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 68638
[CSSRegions] Make RenderFlowThread track if it has variable width regions
https://bugs.webkit.org/show_bug.cgi?id=68638
Summary
[CSSRegions] Make RenderFlowThread track if it has variable width regions
Dave Hyatt
Reported
2011-09-22 10:52:39 PDT
Make RenderFlowThread track if it has variable width regions. This will be relevant for speed as we begin adding code to shrink lines and do custom painting of blocks, etc.
Attachments
Patch
(3.86 KB, patch)
2011-09-22 10:54 PDT
,
Dave Hyatt
no flags
Details
Formatted Diff
Diff
Patch
(3.86 KB, patch)
2011-09-22 10:55 PDT
,
Dave Hyatt
mitz: review+
Details
Formatted Diff
Diff
Show Obsolete
(1)
View All
Add attachment
proposed patch, testcase, etc.
Dave Hyatt
Comment 1
2011-09-22 10:54:15 PDT
Created
attachment 108359
[details]
Patch
Dave Hyatt
Comment 2
2011-09-22 10:55:15 PDT
Created
attachment 108361
[details]
Patch
mitz
Comment 3
2011-09-22 10:58:31 PDT
Comment on
attachment 108361
[details]
Patch r=me I think the naming can be improved, “variable widths” seems like the width can change, maybe you can use “nonUniformLogicalWidth” or flip it and call it “regionsHaveUniformLogicalWidth”.
Adam Roben (:aroben)
Comment 4
2011-09-22 10:59:17 PDT
Comment on
attachment 108361
[details]
Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=108361&action=review
> Source/WebCore/rendering/RenderFlowThread.cpp:308 > + m_hasValidRegions = false; > + LayoutUnit previousRegionLogicalWidth = 0;
Seems like you need to set m_regionsHaveVariableLogicalWidth to false here, too. Otherwise once it goes true it will never become false again!
> Source/WebCore/rendering/RenderFlowThread.h:108 > + bool regionsHaveVariableLogicalWidth() const { return m_regionsHaveVariableLogicalWidth; }
Seems like this isn't needed yet, but maybe you intend to use it in the future?
Dave Hyatt
Comment 5
2011-09-22 11:02:27 PDT
(In reply to
comment #4
)
> Seems like you need to set m_regionsHaveVariableLogicalWidth to false here, too. Otherwise once it goes true it will never become false again! >
Nice catch! Wouldn't be a correctness issue, but would be a performance one!
Dave Hyatt
Comment 6
2011-09-22 11:11:31 PDT
Fixed in
r95740
.
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