WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED FIXED
Bug 82052
[Qt] REGRESSION(
r111480
): It made svg/as-image and svg/as-objects tests assert
https://bugs.webkit.org/show_bug.cgi?id=82052
Summary
[Qt] REGRESSION(r111480): It made svg/as-image and svg/as-objects tests assert
Csaba Osztrogonác
Reported
2012-03-23 06:48:23 PDT
You can easily reproduce it: $Tools/Scripts/old-run-webkit-tests --debug svg/as-image first run: svg/as-image/img-preserveAspectRatio-support-2.html -> crashed svg/as-image/svg-image-leak-loader.html -> crashed second run: svg/as-image/animated-svg-as-image.html -> crashed svg/as-image/svg-image-leak-loader.html -> crashed ... ASSERTION FAILED: !needsLayout() ../../../../Source/WebCore/page/FrameView.cpp(2958) : virtual void WebCore::FrameView::paintContents(WebCore::GraphicsContext*, const WebCore::IntRect&) 1 0x7f2bc7c28b27 /home/oszi/WebKit/WebKitBuild/Debug/lib/libQtWebKit.so.4(WebCore::FrameView::paintContents(WebCore::GraphicsContext*, WebCore::IntRect const&)+0x25f) [0x7f2bc7c28b27] 2 0x7f2bc7d2b811 /home/oszi/WebKit/WebKitBuild/Debug/lib/libQtWebKit.so.4(WebCore::ScrollView::paint(WebCore::GraphicsContext*, WebCore::IntRect const&)+0x4ab) [0x7f2bc7d2b811] 3 0x7f2bc8155d89 /home/oszi/WebKit/WebKitBuild/Debug/lib/libQtWebKit.so.4(WebCore::SVGImage::draw(WebCore::GraphicsContext*, WebCore::FloatRect const&, WebCore::FloatRect const&, WebCore::ColorSpace, WebCore::CompositeOperator)+0x3a3) [0x7f2bc8155d89] 4 0x7f2bc81558f3 /home/oszi/WebKit/WebKitBuild/Debug/lib/libQtWebKit.so.4(WebCore::SVGImage::drawSVGToImageBuffer(WebCore::ImageBuffer*, WebCore::IntSize const&, float, WebCore::SVGImage::ShouldClearBuffer)+0x321) [0x7f2bc81558f3] 5 0x7f2bc8158bf9 /home/oszi/WebKit/WebKitBuild/Debug/lib/libQtWebKit.so.4(WebCore::SVGImageCache::redraw()+0x13b) [0x7f2bc8158bf9] 6 0x7f2bc8158a87 /home/oszi/WebKit/WebKitBuild/Debug/lib/libQtWebKit.so.4(WebCore::SVGImageCache::imageContentChanged()+0x113) [0x7f2bc8158a87] 7 0x7f2bc7b06e9e /home/oszi/WebKit/WebKitBuild/Debug/lib/libQtWebKit.so.4(WebCore::CachedImage::changedInRect(WebCore::Image const*, WebCore::IntRect const&)+0x7e) [0x7f2bc7b06e9e] 8 0x7f2bc815490b /home/oszi/WebKit/WebKitBuild/Debug/lib/libQtWebKit.so.4(WebCore::SVGImageChromeClient::invalidateContentsAndRootView(WebCore::IntRect const&, bool)+0x93) [0x7f2bc815490b] 9 0x7f2bc7bbe94c /home/oszi/WebKit/WebKitBuild/Debug/lib/libQtWebKit.so.4(WebCore::Chrome::invalidateContentsAndRootView(WebCore::IntRect const&, bool)+0x42) [0x7f2bc7bbe94c] 10 0x7f2bc7d2abc9 /home/oszi/WebKit/WebKitBuild/Debug/lib/libQtWebKit.so.4(WebCore::ScrollView::repaintContentRectangle(WebCore::IntRect const&, bool)+0x19d) [0x7f2bc7d2abc9] 11 0x7f2bc7c2442c /home/oszi/WebKit/WebKitBuild/Debug/lib/libQtWebKit.so.4(WebCore::FrameView::doDeferredRepaints()+0x19e) [0x7f2bc7c2442c] 12 0x7f2bc7c24184 /home/oszi/WebKit/WebKitBuild/Debug/lib/libQtWebKit.so.4(WebCore::FrameView::endDeferredRepaints()+0x138) [0x7f2bc7c24184] 13 0x7f2bc7c21269 /home/oszi/WebKit/WebKitBuild/Debug/lib/libQtWebKit.so.4(WebCore::FrameView::layout(bool)+0xcbb) [0x7f2bc7c21269] 14 0x7f2bc7c24783 /home/oszi/WebKit/WebKitBuild/Debug/lib/libQtWebKit.so.4(WebCore::FrameView::layoutTimerFired(WebCore::Timer<WebCore::FrameView>*)+0x21) [0x7f2bc7c24783] 15 0x7f2bc7c31518 /home/oszi/WebKit/WebKitBuild/Debug/lib/libQtWebKit.so.4(WebCore::Timer<WebCore::FrameView>::fired()+0x6e) [0x7f2bc7c31518] 16 0x7f2bc7d4a3e2 /home/oszi/WebKit/WebKitBuild/Debug/lib/libQtWebKit.so.4(WebCore::ThreadTimers::sharedTimerFiredInternal()+0xc6) [0x7f2bc7d4a3e2] 17 0x7f2bc7d4a319 /home/oszi/WebKit/WebKitBuild/Debug/lib/libQtWebKit.so.4(WebCore::ThreadTimers::sharedTimerFired()+0x19) [0x7f2bc7d4a319] 18 0x7f2bc801ad1a /home/oszi/WebKit/WebKitBuild/Debug/lib/libQtWebKit.so.4(WebCore::SharedTimerQt::timerEvent(QTimerEvent*)+0x6a) [0x7f2bc801ad1a] 19 0x7f2bc0c0e519 /usr/local/Trolltech/Qt-4.8.0/lib/libQtCore.so.4(QObject::event(QEvent*)+0xa9) [0x7f2bc0c0e519] 20 0x7f2bc14ca19c /usr/local/Trolltech/Qt-4.8.0/lib/libQtGui.so.4(QApplicationPrivate::notify_helper(QObject*, QEvent*)+0xac) [0x7f2bc14ca19c] 21 0x7f2bc14d106d /usr/local/Trolltech/Qt-4.8.0/lib/libQtGui.so.4(QApplication::notify(QObject*, QEvent*)+0x13d) [0x7f2bc14d106d] 22 0x7f2bc0bfc0ec /usr/local/Trolltech/Qt-4.8.0/lib/libQtCore.so.4(QCoreApplication::notifyInternal(QObject*, QEvent*)+0x8c) [0x7f2bc0bfc0ec] 23 0x7f2bc0c3052e /usr/local/Trolltech/Qt-4.8.0/lib/libQtCore.so.4(+0x1cf52e) [0x7f2bc0c3052e] 24 0x7f2bc0c2d47d /usr/local/Trolltech/Qt-4.8.0/lib/libQtCore.so.4(+0x1cc47d) [0x7f2bc0c2d47d] 25 0x7f2bc2e486f2 /lib/libglib-2.0.so.0(g_main_context_dispatch+0x1f2) [0x7f2bc2e486f2] 26 0x7f2bc2e4c568 /lib/libglib-2.0.so.0(+0x42568) [0x7f2bc2e4c568] 27 0x7f2bc2e4c71c /lib/libglib-2.0.so.0(g_main_context_iteration+0x6c) [0x7f2bc2e4c71c] 28 0x7f2bc0c2d163 /usr/local/Trolltech/Qt-4.8.0/lib/libQtCore.so.4(QEventDispatcherGlib::processEvents(QFlags<QEventLoop::ProcessEventsFlag>)+0x73) [0x7f2bc0c2d163] 29 0x7f2bc1576a9e /usr/local/Trolltech/Qt-4.8.0/lib/libQtGui.so.4(+0x2c2a9e) [0x7f2bc1576a9e] 30 0x7f2bc0bfaca2 /usr/local/Trolltech/Qt-4.8.0/lib/libQtCore.so.4(QEventLoop::processEvents(QFlags<QEventLoop::ProcessEventsFlag>)+0x32) [0x7f2bc0bfaca2] 31 0x7f2bc0bfb114 /usr/local/Trolltech/Qt-4.8.0/lib/libQtCore.so.4(QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>)+0x164) [0x7f2bc0bfb114]
Attachments
Add attachment
proposed patch, testcase, etc.
