Round near zero value in rotation matrix to zero

Round near zero value in rotation matrix to zero
 This seems to be the root cause of computation errors.

BUG=222483, 223983
TEST=covered by test

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=194007

[Ian Vollick, 2013-04-11 16:51:25]

On 2013/04/11 16:29:14, oshima wrote:

I wonder if we can fix gfx::Transform::Rotate to treat rotations of n*90 degrees
specially. We could certainly store some static, constant matrices for these
rotations (that have no numeric error) and apply them in these cases rather than
going through cos or sin.

[oshima, 2013-04-11 18:14:46]

On 2013/04/11 16:51:25, vollick wrote:
> On 2013/04/11 16:29:14, oshima wrote:
> 
> I wonder if we can fix gfx::Transform::Rotate to treat rotations of n*90
degrees
> specially. We could certainly store some static, constant matrices for these
> rotations (that have no numeric error) and apply them in these cases rather
than
> going through cos or sin.

I think that's reasonable, but I'd like to do that once this is merged to m27.
This is less-risky and easy for TPM to approve I think.

[Jun Mukai, 2013-04-11 18:20:37]

On 2013/04/11 18:14:46, oshima wrote:
> On 2013/04/11 16:51:25, vollick wrote:
> > On 2013/04/11 16:29:14, oshima wrote:
> > 
> > I wonder if we can fix gfx::Transform::Rotate to treat rotations of n*90
> degrees
> > specially. We could certainly store some static, constant matrices for these
> > rotations (that have no numeric error) and apply them in these cases rather
> than
> > going through cos or sin.
> 
> I think that's reasonable, but I'd like to do that once this is merged to m27.
> This is less-risky and easy for TPM to approve I think.

That makes sense, but the bug you pointed is marked as M28. Is there real buggy
behaviors reported for M27?

[oshima, 2013-04-11 19:27:56]

On 2013/04/11 18:20:37, Jun Mukai wrote:
> On 2013/04/11 18:14:46, oshima wrote:
> > On 2013/04/11 16:51:25, vollick wrote:
> > > On 2013/04/11 16:29:14, oshima wrote:
> > > 
> > > I wonder if we can fix gfx::Transform::Rotate to treat rotations of n*90
> > degrees
> > > specially. We could certainly store some static, constant matrices for
these
> > > rotations (that have no numeric error) and apply them in these cases
rather
> > than
> > > going through cos or sin.
> > 
> > I think that's reasonable, but I'd like to do that once this is merged to
m27.
> > This is less-risky and easy for TPM to approve I think.
> 
> That makes sense, but the bug you pointed is marked as M28. Is there real
buggy
> behaviors reported for M27?

I'll file a separate bug to cover all changes necessary. (not all changes in
this bug is applicable for m27)

[oshima, 2013-04-12 01:13:15]

On 2013/04/11 19:27:56, oshima wrote:
> On 2013/04/11 18:20:37, Jun Mukai wrote:
> > On 2013/04/11 18:14:46, oshima wrote:
> > > On 2013/04/11 16:51:25, vollick wrote:
> > > > On 2013/04/11 16:29:14, oshima wrote:
> > > > 
> > > > I wonder if we can fix gfx::Transform::Rotate to treat rotations of n*90
> > > degrees
> > > > specially. We could certainly store some static, constant matrices for
> these
> > > > rotations (that have no numeric error) and apply them in these cases
> rather
> > > than
> > > > going through cos or sin.
> > > 
> > > I think that's reasonable, but I'd like to do that once this is merged to
> m27.
> > > This is less-risky and easy for TPM to approve I think.
> > 
> > That makes sense, but the bug you pointed is marked as M28. Is there real
> buggy
> > behaviors reported for M27?
> 
> I'll file a separate bug to cover all changes necessary. (not all changes in
> this bug is applicable for m27)
Actually I should just use existing one (crbug.com/223983). I updated the bug.
PTL.

[Jun Mukai, 2013-04-12 01:19:36]

lgtm

[commit-bot: I haz the power, 2013-04-12 01:33:25]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/oshima@chromium.org/14142011/2001

[commit-bot: I haz the power, 2013-04-12 02:15:06]

Sorry for I got bad news for ya.
Compile failed with a clobber build on win7_aura.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win7_aura&...
Your code is likely broken or HEAD is junk. Please ensure your
code is not broken then alert the build sheriffs.
Look at the try server FAQ for more details.

[commit-bot: I haz the power, 2013-04-12 13:51:50]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/oshima@chromium.org/14142011/24001

[commit-bot: I haz the power, 2013-04-12 20:48:51]

Change committed as 194007