check bounds of each cubic segment against clip

check bounds of each cubic segment against clip

Timing against complex svg image (map),
no tiling: no change or very slightly faster
tiling (4x4): went from 3x slower (than untiled) to 2x slower

no gold changes expected

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1532733002

Committed: https://skia.googlesource.com/skia/+/70509762c88df911c58c3984e6b1e673b5ecaeac

[reed1, 2015-12-16 21:47:12]

Description was changed from

==========
check bounds of each cubic segment against clip

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

to

==========
check bounds of each cubic segment against clip

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

[reed1, 2015-12-17 13:59:23]

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

[commit-bot: I haz the power, 2015-12-17 13:59:30]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1532733002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1532733002/20001

[reed1, 2015-12-17 13:59:34]

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

[reed1, 2015-12-17 14:00:42]

Description was changed from

==========
check bounds of each cubic segment against clip

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

to

==========
check bounds of each cubic segment against clip

Timing against complex svg image (map),
no tiling: no change or very slightly faster
tiling (4x4): went from 3x slower (than untiled) to 2x slower

no gold changes expected

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

[commit-bot: I haz the power, 2015-12-17 14:13:19]

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

[commit-bot: I haz the power, 2015-12-17 14:13:20]

Dry run: This issue passed the CQ dry run.

[mtklein, 2015-12-17 14:58:16]

https://codereview.chromium.org/1532733002/diff/20001/src/core/SkEdgeClipper.cpp
File src/core/SkEdgeClipper.cpp (right):

https://codereview.chromium.org/1532733002/diff/20001/src/core/SkEdgeClipper....
src/core/SkEdgeClipper.cpp:361: static bool quick_reject_in_y(const SkPoint
pts[4], const SkRect& clip) {
Just for fun, let's compare speed with

  SkScalar min = SkTMin(SkTMin(pts[0].fY, pts[1].fY), SkTMin(pts[2].fY,
pts[3].fY));
  SkScalar max = SkTMax(SkTMax(pts[0].fY, pts[1].fY), SkTMax(pts[2].fY,
pts[3].fY));
  return min > clip.bottom() || max < clip.top();

?

If it's roughly the same speed on your desktop, I'd go with that.  Hopping back
and forth between scalar and vector code isn't cheap on ARM, nor is allTrue().

https://codereview.chromium.org/1532733002/diff/20001/src/core/SkScan_Hairlin...
File src/core/SkScan_Hairline.cpp (right):

https://codereview.chromium.org/1532733002/diff/20001/src/core/SkScan_Hairlin...
src/core/SkScan_Hairline.cpp:324: Sk2s min = Sk2s::Load(&pts[0].fX);
This might be slightly faster to do as a tree of min(min(0,1), min(2,3)).  The
critical path is one op shorter.

https://codereview.chromium.org/1532733002/diff/20001/src/core/SkScan_Hairlin...
src/core/SkScan_Hairline.cpp:331: return { min.kth<0>(), min.kth<1>(),
max.kth<0>(), max.kth<1>() };
We might want to double check this is the same as
  SkRect bounds;
  min.store(&bounds.fLeft);
  max.store(&bounds.fRight);
  return bounds;

[caryclark, 2015-12-17 14:59:20]

lgtm

[reed1, 2015-12-17 15:54:52]

The CQ bit was checked by reed@google.com

[commit-bot: I haz the power, 2015-12-17 15:55:04]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1532733002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1532733002/20001

[commit-bot: I haz the power, 2015-12-17 15:55:46]

Description was changed from

==========
check bounds of each cubic segment against clip

Timing against complex svg image (map),
no tiling: no change or very slightly faster
tiling (4x4): went from 3x slower (than untiled) to 2x slower

no gold changes expected

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

to

==========
check bounds of each cubic segment against clip

Timing against complex svg image (map),
no tiling: no change or very slightly faster
tiling (4x4): went from 3x slower (than untiled) to 2x slower

no gold changes expected

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/+/70509762c88df911c58c3984e6b1e673b5ecaeac
==========

[commit-bot: I haz the power, 2015-12-17 15:55:47]

Committed patchset #2 (id:20001) as
https://skia.googlesource.com/skia/+/70509762c88df911c58c3984e6b1e673b5ecaeac