Bug 148157

Summary: [JSC] Optimize more cases of something-compared-to-null/undefined
Product: WebKit Reporter: Benjamin Poulain <benjamin>
Component: New BugsAssignee: Benjamin Poulain <benjamin>
Status: RESOLVED FIXED    
Severity: Normal CC: fpizlo
Priority: P2    
Version: WebKit Nightly Build   
Hardware: Unspecified   
OS: Unspecified   
Attachments:
Description Flags
Patch
none
Patch ggaren: review+

Benjamin Poulain
Reported 2015-08-18 18:44:39 PDT
[JSC] Optimize more cases of something-compared-to-null
Attachments
Patch (54.45 KB, patch)
2015-08-18 19:04 PDT, Benjamin Poulain 		no flags
DetailsFormatted DiffDiff
Patch (54.58 KB, patch)
2015-08-18 19:23 PDT, Benjamin Poulain 		ggaren: review+
DetailsFormatted DiffDiff
Show Obsolete (1) View All Add attachment proposed patch, testcase, etc.
Benjamin Poulain
Comment 1 2015-08-18 19:04:00 PDT
Created attachment 259341 [details] Patch
Benjamin Poulain
Comment 2 2015-08-18 19:21:42 PDT
Conf#1 Conf#2 SunSpider: 3d-cube 4.3320+-0.0587 4.1412+-0.4001 might be 1.0461x faster 3d-morph 4.7820+-0.0983 4.7555+-0.0256 3d-raytrace 4.7429+-0.1841 ? 4.7596+-0.3667 ? access-binary-trees 1.7972+-0.0826 1.7513+-0.0463 might be 1.0262x faster access-fannkuch 4.8633+-0.0609 ? 4.9108+-0.1503 ? access-nbody 2.2955+-0.2588 2.2356+-0.0662 might be 1.0268x faster access-nsieve 2.7100+-0.0253 ? 2.7568+-0.0653 ? might be 1.0173x slower bitops-3bit-bits-in-byte 1.3252+-0.0479 ? 1.3259+-0.0522 ? bitops-bits-in-byte 3.4797+-0.1449 3.2645+-0.4562 might be 1.0659x faster bitops-bitwise-and 1.8219+-0.0141 1.8130+-0.0175 bitops-nsieve-bits 2.7290+-0.1214 ? 2.7855+-0.2066 ? might be 1.0207x slower controlflow-recursive 1.8038+-0.0341 ? 1.8857+-0.0777 ? might be 1.0454x slower crypto-aes 3.4129+-0.1156 ? 3.4209+-0.2994 ? crypto-md5 2.1389+-0.0635 ? 2.2063+-0.1436 ? might be 1.0315x slower crypto-sha1 2.2618+-0.5480 2.1058+-0.1946 might be 1.0741x faster date-format-tofte 6.3775+-0.3880 6.1036+-0.1890 might be 1.0449x faster date-format-xparb 4.0027+-0.1212 ? 4.0540+-0.1840 ? might be 1.0128x slower math-cordic 2.5290+-0.0755 2.5284+-0.0775 math-partial-sums 4.9065+-0.1698 ? 4.9472+-0.2322 ? math-spectral-norm 1.6378+-0.0578 ? 1.6553+-0.1024 ? might be 1.0107x slower regexp-dna 5.8759+-0.1207 ? 5.9268+-0.1677 ? string-base64 3.9315+-0.3277 3.8875+-0.1517 might be 1.0113x faster string-fasta 5.3929+-0.0861 5.2944+-0.1803 might be 1.0186x faster string-tagcloud 7.5638+-0.2792 7.3725+-0.1272 might be 1.0260x faster string-unpack-code 17.2087+-0.5402 ? 17.4806+-1.0212 ? might be 1.0158x slower string-validate-input 4.2167+-0.2597 4.1940+-0.2764 <arithmetic> 4.1592+-0.0470 4.1370+-0.0261 might be 1.0054x faster Conf#1 Conf#2 LongSpider: 3d-cube 764.2908+-5.7906 ? 769.5660+-11.9344 ? 3d-morph 1448.6265+-3.7010 ? 1451.1602+-7.9621 ? 3d-raytrace 576.5409+-5.8986 575.7640+-3.8432 access-binary-trees 880.0142+-27.3949 868.5279+-12.1178 might be 1.0132x faster access-fannkuch 264.7452+-11.6713 260.5678+-1.8838 might be 1.0160x faster access-nbody 489.0494+-0.2718 ! 491.2237+-1.4296 ! definitely 1.0044x slower access-nsieve 476.1485+-4.9304 475.1461+-3.1936 bitops-3bit-bits-in-byte 36.3306+-0.3871 36.2047+-0.4519 bitops-bits-in-byte 71.7745+-1.0244 71.3761+-1.7227 bitops-nsieve-bits 370.2285+-2.4953 368.9565+-0.9128 controlflow-recursive 429.2131+-0.5509 ? 430.1453+-2.2431 ? crypto-aes 559.8577+-10.6826 ? 562.0433+-8.2789 ? crypto-md5 474.9011+-2.3454 473.6127+-10.3663 crypto-sha1 588.6923+-8.0143 586.9300+-3.4435 date-format-tofte 477.6645+-4.2640 ? 483.3655+-9.4265 ? might be 1.0119x slower date-format-xparb 597.7426+-11.4554 596.8495+-5.8417 hash-map 147.9409+-2.6607 147.2597+-2.1779 math-cordic 467.5943+-1.0007 ? 468.8848+-2.1121 ? math-partial-sums 452.9103+-0.7556 ^ 449.0076+-0.3915 ^ definitely 1.0087x faster math-spectral-norm 530.1084+-2.9211 529.0294+-0.3170 string-base64 339.6539+-3.0193 ? 342.1121+-4.4725 ? string-fasta 360.5400+-5.4084 359.2713+-3.7712 string-tagcloud 169.2455+-2.4820 168.0764+-0.5958 <geometric> 380.3246+-0.8561 379.7191+-1.0232 might be 1.0016x faster Conf#1 Conf#2 V8Spider: crypto 45.9628+-1.5058 ? 46.3597+-2.1320 ? deltablue 84.1795+-3.2171 ? 84.7140+-7.2355 ? earley-boyer 38.4523+-1.2901 ? 40.6664+-3.9299 ? might be 1.0576x slower raytrace 30.5253+-1.6546 ! 