Adding a new SkSurface factory for generating surfaces from the scratch texture pool.

Adding a new SkSurface factory for generating surfaces from the scratch texture pool.

TEST=Surface unit test
BUG=crbug.com/351798

Committed: http://code.google.com/p/skia/source/detail?r=13864

[Justin Novosad, 2014-03-19 18:59:08]

PTAL

[bsalomon, 2014-03-19 19:15:37]

https://codereview.chromium.org/201153023/diff/10009/include/gpu/SkGpuDevice.h
File include/gpu/SkGpuDevice.h (right):

https://codereview.chromium.org/201153023/diff/10009/include/gpu/SkGpuDevice....
include/gpu/SkGpuDevice.h:67: SkGpuDevice(GrContext*, GrRenderTarget*, unsigned
flags = 0);
I'm wondering if this and the below one are necessary at all anymore. AFAIK
Blink makes all their gpu devices via surface (though I'm not sure about image
filters). onCreateDevice will use this after this change but it seems like it
could use Create(GrSurface*, unsigned) instead. It seems like it should do that
anyway, even if these can't be removed yet, so that we don't add a call to a
deprecated function.

[reed1, 2014-03-19 19:18:56]

If we're going to make a new PUBLIC factory, lets carefully define why a client
would choose one over the other, and what promises we're making as to how the
two surface types will behave (differently).

[Justin Novosad, 2014-03-19 20:01:41]

On 2014/03/19 19:18:56, reed1 wrote:
> If we're going to make a new PUBLIC factory, lets carefully define why a
client
> would choose one over the other, and what promises we're making as to how the
> two surface types will behave (differently).

Done.

Also applied Brian's suggestion.

[bsalomon, 2014-03-19 20:10:42]

lgtm with suggestion about comment

https://codereview.chromium.org/201153023/diff/30001/include/core/SkSurface.h
File include/core/SkSurface.h (right):

https://codereview.chromium.org/201153023/diff/30001/include/core/SkSurface.h...
include/core/SkSurface.h:79: *  managed by the GrContext. The scratch texture
pool serves the purpose
I'd add something about it counting against the GrContext's overall resource
budget.

[reed1, 2014-03-19 20:25:54]

lgtm -- defers to brian

[Justin Novosad, 2014-03-19 20:35:03]

The CQ bit was checked by junov@chromium.org

[commit-bot: I haz the power, 2014-03-19 20:35:09]

CQ is trying da patch. Follow status at
https://skia-tree-status.appspot.com/cq/junov@chromium.org/201153023/40001

[commit-bot: I haz the power, 2014-03-19 20:40:50]

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

[commit-bot: I haz the power, 2014-03-19 20:40:51]

Retried try job too often on Build-Mac10.8-Clang-x86-Release-Trybot for step(s)
BuildSkiaLib, BuildTests
http://108.170.219.164:10117/buildstatus?builder=Build-Mac10.8-Clang-x86-Rele...

[Justin Novosad, 2014-03-19 20:43:15]

The CQ bit was checked by junov@chromium.org

[commit-bot: I haz the power, 2014-03-19 20:43:21]

CQ is trying da patch. Follow status at
https://skia-tree-status.appspot.com/cq/junov@chromium.org/201153023/60001

[commit-bot: I haz the power, 2014-03-19 21:19:20]

Change committed as 13864

[bungeman-skia, 2014-03-19 23:23:26]

https://codereview.chromium.org/201153023/diff/60001/src/gpu/SkGpuDevice.cpp
File src/gpu/SkGpuDevice.cpp (right):

https://codereview.chromium.org/201153023/diff/60001/src/gpu/SkGpuDevice.cpp#...
src/gpu/SkGpuDevice.cpp:212: SkPixelRef* pr = SkNEW_ARGS(SkGrPixelRef, (info,
surface, flags & kCached_Flag));
This is causing the Windows bots to fail about conversion to bool. This should
be SkToBool(...)

[bungeman-skia, 2014-03-19 23:27:52]

On 2014/03/19 23:23:26, bungeman1 wrote:
> https://codereview.chromium.org/201153023/diff/60001/src/gpu/SkGpuDevice.cpp
> File src/gpu/SkGpuDevice.cpp (right):
> 
>
https://codereview.chromium.org/201153023/diff/60001/src/gpu/SkGpuDevice.cpp#...
> src/gpu/SkGpuDevice.cpp:212: SkPixelRef* pr = SkNEW_ARGS(SkGrPixelRef, (info,
> surface, flags & kCached_Flag));
> This is causing the Windows bots to fail about conversion to bool. This should
> be SkToBool(...)

Since the change is trivial, committed r13870.