Fix caching of sample locations

Fix caching of sample locations

The original caching logic for sample locations wishfully assumed that
the GPU would always use the same sample pattern for render targets
that had the same number of samples. It turns out we can't rely on
that. This change improves the caching logic to handle mismatched
simple patterns with the same count, and adds a unit test that
emulates different sample patterns observed on real hardware.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2111423002

Committed: https://skia.googlesource.com/skia/+/09d49a3bfe2d1e652a648ce1ea0962b38d10d166
Committed: https://skia.googlesource.com/skia/+/0d28e574ac73fef8bf75cab083ffe23f2d8860a1

[csmartdalton, 2016-07-01 06:42:06]

Description was changed from

==========
Fix caching of sample locations

The original caching logic for sample locations wishfully assumed that
the GPU would always use the same sample pattern for render targets
with the same number of samples. It turns out we can't rely on this.
This change improves the caching logic to handle mismatched simple
patterns with the same count, and adds a unit test that emulates
different sample patterns observed on real hardware.

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

to

==========
Fix caching of sample locations

The original caching logic for sample locations wishfully assumed that
the GPU would always use the same sample pattern for render targets
with the same number of samples. It turns out we can't rely on this.
This change improves the caching logic to handle mismatched simple
patterns with the same count, and adds a unit test that emulates
different sample patterns observed on real hardware.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2111423002
==========

[csmartdalton, 2016-07-01 06:45:19]

Description was changed from

==========
Fix caching of sample locations

The original caching logic for sample locations wishfully assumed that
the GPU would always use the same sample pattern for render targets
with the same number of samples. It turns out we can't rely on this.
This change improves the caching logic to handle mismatched simple
patterns with the same count, and adds a unit test that emulates
different sample patterns observed on real hardware.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2111423002
==========

to

==========
Fix caching of sample locations

The original caching logic for sample locations wishfully assumed that
the GPU would always use the same sample pattern for render targets
that had the same number of samples. It turns out we can't rely on
this.  This change improves the caching logic to handle mismatched
simple patterns with the same count, and adds a unit test that
emulates different sample patterns observed on real hardware.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2111423002
==========

[csmartdalton, 2016-07-01 06:48:51]

The CQ bit was checked by csmartdalton@google.com

[csmartdalton, 2016-07-01 06:48:54]

The CQ bit was unchecked by csmartdalton@google.com

[commit-bot: I haz the power, 2016-07-01 06:48:55]

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, 2016-07-01 06:48:56]

Note for Reviewers:
The CQ is waiting for an approval.
If you believe that the CL is not ready yet, or if you would like to L-G-T-M
with comments then please uncheck the CQ checkbox.

Waiting for LGTM from valid reviewer(s) till 2016-07-01 12:48 UTC

[csmartdalton, 2016-07-01 06:49:06]

The CQ bit was unchecked by csmartdalton@google.com

[csmartdalton, 2016-07-01 06:49:34]

csmartdalton@google.com changed reviewers:
+ bsalomon@google.com

[bsalomon, 2016-07-01 13:17:30]

Did you forget to git add the unit test file? Out of curiousity, what device was
producing different sample locations?

[csmartdalton, 2016-07-01 15:40:24]

On 2016/07/01 13:17:30, bsalomon wrote:
> Did you forget to git add the unit test file? Out of curiousity, what device
was
> producing different sample locations?

Oops. Added the unit test. I've see different sample locations on both devices I
work from. The mac is different on-screen vs offscreen, and NVIDIA is different
at 16x with msaa vs mixed samples. (This is all documented in the file I forgot
to upload.)

[csmartdalton, 2016-07-01 15:44:29]

Description was changed from

==========
Fix caching of sample locations

The original caching logic for sample locations wishfully assumed that
the GPU would always use the same sample pattern for render targets
that had the same number of samples. It turns out we can't rely on
this.  This change improves the caching logic to handle mismatched
simple patterns with the same count, and adds a unit test that
emulates different sample patterns observed on real hardware.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2111423002
==========

to

==========
Fix caching of sample locations

The original caching logic for sample locations wishfully assumed that
the GPU would always use the same sample pattern for render targets
that had the same number of samples. It turns out we can't rely on
that. This change improves the caching logic to handle mismatched
simple patterns with the same count, and adds a unit test that
emulates different sample patterns observed on real hardware.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2111423002
==========

[bsalomon, 2016-07-01 18:38:42]

lgtm

[csmartdalton, 2016-07-01 18:49:11]

The CQ bit was checked by csmartdalton@google.com

[commit-bot: I haz the power, 2016-07-01 18:49:21]

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, 2016-07-01 19:05:10]

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

[commit-bot: I haz the power, 2016-07-01 19:05:10]

Try jobs failed on following builders:
  Build-Win-MSVC-x86_64-Debug-Trybot on master.client.skia.compile (JOB_FAILED,
http://build.chromium.org/p/client.skia.compile/builders/Build-Win-MSVC-x86_6...)

