ? might be 1.0111x slower to-int32-boolean 20.5610+-0.5051 20.5512+-0.2696 try-catch-get-by-val-cloned-arguments 19.3025+-0.3193 ? 19.9927+-0.8341 ? might be 1.0358x slower try-catch-get-by-val-direct-arguments 8.6143+-0.4899 8.5870+-0.5895 try-catch-get-by-val-scoped-arguments 10.4035+-0.3539 10.2811+-0.5264 might be 1.0119x faster typed-array-get-set-by-val-profiling 43.9748+-3.5791 41.6689+-1.8365 might be 1.0553x faster undefined-property-access 428.9783+-5.3646 426.4598+-3.5036 undefined-test 4.4958+-0.0670 4.4320+-0.1847 might be 1.0144x faster unprofiled-licm 27.3765+-0.8382 27.1072+-0.8680 varargs-call 17.5976+-0.3500 ? 17.6668+-0.2285 ? varargs-construct-inline 29.6215+-0.5411 ? 29.6700+-0.8032 ? varargs-construct 30.2044+-0.2552 29.9277+-0.1898 varargs-inline 10.4577+-0.1858 10.3757+-0.1519 varargs-strict-mode 12.7011+-0.3044 ? 12.8185+-0.9364 ? varargs 12.5972+-0.1193 ? 12.6832+-0.3788 ? weird-inlining-const-prop 2.8961+-0.2155 2.8811+-0.1620 <geometric> 10.8489+-0.0290 10.8222+-0.0358 might be 1.0025x faster Conf#1 Conf#2 AsmBench: bigfib.cpp 672.6619+-9.4325 670.7493+-10.1379 cray.c 623.6575+-3.1004 621.6807+-2.0942 dry.c 654.0922+-1.5307 ? 654.1190+-0.7919 ? FloatMM.c 956.4980+-2.2288 955.2708+-1.0161 gcc-loops.cpp 5870.7053+-13.7207 5869.2598+-11.0082 n-body.c 1669.3310+-2.5898 1667.9983+-2.1012 Quicksort.c 569.7831+-4.9455 ? 572.7227+-2.4949 ? stepanov_container.cpp 4888.4510+-29.5120 ? 4888.7488+-39.5006 ? Towers.c 374.9509+-2.0877 371.8025+-1.5313 <geometric> 1116.0438+-1.9729 1114.6153+-2.6965 might be 1.0013x faster Conf#1 Conf#2 CompressionBench: huffman 499.2209+-2.1567 498.2756+-3.8201 arithmetic-simple 527.7750+-1.2775 526.3882+-0.8845 arithmetic-precise 402.1168+-2.2626 401.3778+-2.7621 arithmetic-complex-precise 401.8900+-1.9853 400.7168+-2.6534 arithmetic-precise-order-0 561.5009+-8.4529 ? 567.2136+-8.0601 ? might be 1.0102x slower arithmetic-precise-order-1 422.9450+-3.2862 422.7391+-4.4674 arithmetic-precise-order-2 472.2582+-4.3108 ? 474.4724+-12.1010 ? arithmetic-simple-order-1 528.9850+-1.9384 ? 529.0212+-1.4707 ? arithmetic-simple-order-2 584.4789+-2.6455 ? 587.0936+-3.1305 ? lz-string 414.3596+-3.4464 ? 414.8075+-3.0352 ? <geometric> 477.1526+-1.1746 ? 477.6559+-1.3156 ? might be 1.0011x slower Conf#1 Conf#2 Geomean of preferred means: <scaled-result> 83.9340+-0.0934 ^ 83.6950+-0.0775 ^ definitely 1.0029x faster

Geoffrey Garen

Comment 3 2015-05-12 12:20:16 PDT

Comment on attachment 252934 [details] Patch r=me

Benjamin Poulain

Comment 4 2015-05-12 14:30:05 PDT

Comment on attachment 252934 [details] Patch Clearing flags on attachment: 252934 Committed r184220: <http://trac.webkit.org/changeset/184220>

Benjamin Poulain

Comment 5 2015-05-12 14:30:09 PDT

All reviewed patches have been landed. Closing bug.

Alexey Proskuryakov

Comment 6 2015-05-12 18:30:38 PDT

The new test fails on Windows: https://build.webkit.org/builders/Apple%20Win%207%20Release%20(Tests)/builds/51783/steps/jscore-test/logs/stdio

Ryosuke Niwa

Comment 7 2015-05-12 22:36:02 PDT

(In reply to comment #6) > The new test fails on Windows: > https://build.webkit.org/builders/Apple%20Win%207%20Release%20(Tests)/builds/ > 51783/steps/jscore-test/logs/stdio Looks like this is a Windows only failure. It's probably revealing an existing bug in jsc on Windows. I'll skip it temporarily to let Benjamin investigate it.

Ryosuke Niwa

Comment 8 2015-05-12 22:40:03 PDT

Filed https://bugs.webkit.org/show_bug.cgi?id=144947 to track the failure.

Alexey Proskuryakov

Comment 9 2015-05-13 12:00:45 PDT

Skipping didn't work, the test still fails.

Ryosuke Niwa

Comment 10 2015-05-13 14:05:57 PDT

(In reply to comment #9) > Skipping didn't work, the test still fails. That's because I skipped a wrong file :( Fixing that now.

Attachments
Patch (9.50 KB, patch) 2015-05-11 21:57 PDT, Benjamin Poulain	no flags	Details Formatted Diff Diff
View All Add attachment proposed patch, testcase, etc.