Rewrite bleed GM bmp/img generators and change oversized texture case

Rewrite bleed GM bmp/img generators and change oversized texture case

Makes it so that the oversized-texture/bmp case doesn't run on the CPU and makes it so that the subrect rendered is adjacent to the texture pad area on the bottom/right edges.

Committed: https://skia.googlesource.com/skia/+/5c1262d7f7e25282808653b9b70f097d2bbfa2fd

[bsalomon, 2015-11-09 16:39:09]

Description was changed from

==========
work in progress - update bleed gm

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

to

==========
Rewrite bleed GM bmp/img generators and change oversized texture case

Makes it so that the oversized-texture/bmp case doesn't run on the CPU and makes
it so that the subrect rendered is adjacent to the texture pad area on the
bottom/right edges.
==========

[bsalomon, 2015-11-09 16:39:09]

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

[bsalomon, 2015-11-09 16:43:17]

This adds the test case we discussed the other day.

[robertphillips, 2015-11-09 17:11:39]

lgtm w a dox request

https://codereview.chromium.org/1424473006/diff/60001/gm/bleed.cpp
File gm/bleed.cpp (right):

https://codereview.chromium.org/1424473006/diff/60001/gm/bleed.cpp#newcode44
gm/bleed.cpp:44: SkASSERT(0 == width % 2 && 0 == height % 2);
I think the don't stomp memory limit is 2x2 but that wouldn't be a very
interesting texture.

SkASSERT(width > 6 && height > 6);

?

https://codereview.chromium.org/1424473006/diff/60001/gm/bleed.cpp#newcode168
gm/bleed.cpp:168: SkASSERT(0 == width % 2 && 0 == height % 2);
Same extra assert here ?

https://codereview.chromium.org/1424473006/diff/60001/gm/bleed.cpp#newcode313
gm/bleed.cpp:313: 
// This GM exercises the behavior of the drawBitmapRect & drawImageRect calls.
Specifically their handling of:
   SrcRectConstraint (bleed vs. no-bleed)
   handling of the sub-region feature of SkBitmap
   handling of the sub-region feature (area-of-interest) of drawBitmap*
   handling of 8888 vs. A8
   (gpu-only) handling of tiled vs. non-tiled drawing)

In particular, we should never see the padding outside of an SkBitmap's
sub-region (green for 8888, 1/4 for alpha).
In some instances we can see the two outer rings outside of the area-of-interest
(i.e., the inner four checks) due to AA or filtering.

?

[bsalomon, 2015-11-09 17:44:11]

https://codereview.chromium.org/1424473006/diff/60001/gm/bleed.cpp
File gm/bleed.cpp (right):

https://codereview.chromium.org/1424473006/diff/60001/gm/bleed.cpp#newcode44
gm/bleed.cpp:44: SkASSERT(0 == width % 2 && 0 == height % 2);
On 2015/11/09 17:11:39, robertphillips wrote:
> I think the don't stomp memory limit is 2x2 but that wouldn't be a very
> interesting texture.
> 
> SkASSERT(width > 6 && height > 6);
> 
> ?

Done.

https://codereview.chromium.org/1424473006/diff/60001/gm/bleed.cpp#newcode168
gm/bleed.cpp:168: SkASSERT(0 == width % 2 && 0 == height % 2);
On 2015/11/09 17:11:39, robertphillips wrote:
> Same extra assert here ?

Done.

https://codereview.chromium.org/1424473006/diff/60001/gm/bleed.cpp#newcode313
gm/bleed.cpp:313: 
On 2015/11/09 17:11:39, robertphillips wrote:
> // This GM exercises the behavior of the drawBitmapRect & drawImageRect calls.
> Specifically their handling of:
>    SrcRectConstraint (bleed vs. no-bleed)
>    handling of the sub-region feature of SkBitmap
>    handling of the sub-region feature (area-of-interest) of drawBitmap*
>    handling of 8888 vs. A8
>    (gpu-only) handling of tiled vs. non-tiled drawing)
> 
> In particular, we should never see the padding outside of an SkBitmap's
> sub-region (green for 8888, 1/4 for alpha).
> In some instances we can see the two outer rings outside of the
area-of-interest
> (i.e., the inner four checks) due to AA or filtering.
> 
> ?
>    

Done

[bsalomon, 2015-11-09 17:44:14]

The CQ bit was checked by bsalomon@google.com

[bsalomon, 2015-11-09 17:44:15]

The patchset sent to the CQ was uploaded after l-g-t-m from
robertphillips@google.com
Link to the patchset: https://codereview.chromium.org/1424473006/#ps70001
(title: "Address comments")

[commit-bot: I haz the power, 2015-11-09 17:44:19]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1424473006/70001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1424473006/70001

[commit-bot: I haz the power, 2015-11-09 17:45:09]

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

[commit-bot: I haz the power, 2015-11-09 17:45:09]

Try jobs failed on following builders:
  Build-Ubuntu-GCC-x86_64-Release-Trybot on client.skia.compile (JOB_FAILED,
http://build.chromium.org/p/client.skia.compile/builders/Build-Ubuntu-GCC-x86...)

[bsalomon, 2015-11-09 17:48:15]

The CQ bit was checked by bsalomon@google.com

[bsalomon, 2015-11-09 17:48:15]

The patchset sent to the CQ was uploaded after l-g-t-m from
robertphillips@google.com
Link to the patchset: https://codereview.chromium.org/1424473006/#ps90001
(title: "update")

[commit-bot: I haz the power, 2015-11-09 17:48:18]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1424473006/90001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1424473006/90001

[commit-bot: I haz the power, 2015-11-09 18:06:09]

Committed patchset #6 (id:90001) as
https://skia.googlesource.com/skia/+/5c1262d7f7e25282808653b9b70f097d2bbfa2fd