REGRESSION (r217197): New Yorker website hangs for a long time on load, lots of blank tiles

• Component: Layout and Rendering
• Status: RESOLVED FIXED

[zalan]

rdar://problem/33505791

[zalan]

Created attachment 316831 [details]
Patch

[Simon Fraser (smfr)]

Comment on attachment 316831 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=316831&action=review

> Source/WebCore/ChangeLog:9
> +        This patch ensures that we report the desktop, non-frame-flattened frame size for media queries.

"for media queries in subframes"

> Source/WebCore/ChangeLog:11
> +        Some websites don't expect the iframes to be expanded and their media query callbacks code
> +        result in never ending layouts. 

I think you should say a little more here about what the site was doing (MQ listener changing content, triggering resize across the MQ boundary).

> Source/WebCore/page/FrameView.cpp:-1387
> -

i want this one

> Source/WebCore/page/FrameView.cpp:5415
> +    if (isFrameFlatteningValidForThisFrame() && !m_frameFlatteningViewSizeForMediaQuery)
> +        m_frameFlatteningViewSizeForMediaQuery = ScrollView::layoutSize();

Maybe do this somewhere you know you are already in a flattened frame layout.

> LayoutTests/ChangeLog:11
> +        * fast/frames/flattening/media-query-growing-content-expected.txt: Added.
> +        * fast/frames/flattening/media-query-growing-content.html: Added.
> +        * fast/frames/flattening/resources/media-query-min-height-with-flattening.html: Added.

Please also test aspect-ratio and orientation media queries.

[zalan]

Created attachment 316888 [details]
Patch

[WebKit Commit Bot]

Comment on attachment 316888 [details]
Patch

Clearing flags on attachment: 316888

Committed r220112: <http://trac.webkit.org/changeset/220112>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[Michael Catanzaro]

Hi Jason, why has r220112 been reverted on the safari-604 and safari-605 branches, but not on trunk? Is there some regression that's somehow specific to those branches...?

[Jason Marcell]

Michael, I have asked Alan to comment.

[zalan]

(In reply to Michael Catanzaro from comment #6)
> Hi Jason, why has r220112 been reverted on the safari-604 and safari-605
> branches, but not on trunk? Is there some regression that's somehow specific
> to those branches...?
See bug 182859, we are trying to fix it properly on trunk now.