webrtc_m130/modules/audio_processing/aec3/echo_canceller3_unittest.cc

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

919 lines
37 KiB
C++
Raw Normal View History

Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
/*
* Copyright (c) 2016 The WebRTC project authors. All Rights Reserved.
*
* Use of this source code is governed by a BSD-style license
* that can be found in the LICENSE file in the root of the source
* tree. An additional intellectual property rights grant can be found
* in the file PATENTS. All contributing project authors may
* be found in the AUTHORS file in the root of the source tree.
*/
#include "modules/audio_processing/aec3/echo_canceller3.h"
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
#include <deque>
#include <memory>
#include <string>
#include <utility>
#include <vector>
#include "modules/audio_processing/aec3/aec3_common.h"
#include "modules/audio_processing/aec3/block_processor.h"
#include "modules/audio_processing/aec3/frame_blocker.h"
#include "modules/audio_processing/aec3/mock/mock_block_processor.h"
#include "modules/audio_processing/audio_buffer.h"
#include "modules/audio_processing/high_pass_filter.h"
#include "modules/audio_processing/utility/cascaded_biquad_filter.h"
#include "rtc_base/strings/string_builder.h"
#include "test/field_trial.h"
#include "test/gmock.h"
#include "test/gtest.h"
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
namespace webrtc {
namespace {
using ::testing::_;
using ::testing::StrictMock;
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
// Populates the frame with linearly increasing sample values for each band,
// with a band-specific offset, in order to allow simple bitexactness
// verification for each band.
void PopulateInputFrame(size_t frame_length,
size_t num_bands,
size_t frame_index,
float* const* frame,
int offset) {
for (size_t k = 0; k < num_bands; ++k) {
for (size_t i = 0; i < frame_length; ++i) {
float value = static_cast<int>(frame_index * frame_length + i) + offset;
frame[k][i] = (value > 0 ? 5000 * k + value : 0);
}
}
}
// Populates the frame with linearly increasing sample values.
void PopulateInputFrame(size_t frame_length,
size_t frame_index,
float* frame,
int offset) {
for (size_t i = 0; i < frame_length; ++i) {
float value = static_cast<int>(frame_index * frame_length + i) + offset;
frame[i] = std::max(value, 0.f);
}
}
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
// Verifies the that samples in the output frame are identical to the samples
// that were produced for the input frame, with an offset in order to compensate
// for buffering delays.
bool VerifyOutputFrameBitexactness(size_t frame_length,
size_t num_bands,
size_t frame_index,
const float* const* frame,
int offset) {
float reference_frame_data[kMaxNumBands][2 * kSubFrameLength];
float* reference_frame[kMaxNumBands];
for (size_t k = 0; k < num_bands; ++k) {
reference_frame[k] = &reference_frame_data[k][0];
}
PopulateInputFrame(frame_length, num_bands, frame_index, reference_frame,
offset);
for (size_t k = 0; k < num_bands; ++k) {
for (size_t i = 0; i < frame_length; ++i) {
if (reference_frame[k][i] != frame[k][i]) {
return false;
}
}
}
return true;
}
bool VerifyOutputFrameBitexactness(rtc::ArrayView<const float> reference,
rtc::ArrayView<const float> frame,
int offset) {
for (size_t k = 0; k < frame.size(); ++k) {
int reference_index = static_cast<int>(k) + offset;
if (reference_index >= 0) {
if (reference[reference_index] != frame[k]) {
return false;
}
}
}
return true;
}
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
// Class for testing that the capture data is properly received by the block
// processor and that the processor data is properly passed to the
// EchoCanceller3 output.
class CaptureTransportVerificationProcessor : public BlockProcessor {
public:
explicit CaptureTransportVerificationProcessor(size_t num_bands) {}
~CaptureTransportVerificationProcessor() override = default;
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
void ProcessCapture(
bool level_change,
bool saturated_microphone_signal,
std::vector<std::vector<std::vector<float>>>* linear_output,
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
std::vector<std::vector<std::vector<float>>>* capture_block) override {}
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
void BufferRender(
const std::vector<std::vector<std::vector<float>>>& block) override {}
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
void UpdateEchoLeakageStatus(bool leakage_detected) override {}
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
void GetMetrics(EchoControl::Metrics* metrics) const override {}
void SetAudioBufferDelay(int delay_ms) override {}
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
private:
RTC_DISALLOW_IMPLICIT_CONSTRUCTORS(CaptureTransportVerificationProcessor);
};
// Class for testing that the render data is properly received by the block
// processor.