33.6824+-0.7479 ! definitely 1.1034x slower regexp 59.6331+-0.6021 59.2955+-1.0740 richards 62.3392+-2.0380 61.7024+-2.2068 might be 1.0103x faster splay 36.0264+-2.0624 ? 36.5447+-3.5053 ? might be 1.0144x slower <geometric> 48.2323+-0.2501 ! 49.3820+-0.2657 ! definitely 1.0238x slower Conf#1 Conf#2 Octane: encrypt 0.16673+-0.00074 0.16628+-0.00137 decrypt 3.13884+-0.01791 ? 3.14155+-0.00658 ? deltablue x2 0.14754+-0.00179 ? 0.15075+-0.00310 ? might be 1.0217x slower earley 0.29852+-0.00908 0.29753+-0.01004 boyer 4.32043+-0.01946 4.29582+-0.06898 navier-stokes x2 4.67272+-0.00515 ? 4.68349+-0.01595 ? raytrace x2 1.05136+-0.04993 ? 1.07392+-0.00854 ? might be 1.0215x slower richards x2 0.10396+-0.00150 ? 0.10416+-0.00350 ? splay x2 0.37900+-0.00565 ? 0.37964+-0.00875 ? regexp x2 23.86281+-1.77487 23.52744+-1.38830 might be 1.0143x faster pdfjs x2 36.12210+-0.33474 ? 36.55328+-1.11246 ? might be 1.0119x slower mandreel x2 43.03230+-3.43075 ? 43.17738+-2.96491 ? gbemu x2 31.15367+-0.65456 ? 31.65615+-0.75394 ? might be 1.0161x slower closure 0.52942+-0.00824 0.52642+-0.00215 jquery 6.67151+-0.09495 6.66472+-0.02003 box2d x2 9.29647+-0.19048 ? 9.30857+-0.11770 ? zlib x2 381.95235+-4.70541 381.33907+-9.37820 typescript x2 619.03943+-9.04748 ? 633.30298+-12.22184 ? might be 1.0230x slower <geometric> 5.45195+-0.05700 ? 5.48099+-0.03562 ? might be 1.0053x slower Conf#1 Conf#2 Kraken: ai-astar 207.651+-3.586 207.220+-2.121 audio-beat-detection 48.834+-1.145 47.795+-0.405 might be 1.0217x faster audio-dft 89.306+-0.260 ? 89.708+-2.059 ? audio-fft 33.514+-0.122 ? 33.547+-0.166 ? audio-oscillator 56.577+-0.415 ? 56.896+-0.168 ? imaging-darkroom 58.961+-0.200 ? 59.002+-0.210 ? imaging-desaturate 46.870+-0.511 ? 46.941+-0.314 ? imaging-gaussian-blur 80.170+-0.633 79.841+-0.247 json-parse-financial 35.736+-0.671 ? 36.647+-0.788 ? might be 1.0255x slower json-stringify-tinderbox 23.577+-0.120 ^ 21.817+-1.045 ^ definitely 1.0807x faster stanford-crypto-aes 38.884+-1.032 ? 39.027+-0.909 ? stanford-crypto-ccm 32.499+-1.813 32.359+-1.343 stanford-crypto-pbkdf2 90.265+-1.128 89.488+-0.366 stanford-crypto-sha256-iterative 35.626+-1.827 34.668+-0.455 might be 1.0277x faster <arithmetic> 62.748+-0.168 62.497+-0.354 might be 1.0040x faster Conf#1 Conf#2 JSRegress: abc-forward-loop-equal 28.1307+-0.5253 28.0436+-0.4766 abc-postfix-backward-loop 28.3713+-0.8836 ? 28.5770+-1.2618 ? abc-simple-backward-loop 27.5651+-0.4050 ? 28.0515+-0.4874 ? might be 1.0176x slower abc-simple-forward-loop 27.6914+-0.5197 ? 27.7676+-0.3670 ? abc-skippy-loop 19.8237+-0.2376 ? 20.0164+-0.1350 ? abs-boolean 2.2292+-0.0439 ? 2.3989+-0.3961 ? might be 1.0761x slower adapt-to-double-divide 15.5211+-0.2814 15.2526+-0.2950 might be 1.0176x faster aliased-arguments-getbyval 1.0295+-0.1216 1.0037+-0.0483 might be 1.0257x faster allocate-big-object 2.3690+-0.1442 2.3231+-0.1403 might be 1.0198x faster arguments-named-and-reflective 10.4975+-0.3473 10.3997+-0.3189 arguments-out-of-bounds 9.2864+-0.1424 9.2345+-0.2147 arguments-strict-mode 9.4885+-0.6197 9.4305+-0.3421 arguments 8.3257+-0.1658 8.2400+-0.0873 might be 1.0104x faster arity-mismatch-inlining 0.7332+-0.0650 0.7123+-0.0282 might be 1.0293x faster array-access-polymorphic-structure 5.3979+-0.0686 ? 5.4042+-0.1169 ? array-nonarray-polymorhpic-access 23.0314+-0.7180 22.7224+-0.5355 might be 1.0136x faster array-prototype-every 71.1091+-0.5605 70.9395+-0.9976 array-prototype-forEach 70.6488+-0.5157 70.4292+-1.0656 array-prototype-map 79.0865+-1.5276 77.4642+-1.1635 might be 1.0209x faster array-prototype-reduce 69.5240+-0.7135 69.1218+-0.5812 array-prototype-reduceRight 70.3754+-1.9091 68.8304+-0.5437 might be 1.0224x faster array-prototype-some 71.1717+-0.5233 70.7686+-1.0447 array-splice-contiguous 19.4446+-1.1562 19.1470+-0.2213 might be 1.0155x faster array-with-double-add 3.1600+-0.2621 3.1224+-0.1048 might be 1.0120x faster array-with-double-increment 2.7946+-0.0529 ? 2.8068+-0.1083 ? array-with-double-mul-add 3.9362+-0.1745 3.8589+-0.1547 might be 1.0200x faster array-with-double-sum 2.9370+-0.0518 2.8911+-0.0220 might be 1.0159x faster array-with-int32-add-sub 5.3571+-0.1835 5.3330+-0.2464 array-with-int32-or-double-sum 2.9456+-0.1056 ? 2.9877+-0.1092 ? might be 1.0143x slower ArrayBuffer-DataView-alloc-large-long-lived 28.2842+-0.5059 ? 28.3325+-0.6576 ? ArrayBuffer-DataView-alloc-long-lived 11.1048+-0.2004 ? 