Replace coversExtraPixels with simpler logic

Replace coversExtraPixels with simpler logic and remove transform workaround

Replace coversExtraPixels with simpler logic by comparing paint invalidation
rect to LayoutRect(paintInvalicationLocation, size()), and if either changed
and they don't equal, do full paint invalidation instead of incremental
invalidation. This covers the following cases in which incremental
invalidation doesn't apply:

- extra pixels covered by enclosingIntRect() during mapping,
- transforms such as rotate and skew,
- overflowing effects.

This also covers an extra case that we did incremental invalidation but
now do full paint invalidation:
- clipped paint invalidation rect when either borer box or clip changes.
Based on layout tests, 1 not a common case, and this is also not an issue of
spv2 because we won't clip visual rects.

BUG=660195
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Committed: https://crrev.com/e785d3b4de4914b3a8b773a93596c299937f3770
Cr-Commit-Position: refs/heads/master@{#428596}

[Xianzhu, 2016-10-28 00:44:34]

Description was changed from

==========
Replace coversExtraPixels with simpler logic

BUG=660195
==========

to

==========
Replace coversExtraPixels with simpler logic

BUG=660195
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2
==========

[Xianzhu, 2016-10-28 00:44:35]

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

[commit-bot: I haz the power, 2016-10-28 00:45:12]

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, 2016-10-28 01:02:37]

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

[commit-bot: I haz the power, 2016-10-28 01:02:38]

Dry run: Try jobs failed on following builders:
  linux_layout_tests_slimming_paint_v2 on master.tryserver.chromium.linux
(JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_layout_te...)

[Xianzhu, 2016-10-28 17:04:17]

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

[commit-bot: I haz the power, 2016-10-28 17:04:37]

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

[Xianzhu, 2016-10-28 17:36:52]

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

[commit-bot: I haz the power, 2016-10-28 17:37:04]

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, 2016-10-28 18:28:45]

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

[commit-bot: I haz the power, 2016-10-28 18:28:48]

Dry run: Try jobs failed on following builders:
  win_chromium_rel_ng on master.tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel_...)

[Xianzhu, 2016-10-28 18:34:26]

Description was changed from

==========
Replace coversExtraPixels with simpler logic

BUG=660195
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2
==========

to

==========
Replace coversExtraPixels with simpler logic and remove transform workaround

Replace coversExtraPixels with simpler logic by comparing paint invalidation
rect to LayoutRect(paintInvalicationLocation, size()), and if either changed
and they don't equal, do full paint invalidation instead of incremental
invalidation. This covers the following cases in which incremental
invalidation doesn't apply:

- extra pixels covered by enclosingIntRect() during mapping,
- transforms such as rotate and skew,
- overflowing effects.

This also covers an extra case that we did incremental invalidation but
now do full paint invalidation:
- clipped paint invalidation rect when either borer box or clip changes.
Based on layout tests, 1 not a common case, and this is also not an issue of
spv2 because we won't clip visual rects.

BUG=660195
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2
==========

[Xianzhu, 2016-10-28 18:57:46]

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

[commit-bot: I haz the power, 2016-10-28 18:58:18]

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

[Xianzhu, 2016-10-28 19:13:12]

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

[commit-bot: I haz the power, 2016-10-28 19:13:24]

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

[Xianzhu, 2016-10-28 19:29:40]

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

[commit-bot: I haz the power, 2016-10-28 19:30:18]

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

[Xianzhu, 2016-10-28 19:39:09]

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

[commit-bot: I haz the power, 2016-10-28 19:39:23]

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, 2016-10-28 20:48:08]

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

[commit-bot: I haz the power, 2016-10-28 20:48:09]

Dry run: Try jobs failed on following builders:
  linux_layout_tests_slimming_paint_v2 on master.tryserver.chromium.linux
(JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_layout_te...)

[Xianzhu, 2016-10-28 21:56:12]

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

[commit-bot: I haz the power, 2016-10-28 21:56:38]

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

[Xianzhu, 2016-10-28 22:03:05]

wangxianzhu@chromium.org changed reviewers:
+ pdr@chromium.org

[Xianzhu, 2016-10-28 22:03:07]

https://codereview.chromium.org/2457023002/diff/120001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/paint/BoxPaintInvalidator.cpp (left):

https://codereview.chromium.org/2457023002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/paint/BoxPaintInvalidator.cpp:95: void
BoxPaintInvalidator::invalidatePaintRectClippedByOldAndNewBounds(
Now this is not needed because we incrementally invalidate only when the bounds
equals translated border box. Previously this excludes clipped overflows from
the paint invalidation rects.

[pdr., 2016-10-28 22:44:01]

This is great--it removes so many special cases.

Where is the post-snapping size comparison occurring now--is that the border box
comparison?

[commit-bot: I haz the power, 2016-10-28 23:10:37]

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

[commit-bot: I haz the power, 2016-10-28 23:10:37]

Dry run: Try jobs failed on following builders:
  mac_chromium_rel_ng on master.tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[pdr., 2016-10-28 23:22:04]

On 2016/10/28 at 22:44:01, pdr. wrote:
> This is great--it removes so many special cases.
> 
> Where is the post-snapping size comparison occurring now--is that the border
box comparison?

Switch that around--is the paint invalidation rect snapped and size() unsnapped,
and by comparing them both you cover all cases?