class RenderTransportVerificationProcessor : public BlockProcessor {
public:
explicit RenderTransportVerificationProcessor(size_t num_bands) {}
~RenderTransportVerificationProcessor() override = default;
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
void ProcessCapture(
bool level_change,
bool saturated_microphone_signal,
std::vector<std::vector<std::vector<float>>>* linear_output,
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
std::vector<std::vector<std::vector<float>>>* capture_block) override {
std::vector<std::vector<std::vector<float>>> render_block =
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
received_render_blocks_.front();
received_render_blocks_.pop_front();
capture_block->swap(render_block);
}
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
void BufferRender(
const std::vector<std::vector<std::vector<float>>>& block) override {
received_render_blocks_.push_back(block);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
}
void UpdateEchoLeakageStatus(bool leakage_detected) override {}
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
void GetMetrics(EchoControl::Metrics* metrics) const override {}
void SetAudioBufferDelay(int delay_ms) override {}
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
private:
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
std::deque<std::vector<std::vector<std::vector<float>>>>
received_render_blocks_;
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
RTC_DISALLOW_IMPLICIT_CONSTRUCTORS(RenderTransportVerificationProcessor);
};
class EchoCanceller3Tester {
public:
explicit EchoCanceller3Tester(int sample_rate_hz)
: sample_rate_hz_(sample_rate_hz),
num_bands_(NumBandsForRate(sample_rate_hz_)),
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
frame_length_(160),
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
fullband_frame_length_(rtc::CheckedDivExact(sample_rate_hz_, 100)),
capture_buffer_(fullband_frame_length_ * 100,
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
1,
fullband_frame_length_ * 100,
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
1,
fullband_frame_length_ * 100,
1),
render_buffer_(fullband_frame_length_ * 100,
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
1,
fullband_frame_length_ * 100,
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
1,
fullband_frame_length_ * 100,
1) {}
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
// Verifies that the capture data is properly received by the block processor
// and that the processor data is properly passed to the EchoCanceller3
// output.
void RunCaptureTransportVerificationTest() {
EchoCanceller3 aec3(
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
EchoCanceller3Config(), sample_rate_hz_, 1, 1,
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
std::unique_ptr<BlockProcessor>(
new CaptureTransportVerificationProcessor(num_bands_)));
for (size_t frame_index = 0; frame_index < kNumFramesToProcess;
++frame_index) {
aec3.AnalyzeCapture(&capture_buffer_);
OptionalBandSplit();
PopulateInputFrame(frame_length_, num_bands_, frame_index,
&capture_buffer_.split_bands(0)[0], 0);
PopulateInputFrame(frame_length_, frame_index,
&render_buffer_.channels()[0][0], 0);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
aec3.AnalyzeRender(&render_buffer_);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
aec3.ProcessCapture(&capture_buffer_, false);
EXPECT_TRUE(VerifyOutputFrameBitexactness(
frame_length_, num_bands_, frame_index,
&capture_buffer_.split_bands(0)[0], -64));
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
}
}
// Test method for testing that the render data is properly received by the
// block processor.
void RunRenderTransportVerificationTest() {
EchoCanceller3 aec3(
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
EchoCanceller3Config(), sample_rate_hz_, 1, 1,
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
std::unique_ptr<BlockProcessor>(
new RenderTransportVerificationProcessor(num_bands_)));
std::vector<std::vector<float>> render_input(1);
std::vector<float> capture_output;
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
for (size_t frame_index = 0; frame_index < kNumFramesToProcess;
++frame_index) {
aec3.AnalyzeCapture(&capture_buffer_);
OptionalBandSplit();
PopulateInputFrame(frame_length_, num_bands_, frame_index,
&capture_buffer_.split_bands(0)[0], 100);
PopulateInputFrame(frame_length_, num_bands_, frame_index,
&render_buffer_.split_bands(0)[0], 0);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
for (size_t k = 0; k < frame_length_; ++k) {
render_input[0].push_back(render_buffer_.split_bands(0)[0][k]);
}
aec3.AnalyzeRender(&render_buffer_);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
aec3.ProcessCapture(&capture_buffer_, false);
for (size_t k = 0; k < frame_length_; ++k) {
capture_output.push_back(capture_buffer_.split_bands(0)[0][k]);
}
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
}
HighPassFilter hp_filter(16000, 1);
hp_filter.Process(&render_input);
EXPECT_TRUE(
VerifyOutputFrameBitexactness(render_input[0], capture_output, -64));
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
}
// Verifies that information about echo path changes are properly propagated
// to the block processor.