11.1649+-0.2542 ? ArrayBuffer-Int32Array-byteOffset 3.2808+-0.0119 ? 3.3147+-0.0817 ? might be 1.0103x slower ArrayBuffer-Int8Array-alloc-large-long-lived 28.9438+-0.8528 ? 28.9451+-0.5636 ? ArrayBuffer-Int8Array-alloc-long-lived-buffer 18.5299+-0.4618 ? 18.7118+-0.5215 ? ArrayBuffer-Int8Array-alloc-long-lived 10.8255+-0.0641 10.7750+-0.3883 ArrayBuffer-Int8Array-alloc 9.1591+-0.0462 ? 9.2383+-0.4424 ? arrowfunction-call 9.9349+-0.2505 ? 9.9401+-0.3139 ? asmjs_bool_bug 6.7295+-0.2035 6.6997+-0.1681 assign-custom-setter-polymorphic 2.0683+-0.0441 ! 2.1362+-0.0137 ! definitely 1.0328x slower assign-custom-setter 2.9457+-0.0848 ? 2.9999+-0.0325 ? might be 1.0184x slower basic-set 6.5880+-0.1586 ? 6.7290+-0.1833 ? might be 1.0214x slower big-int-mul 3.1361+-0.0043 ? 3.1945+-0.0999 ? might be 1.0186x slower boolean-test 2.6879+-0.1165 2.6564+-0.0479 might be 1.0118x faster branch-fold 3.3031+-0.0138 ? 3.3840+-0.1655 ? might be 1.0245x slower branch-on-string-as-boolean 15.4325+-0.3059 15.3499+-0.2670 by-val-generic 5.5905+-0.0801 5.5674+-0.2210 call-spread-apply 25.1209+-2.4392 24.9460+-1.8824 call-spread-call 19.4127+-0.3988 ? 20.7404+-1.6375 ? might be 1.0684x slower captured-assignments 0.3709+-0.0121 ? 0.3927+-0.0368 ? might be 1.0587x slower cast-int-to-double 4.6940+-0.0742 4.6434+-0.2664 might be 1.0109x faster cell-argument 6.1630+-0.1405 6.0806+-0.2584 might be 1.0136x faster cfg-simplify 2.5312+-0.1485 ? 2.5318+-0.0941 ? chain-getter-access 7.6668+-0.0711 ? 7.6726+-0.1124 ? cmpeq-obj-to-obj-other 10.6282+-2.2212 ? 11.4427+-1.1495 ? might be 1.0766x slower constant-test 4.4823+-0.0849 ? 4.5118+-0.1505 ? create-lots-of-functions 10.3473+-0.2684 ? 10.7435+-1.0656 ? might be 1.0383x slower cse-new-array-buffer 2.1180+-0.2463 2.1055+-0.1409 cse-new-array 2.0957+-0.1220 ? 2.1797+-0.1979 ? might be 1.0401x slower DataView-custom-properties 33.1803+-0.5985 32.8547+-0.9285 delay-tear-off-arguments-strictmode 11.4208+-0.2331 ? 11.6694+-0.2928 ? might be 1.0218x slower deltablue-varargs 155.1306+-3.0487 ? 156.3932+-3.2080 ? destructuring-arguments 155.3320+-2.3028 ? 156.4148+-0.3525 ? destructuring-parameters-overridden-by-function 0.3864+-0.0141 ? 0.4189+-0.0595 ? might be 1.0842x slower destructuring-swap 4.3851+-0.1441 4.3820+-0.0657 direct-arguments-getbyval 1.0412+-0.1550 0.9907+-0.0412 might be 1.0509x faster div-boolean-double 4.9628+-0.0321 4.9142+-0.1934 div-boolean 7.7978+-0.6165 7.5744+-0.0252 might be 1.0295x faster double-get-by-val-out-of-bounds 3.9719+-0.0568 3.9160+-0.1274 might be 1.0143x faster double-pollution-getbyval 8.1218+-0.1501 8.0888+-0.1350 double-pollution-putbyoffset 3.3989+-0.0495 ? 3.4218+-0.0485 ? double-real-use 22.7996+-0.7106 22.5508+-0.3103 might be 1.0110x faster double-to-int32-typed-array-no-inline 1.8984+-0.1940 1.8184+-0.0575 might be 1.0440x faster double-to-int32-typed-array 1.6630+-0.1542 1.6476+-0.1225 double-to-uint32-typed-array-no-inline 1.9619+-0.1876 1.8663+-0.0818 might be 1.0512x faster double-to-uint32-typed-array 1.7013+-0.0698 1.6619+-0.0278 might be 1.0237x faster elidable-new-object-dag 37.5438+-0.6691 37.4698+-0.9780 elidable-new-object-roflcopter 36.1979+-1.1617 ? 36.2987+-0.6831 ? elidable-new-object-then-call 33.8806+-1.0882 ? 33.9935+-0.7506 ? elidable-new-object-tree 41.3449+-0.7828 40.9897+-0.5409 empty-string-plus-int 4.5848+-0.2385 4.5825+-0.2458 emscripten-cube2hash 24.9752+-0.4855 ^ 22.5428+-0.7636 ^ definitely 1.1079x faster exit-length-on-plain-object 11.2985+-0.1868 ? 11.5306+-0.5528 ? might be 1.0205x slower external-arguments-getbyval 0.9941+-0.0281 0.9755+-0.0337 might be 1.0190x faster external-arguments-putbyval 1.8720+-0.0353 1.8695+-0.0798 fixed-typed-array-storage-var-index 1.1119+-0.1133 1.0455+-0.0083 might be 1.0636x faster fixed-typed-array-storage 0.7607+-0.0231 ? 0.8263+-0.1761 ? might be 1.0863x slower Float32Array-matrix-mult 3.4659+-0.0719 ? 3.5764+-0.1873 ? might be 1.0319x slower Float32Array-to-Float64Array-set 44.7214+-0.4520 ? 44.7829+-0.3646 ? Float64Array-alloc-long-lived 62.8608+-0.9082 62.8035+-1.2982 Float64Array-to-Int16Array-set 55.1075+-2.7049 53.9890+-0.5053 might be 1.0207x faster fold-double-to-int 11.4140+-0.5357 11.2550+-0.1672 might be 1.0141x faster fold-get-by-id-to-multi-get-by-offset-rare-int 10.3053+-2.2968 ? 