SkBitmapProcState rounding bias

SkBitmapProcState rounding bias

Epsilon bias to keep bitmap sample rounding consistent with geometry
rounding.

Also update the GM to draw an outer border + drop uninteresting
scales in favor of negative scale variants.

BUG=skia:4680, skia:4649
R=reed@google.com,caryclark@google.com,mtklein@google.com

Committed: https://skia.googlesource.com/skia/+/aed4d32de937a8f2733924eb848cc287e8d04087

[f(malita), 2015-12-14 14:47:55]

Description was changed from

==========
SkBitmapProcState rounding bias

Epsilon bias to keep bitmap sample rounding consistent with geometry
rounding.

BUG=skia:4680
R=reed@google.com,caryclark@google.com,mtklein@google.com
==========

to

==========
SkBitmapProcState rounding bias

Epsilon bias to keep bitmap sample rounding consistent with geometry
rounding.

BUG=skia:4680,skia:4649
R=reed@google.com,caryclark@google.com,mtklein@google.com
==========

[f(malita), 2015-12-14 20:32:55]

Description was changed from

==========
SkBitmapProcState rounding bias

Epsilon bias to keep bitmap sample rounding consistent with geometry
rounding.

BUG=skia:4680,skia:4649
R=reed@google.com,caryclark@google.com,mtklein@google.com
==========

to

==========
SkBitmapProcState rounding bias

Epsilon bias to keep bitmap sample rounding consistent with geometry
rounding.

Also update the GM to draw an outer border + drop uninteresting
scales in favor of negative scale variants.

BUG=skia:4680,skia:4649
R=reed@google.com,caryclark@google.com,mtklein@google.com
==========

[f(malita), 2015-12-14 21:41:16]

Mostly a starting point, as I don't think I'm covering all the bitmap proc
shaders.

Strangely, this used to fix http://crbug.com/539496 on my work computer - but on
my home workstation it only helps when in clamp mode.  Still digging, but
sharing for early feedback.

[f(malita), 2015-12-14 21:41:45]

On 2015/12/14 21:41:16, f(malita) wrote:
> Mostly a starting point, as I don't think I'm covering all the bitmap proc
> shaders.
> 
> Strangely, this used to fix http://crbug.com/539496 on my work computer - but
on
> my home workstation it only helps when in clamp mode.  Still digging, but
> sharing for early feedback.

updated GM: http://i.imgur.com/GHfls3L.png

[reed1, 2015-12-14 21:50:07]

lgtm

we could land the gm first, if there is any snag on the whole CL.

https://codereview.chromium.org/1527633002/diff/20001/src/core/SkBitmapProcSt...
File src/core/SkBitmapProcState.h (right):

https://codereview.chromium.org/1527633002/diff/20001/src/core/SkBitmapProcSt...
src/core/SkBitmapProcState.h:33: // return scale >= 0 ? -1 : 0;
seems like this >= version might be faster (certainly easier to read)

return -(scale > 0);

[f(malita), 2015-12-15 14:34:13]

On 2015/12/14 21:50:07, reed1 wrote:
> lgtm
> 
> we could land the gm first, if there is any snag on the whole CL.

The Chromium guard landed.

I'll try to figure out what else might be missing before removing the guard, but
I believe this part is good to go.

https://codereview.chromium.org/1527633002/diff/20001/src/core/SkBitmapProcSt...
File src/core/SkBitmapProcState.h (right):

https://codereview.chromium.org/1527633002/diff/20001/src/core/SkBitmapProcSt...
src/core/SkBitmapProcState.h:33: // return scale >= 0 ? -1 : 0;
On 2015/12/14 21:50:07, reed1 wrote:
> seems like this >= version might be faster (certainly easier to read)
> 
> return -(scale > 0);

Done.

[f(malita), 2015-12-15 14:34:24]

The CQ bit was checked by fmalita@chromium.org

[f(malita), 2015-12-15 14:34:24]

The patchset sent to the CQ was uploaded after l-g-t-m from reed@google.com
Link to the patchset: https://codereview.chromium.org/1527633002/#ps40001
(title: "simplified expr")

[commit-bot: I haz the power, 2015-12-15 14:34:30]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1527633002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1527633002/40001

[commit-bot: I haz the power, 2015-12-15 14:48:54]

Description was changed from

==========
SkBitmapProcState rounding bias

Epsilon bias to keep bitmap sample rounding consistent with geometry
rounding.

Also update the GM to draw an outer border + drop uninteresting
scales in favor of negative scale variants.

BUG=skia:4680,skia:4649
R=reed@google.com,caryclark@google.com,mtklein@google.com
==========

to

==========
SkBitmapProcState rounding bias

Epsilon bias to keep bitmap sample rounding consistent with geometry
rounding.

Also update the GM to draw an outer border + drop uninteresting
scales in favor of negative scale variants.

BUG=skia:4680,skia:4649
R=reed@google.com,caryclark@google.com,mtklein@google.com

Committed:
https://skia.googlesource.com/skia/+/aed4d32de937a8f2733924eb848cc287e8d04087
==========

[commit-bot: I haz the power, 2015-12-15 14:48:54]

Committed patchset #3 (id:40001) as
https://skia.googlesource.com/skia/+/aed4d32de937a8f2733924eb848cc287e8d04087