Add new filter repaint tests for underinvalidation

Add new filter repaint tests for underinvalidation

This patch adds two new (currently failing) filter repaint tests to show
how we improperly repaint turbulence and image filters. These two tests
should help catch progressions in the future.

BUG=598051, 640264
Committed: https://crrev.com/873f27d84aaf55a3ee7ccfb0583639464efaf7bf
Cr-Commit-Position: refs/heads/master@{#418347}

[pdr., 2016-09-02 22:08:31]

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

[pdr., 2016-09-02 22:09:29]

Lets try this again, now with
https://crrev.com/730c50d75791af41558efece2aa560a0c2fc97e9 fixed.

[pdr., 2016-09-02 22:09:32]

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

[commit-bot: I haz the power, 2016-09-02 22:10:06]

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, 2016-09-02 23:27:46]

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

[commit-bot: I haz the power, 2016-09-02 23:27:47]

Dry run: This issue passed the CQ dry run.

[Stephen White, 2016-09-13 17:27:42]

LGTM

Some notes:

I was wondering if this test should be using text-based-repaint.js style of
reftest in paint/invalidation/filters/effect-reference-*,
since I thought that this test was testing the same issue as those tests do (but
not for feimage / feturbulence). So I tried rewriting 
this test using the that harness, and it failed to repro. It definitely repros
in your test, though. So I wondered
what that harness is testing, if anything. I tried removing
FEGaussianBlur::mapRect(), and
paint/invalidation/filters/effect-reference-repaint-gaussianblur.html
fails, so at least it's still testing what it was designed to test (whew).

Anyway all that to say, I'm not sure what the bug is here, but it doesn't seem
to be the same as the mapRect() issue for feGaussianBlur and friends.

[pdr., 2016-09-13 17:59:38]

On 2016/09/13 at 17:27:42, senorblanco wrote:
> LGTM
> 
> Some notes:
> 
> I was wondering if this test should be using text-based-repaint.js style of
reftest in paint/invalidation/filters/effect-reference-*,
> since I thought that this test was testing the same issue as those tests do
(but not for feimage / feturbulence). So I tried rewriting 
> this test using the that harness, and it failed to repro. It definitely repros
in your test, though. So I wondered
> what that harness is testing, if anything. I tried removing
FEGaussianBlur::mapRect(), and
paint/invalidation/filters/effect-reference-repaint-gaussianblur.html
> fails, so at least it's still testing what it was designed to test (whew).
> 
> Anyway all that to say, I'm not sure what the bug is here, but it doesn't seem
to be the same as the mapRect() issue for feGaussianBlur and friends.

This test harness draws the pixels and verifies they are within the bounds we
claim. I think this bug is that we should be clipping in the css codepath but we
don't.

[pdr., 2016-09-13 17:59:46]

The CQ bit was checked by pdr@chromium.org

[commit-bot: I haz the power, 2016-09-13 18:00:24]

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, 2016-09-13 18:33:25]

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

[commit-bot: I haz the power, 2016-09-13 18:33:26]

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., 2016-09-13 18:56:18]

The CQ bit was checked by pdr@chromium.org

[commit-bot: I haz the power, 2016-09-13 18:56:40]

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, 2016-09-13 20:12:04]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-09-13 20:13:21]

Description was changed from

==========
Add new filter repaint tests for underinvalidation

This patch adds two new (currently failing) filter repaint tests to show
how we improperly repaint turbulence and image filters. These two tests
should help catch progressions in the future.

BUG=598051,640264
==========

to

==========
Add new filter repaint tests for underinvalidation

This patch adds two new (currently failing) filter repaint tests to show
how we improperly repaint turbulence and image filters. These two tests
should help catch progressions in the future.

BUG=598051,640264

Committed: https://crrev.com/873f27d84aaf55a3ee7ccfb0583639464efaf7bf
Cr-Commit-Position: refs/heads/master@{#418347}
==========

[commit-bot: I haz the power, 2016-09-13 20:13:22]

Patchset 1 (id:??) landed as
https://crrev.com/873f27d84aaf55a3ee7ccfb0583639464efaf7bf
Cr-Commit-Position: refs/heads/master@{#418347}