Wednesday, 19 April 2006

Safari crashes

Safari crashes in exactly one situation in my daily browsing habits: while rendering macworld.com. It’s totally reproducible and happens virtually every day.

Macworld is the most popular US Mac magazine. Safari is the most popular Mac Web browser. Does this seem completely bizarre to anyone else?

* 7:41 PM | Computing, Web, Mac OS X

4 comments on “Safari crashes”

  1. Jussi
    19 April 2006 | 9:39 PM

    Yes, it does seem a bit bizarre. I have not been seeing that on my system and I too surf at Macworld almost every day.

    Have you investigated the crashlogs, is there anything that might give a clue why that is happening? If so and you have the time you probably should report the bug to the Bugzilla of WebKit Project.

  2. Nicholas Riley
    20 April 2006 | 12:34 AM

    Here’s the crash log:

    Exception: EXC_BAD_ACCESS (0×0001)
    Codes: KERN_PROTECTION_FAILURE (0×0002) at 0×00028038

    Thread 0 Crashed:
    0 com.apple.WebCore 0×95893818 khtml::RenderStyle::RenderStyle[unified]() + 36
    1 com.apple.WebCore 0×958a48a0 khtml::CSSStyleSelector::styleForElement(DOM::ElementImpl*, khtml::RenderStyle*, bool) + 284
    2 com.apple.WebCore 0×958a4670 DOM::NodeImpl::createRendererIfNeeded() + 108
    3 com.apple.WebCore 0×958a45ec DOM::ElementImpl::attach() + 24
    4 com.apple.WebCore 0×958a367c KHTMLParser::insertNode(DOM::NodeImpl*, bool) + 2444
    5 com.apple.WebCore 0×958a17f0 KHTMLParser::parseToken(khtml::Token*) + 620
    6 com.apple.WebCore 0×9589e898 khtml::HTMLTokenizer::processToken() + 460
    7 com.apple.WebCore 0×958a01fc khtml::HTMLTokenizer::parseTag(khtml::TokenizerString&) + 6296
    8 com.apple.WebCore 0×9589e118 khtml::HTMLTokenizer::write(khtml::TokenizerString const&, bool) + 928
    9 com.apple.WebCore 0×958d855c khtml::HTMLTokenizer::timerEvent(QTimerEvent*) + 252
    10 com.apple.WebCore 0×958d8414 -[KWQObjectTimerTarget sendTimerEvent] + 80
    11 com.apple.Foundation 0×928e949c __NSFireTimer + 116
    12 com.apple.CoreFoundation 0×90770aec __CFRunLoopDoTimer + 184
    13 com.apple.CoreFoundation 0×9075d464 __CFRunLoopRun + 1680
    14 com.apple.CoreFoundation 0×9075ca18 CFRunLoopRunSpecific + 268
    15 com.apple.HIToolbox 0×9318e1e0 RunCurrentEventLoopInMode + 264
    16 com.apple.HIToolbox 0×9318d874 ReceiveNextEventCommon + 380
    17 com.apple.HIToolbox 0×9318d6e0 BlockUntilNextEventMatchingListInMode + 96
    18 com.apple.AppKit 0×9368c104 _DPSNextEvent + 384
    19 com.apple.AppKit 0×9368bdc8 -[NSApplication nextEventMatchingMask:untilDate:inMode:dequeue:] + 116
    20 com.apple.Safari 0×000072d4 0×1000 + 25300
    21 com.apple.AppKit 0×9368830c -[NSApplication run] + 472
    22 com.apple.AppKit 0×93778e68 NSApplicationMain + 452
    23 com.apple.Safari 0×0005cfdc 0×1000 + 376796
    24 com.apple.Safari 0×0005ce80 0×1000 + 376448

    Unfortunately, just then I couldn’t reproduce it (it usually happens when I load multiple pages into tabs in the background, then do something else). Guess I’ll have to put some more time into it.

  3. Paul D. Buck
    3 June 2006 | 9:31 AM

    The problem is not site specific.

    It is not even operation specific (as near as I can tell), meaning, I can get it when clicking on *ANY* new link, after a zip file finishes un-zippping, etc.

    In that this is happening to others I can now eliminate *MY* hardware as being the cause.

    I have not been using tabs so that is not specifically it either. Sigh. I have been reporting each and every occurrence if for no other reason than to “bug” them with a high problem count. I can only wish others are doing the same …

  4. Paul D. Buck
    3 June 2006 | 9:32 AM

    Drat, forgot to check the check box! got it this time …

Leave a reply