I forgot to upload my microbenchmarks in the actual patch.
Created attachment 384097 [details] patch
Comment on attachment 384097 [details] patch View in context: https://bugs.webkit.org/attachment.cgi?id=384097&action=review r=me. > Source/JavaScriptCore/ChangeLog:24 > + between a GetByStatus saying "we're a slow path" and if we actually observed the StructureStubInfo > + go to the slow path. In the DFG/FTL, we only forgo emitting a get by val IC when observing a Maybe it's just me, but I found this a bit confusing throughout the patch. It's kinda weird that we say it's a slow path if the slow path wasn't taken, but I'm not sure what would be a better alternative, so I'll leave it to you to decide whether it makes sense or not.
Created attachment 384111 [details] patch for landing
Created attachment 384113 [details] patch for landing
Comment on attachment 384113 [details] patch for landing Clearing flags on attachment: 384113 Committed r252763: <https://trac.webkit.org/changeset/252763>
All reviewed patches have been landed. Closing bug.
<rdar://problem/57419422>