// The cases tested are:
// -That no set echo path change flags are received when there is no echo path
// change.
// -That set echo path change flags are received and continues to be received
// as long as echo path changes are flagged.
// -That set echo path change flags are no longer received when echo path
// change events stop being flagged.
enum class EchoPathChangeTestVariant { kNone, kOneSticky, kOneNonSticky };
void RunEchoPathChangeVerificationTest(
EchoPathChangeTestVariant echo_path_change_test_variant) {
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
constexpr size_t kNumFullBlocksPerFrame = 160 / kBlockSize;
constexpr size_t kExpectedNumBlocksToProcess =
(kNumFramesToProcess * 160) / kBlockSize;
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
std::unique_ptr<testing::StrictMock<webrtc::test::MockBlockProcessor>>
block_processor_mock(
new StrictMock<webrtc::test::MockBlockProcessor>());
EXPECT_CALL(*block_processor_mock, BufferRender(_))
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
.Times(kExpectedNumBlocksToProcess);
EXPECT_CALL(*block_processor_mock, UpdateEchoLeakageStatus(_)).Times(0);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
switch (echo_path_change_test_variant) {
case EchoPathChangeTestVariant::kNone:
EXPECT_CALL(*block_processor_mock, ProcessCapture(false, _, _, _))
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
.Times(kExpectedNumBlocksToProcess);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
break;
case EchoPathChangeTestVariant::kOneSticky:
EXPECT_CALL(*block_processor_mock, ProcessCapture(true, _, _, _))
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
.Times(kExpectedNumBlocksToProcess);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
break;
case EchoPathChangeTestVariant::kOneNonSticky:
EXPECT_CALL(*block_processor_mock, ProcessCapture(true, _, _, _))
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
.Times(kNumFullBlocksPerFrame);
EXPECT_CALL(*block_processor_mock, ProcessCapture(false, _, _, _))
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
.Times(kExpectedNumBlocksToProcess - kNumFullBlocksPerFrame);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
break;
}
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
EchoCanceller3 aec3(EchoCanceller3Config(), sample_rate_hz_, 1, 1,
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
std::move(block_processor_mock));
for (size_t frame_index = 0; frame_index < kNumFramesToProcess;
++frame_index) {
bool echo_path_change = false;
switch (echo_path_change_test_variant) {
case EchoPathChangeTestVariant::kNone:
break;
case EchoPathChangeTestVariant::kOneSticky:
echo_path_change = true;
break;
case EchoPathChangeTestVariant::kOneNonSticky:
if (frame_index == 0) {
echo_path_change = true;
}
break;
}
aec3.AnalyzeCapture(&capture_buffer_);
OptionalBandSplit();
PopulateInputFrame(frame_length_, num_bands_, frame_index,
&capture_buffer_.split_bands(0)[0], 0);
PopulateInputFrame(frame_length_, frame_index,
&render_buffer_.channels()[0][0], 0);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
aec3.AnalyzeRender(&render_buffer_);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
aec3.ProcessCapture(&capture_buffer_, echo_path_change);
}
}
// Test for verifying that echo leakage information is being properly passed
// to the processor.
// The cases tested are:
// -That no method calls are received when they should not.
// -That false values are received each time they are flagged.
// -That true values are received each time they are flagged.
// -That a false value is received when flagged after a true value has been
// flagged.
