Improve implementation of gfx::Transform::IsBackFaceVisible

Improve implementation of gfx::Transform::IsBackFaceVisible

The prior implementation did an entire inverse() operation when not needed.
The new version does much less computation taking advantage of assumptions to
shortcut the math.

BUG=163062
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=170402

[shawnsingh, 2012-11-29 23:17:22]

+danakj@ for general review

+sky@ because we had some questions about how to wrap the computation of
cofactor33

FYI unit test coverage exists in cc/math_util_unittest.cc and the intention is
to migrate that later.

[danakj, 2012-11-29 23:39:00]

LGTM+nit if sky@ is okay with the comma squishing. I'm not sure what style
exceptions are legit like that.

https://codereview.chromium.org/11411270/diff/1/ui/gfx/transform.cc
File ui/gfx/transform.cc (right):

https://codereview.chromium.org/11411270/diff/1/ui/gfx/transform.cc#newcode293
ui/gfx/transform.cc:293: if (std::abs(determinant) < kTooSmallForDeterminant)
<=

[sky, 2012-11-29 23:58:35]

https://codereview.chromium.org/11411270/diff/6001/ui/gfx/transform.cc
File ui/gfx/transform.cc (right):

https://codereview.chromium.org/11411270/diff/6001/ui/gfx/transform.cc#newcod...
ui/gfx/transform.cc:297: matrix_.getDouble(0,0) * matrix_.getDouble(1,1) *
matrix_.getDouble(3,3) +
It should be 4 space indented and spaces after the commas.

[shawnsingh, 2012-11-30 00:11:09]

Thanks sky@, latest patch fixes style.   I'll wait for you to LGTM before
landing.

[danakj, 2012-11-30 00:13:13]

LGTM thanks.

[commit-bot: I haz the power, 2012-11-30 00:19:56]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/shawnsingh@chromium.org/11411270/6002

[commit-bot: I haz the power, 2012-11-30 05:21:00]

Change committed as 170402