Revert[2] of add ImageShader, sharing code with its Bitmap cousin

Revert[2] of add ImageShader, sharing code with its Bitmap cousin

Adjustment from previous land : runtime check for npot-support when texture is requested for tiling.

(patchset #10 id:180001 of https://codereview.chromium.org/1342113002/ )

This reverts commit f2608513626264459a00388537175600b515cae2.

BUG=skia:4365
Committed: https://skia.googlesource.com/skia/+/fe05707c356d2a8c2c69222591d4cafbd456b4e5

[reed1, 2015-09-18 20:17:07]

reed@google.com changed reviewers:
+ fmalita@chromium.org

[reed1, 2015-09-18 21:40:15]

Tegra3 help : we are failing in a unittest that tries to

1. draw to a render-target surface
2. snap into an image
3. create a shader from that
4. draw the shader (via drawPaint) into another render-target surface
5. read back both sets of pixels, and assert they are equal

... they are not, but only failing on Tegra3 it seems (Nexus 7)

[reed1, 2015-09-18 23:52:05]

Looks like the dst is just opaque-black. Any guesses?

[reed1, 2015-09-18 23:58:39]

SkPaint paint;
    paint.setShader(sourceShader);

    SkCanvas* destinationCanvas = destinationSurface->getCanvas();
    destinationCanvas->clear(SK_ColorTRANSPARENT);
    destinationCanvas->drawPaint(paint);
    destinationCanvas->readPixels(...);


Do we need to flush() or something before calling readPixels?

Interesting that we're getting Black from the readPixels, and not transparent
nor the shader's color.

[reed1, 2015-09-19 15:02:37]

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

[reed1, 2015-09-19 15:02:38]

GPU guys -- need help.

ImageNewShaderTest.cpp

void gpuToGpu(skiatest::Reporter* reporter, GrContext* context) {
    SkImageInfo info = SkImageInfo::MakeN32Premul(5, 5);

    SkAutoTUnref<SkSurface> sourceSurface(
        SkSurface::NewRenderTarget(context, SkSurface::kNo_Budgeted, info));
    SkAutoTUnref<SkSurface> destinationSurface(
        SkSurface::NewRenderTarget(context, SkSurface::kNo_Budgeted, info));

    runShaderTest(reporter, sourceSurface.get(), destinationSurface.get(), info,
"gpu");
}

This guy is failing, only on Tegra 3 Nexus 7.
The dstSurface is opaque-black, instead of having been drawn into by the image
shader.

[robertphillips, 2015-09-21 13:22:34]

On 2015/09/18 23:58:39, reed1 wrote:
> SkPaint paint;
>     paint.setShader(sourceShader);
> 
>     SkCanvas* destinationCanvas = destinationSurface->getCanvas();
>     destinationCanvas->clear(SK_ColorTRANSPARENT);
>     destinationCanvas->drawPaint(paint);
>     destinationCanvas->readPixels(...);
> 
> 
> Do we need to flush() or something before calling readPixels?
> 
> Interesting that we're getting Black from the readPixels, and not transparent
> nor the shader's color.

The readPixels should perform all the required flushing.

[reed1, 2015-09-21 19:29:32]

The CQ bit was checked by reed@google.com to run a CQ dry run

[commit-bot: I haz the power, 2015-09-21 19:29:45]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1352293002/120001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1352293002/120001

[reed1, 2015-09-21 19:30:11]

The CQ bit was unchecked by reed@google.com

[reed1, 2015-09-21 19:31:03]

testing w/ modified test (see skbug.com/4365) If this passes, plan to land.

[reed1, 2015-09-21 21:02:06]

Ready to re-land, with work-around in SkImage_Gpu for npot-tiling support being
checked at runtime.

PTAL

https://codereview.chromium.org/1352293002/diff/140001/tests/ImageNewShaderTe...
File tests/ImageNewShaderTest.cpp (right):

https://codereview.chromium.org/1352293002/diff/140001/tests/ImageNewShaderTe...
tests/ImageNewShaderTest.cpp:132: // if we change these to 5,5, this fails on
Nexus 7 w/ Tegra 3. See skbug.com/4365
Will remove this comment.

[reed1, 2015-09-21 21:02:11]

The CQ bit was checked by reed@google.com

[commit-bot: I haz the power, 2015-09-21 21:02:27]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1352293002/160001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1352293002/160001

[commit-bot: I haz the power, 2015-09-21 21:02:29]

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 2015-09-22 03:02 UTC

[reed2, 2015-09-21 22:24:25]

The CQ bit was unchecked by reed@chromium.org

[reed2, 2015-09-21 23:29:01]

The CQ bit was checked by reed@chromium.org

[commit-bot: I haz the power, 2015-09-21 23:29:12]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1352293002/160001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1352293002/160001

[f(malita), 2015-09-22 00:25:13]

Reland L G T M, deferring to Robert/Brian for GPU.

[commit-bot: I haz the power, 2015-09-22 03:02:07]

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

[commit-bot: I haz the power, 2015-09-22 03:02:08]

No LGTM from a valid reviewer yet. Please ask for an LGTM from a
full Skia committer

[bsalomon, 2015-09-22 13:52:30]

lgtm, we should add GrMakeTextureTileSafe() to SkGr.

[reed1, 2015-09-25 06:38:18]

The CQ bit was checked by reed@google.com to run a CQ dry run

[commit-bot: I haz the power, 2015-09-25 06:38:33]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1352293002/180001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1352293002/180001

[reed1, 2015-09-25 06:41:39]

The CQ bit was unchecked by reed@google.com

[reed1, 2015-09-25 09:25:45]

The CQ bit was checked by reed@google.com

[reed1, 2015-09-25 09:25:45]

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/1352293002/#ps200001
(title: " ")

[commit-bot: I haz the power, 2015-09-25 09:25:57]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1352293002/200001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1352293002/200001

[commit-bot: I haz the power, 2015-09-25 09:31:16]

Committed patchset #11 (id:200001) as
https://skia.googlesource.com/skia/+/fe05707c356d2a8c2c69222591d4cafbd456b4e5

[reed1, 2015-09-28 17:26:35]

A revert of this CL (patchset #11 id:200001) has been created in
https://codereview.chromium.org/1375673002/ by reed@google.com.

The reason for reverting is: guess: Need to handle NPOT support in shader case.