Bug 156815 – [JSC] DFG should not generate two jumps when the target of DoubleBranch is the next block

• Status: RESOLVED
• Resolution: FIXED
• Component: New Bugs

[Benjamin Poulain]

[JSC] DFG should not generate two jumps when the target of DoubleBranch is the next block

[Benjamin Poulain]

Created attachment 276857 [details]
Patch

[Benjamin Poulain]

Minor improvement for the Math parts of Sunspider:

                                    Conf#1                    Conf#2                                      

3d-cube                         5.0044+-0.1016            4.9578+-0.0642        
3d-morph                        5.0955+-0.0679            5.0768+-0.0793        
3d-raytrace                     5.2357+-0.0917     ?      5.2715+-0.1770        ?
access-binary-trees             2.1464+-0.0720            2.1233+-0.0712          might be 1.0109x faster
access-fannkuch                 5.9876+-0.2665            5.8314+-0.1054          might be 1.0268x faster
access-nbody                    2.5794+-0.0468     ?      2.5806+-0.0796        ?
access-nsieve                   3.0613+-0.0721     ?      3.0741+-0.1351        ?
bitops-3bit-bits-in-byte        1.1289+-0.0177     ?      1.1659+-0.0537        ? might be 1.0328x slower
bitops-bits-in-byte             2.8369+-0.1171     ?      2.8464+-0.0572        ?
bitops-bitwise-and              2.1124+-0.0546            2.0740+-0.0438          might be 1.0185x faster
bitops-nsieve-bits              3.1444+-0.0497            3.1089+-0.0373          might be 1.0114x faster
controlflow-recursive           2.3694+-0.0715     ?      2.3979+-0.0859        ? might be 1.0120x slower
crypto-aes                      4.2285+-0.1511            4.0840+-0.0410          might be 1.0354x faster
crypto-md5                      2.4494+-0.0453     ?      2.4816+-0.0780        ? might be 1.0131x slower
crypto-sha1                     2.4393+-0.0886            2.3291+-0.0474          might be 1.0473x faster
date-format-tofte               6.5690+-0.0819     ?      6.6226+-0.2051        ?
date-format-xparb               4.6600+-0.0957            4.6598+-0.1215        
math-cordic                     2.8995+-0.1376     ?      2.9233+-0.1012        ?
math-partial-sums               5.0460+-0.2066            4.9070+-0.1101          might be 1.0283x faster
math-spectral-norm              1.9898+-0.0282     ?      2.0161+-0.0468        ? might be 1.0132x slower
regexp-dna                      6.4846+-0.1613     ?      6.5029+-0.1323        ?
string-base64                   4.6821+-0.1017     ?      4.7872+-0.1703        ? might be 1.0224x slower
string-fasta                    5.8401+-0.1071     ?      5.9178+-0.1276        ? might be 1.0133x slower
string-tagcloud                 8.2384+-0.2066            8.1565+-0.1399          might be 1.0101x faster
string-unpack-code             19.4229+-0.7048           19.3475+-0.5457        
string-validate-input           4.4637+-0.1204            4.3795+-0.0947          might be 1.0192x faster

<arithmetic>                    4.6198+-0.0369            4.6009+-0.0268          might be 1.0041x faster

[Mark Lam]

Comment on attachment 276857 [details]
Patch

r=me

[WebKit Commit Bot]

Comment on attachment 276857 [details]
Patch

Clearing flags on attachment: 276857

Committed r199796: <http://trac.webkit.org/changeset/199796>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.