10.3747+-1.2673 ? fold-get-by-id-to-multi-get-by-offset 9.2518+-0.7121 ? 9.8688+-1.2182 ? might be 1.0667x slower fold-multi-get-by-offset-to-get-by-offset 9.0374+-2.0844 8.8925+-0.8522 might be 1.0163x faster fold-multi-get-by-offset-to-poly-get-by-offset 8.6640+-0.6937 8.6082+-1.0407 fold-multi-put-by-offset-to-poly-put-by-offset 8.4756+-0.4555 8.1950+-1.0046 might be 1.0342x faster fold-multi-put-by-offset-to-put-by-offset 4.3945+-1.0609 3.9868+-1.0151 might be 1.1023x faster fold-multi-put-by-offset-to-replace-or-transition-put-by-offset 8.2040+-0.7673 ? 8.4639+-0.8969 ? might be 1.0317x slower fold-put-by-id-to-multi-put-by-offset 9.7743+-1.6612 8.8141+-0.3255 might be 1.1089x faster fold-put-structure 5.9258+-1.5907 4.3293+-0.7145 might be 1.3688x faster for-of-iterate-array-entries 10.9377+-0.4700 10.5499+-0.4384 might be 1.0368x faster for-of-iterate-array-keys 3.3280+-0.0624 3.2230+-0.1599 might be 1.0326x faster for-of-iterate-array-values 3.1405+-0.1309 ? 3.2747+-0.4905 ? might be 1.0427x slower fround 18.1347+-0.1569 17.4070+-0.8415 might be 1.0418x faster ftl-library-inlining-dataview 53.2344+-0.3773 ? 54.2917+-0.7936 ? might be 1.0199x slower ftl-library-inlining 104.3997+-0.5498 ? 104.6085+-0.6766 ? function-call 10.1235+-0.3412 10.1002+-0.3815 function-dot-apply 1.8473+-0.0370 1.8124+-0.0425 might be 1.0193x faster function-test 2.4758+-0.0749 2.4536+-0.0736 function-with-eval 100.2545+-2.4707 99.1455+-1.8847 might be 1.0112x faster gcse-poly-get-less-obvious 13.2332+-0.1888 ? 13.2620+-0.1804 ? gcse-poly-get 15.0040+-0.3376 ? 15.0042+-0.2826 ? gcse 3.5072+-0.0413 ? 3.6536+-0.2524 ? might be 1.0417x slower get-by-id-bimorphic-check-structure-elimination-simple 2.3627+-0.0188 ? 2.4057+-0.0589 ? might be 1.0182x slower get-by-id-bimorphic-check-structure-elimination 5.2856+-0.0694 ? 5.3297+-0.1092 ? get-by-id-chain-from-try-block 5.0725+-0.0822 ? 5.1132+-0.0994 ? get-by-id-check-structure-elimination 4.0588+-0.1990 ? 4.0662+-0.0880 ? get-by-id-proto-or-self 14.8710+-3.4879 13.6914+-0.4935 might be 1.0862x faster get-by-id-quadmorphic-check-structure-elimination-simple 2.6988+-0.1670 2.6925+-0.1788 get-by-id-self-or-proto 14.2288+-2.5867 13.5596+-0.3787 might be 1.0494x faster get-by-val-out-of-bounds 3.8060+-0.1025 3.6995+-0.0879 might be 1.0288x faster get-by-val-with-string-bimorphic-check-structure-elimination-simple 2.7106+-0.4624 2.5845+-0.1083 might be 1.0488x faster get-by-val-with-string-bimorphic-check-structure-elimination 5.4758+-0.0398 ? 5.5474+-0.1350 ? might be 1.0131x slower get-by-val-with-string-chain-from-try-block 5.2258+-0.1804 ? 5.2330+-0.1063 ? get-by-val-with-string-check-structure-elimination 4.7305+-0.1112 4.7122+-0.0401 get-by-val-with-string-proto-or-self 13.5417+-0.2486 ? 14.0054+-1.3285 ? might be 1.0342x slower get-by-val-with-string-quadmorphic-check-structure-elimination-simple 2.8921+-0.0438 ? 2.9347+-0.1000 ? might be 1.0147x slower get-by-val-with-string-self-or-proto 13.6237+-0.5063 ? 14.1215+-1.1678 ? might be 1.0365x slower get-by-val-with-symbol-bimorphic-check-structure-elimination-simple 2.9509+-0.5143 2.8294+-0.1112 might be 1.0429x faster get-by-val-with-symbol-bimorphic-check-structure-elimination 11.5400+-0.2291 ? 13.3361+-5.5022 ? might be 1.1556x slower get-by-val-with-symbol-chain-from-try-block 5.3015+-0.1655 5.2040+-0.0863 might be 1.0187x faster get-by-val-with-symbol-check-structure-elimination 10.4081+-0.2300 ? 10.5082+-0.3519 ? get-by-val-with-symbol-proto-or-self 13.6393+-0.5944 ? 14.3726+-1.6124 ? might be 1.0538x slower get-by-val-with-symbol-quadmorphic-check-structure-elimination-simple 3.6711+-0.2403 3.6393+-0.1309 get-by-val-with-symbol-self-or-proto 13.6231+-0.4803 ? 14.6212+-1.3898 ? might be 1.0733x slower get_callee_monomorphic 2.4030+-0.1001 2.3585+-0.0453 might be 1.0188x faster get_callee_polymorphic 2.9083+-0.1037 ? 2.9111+-0.0161 ? getter-no-activation 4.5791+-0.2079 ? 4.5941+-0.3525 ? getter-prototype 9.5087+-0.4955 9.3455+-0.1155 might be 1.0175x faster getter-richards 117.1927+-6.3313 111.6265+-14.8306 might be 1.0499x faster getter 4.9160+-0.2013 ? 5.8914+-1.1318 ? might be 1.1984x slower global-object-access-with-mutating-structure 5.0530+-0.0813 ? 5.2035+-0.0811 ? might be 1.0298x slower global-var-const-infer-fire-from-opt 0.7433+-0.1141 ? 0.7505+-0.0873 ? global-var-const-infer 0.5425+-0.0613 ? 