Fix scaling issue with distance field text.

Fix scaling issue with distance field text.

Picks the correct distance field size based on both the text size and
the max matrix scale. Adjusts the matrix scale if non-unity. Also adds
GM for verifying proper distance field scaling.

BUG=skia:2928
Committed: https://skia.googlesource.com/skia/+/9564ce60a657acce89fb956deb8645b324eaad1e

[jvanverth1, 2014-09-12 17:45:40]

jvanverth@google.com changed reviewers:
+ bsalomon@google.com, joshualitt@google.com

[bsalomon, 2014-09-15 13:33:19]

lgtm w/ some questions

https://codereview.chromium.org/568843002/diff/60001/gm/dftext.cpp
File gm/dftext.cpp (right):

https://codereview.chromium.org/568843002/diff/60001/gm/dftext.cpp#newcode23
gm/dftext.cpp:23: return kSkipTiled_Flag | kGPUOnly_Flag;
why the skip tile?

Have you checked that it draws similar to the raster or non-DF code path?

https://codereview.chromium.org/568843002/diff/60001/gm/dftext.cpp#newcode50
gm/dftext.cpp:50: paint.setDistanceFieldTextTEMP(true);
Is the surface method of triggering DF more permanent? If so, maybe make a
surface, draw to it, snap and draw it back to canvas?

https://codereview.chromium.org/568843002/diff/60001/src/gpu/GrDistanceFieldT...
File src/gpu/GrDistanceFieldTextContext.cpp (right):

https://codereview.chromium.org/568843002/diff/60001/src/gpu/GrDistanceFieldT...
src/gpu/GrDistanceFieldTextContext.cpp:460: // TODO: do we need to do this if
we're scaling down (i.e. maxScale < 1)?
Should your GM test this?

[jvanverth1, 2014-09-15 13:43:53]

https://codereview.chromium.org/568843002/diff/60001/gm/dftext.cpp
File gm/dftext.cpp (right):

https://codereview.chromium.org/568843002/diff/60001/gm/dftext.cpp#newcode23
gm/dftext.cpp:23: return kSkipTiled_Flag | kGPUOnly_Flag;
On 2014/09/15 13:33:19, bsalomon wrote:
> why the skip tile?

I copied another GM that had it. I can remove it.

> Have you checked that it draws similar to the raster or non-DF code path?

Yes, it looks similar.

https://codereview.chromium.org/568843002/diff/60001/gm/dftext.cpp#newcode50
gm/dftext.cpp:50: paint.setDistanceFieldTextTEMP(true);
On 2014/09/15 13:33:19, bsalomon wrote:
> Is the surface method of triggering DF more permanent? If so, maybe make a
> surface, draw to it, snap and draw it back to canvas?

Will do.

https://codereview.chromium.org/568843002/diff/60001/src/gpu/GrDistanceFieldT...
File src/gpu/GrDistanceFieldTextContext.cpp (right):

https://codereview.chromium.org/568843002/diff/60001/src/gpu/GrDistanceFieldT...
src/gpu/GrDistanceFieldTextContext.cpp:460: // TODO: do we need to do this if
we're scaling down (i.e. maxScale < 1)?
On 2014/09/15 13:33:19, bsalomon wrote:
> Should your GM test this?

Yes, probably :)

[jvanverth1, 2014-09-16 12:36:25]

The CQ bit was checked by jvanverth@google.com

[commit-bot: I haz the power, 2014-09-16 12:37:07]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patchset/568843002/120001

[commit-bot: I haz the power, 2014-09-16 12:45:24]

Committed patchset #7 (id:120001) as 9564ce60a657acce89fb956deb8645b324eaad1e