Generic 4f gradient T sampler fallback

Generic 4f gradient T sampler fallback

Add a generic T sampler fallback impl which uses T series produced by
subclasses mapTs() overrides.  The fallback path uses the same interval
structures as the current optimized linear4f impl, but always sorted
in stop order (never inverted to match dx/increasing x order).

Enable the new mechanism for 4f linear w/ perspective.

Other boring changes:

  * relocate the interval builder (back) to the base class
  * add a private header for shared templates

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1783823002

Committed: https://skia.googlesource.com/skia/+/7e6fcf890a1a63249136b8e6d9f4d5a606ef7508

[f(malita), 2016-03-10 15:28:47]

Description was changed from

==========
[WIP] Generic gradient sampler

BUG=skia:
==========

to

==========
[WIP] Generic gradient sampler

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[f(malita), 2016-03-10 17:03:35]

Description was changed from

==========
[WIP] Generic gradient sampler

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Generic 4f gradient T sampler fallback

Add a generic T sampler fallback impl which uses T series produced by
subclasses mapTs() overrides.

Enable the new mechanism for 4f linear w/ perspective.


BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[f(malita), 2016-03-10 17:03:35]

fmalita@chromium.org changed reviewers:
+ reed@google.com

[f(malita), 2016-03-10 17:16:10]

Description was changed from

==========
Generic 4f gradient T sampler fallback

Add a generic T sampler fallback impl which uses T series produced by
subclasses mapTs() overrides.

Enable the new mechanism for 4f linear w/ perspective.


BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Generic 4f gradient T sampler fallback

Add a generic T sampler fallback impl which uses T series produced by
subclasses mapTs() overrides.

Enable the new mechanism for 4f linear w/ perspective.

Other boring changes:

  * relocate the interval builder (back) to the base class
  * add a private header for shared templates

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[f(malita), 2016-03-10 17:18:07]

Description was changed from

==========
Generic 4f gradient T sampler fallback

Add a generic T sampler fallback impl which uses T series produced by
subclasses mapTs() overrides.

Enable the new mechanism for 4f linear w/ perspective.

Other boring changes:

  * relocate the interval builder (back) to the base class
  * add a private header for shared templates

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Generic 4f gradient T sampler fallback

Add a generic T sampler fallback impl which uses T series produced by
subclasses mapTs() overrides.  The fallback path uses the same interval
structures as the current optimized linear4f impl, but always sorted
in stop order (never inverted to match dx/increasing x order).

Enable the new mechanism for 4f linear w/ perspective.

Other boring changes:

  * relocate the interval builder (back) to the base class
  * add a private header for shared templates

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[f(malita), 2016-03-10 17:23:43]

There's a lot of noise in this one due to the interval builder relocation
(changed my mind again, doh), but the new bits are not too bulky:

LinearGradient4fContext::mapTs()
GradientShaderBase4fContext::shadeSpan() & friends
GradientShaderBase4fContext::TSampler

[f(malita), 2016-03-10 17:23:49]

The CQ bit was checked by fmalita@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-03-10 17:23:56]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1783823002/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1783823002/80001

[commit-bot: I haz the power, 2016-03-10 17:34:30]

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

[commit-bot: I haz the power, 2016-03-10 17:34:31]

Dry run: This issue passed the CQ dry run.

[reed1, 2016-03-10 17:49:37]

https://codereview.chromium.org/1783823002/diff/80001/src/effects/gradients/S...
File src/effects/gradients/Sk4fGradientBase.cpp (right):

https://codereview.chromium.org/1783823002/diff/80001/src/effects/gradients/S...
src/effects/gradients/Sk4fGradientBase.cpp:34: return t - SkScalarFloorToInt(t);
SkScalarFloorToScalar ?

https://codereview.chromium.org/1783823002/diff/80001/src/effects/gradients/S...
src/effects/gradients/Sk4fGradientBase.cpp:40: return t - SkScalarFloorToInt(t /
2) * 2;
ditto

https://codereview.chromium.org/1783823002/diff/80001/src/effects/gradients/S...
File src/effects/gradients/Sk4fGradientBase.h (right):

https://codereview.chromium.org/1783823002/diff/80001/src/effects/gradients/S...
src/effects/gradients/Sk4fGradientBase.h:64: template <typename DstType, bool
premul>
does this mean you will apply alpha to the colors as part of your processing, or
that your inputs have already had that done, and you will do nothing in your
processing?

