Fix bug in GrCachedLayer reuse

Fix bug in GrCachedLayer reuse

In the new MultiPictureDraw tests a single hoisted layer is reused multiple times. The previous plot locking scheme allowed GrCachedLayer objects to be aggressively deleted prematurely leaving the reusing GrHoistedLayer objects with dangling pointers.

This CL changes adds a new pseudo-ref to GrCachedLayer. (It can't be a real ref since the cached layers aren't deleted when it goes to 0).

Committed: https://skia.googlesource.com/skia/+/5c481666c9678f43e039ad605457be3854cf8de3

Committed: https://skia.googlesource.com/skia/+/7bb9ed756e8663afe68e1a5fc680d57f83a31fea

[robertphillips, 2014-10-09 16:47:47]

robertphillips@google.com changed reviewers:
+ jvanverth@google.com

[jvanverth1, 2014-10-09 19:11:33]

lgtm

[robertphillips, 2014-10-09 19:12:39]

The CQ bit was checked by robertphillips@google.com

[commit-bot: I haz the power, 2014-10-09 19:13:15]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/640323002/80001

[robertphillips, 2014-10-09 19:29:34]

The CQ bit was unchecked by robertphillips@google.com

[robertphillips, 2014-10-09 19:29:35]

The CQ bit was checked by robertphillips@google.com

[commit-bot: I haz the power, 2014-10-09 19:29:46]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/640323002/80001

[commit-bot: I haz the power, 2014-10-09 19:30:14]

Committed patchset #3 (id:80001) as 5c481666c9678f43e039ad605457be3854cf8de3

[robertphillips, 2014-10-09 19:45:57]

A revert of this CL (patchset #3 id:80001) has been created in
https://codereview.chromium.org/643673004/ by robertphillips@google.com.

The reason for reverting is: Failing unit tests.

[robertphillips, 2014-10-10 17:44:35]

PTAL. This uses a different technique and doesn't conflate locking in the
resource cache with layers being reused multiple times.

[jvanverth1, 2014-10-10 18:21:44]

lgtm + comment

https://codereview.chromium.org/640323002/diff/300001/src/gpu/GrLayerCache.cpp
File src/gpu/GrLayerCache.cpp (right):

https://codereview.chromium.org/640323002/diff/300001/src/gpu/GrLayerCache.cp...
src/gpu/GrLayerCache.cpp:47: // but not used.
This comment is a little confusing. Maybe "There is a brief time where a layer
can be locked, but not used, so we can only verify that fUses >= 0"? Not sure
that's better.

[robertphillips, 2014-10-10 18:37:43]

https://codereview.chromium.org/640323002/diff/300001/src/gpu/GrLayerCache.cpp
File src/gpu/GrLayerCache.cpp (right):

https://codereview.chromium.org/640323002/diff/300001/src/gpu/GrLayerCache.cp...
src/gpu/GrLayerCache.cpp:47: // but not used.
On 2014/10/10 18:21:44, jvanverth1 wrote:
> This comment is a little confusing. Maybe "There is a brief time where a layer
> can be locked, but not used, so we can only verify that fUses >= 0"? Not sure
> that's better.

Done.

[robertphillips, 2014-10-10 18:37:46]

The CQ bit was checked by robertphillips@google.com

[commit-bot: I haz the power, 2014-10-10 18:38:19]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/640323002/350001

[commit-bot: I haz the power, 2014-10-10 18:38:55]

Committed patchset #8 (id:350001) as 7bb9ed756e8663afe68e1a5fc680d57f83a31fea