enum class EchoLeakageTestVariant {
kNone,
kFalseSticky,
kTrueSticky,
kTrueNonSticky
};
void RunEchoLeakageVerificationTest(
EchoLeakageTestVariant leakage_report_variant) {
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
constexpr size_t kExpectedNumBlocksToProcess =
(kNumFramesToProcess * 160) / kBlockSize;
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
std::unique_ptr<testing::StrictMock<webrtc::test::MockBlockProcessor>>
block_processor_mock(
new StrictMock<webrtc::test::MockBlockProcessor>());
EXPECT_CALL(*block_processor_mock, BufferRender(_))
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
.Times(kExpectedNumBlocksToProcess);
EXPECT_CALL(*block_processor_mock, ProcessCapture(_, _, _, _))
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
.Times(kExpectedNumBlocksToProcess);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
switch (leakage_report_variant) {
case EchoLeakageTestVariant::kNone:
EXPECT_CALL(*block_processor_mock, UpdateEchoLeakageStatus(_)).Times(0);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
break;
case EchoLeakageTestVariant::kFalseSticky:
EXPECT_CALL(*block_processor_mock, UpdateEchoLeakageStatus(false))
.Times(1);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
break;
case EchoLeakageTestVariant::kTrueSticky:
EXPECT_CALL(*block_processor_mock, UpdateEchoLeakageStatus(true))
.Times(1);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
break;
case EchoLeakageTestVariant::kTrueNonSticky: {
::testing::InSequence s;
EXPECT_CALL(*block_processor_mock, UpdateEchoLeakageStatus(true))
.Times(1);
EXPECT_CALL(*block_processor_mock, UpdateEchoLeakageStatus(false))
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
.Times(kNumFramesToProcess - 1);
} break;
}
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
EchoCanceller3 aec3(EchoCanceller3Config(), sample_rate_hz_, 1, 1,
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
std::move(block_processor_mock));
for (size_t frame_index = 0; frame_index < kNumFramesToProcess;
++frame_index) {
switch (leakage_report_variant) {
case EchoLeakageTestVariant::kNone:
break;
case EchoLeakageTestVariant::kFalseSticky:
if (frame_index == 0) {
aec3.UpdateEchoLeakageStatus(false);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
}
break;
case EchoLeakageTestVariant::kTrueSticky:
if (frame_index == 0) {
aec3.UpdateEchoLeakageStatus(true);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
}
break;
case EchoLeakageTestVariant::kTrueNonSticky:
if (frame_index == 0) {
aec3.UpdateEchoLeakageStatus(true);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
} else {
aec3.UpdateEchoLeakageStatus(false);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
}
break;
}
aec3.AnalyzeCapture(&capture_buffer_);
OptionalBandSplit();
PopulateInputFrame(frame_length_, num_bands_, frame_index,
&capture_buffer_.split_bands(0)[0], 0);
PopulateInputFrame(frame_length_, frame_index,
&render_buffer_.channels()[0][0], 0);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
aec3.AnalyzeRender(&render_buffer_);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
aec3.ProcessCapture(&capture_buffer_, false);
}
}
// This verifies that saturation information is properly passed to the
// BlockProcessor.
// The cases tested are:
// -That no saturation event is passed to the processor if there is no
// saturation.
// -That one frame with one negative saturated sample value is reported to be
// saturated and that following non-saturated frames are properly reported as
// not being saturated.
// -That one frame with one positive saturated sample value is reported to be
// saturated and that following non-saturated frames are properly reported as
// not being saturated.
enum class SaturationTestVariant { kNone, kOneNegative, kOnePositive };
void RunCaptureSaturationVerificationTest(
SaturationTestVariant saturation_variant) {
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
const size_t kNumFullBlocksPerFrame = 160 / kBlockSize;
const size_t kExpectedNumBlocksToProcess =
(kNumFramesToProcess * 160) / kBlockSize;
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
std::unique_ptr<testing::StrictMock<webrtc::test::MockBlockProcessor>>
block_processor_mock(
new StrictMock<webrtc::test::MockBlockProcessor>());
EXPECT_CALL(*block_processor_mock, BufferRender(_))
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
.Times(kExpectedNumBlocksToProcess);
EXPECT_CALL(*block_processor_mock, UpdateEchoLeakageStatus(_)).Times(0);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
switch (saturation_variant) {
case SaturationTestVariant::kNone:
EXPECT_CALL(*block_processor_mock, ProcessCapture(_, false, _, _))
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
.Times(kExpectedNumBlocksToProcess);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
break;
case SaturationTestVariant::kOneNegative: {
::testing::InSequence s;
EXPECT_CALL(*block_processor_mock, ProcessCapture(_, true, _, _))
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
.Times(kNumFullBlocksPerFrame);
EXPECT_CALL(*block_processor_mock, ProcessCapture(_, false, _, _))
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