Csaba Osztrogonác
Comment 1
2012-03-23 07:01:24 PDT
Similar bug:
https://bugs.webkit.org/show_bug.cgi?id=79631
Csaba Osztrogonác
Comment 2
2012-03-23 07:02:15 PDT
tests skipped -
http://trac.webkit.org/changeset/111857
Csaba Osztrogonác
Comment 3
2012-03-28 00:35:08 PDT
***
Bug 82434
has been marked as a duplicate of this bug. ***
Csaba Osztrogonác
Comment 4
2012-03-28 00:36:36 PDT
Tim, Hajime, could you check and fix the regression you caused, please?
Csaba Osztrogonác
Comment 5
2012-03-28 00:37:09 PDT
New failing tests skipped by
r112370
Kristóf Kosztyó
Comment 6
2012-03-28 02:57:36 PDT
the svg/as-object tests also start to assert I skipped them in
r112378
http://trac.webkit.org/changeset/112378
Csaba Osztrogonác
Comment 7
2012-03-30 05:12:22 PDT
Set priority to P1, because crashes/asserts are always serious problems. (But I don't think if anybody is interested in fixing bugs ...)
Zoltan Arvai
Comment 8
2012-03-30 06:40:18 PDT
Other asserting test skipped in
r112662
: svg/zoom/page/zoom-mask-with-percentages.svg
http://trac.webkit.org/changeset/112662
Tim Horton
Comment 9
2012-03-30 10:57:37 PDT
(In reply to
comment #7
)
> Set priority to P1, because crashes/asserts are always serious problems. (But I don't think if anybody is interested in fixing bugs ...)
I'll take it, since I've seen reports of it happening on Mac too, intermittently. Ugh.
Tim Horton
Comment 10
2012-04-05 11:06:06 PDT
(In reply to
comment #9
)
> (In reply to
comment #7
) > > Set priority to P1, because crashes/asserts are always serious problems. (But I don't think if anybody is interested in fixing bugs ...) > > I'll take it, since I've seen reports of it happening on Mac too, intermittently. Ugh.
Stephen managed to make this reproduce and (thankfully) fixed it! See if
http://trac.webkit.org/changeset/113323
helps.
Csaba Osztrogonác
Comment 11
2012-04-06 03:08:37 PDT
(In reply to
comment #10
)
> (In reply to
comment #9
) > > (In reply to
comment #7
) > > > Set priority to P1, because crashes/asserts are always serious problems. (But I don't think if anybody is interested in fixing bugs ...) > > > > I'll take it, since I've seen reports of it happening on Mac too, intermittently. Ugh. > > Stephen managed to make this reproduce and (thankfully) fixed it! > > See if
http://trac.webkit.org/changeset/113323
helps.
I unskipped the tests in
http://trac.webkit.org/changeset/113419
But leave the bug open for 1-2 days to make sure if these tests don't crash anymore.
Csaba Osztrogonác
Comment 12
2012-05-14 03:50:55 PDT
They don't crash long time ago, so we can close it now.
Note
You need to
log in
before you can comment on or make changes to this bug.
Top of Page
Format For Printing
XML
Clone This Bug