0.5515+-0.1036 ? might be 1.0165x slower HashMap-put-get-iterate-keys 24.8284+-2.4094 24.6608+-2.0885 HashMap-put-get-iterate 28.1352+-1.6546 26.8884+-0.9686 might be 1.0464x faster HashMap-string-put-get-iterate 24.6302+-1.7032 ? 25.7580+-2.2963 ? might be 1.0458x slower hoist-make-rope 9.4874+-0.3702 9.4481+-1.4359 hoist-poly-check-structure-effectful-loop 3.6766+-0.0387 ? 3.8304+-0.2106 ? might be 1.0418x slower hoist-poly-check-structure 3.0370+-0.1552 3.0123+-0.0645 imul-double-only 6.7993+-0.8869 6.6915+-0.4266 might be 1.0161x faster imul-int-only 8.8172+-0.6548 8.3345+-1.1870 might be 1.0579x faster imul-mixed 6.5760+-0.1374 6.3395+-0.5516 might be 1.0373x faster in-four-cases 15.8185+-0.3080 ? 16.1951+-0.6874 ? might be 1.0238x slower in-one-case-false 9.2048+-0.5969 9.1288+-0.4980 in-one-case-true 8.9587+-0.6720 8.8094+-0.0713 might be 1.0169x faster in-two-cases 9.4320+-1.1630 ? 9.5925+-1.2098 ? might be 1.0170x slower indexed-properties-in-objects 2.4990+-0.0141 2.4952+-0.0285 infer-closure-const-then-mov-no-inline 2.8655+-0.1295 ? 2.9010+-0.1548 ? might be 1.0124x slower infer-closure-const-then-mov 16.8227+-0.2984 ? 17.0471+-0.3087 ? might be 1.0133x slower infer-closure-const-then-put-to-scope-no-inline 10.3746+-0.1244 10.3267+-0.1732 infer-closure-const-then-put-to-scope 20.0470+-0.8221 19.9337+-0.4564 infer-closure-const-then-reenter-no-inline 47.7305+-0.3026 ? 47.7427+-0.4830 ? infer-closure-const-then-reenter 20.4565+-0.8691 ? 21.1375+-0.3247 ? might be 1.0333x slower infer-constant-global-property 3.1286+-0.1161 ? 3.1580+-0.0932 ? infer-constant-property 2.4235+-0.1814 2.4199+-0.0893 infer-one-time-closure-ten-vars 8.0407+-0.5398 ? 8.0895+-0.3520 ? infer-one-time-closure-two-vars 7.6534+-0.1526 ? 7.6945+-0.2571 ? infer-one-time-closure 7.7328+-0.3340 7.5710+-0.0394 might be 1.0214x faster infer-one-time-deep-closure 12.1055+-0.0925 ? 12.1179+-0.3840 ? inline-arguments-access 3.7010+-0.0672 ? 3.8995+-0.2871 ? might be 1.0536x slower inline-arguments-aliased-access 3.6990+-0.0447 ? 3.7096+-0.0801 ? inline-arguments-local-escape 3.8208+-0.0987 3.7451+-0.0350 might be 1.0202x faster inline-get-scoped-var 4.4369+-0.1237 ? 4.4574+-0.2847 ? inlined-put-by-id-transition 10.7570+-0.5446 ? 10.9053+-0.6965 ? might be 1.0138x slower int-or-other-abs-then-get-by-val 4.4870+-0.1873 4.3750+-0.1674 might be 1.0256x faster int-or-other-abs-zero-then-get-by-val 14.9845+-0.1189 14.7850+-0.2692 might be 1.0135x faster int-or-other-add-then-get-by-val 3.7150+-0.1635 ? 3.7174+-0.1272 ? int-or-other-add 4.5416+-0.1412 4.5119+-0.0670 int-or-other-div-then-get-by-val 3.3738+-0.0979 ? 3.3963+-0.1342 ? int-or-other-max-then-get-by-val 3.6537+-0.2204 3.6505+-0.2019 int-or-other-min-then-get-by-val 3.6807+-0.2707 3.6400+-0.2136 might be 1.0112x faster int-or-other-mod-then-get-by-val 3.2963+-0.2867 3.1781+-0.0542 might be 1.0372x faster int-or-other-mul-then-get-by-val 3.2392+-0.0096 ? 3.2794+-0.0332 ? might be 1.0124x slower int-or-other-neg-then-get-by-val 4.1825+-0.2100 4.0921+-0.1515 might be 1.0221x faster int-or-other-neg-zero-then-get-by-val 14.8413+-0.1135 ? 14.8625+-0.1777 ? int-or-other-sub-then-get-by-val 3.7227+-0.1359 3.6203+-0.0244 might be 1.0283x faster int-or-other-sub 3.0823+-0.0835 ? 3.0973+-0.0660 ? int-overflow-local 4.1581+-0.3309 3.8074+-0.1160 might be 1.0921x faster Int16Array-alloc-long-lived 44.4243+-1.1790 44.1898+-0.7822 Int16Array-bubble-sort-with-byteLength 16.6092+-0.7520 16.4497+-0.4494 Int16Array-bubble-sort 16.4949+-0.1125 ? 17.0073+-1.6150 ? might be 1.0311x slower Int16Array-load-int-mul 1.2758+-0.0278 ? 1.2832+-0.0616 ? Int16Array-to-Int32Array-set 43.7805+-0.5413 ? 43.8387+-0.1314 ? Int32Array-alloc-large 10.9038+-0.9219 10.8234+-0.2259 Int32Array-alloc-long-lived 50.2711+-1.2289 ? 51.1222+-1.3229 ? might be 1.0169x slower Int32Array-alloc 2.7811+-0.1059 ? 2.9445+-0.3312 ? might be 1.0587x slower Int32Array-Int8Array-view-alloc 5.5891+-0.1493 ? 5.7365+-0.3419 ? might be 1.0264x slower int52-spill 4.1675+-0.1851 ? 4.1883+-0.2040 ? Int8Array-alloc-long-lived 39.3405+-1.3145 38.7850+-0.5137 might be 1.0143x faster Int8Array-load-with-byteLength 3.0730+-0.0253 ? 3.1763+-0.2103 ? might be 1.0336x slower Int8Array-load 3.1161+-0.1131 3.1064+-0.0245 integer-divide 9.7459+-0.1125 9.7088+-0.1826 integer-modulo 1.3969+-0.0129 1.3927+-0.0851 is-boolean-fold-tricky 3.4763+-0.1166 ? 