.Times(kExpectedNumBlocksToProcess - kNumFullBlocksPerFrame);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
} break;
case SaturationTestVariant::kOnePositive: {
::testing::InSequence s;
EXPECT_CALL(*block_processor_mock, ProcessCapture(_, true, _, _))
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
.Times(kNumFullBlocksPerFrame);
EXPECT_CALL(*block_processor_mock, ProcessCapture(_, false, _, _))
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
.Times(kExpectedNumBlocksToProcess - kNumFullBlocksPerFrame);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
} break;
}
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
EchoCanceller3 aec3(EchoCanceller3Config(), sample_rate_hz_, 1, 1,
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
std::move(block_processor_mock));
for (size_t frame_index = 0; frame_index < kNumFramesToProcess;
++frame_index) {
for (int k = 0; k < fullband_frame_length_; ++k) {
capture_buffer_.channels()[0][k] = 0.f;
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
}
switch (saturation_variant) {
case SaturationTestVariant::kNone:
break;
case SaturationTestVariant::kOneNegative:
if (frame_index == 0) {
capture_buffer_.channels()[0][10] = -32768.f;
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
}
break;
case SaturationTestVariant::kOnePositive:
if (frame_index == 0) {
capture_buffer_.channels()[0][10] = 32767.f;
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
}
break;
}
aec3.AnalyzeCapture(&capture_buffer_);
OptionalBandSplit();
PopulateInputFrame(frame_length_, num_bands_, frame_index,
&capture_buffer_.split_bands(0)[0], 0);
PopulateInputFrame(frame_length_, num_bands_, frame_index,
&render_buffer_.split_bands(0)[0], 0);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
aec3.AnalyzeRender(&render_buffer_);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
aec3.ProcessCapture(&capture_buffer_, false);
}
}
// This test verifies that the swapqueue is able to handle jitter in the
// capture and render API calls.
void RunRenderSwapQueueVerificationTest() {
const EchoCanceller3Config config;
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
EchoCanceller3 aec3(
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
config, sample_rate_hz_, 1, 1,
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
std::unique_ptr<BlockProcessor>(
new RenderTransportVerificationProcessor(num_bands_)));
std::vector<std::vector<float>> render_input(1);
std::vector<float> capture_output;
for (size_t frame_index = 0; frame_index < kRenderTransferQueueSizeFrames;
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
++frame_index) {
if (sample_rate_hz_ > 16000) {
render_buffer_.SplitIntoFrequencyBands();
}
PopulateInputFrame(frame_length_, num_bands_, frame_index,
&render_buffer_.split_bands(0)[0], 0);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
if (sample_rate_hz_ > 16000) {
render_buffer_.SplitIntoFrequencyBands();
}
for (size_t k = 0; k < frame_length_; ++k) {
render_input[0].push_back(render_buffer_.split_bands(0)[0][k]);
}
aec3.AnalyzeRender(&render_buffer_);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
}
for (size_t frame_index = 0; frame_index < kRenderTransferQueueSizeFrames;
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
++frame_index) {
aec3.AnalyzeCapture(&capture_buffer_);
if (sample_rate_hz_ > 16000) {
capture_buffer_.SplitIntoFrequencyBands();
}
PopulateInputFrame(frame_length_, num_bands_, frame_index,
&capture_buffer_.split_bands(0)[0], 0);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
aec3.ProcessCapture(&capture_buffer_, false);
for (size_t k = 0; k < frame_length_; ++k) {
capture_output.push_back(capture_buffer_.split_bands(0)[0][k]);
}
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
}
HighPassFilter hp_filter(16000, 1);
hp_filter.Process(&render_input);
EXPECT_TRUE(
VerifyOutputFrameBitexactness(render_input[0], capture_output, -64));
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
}
// This test verifies that a buffer overrun in the render swapqueue is
// properly reported.
void RunRenderPipelineSwapQueueOverrunReturnValueTest() {
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
EchoCanceller3 aec3(EchoCanceller3Config(), sample_rate_hz_, 1, 1);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
constexpr size_t kRenderTransferQueueSize = 30;
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
for (size_t k = 0; k < 2; ++k) {
for (size_t frame_index = 0; frame_index < kRenderTransferQueueSize;
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
++frame_index) {
if (sample_rate_hz_ > 16000) {
render_buffer_.SplitIntoFrequencyBands();
}
PopulateInputFrame(frame_length_, frame_index,
&render_buffer_.channels()[0][0], 0);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
aec3.AnalyzeRender(&render_buffer_);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
}
}
}
#if RTC_DCHECK_IS_ON && GTEST_HAS_DEATH_TEST && !defined(WEBRTC_ANDROID)
// Verifies the that the check for the number of bands in the AnalyzeRender
// input is correct by adjusting the sample rates of EchoCanceller3 and the
// input AudioBuffer to have a different number of bands.
