WebKit2: Enable serializing of data types needed for cross-process accelerated compositing

[Noam Rosenthal]

This is the first part of upstreaming the cross-process accelerated-compositing implementation we use with Qt.
To enable accelerated compositing on the UI process, we need to serialize the following:
* TransformationMatrix
* TimingFunction
* TransformOperation[s]
* Animation
* The GraphicsLayer data

As a first measure, we'll enable ArgumentCoders for relevant existing data types.

[Noam Rosenthal]

Created attachment 95287 [details]
Patch 1: enable Length/TransformationMatrix via SimpleArgumentCoder

[Noam Rosenthal]

Created attachment 95289 [details]
Patch 2: Serialize TimingFunction

[Noam Rosenthal]

Created attachment 95304 [details]
Patch 3: ArgumentCoder for WebCore::Animation

[WebKit Commit Bot]

Comment on attachment 95304 [details]
Patch 3: ArgumentCoder for WebCore::Animation

Attachment 95304 [details] did not pass mac-ews (mac):
Output: http://queues.webkit.org/results/8747583

[Noam Rosenthal]

(In reply to comment #4)
> (From update of attachment 95304 [details])
> Attachment 95304 [details] did not pass mac-ews (mac):
> Output: http://queues.webkit.org/results/8747583

This relies on the other patches to compile. It should still be reviewable.

[Noam Rosenthal]

Created attachment 95320 [details]
Patch 4: TransformOperation[s]

[Adam Roben (:aroben)]

Comment on attachment 95287 [details]
Patch 1: enable Length/TransformationMatrix via SimpleArgumentCoder

View in context: https://bugs.webkit.org/attachment.cgi?id=95287&action=review

> Source/WebKit2/Scripts/webkit2/messages.py:447
> +        'WebCore::Length': '<WebCore/Length.h>',

I thought the scripts knew how to infer this. My understanding was that headers_for_type only needed to be modified if the header did not match the namespace/class name.

[Noam Rosenthal]

(In reply to comment #7)
> (From update of attachment 95287 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=95287&action=review
> 
> > Source/WebKit2/Scripts/webkit2/messages.py:447
> > +        'WebCore::Length': '<WebCore/Length.h>',
> 
> I thought the scripts knew how to infer this. My understanding was that headers_for_type only needed to be modified if the header did not match the namespace/class name.

This is needed, otherwise the headers don't get copied to the include/WebCore directory.

[Adam Roben (:aroben)]

Comment on attachment 95287 [details]
Patch 1: enable Length/TransformationMatrix via SimpleArgumentCoder

View in context: https://bugs.webkit.org/attachment.cgi?id=95287&action=review

>>> Source/WebKit2/Scripts/webkit2/messages.py:447
>>> +        'WebCore::Length': '<WebCore/Length.h>',
>> 
>> I thought the scripts knew how to infer this. My understanding was that headers_for_type only needed to be modified if the header did not match the namespace/class name.
> 
> This is needed, otherwise the headers don't get copied to the include/WebCore directory.

How does this script affect what headers get copied? All it's supposed to do is write generated source files.

[Noam Rosenthal]

> > This is needed, otherwise the headers don't get copied to the include/WebCore directory.
> 
> How does this script affect what headers get copied? All it's supposed to do is write generated source files.

OK, It's needed in the script but it can be added to a different part of the script (not to special_cases). I'll amend before committing.

[Noam Rosenthal]

> > This is needed, otherwise the headers don't get copied to the include/WebCore directory.
> 
> How does this script affect what headers get copied? All it's supposed to do is write generated source files.

OK, It's needed in the script but it can be added to a different part of the script (not to special_cases). I'll amend before committing.

[Noam Rosenthal]

Created attachment 95349 [details]
Patch 1: enable Length/TransformationMatrix via SimpleArgumentCoder

[Noam Rosenthal]

Created attachment 95350 [details]
Patch 2: Serialize TimingFunction

[Noam Rosenthal]

Created attachment 95354 [details]
Patch 3: ArgumentCoder for WebCore::Animation

[WebKit Commit Bot]

Comment on attachment 95350 [details]
Patch 2: Serialize TimingFunction

Rejecting attachment 95350 [details] from commit-queue.

Failed to run "['./Tools/Scripts/webkit-patch', '--status-host=queues.webkit.org', '--bot-id=cr-jail-8', 'apply-..." exit_code: 2

Last 500 characters of output:
tching file Source/WebKit2/Scripts/webkit2/messages.py
Hunk #1 FAILED at 258.
1 out of 1 hunk FAILED -- saving rejects to file Source/WebKit2/Scripts/webkit2/messages.py.rej
patching file Source/WebKit2/Shared/WebCoreArgumentCoders.h
Hunk #1 FAILED at 48.
Hunk #2 succeeded at 481 (offset -5 lines).
1 out of 2 hunks FAILED -- saving rejects to file Source/WebKit2/Shared/WebCoreArgumentCoders.h.rej

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/8747754

[WebKit Commit Bot]

Comment on attachment 95349 [details]
Patch 1: enable Length/TransformationMatrix via SimpleArgumentCoder

Rejecting attachment 95349 [details] from commit-queue.

Failed to run "['./Tools/Scripts/webkit-patch', '--status-host=queues.webkit.org', '--bot-id=cr-jail-7', 'land-a..." exit_code: 1

Last 500 characters of output:
pe=xml
Processing 1 patch from 1 bug.
Cleaning working directory
Updating working directory
Processing patch 95349 from bug 61694.
Simon Haussman found in /mnt/git/webkit-commit-queue/Source/WebKit2/ChangeLog does not appear to be a valid reviewer according to committers.py.
ERROR: /mnt/git/webkit-commit-queue/Source/WebKit2/ChangeLog neither lists a valid reviewer nor contains the string "Unreviewed" or "Rubber stamp" (case insensitive).
Updating OpenSource
Current branch master is up to date.