[csmartdalton, 2016-07-04 22:44:19]

The CQ bit was checked by csmartdalton@google.com

[csmartdalton, 2016-07-04 22:44:19]

The patchset sent to the CQ was uploaded after l-g-t-m from bsalomon@google.com
Link to the patchset: https://codereview.chromium.org/2111423002/#ps40001
(title: "msvc warnings")

[commit-bot: I haz the power, 2016-07-04 22:44:21]

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, 2016-07-04 22:55:21]

Description was changed from

==========
Fix caching of sample locations

The original caching logic for sample locations wishfully assumed that
the GPU would always use the same sample pattern for render targets
that had the same number of samples. It turns out we can't rely on
that. This change improves the caching logic to handle mismatched
simple patterns with the same count, and adds a unit test that
emulates different sample patterns observed on real hardware.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2111423002
==========

to

==========
Fix caching of sample locations

The original caching logic for sample locations wishfully assumed that
the GPU would always use the same sample pattern for render targets
that had the same number of samples. It turns out we can't rely on
that. This change improves the caching logic to handle mismatched
simple patterns with the same count, and adds a unit test that
emulates different sample patterns observed on real hardware.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2111423002

Committed:
https://skia.googlesource.com/skia/+/09d49a3bfe2d1e652a648ce1ea0962b38d10d166
==========

[commit-bot: I haz the power, 2016-07-04 22:55:22]

Committed patchset #3 (id:40001) as
https://skia.googlesource.com/skia/+/09d49a3bfe2d1e652a648ce1ea0962b38d10d166

[commit-bot: I haz the power, 2016-07-04 22:55:25]

CQ bit was unchecked.

[rmistry, 2016-07-05 00:44:50]

A revert of this CL (patchset #3 id:40001) has been created in
https://codereview.chromium.org/2120403002/ by rmistry@google.com.

The reason for reverting is: Seems to be causing Chromium roll failures:
* https://codereview.chromium.org/2120373003
* https://codereview.chromium.org/2117193002
* https://codereview.chromium.org/2124653002.

[csmartdalton, 2016-07-05 18:21:50]

Description was changed from

==========
Fix caching of sample locations

The original caching logic for sample locations wishfully assumed that
the GPU would always use the same sample pattern for render targets
that had the same number of samples. It turns out we can't rely on
that. This change improves the caching logic to handle mismatched
simple patterns with the same count, and adds a unit test that
emulates different sample patterns observed on real hardware.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2111423002

Committed:
https://skia.googlesource.com/skia/+/09d49a3bfe2d1e652a648ce1ea0962b38d10d166
==========

to

==========
Fix caching of sample locations

The original caching logic for sample locations wishfully assumed that
the GPU would always use the same sample pattern for render targets
that had the same number of samples. It turns out we can't rely on
that. This change improves the caching logic to handle mismatched
simple patterns with the same count, and adds a unit test that
emulates different sample patterns observed on real hardware.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2111423002

Committed:
https://skia.googlesource.com/skia/+/09d49a3bfe2d1e652a648ce1ea0962b38d10d166
==========

[csmartdalton, 2016-07-05 23:38:18]

csmartdalton@google.com changed reviewers:
+ rmistry@google.com

[csmartdalton, 2016-07-05 23:38:18]

it looks like a few of the chrome bots don't have map::emplace.

[csmartdalton, 2016-07-05 23:39:25]

(Is there a home grown tree map within skia that I should be using instead?)

[csmartdalton, 2016-07-05 23:39:25]

(Is there a home grown tree map within skia that I should be using instead?)

[csmartdalton, 2016-07-06 05:26:52]

Patchset #4 (id:60001) has been deleted

[csmartdalton, 2016-07-06 14:53:47]

I deleted my latest patch set last night thinking I would do this another way.
But the new way ended up being worse in my opinion, so here is the deleted patch
set back. Some chrome bots don't have map::emplace so now we use insert instead.

(Is there a different tree map I should be using instead?)

[bsalomon, 2016-07-06 16:16:44]

I don't know of an equivalent Sk data structure. lgtm.

[csmartdalton, 2016-07-06 16:18:07]

The CQ bit was checked by csmartdalton@google.com

[commit-bot: I haz the power, 2016-07-06 16:18:16]

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, 2016-07-06 16:59:46]

Description was changed from

==========
Fix caching of sample locations

The original caching logic for sample locations wishfully assumed that
the GPU would always use the same sample pattern for render targets
that had the same number of samples. It turns out we can't rely on
that. This change improves the caching logic to handle mismatched
simple patterns with the same count, and adds a unit test that
emulates different sample patterns observed on real hardware.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2111423002

Committed:
https://skia.googlesource.com/skia/+/09d49a3bfe2d1e652a648ce1ea0962b38d10d166
==========

to

==========
Fix caching of sample locations

The original caching logic for sample locations wishfully assumed that
the GPU would always use the same sample pattern for render targets
that had the same number of samples. It turns out we can't rely on
that. This change improves the caching logic to handle mismatched
simple patterns with the same count, and adds a unit test that
emulates different sample patterns observed on real hardware.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2111423002

Committed:
https://skia.googlesource.com/skia/+/09d49a3bfe2d1e652a648ce1ea0962b38d10d166
Committed:
https://skia.googlesource.com/skia/+/0d28e574ac73fef8bf75cab083ffe23f2d8860a1
==========

[commit-bot: I haz the power, 2016-07-06 16:59:47]

Committed patchset #4 (id:80001) as
https://skia.googlesource.com/skia/+/0d28e574ac73fef8bf75cab083ffe23f2d8860a1

[commit-bot: I haz the power, 2016-07-06 16:59:55]

CQ bit was unchecked.