void RunAnalyzeRenderNumBandsCheckVerification() {
// Set aec3_sample_rate_hz to be different from sample_rate_hz_ in such a
// way that the number of bands for the rates are different.
const int aec3_sample_rate_hz = sample_rate_hz_ == 48000 ? 32000 : 48000;
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
EchoCanceller3 aec3(EchoCanceller3Config(), aec3_sample_rate_hz, 1, 1);
PopulateInputFrame(frame_length_, 0, &render_buffer_.channels_f()[0][0], 0);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
EXPECT_DEATH(aec3.AnalyzeRender(&render_buffer_), "");
}
// Verifies the that the check for the number of bands in the ProcessCapture
// input is correct by adjusting the sample rates of EchoCanceller3 and the
// input AudioBuffer to have a different number of bands.
void RunProcessCaptureNumBandsCheckVerification() {
// Set aec3_sample_rate_hz to be different from sample_rate_hz_ in such a
// way that the number of bands for the rates are different.
const int aec3_sample_rate_hz = sample_rate_hz_ == 48000 ? 32000 : 48000;
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
EchoCanceller3 aec3(EchoCanceller3Config(), aec3_sample_rate_hz, 1, 1);
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
PopulateInputFrame(frame_length_, num_bands_, 0,
&capture_buffer_.split_bands_f(0)[0], 100);
EXPECT_DEATH(aec3.ProcessCapture(&capture_buffer_, false), "");
}
#endif
private:
void OptionalBandSplit() {
if (sample_rate_hz_ > 16000) {
capture_buffer_.SplitIntoFrequencyBands();
render_buffer_.SplitIntoFrequencyBands();
}
}
static constexpr size_t kNumFramesToProcess = 20;
const int sample_rate_hz_;
const size_t num_bands_;
const size_t frame_length_;
const int fullband_frame_length_;
AudioBuffer capture_buffer_;
AudioBuffer render_buffer_;
RTC_DISALLOW_IMPLICIT_CONSTRUCTORS(EchoCanceller3Tester);
};
std::string ProduceDebugText(int sample_rate_hz) {
rtc::StringBuilder ss;
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
ss << "Sample rate: " << sample_rate_hz;
return ss.Release();
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
}
std::string ProduceDebugText(int sample_rate_hz, int variant) {
rtc::StringBuilder ss;
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
ss << "Sample rate: " << sample_rate_hz << ", variant: " << variant;
return ss.Release();
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
}
} // namespace
TEST(EchoCanceller3Buffering, CaptureBitexactness) {
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
for (auto rate : {16000, 32000, 48000}) {
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
SCOPED_TRACE(ProduceDebugText(rate));
EchoCanceller3Tester(rate).RunCaptureTransportVerificationTest();
}
}
TEST(EchoCanceller3Buffering, RenderBitexactness) {
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
for (auto rate : {16000, 32000, 48000}) {
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
SCOPED_TRACE(ProduceDebugText(rate));
EchoCanceller3Tester(rate).RunRenderTransportVerificationTest();
}
}
TEST(EchoCanceller3Buffering, RenderSwapQueue) {
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
EchoCanceller3Tester(16000).RunRenderSwapQueueVerificationTest();
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
}
TEST(EchoCanceller3Buffering, RenderSwapQueueOverrunReturnValue) {
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
for (auto rate : {16000, 32000, 48000}) {
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
SCOPED_TRACE(ProduceDebugText(rate));
EchoCanceller3Tester(rate)
.RunRenderPipelineSwapQueueOverrunReturnValueTest();
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
}
}
TEST(EchoCanceller3Messaging, CaptureSaturation) {
auto variants = {EchoCanceller3Tester::SaturationTestVariant::kNone,
EchoCanceller3Tester::SaturationTestVariant::kOneNegative,
EchoCanceller3Tester::SaturationTestVariant::kOnePositive};
