speed up maprect for scale+trans case

speed up maprect for scale+trans case

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2111703002

Committed: https://skia.googlesource.com/skia/+/47df89ebfd548de2ccd91387e1d7eee418b65f96

[reed1, 2016-06-29 22:08:37]

Description was changed from

==========
speed up maprect for scale+trans case

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

to

==========
speed up maprect for scale+trans case

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2111703002
==========

[reed1, 2016-06-30 12:15:41]

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

[commit-bot: I haz the power, 2016-06-30 12:15:48]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-06-30 12:34:01]

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

[commit-bot: I haz the power, 2016-06-30 12:34:01]

Dry run: This issue passed the CQ dry run.

[reed1, 2016-06-30 12:39:48]

reed@google.com changed reviewers:
+ herb@google.com, mtklein@google.com

[reed1, 2016-06-30 12:39:48]

two versions of the matrix setup:

scaler one is faster on nexus_7 (18 -vs- 20)
closer to the same speed on mac

[reed1, 2016-06-30 12:39:53]

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

[commit-bot: I haz the power, 2016-06-30 12:40:01]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[reed1, 2016-06-30 12:48:22]

reed@google.com changed reviewers:
+ msarett@google.com

[commit-bot: I haz the power, 2016-06-30 13:05:18]

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

[commit-bot: I haz the power, 2016-06-30 13:05:18]

Dry run: This issue passed the CQ dry run.

[mtklein, 2016-06-30 13:15:40]

lgtm

https://codereview.chromium.org/2111703002/diff/40001/src/core/SkMatrix.cpp
File src/core/SkMatrix.cpp (right):

https://codereview.chromium.org/2111703002/diff/40001/src/core/SkMatrix.cpp#n...
src/core/SkMatrix.cpp:1104: #if 1
You're saying this enabled #if case is faster than the #else?

If so, good that the simple one works best, and good to have tried the #else
experiment.  I don't think there's anything else to try here for now.

https://codereview.chromium.org/2111703002/diff/40001/src/core/SkMatrix.cpp#n...
src/core/SkMatrix.cpp:1118: Sk4f result = Sk4f::Load(&src.fLeft) * scale +
trans;
I'd personally rename result to ltrb and flipped to rblt... makes it just a
little easier to see what's going on.

[reed1, 2016-06-30 13:20:52]

https://codereview.chromium.org/2111703002/diff/40001/src/core/SkMatrix.cpp
File src/core/SkMatrix.cpp (right):

https://codereview.chromium.org/2111703002/diff/40001/src/core/SkMatrix.cpp#n...
src/core/SkMatrix.cpp:1104: #if 1
On 2016/06/30 13:15:39, mtklein wrote:
> You're saying this enabled #if case is faster than the #else?
> 
> If so, good that the simple one works best, and good to have tried the #else
> experiment.  I don't think there's anything else to try here for now.

Done.

https://codereview.chromium.org/2111703002/diff/40001/src/core/SkMatrix.cpp#n...
src/core/SkMatrix.cpp:1118: Sk4f result = Sk4f::Load(&src.fLeft) * scale +
trans;
On 2016/06/30 13:15:39, mtklein wrote:
> I'd personally rename result to ltrb and flipped to rblt... makes it just a
> little easier to see what's going on.

Done.

[reed1, 2016-06-30 13:22:58]

The CQ bit was checked by reed@google.com

[reed1, 2016-06-30 13:22:58]

The patchset sent to the CQ was uploaded after l-g-t-m from mtklein@google.com
Link to the patchset: https://codereview.chromium.org/2111703002/#ps80001
(title: "apply comments from #14")

[commit-bot: I haz the power, 2016-06-30 13:23:08]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[msarett, 2016-06-30 13:23:53]

lgtm

[mtklein, 2016-06-30 13:24:56]

https://codereview.chromium.org/2111703002/diff/60001/src/core/SkMatrix.cpp
File src/core/SkMatrix.cpp (right):

https://codereview.chromium.org/2111703002/diff/60001/src/core/SkMatrix.cpp#n...
src/core/SkMatrix.cpp:1123: // We can extract either pair [0,1] or [2,3] from
min and max and be correct, but on
Neat.

[commit-bot: I haz the power, 2016-06-30 13:38:59]

Description was changed from

==========
speed up maprect for scale+trans case

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2111703002
==========

to

==========
speed up maprect for scale+trans case

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2111703002

Committed:
https://skia.googlesource.com/skia/+/47df89ebfd548de2ccd91387e1d7eee418b65f96
==========

[commit-bot: I haz the power, 2016-06-30 13:39:00]

Committed patchset #5 (id:80001) as
https://skia.googlesource.com/skia/+/47df89ebfd548de2ccd91387e1d7eee418b65f96