3.5128+-0.1646 ? might be 1.0105x slower is-boolean-fold 2.3828+-0.0137 ? 2.4127+-0.0508 ? might be 1.0126x slower is-function-fold-tricky-internal-function 9.2090+-0.1823 9.1898+-0.1886 is-function-fold-tricky 3.7667+-0.0428 ? 3.8735+-0.0868 ? might be 1.0284x slower is-function-fold 2.4218+-0.1065 ? 2.4487+-0.0807 ? might be 1.0111x slower is-number-fold-tricky 3.7835+-0.2021 3.6926+-0.0201 might be 1.0246x faster is-number-fold 2.3909+-0.0206 2.3827+-0.0152 is-object-or-null-fold-functions 2.4260+-0.0328 ? 2.4501+-0.0555 ? is-object-or-null-fold-less-tricky 3.7793+-0.1053 ? 3.8517+-0.1584 ? might be 1.0192x slower is-object-or-null-fold-tricky 4.8323+-0.1695 4.8011+-0.1697 is-object-or-null-fold 2.4348+-0.0924 ? 2.4578+-0.1135 ? is-object-or-null-trickier-function 3.7950+-0.0928 ? 3.8434+-0.1311 ? might be 1.0128x slower is-object-or-null-trickier-internal-function 9.6400+-0.1860 ? 9.7579+-0.3854 ? might be 1.0122x slower is-object-or-null-tricky-function 3.8556+-0.1218 3.7738+-0.0408 might be 1.0217x faster is-object-or-null-tricky-internal-function 7.1075+-0.0527 ? 7.1138+-0.0949 ? is-string-fold-tricky 3.7090+-0.0535 ? 3.7397+-0.1000 ? is-string-fold 2.3805+-0.0156 ? 2.4266+-0.1123 ? might be 1.0194x slower is-undefined-fold-tricky 3.0208+-0.0320 ? 3.0745+-0.1151 ? might be 1.0178x slower is-undefined-fold 2.3977+-0.0139 ? 2.4587+-0.1098 ? might be 1.0255x slower JSONP-negative-0 0.2300+-0.0219 0.2158+-0.0066 might be 1.0656x faster large-int-captured 3.7728+-0.0670 ? 4.0264+-0.7473 ? might be 1.0672x slower large-int-neg 13.1509+-0.1698 ? 13.1954+-0.2603 ? large-int 12.6638+-0.2029 12.4351+-0.2757 might be 1.0184x faster load-varargs-elimination 20.6477+-0.4012 ^ 19.4882+-0.2508 ^ definitely 1.0595x faster logical-not-weird-types 2.6624+-0.1358 2.6383+-0.0942 logical-not 3.9583+-0.2867 3.8461+-0.1325 might be 1.0292x faster lots-of-fields 9.0632+-0.2842 9.0078+-0.4183 make-indexed-storage 2.4125+-0.2981 ? 2.4687+-0.0709 ? might be 1.0233x slower make-rope-cse 4.1198+-0.1817 4.0156+-0.0796 might be 1.0260x faster marsaglia-larger-ints 29.9467+-0.6955 29.6468+-0.3447 might be 1.0101x faster marsaglia-osr-entry 19.9342+-0.3525 19.8265+-0.2153 math-with-out-of-bounds-array-values 19.8613+-0.3092 ? 20.0020+-0.8370 ? max-boolean 2.4799+-0.0446 2.4730+-0.1327 method-on-number 16.0612+-1.3556 15.3971+-1.9407 might be 1.0431x faster min-boolean 2.4383+-0.0839 ? 2.5698+-0.4645 ? might be 1.0539x slower minus-boolean-double 2.8248+-0.0909 2.7959+-0.0254 might be 1.0103x faster minus-boolean 2.1680+-0.1215 2.1406+-0.0751 might be 1.0128x faster misc-strict-eq 29.5515+-0.4710 28.9578+-0.9018 might be 1.0205x faster mod-boolean-double 10.3306+-0.0741 ! 10.4968+-0.0921 ! definitely 1.0161x slower mod-boolean 7.7781+-0.0726 ? 7.8437+-0.2206 ? mul-boolean-double 3.3920+-0.1833 3.3356+-0.0885 might be 1.0169x faster mul-boolean 2.5851+-0.1210 2.5513+-0.0198 might be 1.0133x faster neg-boolean 2.8578+-0.0778 ? 2.8625+-0.1216 ? negative-zero-divide 0.2944+-0.0214 ? 0.3206+-0.0639 ? might be 1.0889x slower negative-zero-modulo 0.3026+-0.0240 ? 0.3222+-0.0400 ? might be 1.0650x slower negative-zero-negate 0.2756+-0.0180 0.2692+-0.0092 might be 1.0238x faster nested-function-parsing 42.0218+-0.4005 ? 42.0792+-2.1488 ? new-array-buffer-dead 99.3793+-2.4754 98.7009+-0.8377 new-array-buffer-push 6.0219+-0.0913 5.9671+-0.0456 new-array-dead 15.0530+-0.3517 ? 15.1920+-0.4470 ? new-array-push 3.3562+-0.0520 ? 3.5501+-0.3670 ? might be 1.0578x slower no-inline-constructor 33.8278+-1.3365 33.2073+-0.5220 might be 1.0187x faster number-test 2.6871+-0.1362 ? 2.7189+-0.1440 ? might be 1.0118x slower object-closure-call 4.4099+-0.0644 ? 4.4618+-0.1142 ? might be 1.0118x slower object-get-own-property-symbols-on-large-array 3.4291+-0.1010 ? 3.5024+-0.0844 ? might be 1.0214x slower object-test 2.4528+-0.0899 2.4512+-0.1404 obvious-sink-pathology-taken 118.2056+-4.6679 117.0830+-2.4729 obvious-sink-pathology 112.2554+-1.7088 112.0320+-0.9510 obviously-elidable-new-object 30.4865+-0.2238 ? 30.9480+-0.7748 ? might be 1.0151x slower plus-boolean-arith 2.2045+-0.1001 2.1542+-0.0131 might be 1.0234x faster plus-boolean-double 2.8821+-0.1394 ? 2.9208+-0.1857 ? might be 1.0134x slower plus-boolean 2.3508+-0.0652 2.3126+-0.0337 might be 1.0165x faster poly-chain-access-different-prototypes-simple 2.9725+-0.0676 ? 