Initial change to move 2D kernel to its own file

Initial change to move 2D kernel to its own file.

BUG=skia:

Committed: https://skia.googlesource.com/skia/+/ac9779234ef7a8cf3d791ab7690ef8c388662836

[joshua.litt, 2014-07-10 21:20:43]

I wasn't entirely sure what to do with SkMatrixConvolutionImageFilter's special
tile modes.  This is just phase 1.

[bsalomon, 2014-07-11 13:07:41]

https://codereview.chromium.org/379253003/diff/60001/src/gpu/effects/GrMatrix...
File src/gpu/effects/GrMatrixConvolutionEffect.h (right):

https://codereview.chromium.org/379253003/diff/60001/src/gpu/effects/GrMatrix...
src/gpu/effects/GrMatrixConvolutionEffect.h:12: #include
"../effects/SkMatrixConvolutionImageFilter.h"
Hmm... we like the effects code to be optional so clients can only build the
subset of it that they require.

We might have to duplicate the enum and have a translation in SkCMIF.cpp (it
could just cast and statically assert that the enum values match).

Also, it might be worth looking at whether GrMCE should be using the
GrTextureDomain helper. It doesn't currently support mirroring, though it could
be extended fairly easily, I believe. (If that is something to pursue I'd argue
for doing it in a separate change).

[joshua.litt, 2014-07-16 16:34:47]

On 2014/07/11 13:07:41, bsalomon wrote:
>
https://codereview.chromium.org/379253003/diff/60001/src/gpu/effects/GrMatrix...
> File src/gpu/effects/GrMatrixConvolutionEffect.h (right):
> 
>
https://codereview.chromium.org/379253003/diff/60001/src/gpu/effects/GrMatrix...
> src/gpu/effects/GrMatrixConvolutionEffect.h:12: #include
> "../effects/SkMatrixConvolutionImageFilter.h"
> Hmm... we like the effects code to be optional so clients can only build the
> subset of it that they require.
> 
> We might have to duplicate the enum and have a translation in SkCMIF.cpp (it
> could just cast and statically assert that the enum values match).
> 
> Also, it might be worth looking at whether GrMCE should be using the
> GrTextureDomain helper. It doesn't currently support mirroring, though it
could
> be extended fairly easily, I believe. (If that is something to pursue I'd
argue
> for doing it in a separate change).

Change incorporated, I will investigate using GrTextureDomainHelper for the next
cl

[bsalomon, 2014-07-16 17:10:44]

lgtm

[joshua.litt, 2014-07-22 16:47:21]

The CQ bit was checked by joshualitt@chromium.org

[commit-bot: I haz the power, 2014-07-22 16:47:33]

CQ is trying da patch. Follow status at

https://skia-tree-status.appspot.com/cq/joshualitt@chromium.org/379253003/160001

[commit-bot: I haz the power, 2014-07-22 16:52:17]

Change committed as ac9779234ef7a8cf3d791ab7690ef8c388662836