RESOLVED FIXED Bug 97080
DFG should not emit PutByVal hole case unless it has to
https://bugs.webkit.org/show_bug.cgi?id=97080
Summary DFG should not emit PutByVal hole case unless it has to
Filip Pizlo
Reported 2012-09-19 00:57:30 PDT
This is useful for two reasons: 1) If the hole case is never hit then we should just emit less code. The hole check becomes a speculation check instead of a jump over some slow path code. 2) It gives us some ability to bloat the hole case code if we need to, for example if it becomes useful for contiguous array inference.
Attachments
work in progress (24.32 KB, patch)
2012-09-19 01:01 PDT, Filip Pizlo
webkit-ews: commit-queue-
the patch (27.84 KB, patch)
2012-09-19 02:44 PDT, Filip Pizlo
ggaren: review+
Filip Pizlo
Comment 1 2012-09-19 01:01:05 PDT
Created attachment 164683 [details] work in progress
Early Warning System Bot
Comment 2 2012-09-19 01:07:57 PDT
Comment on attachment 164683 [details] work in progress Attachment 164683 [details] did not pass qt-ews (qt): Output: http://queues.webkit.org/results/13903416
Early Warning System Bot
Comment 3 2012-09-19 01:08:19 PDT
Comment on attachment 164683 [details] work in progress Attachment 164683 [details] did not pass qt-wk2-ews (qt): Output: http://queues.webkit.org/results/13898436
Gyuyoung Kim
Comment 4 2012-09-19 01:10:50 PDT
Comment on attachment 164683 [details] work in progress Attachment 164683 [details] did not pass efl-ews (efl): Output: http://queues.webkit.org/results/13892465
Build Bot
Comment 5 2012-09-19 01:26:57 PDT
Comment on attachment 164683 [details] work in progress Attachment 164683 [details] did not pass win-ews (win): Output: http://queues.webkit.org/results/13899447
Filip Pizlo
Comment 6 2012-09-19 02:44:32 PDT
Created attachment 164699 [details] the patch
Geoffrey Garen
Comment 7 2012-09-19 08:43:12 PDT
Comment on attachment 164699 [details] the patch r=me
Filip Pizlo
Comment 8 2012-09-19 14:45:00 PDT
Note You need to log in before you can comment on or make changes to this bug.