Fixed the render queue item size preallocation and verification, moved constant declarations, remov…

Fixed the render queue item size preallocation and verification, moved constant declarations, removed redundant queue allocation

BUG=

Committed: https://crrev.com/2446e5a2de9d5d78117a02ff2b206194bb8f4e8e
Cr-Commit-Position: refs/heads/master@{#10689}

[peah-webrtc, 2015-11-17 10:55:35]

peah@webrtc.org changed reviewers:
+ solenberg@webrtc.org

[peah-webrtc, 2015-11-17 10:55:35]

This is a CL fixing the issues found by @solenberg.

The fix does the following for the echo canceller, echo canceller mobile, and
adaptive gain controller render queue implementations:
1) Remove constant declarations from the .cc file and fix the usage of these
constants in a function using call by reference by applying a static_cast.
2) Removes the redundant queue allocation done in the constructor.
3) Fixes the issue in the render queue item size preallocation and verification
which caused the whole queue to be reallocated during the first 100 frames of
processing.

[peah-webrtc, 2015-11-17 10:56:15]

Description was changed from

==========
Fixed the render queue item size preallocation and verification, moved constant
declarations, removed redundant queue allocation

BUG=
==========

to

==========
Fixed the render queue item size preallocation and verification, moved constant
declarations, removed redundant queue allocation

BUG=
==========

[peah-webrtc, 2015-11-17 10:56:15]

peah@webrtc.org changed reviewers:
+ henrik.lundin@webrtc.org

[the sun, 2015-11-17 13:44:37]

https://codereview.webrtc.org/1454683002/diff/1/webrtc/modules/audio_processi...
File webrtc/modules/audio_processing/echo_cancellation_impl.cc (right):

https://codereview.webrtc.org/1454683002/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/echo_cancellation_impl.cc:383:
std::max<size_t>(static_cast<size_t>(kAllowedValuesOfSamplesPerFrame1),
It seems wasteful to me to do this at runtime. Why not keep one constant (in the
anonymous namespace above) and add a comment that it is the larger of the two
possbilities?

https://codereview.webrtc.org/1454683002/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/echo_cancellation_impl.cc:391: if
(new_render_queue_element_max_size > render_queue_element_max_size_) {
super nit: it's a little nicer to keep the operands in the same order as the
assignment below.

https://codereview.webrtc.org/1454683002/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/echo_cancellation_impl.cc:398:
kMaxNumFramesToBuffer, template_queue_element,
Move kMaxNumFramesToBuffer into the anonymous namespace above instead.

[peah-webrtc, 2015-11-18 05:53:46]

https://codereview.webrtc.org/1454683002/diff/1/webrtc/modules/audio_processi...
File webrtc/modules/audio_processing/echo_cancellation_impl.cc (right):

https://codereview.webrtc.org/1454683002/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/echo_cancellation_impl.cc:383:
std::max<size_t>(static_cast<size_t>(kAllowedValuesOfSamplesPerFrame1),
On 2015/11/17 13:44:37, the sun wrote:
> It seems wasteful to me to do this at runtime. Why not keep one constant (in
the
> anonymous namespace above) and add a comment that it is the larger of the two
> possbilities?

Good point! 

Done.

https://codereview.webrtc.org/1454683002/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/echo_cancellation_impl.cc:391: if
(new_render_queue_element_max_size > render_queue_element_max_size_) {
On 2015/11/17 13:44:37, the sun wrote:
> super nit: it's a little nicer to keep the operands in the same order as the
> assignment below.

Agree. And the comparison is also more readable in itself done in that
direction.

Done.

https://codereview.webrtc.org/1454683002/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/echo_cancellation_impl.cc:398:
kMaxNumFramesToBuffer, template_queue_element,
On 2015/11/17 13:44:37, the sun wrote:
> Move kMaxNumFramesToBuffer into the anonymous namespace above instead.

Done.

[the sun, 2015-11-18 08:46:07]

lgtm

[hlundin-webrtc, 2015-11-18 09:11:46]

lgtm

[peah-webrtc, 2015-11-18 10:32:28]

The CQ bit was checked by peah@webrtc.org

[peah-webrtc, 2015-11-18 10:32:29]

The patchset sent to the CQ was uploaded after l-g-t-m from
solenberg@webrtc.org, henrik.lundin@webrtc.org
Link to the patchset: https://codereview.webrtc.org/1454683002/#ps40001 (title:
"Merge with latest changes in master")

[commit-bot: I haz the power, 2015-11-18 10:32:35]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1454683002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1454683002/40001

[commit-bot: I haz the power, 2015-11-18 14:11:17]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-11-18 14:11:25]

Patchset 3 (id:??) landed as
https://crrev.com/2446e5a2de9d5d78117a02ff2b206194bb8f4e8e
Cr-Commit-Position: refs/heads/master@{#10689}