[Webkit-unassigned] [Bug 79899] DFG BasicBlocks should not require that their nodes have contiguous indices in the graph
bugzilla-daemon at webkit.org
bugzilla-daemon at webkit.org
Wed Feb 29 19:20:43 PST 2012
https://bugs.webkit.org/show_bug.cgi?id=79899
Yuqiang Xian <yuqiang.xian at intel.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Attachment #129448|0 |1
is obsolete| |
Attachment #129614| |review?
Flag| |
--- Comment #5 from Yuqiang Xian <yuqiang.xian at intel.com> 2012-02-29 19:20:42 PST ---
Created an attachment (id=129614)
--> (https://bugs.webkit.org/attachment.cgi?id=129614&action=review)
Patch for review
Filip, thanks for the clarification.
Here's the patch for review.
Performance result is here as well (almost neutral):
Collected 12 samples per benchmark/VM, with 4 VM invocations per benchmark. Emitted a call to gc() between sample
measurements. Used 1 benchmark iteration per VM invocation for warm-up. Used the jsc-specific preciseTime()
function to get microsecond-level timing. Reporting benchmark execution times with 95% confidence intervals in
milliseconds.
ToT 79899
SunSpider:
3d-cube 8.1317+-0.0753 ? 8.1812+-0.0359 ?
3d-morph 9.6125+-0.0396 ? 9.6248+-0.0347 ?
3d-raytrace 10.4898+-0.1458 10.4073+-0.0569
access-binary-trees 2.4043+-0.0633 ? 2.4274+-0.0725 ?
access-fannkuch 10.7972+-0.9117 10.3027+-0.0398 might be 1.0480x faster
access-nbody 5.4502+-0.0471 ? 5.4977+-0.0513 ?
access-nsieve 4.0119+-0.0407 4.0100+-0.0545
bitops-3bit-bits-in-byte 1.2555+-0.0202 1.2482+-0.0243
bitops-bits-in-byte 4.6557+-0.0646 4.6193+-0.0646
bitops-bitwise-and 4.2758+-0.0490 ? 4.3005+-0.0636 ?
bitops-nsieve-bits 4.2680+-0.0539 4.2642+-0.0477
controlflow-recursive 2.8596+-0.0376 2.8366+-0.0324
crypto-aes 9.5863+-0.1921 ? 9.6874+-0.1197 ? might be 1.0105x slower
crypto-md5 3.3988+-0.0716 ? 3.4228+-0.0502 ?
crypto-sha1 2.7206+-0.0500 ? 2.7903+-0.0403 ? might be 1.0256x slower
date-format-tofte 13.6587+-1.9276 12.7911+-0.0741 might be 1.0678x faster
date-format-xparb 11.9622+-0.1335 ? 12.1260+-0.1427 ? might be 1.0137x slower
math-cordic 4.0385+-0.0675 ? 4.0777+-0.0784 ?
math-partial-sums 14.8533+-0.0333 ? 15.5325+-1.5644 ? might be 1.0457x slower
math-spectral-norm 2.7100+-0.0501 2.6963+-0.0447
regexp-dna 9.3486+-0.0752 ? 9.4206+-0.0760 ?
string-base64 5.6182+-0.0711 ? 5.7111+-0.1056 ? might be 1.0165x slower
string-fasta 9.4383+-0.0563 9.4094+-0.0468
string-tagcloud 16.9364+-1.9261 16.1189+-0.0514 might be 1.0507x faster
string-unpack-code 27.5010+-0.2195 27.4682+-0.0478
string-validate-input 8.3231+-0.0630 ? 8.3302+-0.0378 ?
<arithmetic> * 8.0118+-0.1202 7.9732+-0.0610 might be 1.0048x faster
<geometric> 6.3050+-0.0562 6.3023+-0.0375 might be 1.0004x faster
<harmonic> 4.8941+-0.0416 ? 4.9019+-0.0366 ? might be 1.0016x slower
ToT 79899
V8:
crypto 94.0062+-0.5066 ? 95.0840+-1.2626 ? might be 1.0115x slower
deltablue 165.9663+-1.3598 ? 166.3503+-1.6042 ?
earley-boyer 121.4067+-2.5646 119.8530+-2.6862 might be 1.0130x faster
raytrace 58.6499+-0.4516 ? 59.4065+-0.4719 ? might be 1.0129x slower
regexp 111.1548+-0.4702 110.9294+-0.4109
richards 191.6663+-1.2047 190.4287+-0.8902
splay 77.4138+-0.9005 77.3018+-0.2461
<arithmetic> 117.1806+-0.5576 117.0505+-0.6485 might be 1.0011x faster
<geometric> * 109.0251+-0.5826 ? 109.0822+-0.6494 ? might be 1.0005x slower
<harmonic> 101.3128+-0.6036 ? 101.5676+-0.6212 ? might be 1.0025x slower
ToT 79899
Kraken:
ai-astar 789.5488+-4.9142 788.0722+-5.0229
audio-beat-detection 237.0390+-0.3647 ? 238.9648+-1.5770 ?
audio-dft 367.7708+-2.5980 ? 378.8661+-9.5683 ? might be 1.0302x slower
audio-fft 151.2529+-0.0775 ? 151.2913+-0.1593 ?
audio-oscillator 344.1697+-1.3283 ? 345.5862+-2.6765 ?
imaging-darkroom 377.7554+-9.3130 375.5139+-9.5351
imaging-desaturate 303.6464+-0.6332 303.4512+-0.6833
imaging-gaussian-blur 506.0675+-0.1486 ? 506.2780+-0.2269 ?
json-parse-financial 85.7818+-0.3590 ! 86.9759+-0.7928 ! definitely 1.0139x slower
json-stringify-tinderbox 106.0411+-0.7759 ? 106.0652+-0.6543 ?
stanford-crypto-aes 105.6489+-0.5791 ? 106.1828+-0.4288 ?
stanford-crypto-ccm 100.3177+-0.8778 99.5192+-0.5648
stanford-crypto-pbkdf2 235.7453+-0.5540 ^ 234.9382+-0.1006 ^ definitely 1.0034x faster
stanford-crypto-sha256-iterative 102.8190+-0.5031 ? 103.1907+-0.5652 ?
<arithmetic> * 272.4003+-1.0940 ? 273.2068+-1.3736 ? might be 1.0030x slower
<geometric> 216.1398+-0.7343 ? 216.8310+-0.8875 ? might be 1.0032x slower
<harmonic> 174.7069+-0.5262 ? 175.2619+-0.5780 ? might be 1.0032x slower
ToT 79899
All benchmarks:
<arithmetic> 103.0250+-0.4143 ? 103.2245+-0.4864 ? might be 1.0019x slower
<geometric> 27.6239+-0.1723 ? 27.6462+-0.1356 ? might be 1.0008x slower
<harmonic> 8.6053+-0.0719 ? 8.6194+-0.0633 ? might be 1.0016x slower
ToT 79899
Geomean of preferred means:
<scaled-result> 61.9631+-0.4216 61.9369+-0.2838 might be 1.0004x faster
--
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
More information about the webkit-unassigned
mailing list