Bug 98065 – Avoid dumping layer properties that differ in tiled drawing mode

• Status: NEW
• Component: WebCore Misc.

[Tim Horton]

There are a few things that only happen in tiled drawing mode:

* the "FrameView layer" has drawsContent and a background color set.
* there is a "TileCache flattening layer" in between the "FrameView layer" and what would ordinarily be its children

We can update the layer tree logging code to minimize the differences to layer tree dumps.

[Tim Horton]

Created attachment 166540 [details]
patch

[WebKit Review Bot]

Attachment 166540 [details] did not pass style-queue:

Failed to run "['Tools/Scripts/check-webkit-style', '--diff-files', u'Source/WebCore/ChangeLog', u'Source/WebCor..." exit_code: 1
Source/WebCore/ChangeLog:16:  Need whitespace between colon and description  [changelog/filechangedescriptionwhitespace] [5]
Source/WebCore/ChangeLog:17:  Need whitespace between colon and description  [changelog/filechangedescriptionwhitespace] [5]
Source/WebCore/ChangeLog:18:  Need whitespace between colon and description  [changelog/filechangedescriptionwhitespace] [5]
Total errors found: 3 in 4 files


If any of these errors are false positives, please file a bug against check-webkit-style.

[Tim Horton]

(In reply to comment #2)
> Attachment 166540 [details] did not pass style-queue:
> 
> Failed to run "['Tools/Scripts/check-webkit-style', '--diff-files', u'Source/WebCore/ChangeLog', u'Source/WebCor..." exit_code: 1
> Source/WebCore/ChangeLog:16:  Need whitespace between colon and description  [changelog/filechangedescriptionwhitespace] [5]
> Source/WebCore/ChangeLog:17:  Need whitespace between colon and description  [changelog/filechangedescriptionwhitespace] [5]
> Source/WebCore/ChangeLog:18:  Need whitespace between colon and description  [changelog/filechangedescriptionwhitespace] [5]
> Total errors found: 3 in 4 files
> 
> 
> If any of these errors are false positives, please file a bug against check-webkit-style.

Silly stylebot, those aren't filechangedescriptions. Maybe I shouldn't use *.

[Simon Fraser (smfr)]

Comment on attachment 166540 [details]
patch

View in context: https://bugs.webkit.org/attachment.cgi?id=166540&action=review

> Source/WebCore/platform/graphics/GraphicsLayer.cpp:84
> +    , m_isTileCacheFlatteningLayer(false)

Sucks to add memory storage for something that's only used for dumping.

How about adding shouldDumpLayer() etc. to GraphicsLayerClient?

[Tim Horton]

(In reply to comment #4)
> (From update of attachment 166540 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=166540&action=review
> 
> > Source/WebCore/platform/graphics/GraphicsLayer.cpp:84
> > +    , m_isTileCacheFlatteningLayer(false)
> 
> Sucks to add memory storage for something that's only used for dumping.

I assumed that changing the little bitfield thing from 12 1 bit members to 13 1 bit members wouldn't increase storage at all. Is that not true?

> How about adding shouldDumpLayer() etc. to GraphicsLayerClient?

That won't help us find the parent of the tile cache flattening layer :P