Anisotropic mipmap fixes

Anisotropic mipmap fixes

1) when selecting a level scale, use max(scaleX, scaleY) instead of
  current sqrt(scaleX * scaleY)

2) track and apply non-uniform fixup scales

R=reed@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1617183004

Committed: https://skia.googlesource.com/skia/+/921d7ac75206740e69fc6bc9893971249a909fe1

[f(malita), 2016-01-22 17:22:31]

Description was changed from

==========
Anisotropic mipmap fixes

1) when selecting a level scale, use max(scaleX, scaleY) instead of
  current sqrt(scaleX * scaleY)

2) track and apply non-uniform fixup scales

R=reed@google.com
==========

to

==========
Anisotropic mipmap fixes

1) when selecting a level scale, use max(scaleX, scaleY) instead of
  current sqrt(scaleX * scaleY)

2) track and apply non-uniform fixup scales

R=reed@google.com
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[f(malita), 2016-01-22 17:24:31]

Not quire ready for landing (missing guards and I haven't checked the overall GM
impact), but this fixes the seaming problem we're looking at.  WDYT?

[f(malita), 2016-01-22 17:24:42]

The CQ bit was checked by fmalita@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-01-22 17:24:53]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1617183004/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1617183004/1

[f(malita), 2016-01-22 17:42:13]

The CQ bit was checked by fmalita@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-01-22 17:42:26]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1617183004/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1617183004/20001

[commit-bot: I haz the power, 2016-01-22 17:56:24]

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

[commit-bot: I haz the power, 2016-01-22 17:56:25]

Dry run: This issue passed the CQ dry run.

[f(malita), 2016-01-22 17:56:51]

Added guard. GM diffs:
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...

I find it reassuring that the largest diffs are now well-aligned with GPU
results :)

PTAL

[reed1, 2016-01-22 19:19:26]

lgtm

Another structural change could be to pass in scale x,y to the miplevel lookup.
That allows..
- the mip code itself to decide t use min
- allows us to (later) handle real anisotropic scales, since we'd have both
scale values.

https://codereview.chromium.org/1617183004/diff/20001/src/core/SkMipMap.cpp
File src/core/SkMipMap.cpp (right):

https://codereview.chromium.org/1617183004/diff/20001/src/core/SkMipMap.cpp#n...
src/core/SkMipMap.cpp:301: levels[i].fScale  =
SkSize::Make(static_cast<float>(width) / src.width(),
SkIntToScalar() is slightly more correct, since now we're assigning to scalars,
not floats.

[f(malita), 2016-01-22 19:28:46]

On 2016/01/22 19:19:26, reed1 wrote:
> Another structural change could be to pass in scale x,y to the miplevel
lookup.
> That allows..
> - the mip code itself to decide t use min
> - allows us to (later) handle real anisotropic scales, since we'd have both
> scale values.

Makes sense, I'll follow up on that.

https://codereview.chromium.org/1617183004/diff/20001/src/core/SkMipMap.cpp
File src/core/SkMipMap.cpp (right):

https://codereview.chromium.org/1617183004/diff/20001/src/core/SkMipMap.cpp#n...
src/core/SkMipMap.cpp:301: levels[i].fScale  =
SkSize::Make(static_cast<float>(width) / src.width(),
On 2016/01/22 19:19:25, reed1 wrote:
> SkIntToScalar() is slightly more correct, since now we're assigning to
scalars,
> not floats.

Done.

[f(malita), 2016-01-22 19:28:52]

The CQ bit was checked by fmalita@chromium.org

[f(malita), 2016-01-22 19:28:52]

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/1617183004/#ps40001
(title: "SkIntToScalar")

[commit-bot: I haz the power, 2016-01-22 19:28:58]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1617183004/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1617183004/40001

[commit-bot: I haz the power, 2016-01-22 19:45:41]

Description was changed from

==========
Anisotropic mipmap fixes

1) when selecting a level scale, use max(scaleX, scaleY) instead of
  current sqrt(scaleX * scaleY)

2) track and apply non-uniform fixup scales

R=reed@google.com
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Anisotropic mipmap fixes

1) when selecting a level scale, use max(scaleX, scaleY) instead of
  current sqrt(scaleX * scaleY)

2) track and apply non-uniform fixup scales

R=reed@google.com
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...

Committed:
https://skia.googlesource.com/skia/+/921d7ac75206740e69fc6bc9893971249a909fe1
==========

[commit-bot: I haz the power, 2016-01-22 19:45:41]

Committed patchset #3 (id:40001) as
https://skia.googlesource.com/skia/+/921d7ac75206740e69fc6bc9893971249a909fe1