[Xianzhu, 2016-10-28 23:25:24]

On 2016/10/28 23:22:04, pdr. wrote:
> On 2016/10/28 at 22:44:01, pdr. wrote:
> > This is great--it removes so many special cases.
> > 
> > Where is the post-snapping size comparison occurring now--is that the border
> box comparison?
> 
> Switch that around--is the paint invalidation rect snapped and size()
unsnapped,
> and by comparing them both you cover all cases?

Exactly. Uploaded a new patch with updated comment about the captured cases.

[pdr., 2016-10-28 23:31:07]

On 2016/10/28 at 23:25:24, wangxianzhu wrote:
> On 2016/10/28 23:22:04, pdr. wrote:
> > On 2016/10/28 at 22:44:01, pdr. wrote:
> > > This is great--it removes so many special cases.
> > > 
> > > Where is the post-snapping size comparison occurring now--is that the
border
> > box comparison?
> > 
> > Switch that around--is the paint invalidation rect snapped and size()
unsnapped,
> > and by comparing them both you cover all cases?
> 
> Exactly. Uploaded a new patch with updated comment about the captured cases.

LGTM! This approach is so much better.

[Xianzhu, 2016-10-29 02:26:11]

The CQ bit was checked by wangxianzhu@chromium.org

[Xianzhu, 2016-10-29 02:26:11]

The patchset sent to the CQ was uploaded after l-g-t-m from pdr@chromium.org
Link to the patchset: https://codereview.chromium.org/2457023002/#ps180001
(title: "Rebaseline")

[commit-bot: I haz the power, 2016-10-29 02:26:25]

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, 2016-10-29 04:11:28]

Description was changed from

==========
Replace coversExtraPixels with simpler logic and remove transform workaround

Replace coversExtraPixels with simpler logic by comparing paint invalidation
rect to LayoutRect(paintInvalicationLocation, size()), and if either changed
and they don't equal, do full paint invalidation instead of incremental
invalidation. This covers the following cases in which incremental
invalidation doesn't apply:

- extra pixels covered by enclosingIntRect() during mapping,
- transforms such as rotate and skew,
- overflowing effects.

This also covers an extra case that we did incremental invalidation but
now do full paint invalidation:
- clipped paint invalidation rect when either borer box or clip changes.
Based on layout tests, 1 not a common case, and this is also not an issue of
spv2 because we won't clip visual rects.

BUG=660195
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2
==========

to

==========
Replace coversExtraPixels with simpler logic and remove transform workaround

Replace coversExtraPixels with simpler logic by comparing paint invalidation
rect to LayoutRect(paintInvalicationLocation, size()), and if either changed
and they don't equal, do full paint invalidation instead of incremental
invalidation. This covers the following cases in which incremental
invalidation doesn't apply:

- extra pixels covered by enclosingIntRect() during mapping,
- transforms such as rotate and skew,
- overflowing effects.

This also covers an extra case that we did incremental invalidation but
now do full paint invalidation:
- clipped paint invalidation rect when either borer box or clip changes.
Based on layout tests, 1 not a common case, and this is also not an issue of
spv2 because we won't clip visual rects.

BUG=660195
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2
==========

[commit-bot: I haz the power, 2016-10-29 04:11:29]

Committed patchset #10 (id:180001)

[commit-bot: I haz the power, 2016-10-29 04:14:51]

Description was changed from

==========
Replace coversExtraPixels with simpler logic and remove transform workaround

Replace coversExtraPixels with simpler logic by comparing paint invalidation
rect to LayoutRect(paintInvalicationLocation, size()), and if either changed
and they don't equal, do full paint invalidation instead of incremental
invalidation. This covers the following cases in which incremental
invalidation doesn't apply:

- extra pixels covered by enclosingIntRect() during mapping,
- transforms such as rotate and skew,
- overflowing effects.

This also covers an extra case that we did incremental invalidation but
now do full paint invalidation:
- clipped paint invalidation rect when either borer box or clip changes.
Based on layout tests, 1 not a common case, and this is also not an issue of
spv2 because we won't clip visual rects.

BUG=660195
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2
==========

to

==========
Replace coversExtraPixels with simpler logic and remove transform workaround

Replace coversExtraPixels with simpler logic by comparing paint invalidation
rect to LayoutRect(paintInvalicationLocation, size()), and if either changed
and they don't equal, do full paint invalidation instead of incremental
invalidation. This covers the following cases in which incremental
invalidation doesn't apply:

- extra pixels covered by enclosingIntRect() during mapping,
- transforms such as rotate and skew,
- overflowing effects.

This also covers an extra case that we did incremental invalidation but
now do full paint invalidation:
- clipped paint invalidation rect when either borer box or clip changes.
Based on layout tests, 1 not a common case, and this is also not an issue of
spv2 because we won't clip visual rects.

BUG=660195
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Committed: https://crrev.com/e785d3b4de4914b3a8b773a93596c299937f3770
Cr-Commit-Position: refs/heads/master@{#428596}
==========

[commit-bot: I haz the power, 2016-10-29 04:14:52]

Patchset 10 (id:??) landed as
https://crrev.com/e785d3b4de4914b3a8b773a93596c299937f3770
Cr-Commit-Position: refs/heads/master@{#428596}