Bug 51359 – Added PageAllocationAligned, a cross-platform abstraction for memory allocations with arbitrary alignment requirements

• Component: New Bugs
• Status: RESOLVED FIXED

[Geoffrey Garen]

Added PageAllocationAligned, a cross-platform abstraction for memory allocations with arbitrary alignment requirements

[Geoffrey Garen]

Created attachment 77055 [details]
Patch

[Geoffrey Garen]

<rdar://problem/8107952>

[Gavin Barraclough]

Comment on attachment 77055 [details]
Patch

might be nice to remove m_trueBase on DARWIN, it's not really needed.  your call, r+

[Geoffrey Garen]

I sat down to remove m_trueBase on DARWIN and realized that it would be more involved than just an #ifdef around the data member -- to satisfy unused parameter warnings, you'd either need to give DARWIN its own constructor, or you'd need to #ifdef out a portion of the shared constructor and #ifdef in an UNUSED_PARAM warning.

My gut says to stick with the simpler solution, since we're only talking about 4-8 bytes in an allocation that's >= 4K.

[Geoffrey Garen]

Created attachment 77063 [details]
Patch

[Geoffrey Garen]

Comment on attachment 77063 [details]
Patch

Here's an updated version of the patch with project file conflicts resolved.

Unfortunately, I'm not going to have time to land this before I leave for vacation. 

I'd sure appreciate it if a WebKit committer (human or robot!) could land this for me.

[WebKit Commit Bot]

Comment on attachment 77063 [details]
Patch

Rejecting attachment 77063 [details] from commit-queue.

Failed to run "['./Tools/Scripts/webkit-patch', '--status-host=queues.webkit.org', '--bot-id=cr-jail-3', 'apply-attachment', '--non-interactive', 77063]" exit_code: 2
Last 500 characters of output:
ned.cpp
patching file JavaScriptCore/wtf/PageAllocationAligned.h
patching file JavaScriptCore/wtf/Platform.h
patching file JavaScriptCore/wtf/wtf.pri
patching file JavaScriptGlue/ChangeLog
patching file JavaScriptGlue/ForwardingHeaders/wtf/PageAllocationAligned.h
patching file WebCore/ChangeLog
Hunk #1 succeeded at 1 with fuzz 3.
patching file WebCore/ForwardingHeaders/wtf/PageAllocationAligned.h

Failed to run "[u'/mnt/git/webkit-commit-queue/Tools/Scripts/svn-apply', u'--force']" exit_code: 1

Full output: http://queues.webkit.org/results/7248073

[WebKit Review Bot]

Attachment 77063 [details] did not build on chromium:
Build output: http://queues.webkit.org/results/7295083

[Early Warning System Bot]

Attachment 77063 [details] did not build on qt:
Build output: http://queues.webkit.org/results/7245069

[WebKit Review Bot]

Attachment 77063 [details] did not build on gtk:
Build output: http://queues.webkit.org/results/7263078

[Build Bot]

Attachment 77063 [details] did not build on win:
Build output: http://queues.webkit.org/results/7249084

[Gavin Barraclough]

Created attachment 77157 [details]
Cleaned up, fixed build errors, cleaned up WINCE extra decommit, re-added check for successful release.

[Oliver Hunt]

Comment on attachment 77157 [details]
Cleaned up, fixed build errors, cleaned up WINCE extra decommit, re-added check for successful release.

r=me

[Gavin Barraclough]

Oh, also, that patch fixed a bug in Geoff's last revision – deallocate() on non-DARWIN platforms was releasing 'size' bytes of memory, but be needed to release the amount we had reserved.

Fixed in r74431.

[WebKit Review Bot]

http://trac.webkit.org/changeset/74431 might have broken SnowLeopard Intel Release (Build)