3.0477+-0.1686 ? might be 1.0253x slower poly-chain-access-different-prototypes 2.6594+-0.0687 ? 2.6886+-0.1960 ? might be 1.0110x slower poly-chain-access-simpler 2.9790+-0.1205 ? 3.0157+-0.1424 ? might be 1.0123x slower poly-chain-access 3.0645+-0.1502 2.9772+-0.0695 might be 1.0293x faster poly-stricteq 48.4321+-0.5676 ? 48.4660+-0.5014 ? polymorphic-array-call 1.0378+-0.0178 ? 1.0446+-0.0531 ? polymorphic-get-by-id 2.5576+-0.0486 ? 2.6830+-0.2654 ? might be 1.0490x slower polymorphic-put-by-id 23.4388+-0.5440 ? 23.8307+-1.0479 ? might be 1.0167x slower polymorphic-structure 12.2604+-0.3139 12.1516+-0.1410 polyvariant-monomorphic-get-by-id 6.4839+-1.3796 ? 6.7169+-1.9236 ? might be 1.0359x slower proto-getter-access 7.7809+-0.1709 7.6189+-0.1120 might be 1.0213x faster prototype-access-with-mutating-prototype 4.9463+-0.1937 ? 5.0630+-0.2276 ? might be 1.0236x slower put-by-id-replace-and-transition 7.4903+-0.1137 ? 7.6229+-0.1605 ? might be 1.0177x slower put-by-id-slightly-polymorphic 2.3905+-0.0160 2.3833+-0.0156 put-by-id 11.4437+-0.1052 ! 11.7938+-0.2237 ! definitely 1.0306x slower put-by-val-direct 0.3046+-0.0292 ? 0.3060+-0.0333 ? put-by-val-large-index-blank-indexing-type 4.8892+-0.1386 ? 4.9377+-0.0489 ? put-by-val-machine-int 2.1345+-0.0241 ? 2.1614+-0.0328 ? might be 1.0126x slower rare-osr-exit-on-local 13.9852+-0.9574 13.5388+-0.1564 might be 1.0330x faster register-pressure-from-osr 15.3957+-0.2457 ? 15.4186+-0.1786 ? repeat-multi-get-by-offset 20.3784+-0.3650 20.3503+-0.1455 setter-prototype 7.2803+-0.1504 ? 7.2916+-0.1250 ? setter 4.9507+-0.0901 ? 5.3691+-0.6632 ? might be 1.0845x slower simple-activation-demo 22.7606+-0.1660 22.6671+-0.1550 simple-getter-access 9.7059+-0.2382 ? 9.7530+-0.3336 ? simple-poly-call-nested 7.3297+-0.6011 ? 8.5966+-0.6791 ? might be 1.1728x slower simple-poly-call 1.1111+-0.0864 1.0692+-0.0409 might be 1.0392x faster sin-boolean 19.0338+-2.6455 17.5510+-2.1123 might be 1.0845x faster singleton-scope 57.7782+-0.3624 57.6447+-0.3950 sink-function 10.5342+-0.3065 ? 10.7287+-0.1497 ? might be 1.0185x slower sink-huge-activation 15.8782+-0.3012 ? 16.0065+-1.0538 ? sinkable-new-object-dag 58.8242+-6.2361 ? 59.5800+-2.7920 ? might be 1.0128x slower sinkable-new-object-taken 42.5032+-0.6262 ? 42.6943+-0.5460 ? sinkable-new-object 32.1003+-0.5976 ? 32.1489+-0.7934 ? slow-array-profile-convergence 2.1802+-0.2555 2.1092+-0.1264 might be 1.0337x faster slow-convergence 2.0793+-0.1092 ? 2.1117+-0.1196 ? might be 1.0156x slower slow-ternaries 16.2823+-0.1998 ? 16.3262+-0.2796 ? sorting-benchmark 15.7859+-0.7355 15.5121+-0.1863 might be 1.0177x faster sparse-conditional 1.0391+-0.0836 1.0102+-0.0157 might be 1.0286x faster splice-to-remove 11.5527+-0.2087 ? 11.5884+-0.1385 ? string-char-code-at 13.3163+-0.1066 ! 15.7315+-0.0808 ! definitely 1.1814x slower string-concat-object 2.1044+-0.1337 ? 2.1439+-0.3408 ? might be 1.0188x slower string-concat-pair-object 2.1172+-0.2588 2.0919+-0.2136 might be 1.0121x faster string-concat-pair-simple 10.5715+-0.3596 10.3104+-0.1269 might be 1.0253x faster string-concat-simple 10.6246+-0.4660 10.3213+-0.2170 might be 1.0294x faster string-cons-repeat 7.1260+-0.2411 ? 7.2795+-0.1290 ? might be 1.0215x slower string-cons-tower 7.3710+-0.1486 ? 7.4099+-0.1343 ? string-equality 14.1112+-0.1446 ? 14.2170+-0.2567 ? string-get-by-val-big-char 6.2004+-0.1633 ? 6.4890+-1.1789 ? might be 1.0465x slower string-get-by-val-out-of-bounds-insane 2.8797+-0.0494 ? 2.9246+-0.1612 ? might be 1.0156x slower string-get-by-val-out-of-bounds 3.7250+-0.1826 ? 3.7355+-0.0855 ? string-get-by-val 2.9686+-0.0771 ^ 2.5738+-0.0264 ^ definitely 1.1534x faster string-hash 1.7173+-0.1756 ? 1.7573+-0.0163 ? might be 1.0233x slower string-long-ident-equality 11.6415+-0.1187 ? 11.6634+-0.1315 ? string-out-of-bounds 10.1749+-0.1802 10.1293+-0.2696 string-repeat-arith 25.4526+-0.2735 25.4291+-0.4812 string-sub 49.5026+-0.5943 49.4460+-0.1404 string-test 2.5120+-0.0311 ? 2.6223+-0.1616 ? might be 1.0439x slower string-var-equality 24.7866+-0.5373 24.5566+-0.5194 structure-hoist-over-transitions 2.2693+-0.2080 2.2115+-0.1624 might be 1.0261x faster substring-concat-weird 36.1452+-0.3454 35.8426+-0.1411 substring-concat 38.8893+-0.3189 ? 