Simplify ColorFilter in the add-free case.

Simplify ColorFilter in the add-free case.
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1938003002

Committed: https://skia.googlesource.com/skia/+/0be0a3bb99549fc75a7d7229dd7ac2f62ec15cf9

[djsollen, 2016-05-02 16:20:02]

Description was changed from

==========
Simplify ColorFilter in the add-free case.
==========

to

==========
Simplify ColorFilter in the add-free case.
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[djsollen, 2016-05-02 16:21:03]

The CQ bit was checked by djsollen@google.com to run a CQ dry run

[commit-bot: I haz the power, 2016-05-02 16:21:11]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1938003002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1938003002/1

[djsollen, 2016-05-02 16:21:15]

djsollen@google.com changed reviewers:
+ ccraik@google.com, reed@google.com

[reed1, 2016-05-02 16:33:31]

nice optimization
lgtm

[commit-bot: I haz the power, 2016-05-02 16:35:19]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-05-02 16:35:20]

Dry run: This issue passed the CQ dry run.

[Chris Craik, 2016-05-02 16:42:22]

Would you consider a similar optimization for ColorMatrixColorFilter? (When it's
a pure scale matrix, and the add vector is zero).

It's not currently being used in recents, but it's a similar pattern.

[djsollen, 2016-05-02 16:44:20]

The CQ bit was checked by djsollen@google.com

[commit-bot: I haz the power, 2016-05-02 16:44:29]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1938003002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1938003002/1

[commit-bot: I haz the power, 2016-05-02 16:45:45]

Description was changed from

==========
Simplify ColorFilter in the add-free case.
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Simplify ColorFilter in the add-free case.
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...

Committed:
https://skia.googlesource.com/skia/+/0be0a3bb99549fc75a7d7229dd7ac2f62ec15cf9
==========

[commit-bot: I haz the power, 2016-05-02 16:45:46]

Committed patchset #1 (id:1) as
https://skia.googlesource.com/skia/+/0be0a3bb99549fc75a7d7229dd7ac2f62ec15cf9

[reed1, 2016-05-02 16:47:45]

On 2016/05/02 16:42:22, Chris Craik wrote:
> Would you consider a similar optimization for ColorMatrixColorFilter? (When
it's
> a pure scale matrix, and the add vector is zero).
> 
> It's not currently being used in recents, but it's a similar pattern.

Sure, if we think that is a common(ish) case. We could specialize the impl to do
less math. Much more work to decide that we could turn it into ModeFilter, since
we'd have to confirm that all of the scale values are rationals of the form
1/255, 2/255, etc.

[Chris Craik, 2016-05-02 17:06:13]

Agree that it's more of an impl optimization thing then - app developers tend to
animate scale in floats without concern for quantization.