[f(malita), 2016-03-10 18:46:10]

https://codereview.chromium.org/1783823002/diff/80001/src/effects/gradients/S...
File src/effects/gradients/Sk4fGradientBase.cpp (right):

https://codereview.chromium.org/1783823002/diff/80001/src/effects/gradients/S...
src/effects/gradients/Sk4fGradientBase.cpp:34: return t - SkScalarFloorToInt(t);
On 2016/03/10 17:49:37, reed1 wrote:
> SkScalarFloorToScalar ?

Done.

https://codereview.chromium.org/1783823002/diff/80001/src/effects/gradients/S...
src/effects/gradients/Sk4fGradientBase.cpp:40: return t - SkScalarFloorToInt(t /
2) * 2;
On 2016/03/10 17:49:37, reed1 wrote:
> ditto

Done.

https://codereview.chromium.org/1783823002/diff/80001/src/effects/gradients/S...
File src/effects/gradients/Sk4fGradientBase.h (right):

https://codereview.chromium.org/1783823002/diff/80001/src/effects/gradients/S...
src/effects/gradients/Sk4fGradientBase.h:64: template <typename DstType, bool
premul>
On 2016/03/10 17:49:37, reed1 wrote:
> does this mean you will apply alpha to the colors as part of your processing,
or
> that your inputs have already had that done, and you will do nothing in your
> processing?

The former: interval colors are in unpremul format, so we have to apply alpha as
part of processing.  Renamed to 'do_premul' to hopefully clarify.

[f(malita), 2016-03-10 18:46:20]

The CQ bit was checked by fmalita@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-03-10 18:46:28]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1783823002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1783823002/100001

[reed1, 2016-03-10 19:01:12]

lgtm

https://codereview.chromium.org/1783823002/diff/100001/src/effects/gradients/...
File src/effects/gradients/Sk4fGradientBase.cpp (right):

https://codereview.chromium.org/1783823002/diff/100001/src/effects/gradients/...
src/effects/gradients/Sk4fGradientBase.cpp:279: if (fColorsArePremul) {
This is clearer now, but it is slightly awkward that we pick the boolean
opposite of the predicate here. That said, not sure that I would change
anything, except possibly using an enum instead of a bool for the template
parameter.

[commit-bot: I haz the power, 2016-03-10 19:05:57]

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

[commit-bot: I haz the power, 2016-03-10 19:05:58]

Dry run: This issue passed the CQ dry run.

[f(malita), 2016-03-10 19:17:19]

https://codereview.chromium.org/1783823002/diff/100001/src/effects/gradients/...
File src/effects/gradients/Sk4fGradientBase.cpp (right):

https://codereview.chromium.org/1783823002/diff/100001/src/effects/gradients/...
src/effects/gradients/Sk4fGradientBase.cpp:279: if (fColorsArePremul) {
On 2016/03/10 19:01:12, reed1 wrote:
> This is clearer now, but it is slightly awkward that we pick the boolean
> opposite of the predicate here. That said, not sure that I would change
> anything, except possibly using an enum instead of a bool for the template
> parameter.

Good idea, I'll do that in a follow-up.

[f(malita), 2016-03-10 19:17:25]

The CQ bit was checked by fmalita@chromium.org

[commit-bot: I haz the power, 2016-03-10 19:17:27]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1783823002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1783823002/100001

[commit-bot: I haz the power, 2016-03-10 19:18:46]

Description was changed from

==========
Generic 4f gradient T sampler fallback

Add a generic T sampler fallback impl which uses T series produced by
subclasses mapTs() overrides.  The fallback path uses the same interval
structures as the current optimized linear4f impl, but always sorted
in stop order (never inverted to match dx/increasing x order).

Enable the new mechanism for 4f linear w/ perspective.

Other boring changes:

  * relocate the interval builder (back) to the base class
  * add a private header for shared templates

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Generic 4f gradient T sampler fallback

Add a generic T sampler fallback impl which uses T series produced by
subclasses mapTs() overrides.  The fallback path uses the same interval
structures as the current optimized linear4f impl, but always sorted
in stop order (never inverted to match dx/increasing x order).

Enable the new mechanism for 4f linear w/ perspective.

Other boring changes:

  * relocate the interval builder (back) to the base class
  * add a private header for shared templates

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...

Committed:
https://skia.googlesource.com/skia/+/7e6fcf890a1a63249136b8e6d9f4d5a606ef7508
==========

[commit-bot: I haz the power, 2016-03-10 19:18:47]

Committed patchset #6 (id:100001) as
https://skia.googlesource.com/skia/+/7e6fcf890a1a63249136b8e6d9f4d5a606ef7508