Change members's order and combine duplicated method for FFTFrame.

Change members's order and combine duplicated method for FFTFrame.

realData, imagData are common members for all platform and solution.
It doesn't need to be under each solution's macro.
Then, additionally wrong style has been fixed.

BUG=411127
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=181442

[KhNo, 2014-09-03 07:29:27]

Patchset #1 (id:1) has been deleted

[KhNo, 2014-09-03 07:38:45]

Patchset #1 (id:20001) has been deleted

[KhNo, 2014-09-03 08:54:59]

Patchset #1 (id:40001) has been deleted

[KhNo, 2014-09-03 09:22:49]

keonho07.kim@samsung.com changed reviewers:
+ rtoy@chromium.org

[KhNo, 2014-09-03 09:24:37]

Hi, Raymond. I have updated FFTFrame files.
1. Removed duplicated and common method to out of marcro.
2. Correct webkit-style.

Please review. I'm back

[Raymond Toy, 2014-09-03 16:42:47]

Looks good overall, but I think the change from vDSP_vsmul to VectorMath::vsmul
is wrong.

I think OSX's vsmul is very fast (and faster than VectorMath::vsmul) so we don't
want to change that, unless you have data that shows vDSP_vsmul is slower. I
think this also means you can't use the same realData and imagData function for
everything.

https://codereview.chromium.org/536843002/diff/60001/Source/platform/audio/ma...
File Source/platform/audio/mac/FFTFrameMac.cpp (right):

https://codereview.chromium.org/536843002/diff/60001/Source/platform/audio/ma...
Source/platform/audio/mac/FFTFrameMac.cpp:120: VectorMath::vsmul(data, 1,
&scale, data, 1, m_FFTSize);
This looks wrong.  Why the change from vDSP_vsmul to VectorMath::vsmul?

[KhNo, 2014-09-04 04:11:36]

Please check my comment.

https://codereview.chromium.org/536843002/diff/60001/Source/platform/audio/ma...
File Source/platform/audio/mac/FFTFrameMac.cpp (left):

https://codereview.chromium.org/536843002/diff/60001/Source/platform/audio/ma...
Source/platform/audio/mac/FFTFrameMac.cpp:158: return m_frame.realp;
.realp and .imap are just pointer of m_realData.data() and m_imagData.data().

https://codereview.chromium.org/536843002/diff/60001/Source/platform/audio/ma...
File Source/platform/audio/mac/FFTFrameMac.cpp (right):

https://codereview.chromium.org/536843002/diff/60001/Source/platform/audio/ma...
Source/platform/audio/mac/FFTFrameMac.cpp:107: VectorMath::vsmul(data, 1,
&scale, scaledData.data(), 1, m_FFTSize);
It is using VectorMath::vsmul since vDSP_vsmul is called inside of
VectorMatch::vsmul by MACOSX macro.

https://codereview.chromium.org/536843002/diff/60001/Source/platform/audio/ma...
Source/platform/audio/mac/FFTFrameMac.cpp:120: VectorMath::vsmul(data, 1,
&scale, data, 1, m_FFTSize);
On 2014/09/03 16:42:47, Raymond Toy wrote:
> This looks wrong.  Why the change from vDSP_vsmul to VectorMath::vsmul?

To make it has unity and code maintenance.

[Raymond Toy, 2014-09-04 17:14:57]

Please file a bug on this as well.

https://codereview.chromium.org/536843002/diff/60001/Source/platform/audio/ma...
File Source/platform/audio/mac/FFTFrameMac.cpp (left):

https://codereview.chromium.org/536843002/diff/60001/Source/platform/audio/ma...
Source/platform/audio/mac/FFTFrameMac.cpp:158: return m_frame.realp;
On 2014/09/04 04:11:36, KhNo wrote:
> .realp and .imap are just pointer of m_realData.data() and m_imagData.data().

Oh, because of line 62-63 in FFTFrame::FFTFrame.

Ok.

https://codereview.chromium.org/536843002/diff/60001/Source/platform/audio/ma...
File Source/platform/audio/mac/FFTFrameMac.cpp (right):

https://codereview.chromium.org/536843002/diff/60001/Source/platform/audio/ma...
Source/platform/audio/mac/FFTFrameMac.cpp:120: VectorMath::vsmul(data, 1,
&scale, data, 1, m_FFTSize);
On 2014/09/04 04:11:36, KhNo wrote:
> On 2014/09/03 16:42:47, Raymond Toy wrote:
> > This looks wrong.  Why the change from vDSP_vsmul to VectorMath::vsmul?
> 
> To make it has unity and code maintenance.

I would prefer this change be done separately.  This has nothing to do with what
the CL is doing.

[KhNo, 2014-09-05 02:31:49]

PTAL.

[KhNo, 2014-09-05 02:37:38]

Patchset #3 (id:100001) has been deleted

[Raymond Toy, 2014-09-05 05:02:20]

LGTM

Thanks for this nice clean up patch!

[KhNo, 2014-09-05 05:09:35]

The CQ bit was checked by keonho07.kim@samsung.com

[commit-bot: I haz the power, 2014-09-05 05:10:00]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/keonho07.kim@samsung.com/536843002/80001

[commit-bot: I haz the power, 2014-09-05 06:13:48]

Committed patchset #2 (id:80001) as 181442