Magnifier gpu path fix

Magnifier gpu path fix

The y axis was inverted for the offset, which should have been computed on the other side of the magnifying rectangle.

BUG=

Committed: http://code.google.com/p/skia/source/detail?r=11883

[sugoi1, 2013-10-18 15:32:00]

https://codereview.chromium.org/28273005/diff/1/src/effects/SkMagnifierImageF...
File src/effects/SkMagnifierImageFilter.cpp (right):

https://codereview.chromium.org/28273005/diff/1/src/effects/SkMagnifierImageF...
src/effects/SkMagnifierImageFilter.cpp:140: builder->fsCodeAppendf("\t\tvec2
zoom_coord = %s + %s * %s;\n",
Changed this line to a MAD

https://codereview.chromium.org/28273005/diff/1/src/effects/SkMagnifierImageF...
src/effects/SkMagnifierImageFilter.cpp:147: builder->fsCodeAppendf("\t\tdelta =
delta * %s;\n", builder->getUniformCStr(fInvInsetVar));
Changed / -> *

https://codereview.chromium.org/28273005/diff/1/src/effects/SkMagnifierImageF...
src/effects/SkMagnifierImageFilter.cpp:253: (texture->height() - (fSrcRect.y() +
fSrcRect.height()));
The offset was on the wrong side of the magnifying rectangle (we couldn't see
it, because the gm is centered)

https://codereview.chromium.org/28273005/diff/1/src/effects/SkMagnifierImageF...
src/effects/SkMagnifierImageFilter.cpp:254: SkScalar invInset = fInset > 0 ?
SkScalarInvert(fInset) : SK_Scalar1;
Invert inset here (same logic as CPU)

[Stephen White, 2013-10-18 15:49:42]

Please add a test case that exercises the y-flip issue, and add the test to
expectations/gm/ignored-tests.txt until it's rebaselined.

https://codereview.chromium.org/28273005/diff/1/src/effects/SkMagnifierImageF...
File src/effects/SkMagnifierImageFilter.cpp (right):

https://codereview.chromium.org/28273005/diff/1/src/effects/SkMagnifierImageF...
src/effects/SkMagnifierImageFilter.cpp:260: invInset * texture->width(),
Might be clearer to switch this around so it looks like a divide. I.e.

texture->width() * invInset

[sugoi1, 2013-10-18 17:23:29]

https://codereview.chromium.org/28273005/diff/1/src/effects/SkMagnifierImageF...
File src/effects/SkMagnifierImageFilter.cpp (right):

https://codereview.chromium.org/28273005/diff/1/src/effects/SkMagnifierImageF...
src/effects/SkMagnifierImageFilter.cpp:260: invInset * texture->width(),
On 2013/10/18 15:49:42, Stephen White wrote:
> Might be clearer to switch this around so it looks like a divide. I.e.
> 
> texture->width() * invInset

Done.

[Stephen White, 2013-10-18 17:51:41]

https://codereview.chromium.org/28273005/diff/50001/src/effects/SkMagnifierIm...
File src/effects/SkMagnifierImageFilter.cpp (right):

https://codereview.chromium.org/28273005/diff/50001/src/effects/SkMagnifierIm...
src/effects/SkMagnifierImageFilter.cpp:133: kVec2f_GrSLType, "uInvZoom");
Not new to this patch, but I just remembered the shader generator will put a 'u'
in front of these variable names automatically at code generation time; please
remove them here.

[sugoi1, 2013-10-21 15:16:58]

https://codereview.chromium.org/28273005/diff/50001/src/effects/SkMagnifierIm...
File src/effects/SkMagnifierImageFilter.cpp (right):

https://codereview.chromium.org/28273005/diff/50001/src/effects/SkMagnifierIm...
src/effects/SkMagnifierImageFilter.cpp:133: kVec2f_GrSLType, "uInvZoom");
On 2013/10/18 17:51:41, Stephen White wrote:
> Not new to this patch, but I just remembered the shader generator will put a
'u'
> in front of these variable names automatically at code generation time; please
> remove them here.

Done.

[Stephen White, 2013-10-21 15:29:42]

LGTM

[commit-bot: I haz the power, 2013-10-21 15:34:39]

CQ is trying da patch. Follow status at
https://skia-tree-status.appspot.com/cq/sugoi@chromium.org/28273005/130001

[commit-bot: I haz the power, 2013-10-21 15:34:41]

Failed to apply patch for expectations/gm/ignored-tests.txt:
While running patch -p1 --forward --force --no-backup-if-mismatch;
  patching file expectations/gm/ignored-tests.txt
  Hunk #1 FAILED at 31.
  1 out of 1 hunk FAILED -- saving rejects to file
expectations/gm/ignored-tests.txt.rej

Patch:       expectations/gm/ignored-tests.txt
Index: expectations/gm/ignored-tests.txt
diff --git a/expectations/gm/ignored-tests.txt
b/expectations/gm/ignored-tests.txt
index
561ac6b4749c3d6aa2d00705a393216a857ee37c..46c7a31a36b353a66373cdcb45071b58848eaf4a
100644
--- a/expectations/gm/ignored-tests.txt
+++ b/expectations/gm/ignored-tests.txt
@@ -31,3 +31,6 @@
 # ('Poppler random failures'): ignore any Poppler failures, for now
 pdf-poppler
 
+# Added by sugoi for https://codereview.chromium.org/28273005/
+imagemagnifier
+

[commit-bot: I haz the power, 2013-10-21 15:43:01]

CQ is trying da patch. Follow status at
https://skia-tree-status.appspot.com/cq/sugoi@chromium.org/28273005/190001

[commit-bot: I haz the power, 2013-10-21 15:59:28]

Change committed as 11883