Bug 14715 – REGRESSION: Hang loading/scrolling iExploder test #521

• Status: NEW
• Component: Layout and Rendering

[David Kilzer (:ddkilzer)]

* SUMMARY
After applying the patch to Bugzilla Bug 14713 (http://bugs.webkit.org/show_bug.cgi?id=14713), running iExploder test #521 causes an out-of-memory error.

* STEPS TO REPRODUCE
1. Launch Safari/WebKit.
2. Open attached test case: 5352867-v2.html

* RESULTS
Safari eventually crashes with an out-of-memory error.  Console output:

Safari(3425,0xa000ed88) malloc: *** vm_allocate(size=1158688768) failed (error code=3)
Safari(3425,0xa000ed88) malloc: *** error: can't allocate region
Safari(3425,0xa000ed88) malloc: *** set a breakpoint in szone_error to debug
Bus error

* REGRESSION
Only tested with a local debug build of WebKit r24513 with Safari 3.0 (522.12) on Mac OS X 10.4.10 (8R218).

* NOTES
<rdar://problem/5352867>

[David Kilzer (:ddkilzer)]

7/22/07 10:51 AM David Kilzer:
Same crash occurs in Safari 3.0 (522.12) public beta (with shipping WebKit).

7/22/07 10:52 AM David Kilzer:
Crash does NOT occur with shipping Safari 2.0.4 (419.3) with original Safari on Mac OS X 10.4.10 (8R218).

[David Kilzer (:ddkilzer)]

Created attachment 15631 [details]
Original iExploder source for test #521

[David Kilzer (:ddkilzer)]

Created attachment 15632 [details]
Test case

[David Kilzer (:ddkilzer)]

With a local debug build of WebKit r25122 with Safari 3 Public Beta v. 3.0.3 (522.12.1) on Mac OS X 10.4.10 (8R218), the test case (Attachment #15632 [details]) now hangs Safari when loading the page and when scrolling the page horizontally.

This behavior does not occur in shipping Safari 2.0.4 with its original WebKit on 10.4.10, so this is still a regression.

[Gavin Sherlock]

This no longer hangs webkit, using r61056 + Safari 5.  CPU usage is very high when scrolling, suggesting further optimization is possible, as nothing in the view actually changes.  As it no longer hangs, I suggest that it be downgraded to a P2, and the title be changed.