Bug 39179 - Shrink SimpleFontData
Summary: Shrink SimpleFontData
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: Layout and Rendering (show other bugs)
Version: 528+ (Nightly build)
Hardware: All OS X 10.5
: P2 Blocker
Assignee: Simon Fraser (smfr)
URL:
Keywords: InRadar
Depends on:
Blocks:
 
Reported: 2010-05-16 09:29 PDT by Simon Fraser (smfr)
Modified: 2010-05-16 13:27 PDT (History)
5 users (show)

See Also:


Attachments
Patch (2.11 KB, patch)
2010-05-16 09:31 PDT, Simon Fraser (smfr)
andersca: review+
Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Simon Fraser (smfr) 2010-05-16 09:29:56 PDT
SimpleFontData is 5632 bytes big, mostly because of the 255 * sizeof(FloatRect) cache, which is only used in the complex code path. We can shrink it by allocating this cache lazily. It's 1536 bytes after this change.
Comment 1 Simon Fraser (smfr) 2010-05-16 09:31:52 PDT
Created attachment 56189 [details]
Patch
Comment 2 Anders Carlsson 2010-05-16 10:07:00 PDT
Comment on attachment 56189 [details]
Patch

r=me
Comment 3 Simon Fraser (smfr) 2010-05-16 10:18:40 PDT
<rdar://problem/7990089>
Comment 4 Simon Fraser (smfr) 2010-05-16 10:25:12 PDT
http://trac.webkit.org/changeset/59577
Comment 5 WebKit Review Bot 2010-05-16 12:04:39 PDT
http://trac.webkit.org/changeset/59577 might have broken GTK Linux 64-bit Debug
Comment 6 Eric Seidel (no email) 2010-05-16 13:09:04 PDT
I'm not sure how this could be possible, but the buildbot says this change caused a Gtk layout test to crash:

http://build.webkit.org/results/GTK%20Linux%2064-bit%20Debug/r59577%20(5945)/fast/workers/storage/execute-sql-args-worker-stderr.txt

ASSERTION FAILED: !protectedObjectCount()
(../../JavaScriptCore/runtime/Collector.cpp:323 void JSC::Heap::freeBlocks())

It's crashed on every build since this was checked in.  It's possible this was an earlier regression which was not noticed earlier due to Gtk build system dependency tracking problems.
Comment 7 Eric Seidel (no email) 2010-05-16 13:09:43 PDT
Nevermind, looks like this test has been flaky for a while:
http://build.webkit.org/results/GTK%20Linux%2064-bit%20Debug/r59574%20(5941)/results.html
Comment 8 Eric Seidel (no email) 2010-05-16 13:27:21 PDT
Filed the false positive as bug 39183.