ASSIGNED226914
DFGPredictionPropagation should be flow-sensitive 
https://bugs.webkit.org/show_bug.cgi?id=226914
Summary DFGPredictionPropagation should be flow-sensitive
Robin Morisset
Reported 2021-06-11 00:15:03 PDT
According to Phil, better predictions should help a lot with some of the hottest functions in the jquery subtest of Speedometer2.
Attachments
WIP (32.91 KB, patch)
2021-06-11 00:15 PDT, Robin Morisset 		no flags
DetailsFormatted DiffDiff
WIP (39.58 KB, patch)
2021-06-11 23:38 PDT, Robin Morisset 		no flags
DetailsFormatted DiffDiff
WIP (39.56 KB, patch)
2021-06-12 08:12 PDT, Robin Morisset 		ews-feeder: commit-queue-
DetailsFormatted DiffDiff
WIP_nearlyDone (57.18 KB, patch)
2021-06-14 18:59 PDT, Robin Morisset 		ews-feeder: commit-queue-
DetailsFormatted DiffDiff
Patch (61.18 KB, patch)
2021-06-15 18:23 PDT, Robin Morisset 		no flags
DetailsFormatted DiffDiff
Patch (68.14 KB, patch)
2021-06-28 10:45 PDT, Robin Morisset 		no flags
DetailsFormatted DiffDiff
Show Obsolete (5) View All Add attachment proposed patch, testcase, etc.
Robin Morisset
Comment 1 2021-06-11 00:15:44 PDT
Created attachment 431182 [details] WIP for EWS testing
Robin Morisset
Comment 2 2021-06-11 23:38:06 PDT
Created attachment 431251 [details] WIP Nearly done, only needs some thorough benchmarking + testing and a Changelog.
Robin Morisset
Comment 3 2021-06-12 08:12:24 PDT
Created attachment 431257 [details] WIP Fixed a bug: I thought that Symbols could only be equal to Symbols, and forgot that Object.toPrimitive can be called and turn one into a Symbol during equality.
Robin Morisset
Comment 4 2021-06-14 18:59:57 PDT
Created attachment 431392 [details] WIP_nearlyDone
Robin Morisset
Comment 5 2021-06-15 18:23:54 PDT
Created attachment 431507 [details] Patch I'm waiting for performance numbers from the A/B bots before asking for reviews, but apart from that this patch should be complete.
Radar WebKit Bug Importer
Comment 6 2021-06-18 00:15:36 PDT
<rdar://problem/79492202>
Robin Morisset
Comment 7 2021-06-28 10:45:17 PDT
Created attachment 432409 [details] Patch Roughly neutral overall, it is a progression on some subtests, most notably on ai-astar (14%), but is a regression on mandrel (12% on first iteration) and a smaller one (1 to 2%) on async-fs and coffeescript-wtb. Coarsening the numerical predictions fixed a massive segmentation regression, but seems to have reduced the size of the wins as well.
Note You need to log in before you can comment on or make changes to this bug.