Full output: http://queues.webkit.org/results/8751503

[Noam Rosenthal]

Created attachment 95360 [details]
Patch 1: enable Length/TransformationMatrix via SimpleArgumentCoder

[Noam Rosenthal]

Created attachment 95366 [details]
Patch 2: Serialize TimingFunction

[WebKit Commit Bot]

Comment on attachment 95366 [details]
Patch 2: Serialize TimingFunction

Rejecting attachment 95366 [details] from commit-queue.

Failed to run "['./Tools/Scripts/webkit-patch', '--status-host=queues.webkit.org', '--bot-id=cr-jail-7', 'apply-..." exit_code: 2

Last 500 characters of output:
tching file Source/WebKit2/Scripts/webkit2/messages.py
Hunk #1 FAILED at 258.
1 out of 1 hunk FAILED -- saving rejects to file Source/WebKit2/Scripts/webkit2/messages.py.rej
patching file Source/WebKit2/Shared/WebCoreArgumentCoders.h
Hunk #1 FAILED at 48.
Hunk #2 succeeded at 481 (offset -5 lines).
1 out of 2 hunks FAILED -- saving rejects to file Source/WebKit2/Shared/WebCoreArgumentCoders.h.rej

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/8753229

[WebKit Commit Bot]

Comment on attachment 95360 [details]
Patch 1: enable Length/TransformationMatrix via SimpleArgumentCoder

Clearing flags on attachment: 95360

Committed r87699: <http://trac.webkit.org/changeset/87699>

[WebKit Review Bot]

Attachment 95366 [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/WebKit2/Shared/WebCoreArgumentCoders.h:499:  Non-label code inside switch statements should be indented.  [whitespace/indent] [4]
Total errors found: 1 in 5 files


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

[Noam Rosenthal]

Created attachment 95377 [details]
Patch for landing

[WebKit Commit Bot]

Comment on attachment 95377 [details]
Patch for landing

Clearing flags on attachment: 95377

Committed r87701: <http://trac.webkit.org/changeset/87701>

[Noam Rosenthal]

Created attachment 95381 [details]
Patch 3: ArgumentCoder for WebCore::Animation

[WebKit Commit Bot]

Comment on attachment 95381 [details]
Patch 3: ArgumentCoder for WebCore::Animation

Clearing flags on attachment: 95381

Committed r87707: <http://trac.webkit.org/changeset/87707>

[Noam Rosenthal]

Created attachment 96206 [details]
Patch 4: TransformOperation[s]

[WebKit Review Bot]

Attachment 96206 [details] did not pass style-queue:

Failed to run "['Tools/Scripts/check-webkit-style', '--diff-files', u'Source/WebKit2/ChangeLog', u'Source/WebKit..." exit_code: 1

Source/WebKit2/Shared/WebCoreArgumentCoders.h:49:  Alphabetical sorting problem.  [build/include_order] [4]
Source/WebKit2/Shared/WebCoreArgumentCoders.h:62:  Alphabetical sorting problem.  [build/include_order] [4]
Total errors found: 2 in 3 files


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

[Noam Rosenthal]

Created attachment 96209 [details]
Patch

[Noam Rosenthal]

Created attachment 96210 [details]
Patch 4: TransformOperation[s]

[WebKit Review Bot]

Comment on attachment 96210 [details]
Patch 4: TransformOperation[s]

Clearing flags on attachment: 96210

Committed r88222: <http://trac.webkit.org/changeset/88222>

[Noam Rosenthal]

Not sure why this (In reply to comment #30)
> (From update of attachment 96210 [details])
> Clearing flags on attachment: 96210
> 
> Committed r88222: <http://trac.webkit.org/changeset/88222>

Not sure why this broke the Mac build; It builds fine for me. It seems like I'm not putting the transform operation headers in the right place in the python script, but I'm not sure where they're supposed to go.

[WebKit Review Bot]

Comment on attachment 96210 [details]
Patch 4: TransformOperation[s]

Attachment 96210 [details] did not pass mac-ews (mac):
Output: http://queues.webkit.org/results/8810032

[Noam Rosenthal]

Created attachment 96440 [details]
Patch

[Noam Rosenthal]

Created attachment 96457 [details]
Patch 4: TransformOperation[s]

[Noam Rosenthal]

(In reply to comment #34)
> Created an attachment (id=96457) [details]
> Patch 4: TransformOperation[s]

This includes some build fixes for Mac.

[WebKit Review Bot]

Comment on attachment 96457 [details]
Patch 4: TransformOperation[s]

Clearing flags on attachment: 96457

Committed r88378: <http://trac.webkit.org/changeset/88378>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.

[Noam Rosenthal]

Not done yet... :)

[Noam Rosenthal]

Created attachment 96469 [details]
Patch

[WebKit Review Bot]

Comment on attachment 96469 [details]
Patch

Clearing flags on attachment: 96469

Committed r88392: <http://trac.webkit.org/changeset/88392>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.

[Noam Rosenthal]

Created attachment 96522 [details]
Last patch: KeyframeValueList coder

[Noam Rosenthal]

Opening for last patch.

[Simon Hausmann]

Comment on attachment 96522 [details]
Last patch: KeyframeValueList coder

r=me

[Igor Trindade Oliveira]

Committed r100789: http://trac.webkit.org/changeset/100789


(In reply to comment #44)
> (From update of attachment 96522 [details])
> r=me

[Noam Rosenthal]

Comment on attachment 96522 [details]
Last patch: KeyframeValueList coder

Committed.

[Noam Rosenthal]

All patches have landed