Bug 57338 - Start to clean up BidiResolver::createBidiRunsForLine so that mere mortals can understand it
Summary: Start to clean up BidiResolver::createBidiRunsForLine so that mere mortals ca...
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: New Bugs (show other bugs)
Version: 528+ (Nightly build)
Hardware: Other OS X 10.5
: P2 Normal
Assignee: Eric Seidel (no email)
URL:
Keywords:
Depends on:
Blocks: 57341
  Show dependency treegraph
 
Reported: 2011-03-29 07:22 PDT by Eric Seidel (no email)
Modified: 2011-03-29 12:00 PDT (History)
4 users (show)

See Also:


Attachments
Patch (5.66 KB, patch)
2011-03-29 07:26 PDT, Eric Seidel (no email)
no flags Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Eric Seidel (no email) 2011-03-29 07:22:59 PDT
Start to clean up BidiResolver::createBidiRunsForLine so that mere mortals can understand it
Comment 1 Eric Seidel (no email) 2011-03-29 07:26:40 PDT
Created attachment 87316 [details]
Patch
Comment 2 Eric Seidel (no email) 2011-03-29 07:28:42 PDT
My understanding is that this code is super performance sensitive.  However, hopefully the magic of the "inline" keyword will make this change a noop for perf.  Then again, it's also possible if the compiler outlines this function things could get faster. :)

In either case the Chromium PLT bots will let us know if any of these re-factorings make things slower (or faster!).

This change should be perf-neutral.  I just bring this up because this stuff is so hot.
Comment 3 Ryosuke Niwa 2011-03-29 09:38:08 PDT
Comment on attachment 87316 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=87316&action=review

> Source/WebCore/platform/text/BidiResolver.h:525
> +inline void BidiResolver<Iterator, Run>::updateStatusLastFromCurrentDirection(WTF::Unicode::Direction dirCurrent)

Nit: Should we rename dirCurrent to currentDirection?
Comment 4 WebKit Commit Bot 2011-03-29 11:57:54 PDT
The commit-queue encountered the following flaky tests while processing attachment 87316 [details]:

fast/workers/storage/use-same-database-in-page-and-workers.html bug 50995 (author: dumi@chromium.org)
The commit-queue is continuing to process your patch.
Comment 5 WebKit Commit Bot 2011-03-29 12:00:38 PDT
Comment on attachment 87316 [details]
Patch

Clearing flags on attachment: 87316

Committed r82297: <http://trac.webkit.org/changeset/82297>
Comment 6 WebKit Commit Bot 2011-03-29 12:00:43 PDT
All reviewed patches have been landed.  Closing bug.