Use draw buffer for texture copy impl.

Use draw buffer for texture copy impl.

Committed: https://skia.googlesource.com/skia/+/e3d4bf234a04e14b6b0f33e11b3e1132b560c145

[bsalomon, 2014-09-22 21:34:44]

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

[robertphillips, 2014-09-23 13:21:54]

lgtm + questions

https://codereview.chromium.org/589143004/diff/1/src/gpu/GrContext.cpp
File src/gpu/GrContext.cpp (right):

https://codereview.chromium.org/589143004/diff/1/src/gpu/GrContext.cpp#newcod...
src/gpu/GrContext.cpp:1556: // this to our clients, though.
Is this flush on your radar?

https://codereview.chromium.org/589143004/diff/1/src/gpu/GrContext.cpp#newcod...
src/gpu/GrContext.cpp:1578: }
ASSERT_OWNED_RESOURCE(dst); ?

https://codereview.chromium.org/589143004/diff/1/src/gpu/GrContext.cpp#newcod...
src/gpu/GrContext.cpp:1585: 
So this prepareToDraw call does a bit of work on the DrawState we're not going
to use. Would it make more sense after the ASR?

https://codereview.chromium.org/589143004/diff/1/src/gpu/GrContext.cpp#newcod...
src/gpu/GrContext.cpp:1743: AutoRestoreEffects* are,
Does it still make sense to pass the acf?

[bsalomon, 2014-09-23 15:08:57]

https://codereview.chromium.org/589143004/diff/1/src/gpu/GrContext.cpp
File src/gpu/GrContext.cpp (right):

https://codereview.chromium.org/589143004/diff/1/src/gpu/GrContext.cpp#newcod...
src/gpu/GrContext.cpp:1556: // this to our clients, though.
On 2014/09/23 13:21:53, robertphillips wrote:
> Is this flush on your radar?

Yes, I think this whole function can be deleted.

https://codereview.chromium.org/589143004/diff/1/src/gpu/GrContext.cpp#newcod...
src/gpu/GrContext.cpp:1578: }
On 2014/09/23 13:21:53, robertphillips wrote:
> ASSERT_OWNED_RESOURCE(dst); ?

Done.

https://codereview.chromium.org/589143004/diff/1/src/gpu/GrContext.cpp#newcod...
src/gpu/GrContext.cpp:1585: 
On 2014/09/23 13:21:53, robertphillips wrote:
> So this prepareToDraw call does a bit of work on the DrawState we're not going
> to use. Would it make more sense after the ASR?

I think there is a larger change coming related to not making GrGpu a GrDT
subclass. Right now this function still applies the override to kYes_

I did just realize that this should turn off the clip bit on GrDrawState. Will
fix that.

https://codereview.chromium.org/589143004/diff/1/src/gpu/GrContext.cpp#newcod...
src/gpu/GrContext.cpp:1743: AutoRestoreEffects* are,
On 2014/09/23 13:21:53, robertphillips wrote:
> Does it still make sense to pass the acf?

I'm not totally sure. I want to revisit this once the new cache is operational.

[bsalomon, 2014-09-23 15:19:45]

The CQ bit was checked by bsalomon@google.com

[commit-bot: I haz the power, 2014-09-23 15:19:59]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/589143004/20001

[commit-bot: I haz the power, 2014-09-23 16:15:06]

Committed patchset #2 (id:20001) as e3d4bf234a04e14b6b0f33e11b3e1132b560c145

[bsalomon, 2014-09-24 13:51:12]

A revert of this CL (patchset #2 id:20001) has been created in
https://codereview.chromium.org/597683008/ by bsalomon@google.com.

The reason for reverting is: Likely blocking the roll..