39.4170+-0.7695 ? might be 1.0136x slower substring 43.4373+-1.3797 ? 43.5703+-0.5629 ? switch-char-constant 2.5798+-0.1472 2.5246+-0.1410 might be 1.0219x faster switch-char 5.6861+-1.5569 ? 5.7839+-1.1761 ? might be 1.0172x slower switch-constant 7.6008+-0.4400 7.5909+-0.2412 switch-string-basic-big-var 12.9896+-0.0942 ? 13.0135+-0.1002 ? switch-string-basic-big 13.7849+-0.1447 ? 13.9345+-0.2110 ? might be 1.0109x slower switch-string-basic-var 12.1566+-0.3900 ? 12.3443+-0.5409 ? might be 1.0154x slower switch-string-basic 11.9726+-0.3665 11.9400+-0.1886 switch-string-big-length-tower-var 16.6581+-0.3343 ? 16.7551+-0.3204 ? switch-string-length-tower-var 12.0955+-0.1430 ? 12.2636+-0.5109 ? might be 1.0139x slower switch-string-length-tower 10.9318+-0.0893 10.9060+-0.0994 switch-string-short 10.9148+-0.0574 ? 11.0245+-0.2145 ? might be 1.0101x slower switch 10.7757+-0.1698 ? 11.1528+-0.6738 ? might be 1.0350x slower tear-off-arguments-simple 2.8628+-0.1102 2.8470+-0.1315 tear-off-arguments 4.1085+-0.3602 3.8113+-0.0528 might be 1.0780x faster temporal-structure 11.2295+-0.2660 ? 11.2422+-0.3417 ? to-int32-boolean 11.5236+-0.2324 ? 11.7186+-0.1727 ? might be 1.0169x slower try-catch-get-by-val-cloned-arguments 13.0929+-0.5567 12.8331+-0.2791 might be 1.0202x faster try-catch-get-by-val-direct-arguments 6.2526+-0.1947 ^ 5.9146+-0.0845 ^ definitely 1.0571x faster try-catch-get-by-val-scoped-arguments 6.8518+-0.2032 ? 6.8911+-0.0858 ? typed-array-get-set-by-val-profiling 25.4507+-0.7346 ? 25.5025+-0.5316 ? undefined-property-access 214.7244+-0.6866 214.2297+-0.7756 undefined-test 2.6466+-0.0795 ? 2.6562+-0.0444 ? unprofiled-licm 13.4975+-0.1479 13.4937+-0.4884 varargs-call 12.7442+-0.2104 12.6175+-0.2306 might be 1.0100x faster varargs-construct-inline 23.6071+-0.5459 23.2943+-0.4000 might be 1.0134x faster varargs-construct 18.9167+-0.4187 18.6028+-0.1908 might be 1.0169x faster varargs-inline 7.9790+-0.2391 7.9091+-0.2478 varargs-strict-mode 8.5623+-0.2645 8.4319+-0.1005 might be 1.0155x faster varargs 8.3916+-0.1661 ? 8.5083+-0.1507 ? might be 1.0139x slower weird-inlining-const-prop 1.9417+-0.3623 1.8581+-0.2318 might be 1.0450x faster <geometric> 7.2156+-0.0342 ? 7.2175+-0.0184 ? might be 1.0003x slower Conf#1 Conf#2 AsmBench: bigfib.cpp 428.4682+-19.3181 ? 441.5974+-7.1162 ? might be 1.0306x slower cray.c 380.6041+-5.0616 380.0106+-2.8254 dry.c 412.0317+-0.8239 ? 413.0728+-1.6410 ? FloatMM.c 662.6547+-1.9008 ? 664.0132+-7.1603 ? gcc-loops.cpp 3313.1998+-17.7356 3307.1614+-14.9700 n-body.c 797.1549+-3.0801 796.4428+-1.8092 Quicksort.c 390.0220+-8.5347 ? 393.8835+-2.7129 ? stepanov_container.cpp 3323.7608+-25.7710 ? 3327.9243+-31.9163 ? Towers.c 225.1711+-1.4582 ? 226.7105+-4.7830 ? <geometric> 687.5052+-5.3035 ? 691.2275+-2.6753 ? might be 1.0054x slower Conf#1 Conf#2 CompressionBench: huffman 48.0900+-2.0320 ? 49.3334+-0.9761 ? might be 1.0259x slower arithmetic-simple 258.9332+-1.5666 ! 261.6910+-1.1370 ! definitely 1.0107x slower arithmetic-precise 235.0129+-3.4424 ? 238.8639+-3.4397 ? might be 1.0164x slower arithmetic-complex-precise 238.1310+-1.2883 237.6783+-3.2221 arithmetic-precise-order-0 272.4985+-1.4502 ? 272.5027+-3.3747 ? arithmetic-precise-order-1 295.0438+-9.2247 293.7849+-9.3492 arithmetic-precise-order-2 338.4001+-7.1616 336.4028+-6.3082 arithmetic-simple-order-1 312.1372+-2.0283 ? 316.5311+-4.6955 ? might be 1.0141x slower arithmetic-simple-order-2 362.0999+-6.9152 ? 369.4825+-13.2239 ? might be 1.0204x slower lz-string 288.0353+-8.2231 ? 288.4854+-15.1277 ? <geometric> 239.3404+-1.0202 ? 241.1609+-1.4670 ? might be 1.0076x slower Conf#1 Conf#2 Geomean of preferred means: <scaled-result> 48.5736+-0.1493 ? 48.7632+-0.0423 ? might be 1.0039x slower
Benjamin Poulain
Comment 3 2015-08-18 19:23:51 PDT
Created attachment 259342 [details] Patch
Geoffrey Garen
Comment 4 2015-08-18 19:36:57 PDT
Comment on attachment 259342 [details] Patch r=me
Filip Pizlo
Comment 5 2015-08-18 19:45:21 PDT
Comment on attachment 259342 [details] Patch Nice! LGTM too.
Benjamin Poulain
Comment 6 2015-08-18 21:05:53 PDT
*** Bug 148045 has been marked as a duplicate of this bug. ***
Benjamin Poulain
Comment 7 2015-08-18 21:09:17 PDT
Committed r188624: <http://trac.webkit.org/changeset/188624>
Note You need to log in before you can comment on or make changes to this bug.