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
for (auto rate : {16000, 32000, 48000}) {
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
for (auto variant : variants) {
SCOPED_TRACE(ProduceDebugText(rate, static_cast<int>(variant)));
EchoCanceller3Tester(rate).RunCaptureSaturationVerificationTest(variant);
}
}
}
TEST(EchoCanceller3Messaging, EchoPathChange) {
auto variants = {
EchoCanceller3Tester::EchoPathChangeTestVariant::kNone,
EchoCanceller3Tester::EchoPathChangeTestVariant::kOneSticky,
EchoCanceller3Tester::EchoPathChangeTestVariant::kOneNonSticky};
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
for (auto rate : {16000, 32000, 48000}) {
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
for (auto variant : variants) {
SCOPED_TRACE(ProduceDebugText(rate, static_cast<int>(variant)));
EchoCanceller3Tester(rate).RunEchoPathChangeVerificationTest(variant);
}
}
}
TEST(EchoCanceller3Messaging, EchoLeakage) {
auto variants = {
EchoCanceller3Tester::EchoLeakageTestVariant::kNone,
EchoCanceller3Tester::EchoLeakageTestVariant::kFalseSticky,
EchoCanceller3Tester::EchoLeakageTestVariant::kTrueSticky,
EchoCanceller3Tester::EchoLeakageTestVariant::kTrueNonSticky};
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
for (auto rate : {16000, 32000, 48000}) {
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
for (auto variant : variants) {
SCOPED_TRACE(ProduceDebugText(rate, static_cast<int>(variant)));
EchoCanceller3Tester(rate).RunEchoLeakageVerificationTest(variant);
}
}
}
// Tests the parameter functionality for the field trial override for the
// default_len parameter.
TEST(EchoCanceller3FieldTrials, Aec3SuppressorEpStrengthDefaultLenOverride) {
EchoCanceller3Config default_config;
EchoCanceller3Config adjusted_config = AdjustConfig(default_config);
ASSERT_EQ(default_config.ep_strength.default_len,
adjusted_config.ep_strength.default_len);
webrtc::test::ScopedFieldTrials field_trials(
"WebRTC-Aec3SuppressorEpStrengthDefaultLenOverride/-0.02/");
adjusted_config = AdjustConfig(default_config);
ASSERT_NE(default_config.ep_strength.default_len,
adjusted_config.ep_strength.default_len);
EXPECT_FLOAT_EQ(-0.02f, adjusted_config.ep_strength.default_len);
}
// Tests the parameter functionality for the field trial override for the
// anti-howling gain.
TEST(EchoCanceller3FieldTrials, Aec3SuppressorAntiHowlingGainOverride) {
EchoCanceller3Config default_config;
EchoCanceller3Config adjusted_config = AdjustConfig(default_config);
ASSERT_EQ(
default_config.suppressor.high_bands_suppression.anti_howling_gain,
adjusted_config.suppressor.high_bands_suppression.anti_howling_gain);
webrtc::test::ScopedFieldTrials field_trials(
"WebRTC-Aec3SuppressorAntiHowlingGainOverride/0.02/");
adjusted_config = AdjustConfig(default_config);
ASSERT_NE(
default_config.suppressor.high_bands_suppression.anti_howling_gain,
adjusted_config.suppressor.high_bands_suppression.anti_howling_gain);
EXPECT_FLOAT_EQ(
0.02f,
adjusted_config.suppressor.high_bands_suppression.anti_howling_gain);
}
// Tests the field trial override for the enforcement of a low active render
// limit.
TEST(EchoCanceller3FieldTrials, Aec3EnforceLowActiveRenderLimit) {
EchoCanceller3Config default_config;
EchoCanceller3Config adjusted_config = AdjustConfig(default_config);
ASSERT_EQ(default_config.render_levels.active_render_limit,
adjusted_config.render_levels.active_render_limit);
webrtc::test::ScopedFieldTrials field_trials(
"WebRTC-Aec3EnforceLowActiveRenderLimit/Enabled/");
adjusted_config = AdjustConfig(default_config);
ASSERT_NE(default_config.render_levels.active_render_limit,
adjusted_config.render_levels.active_render_limit);
EXPECT_FLOAT_EQ(50.f, adjusted_config.render_levels.active_render_limit);
}
// Testing the field trial-based override of the suppressor parameters for a
// joint passing of all parameters.
TEST(EchoCanceller3FieldTrials, Aec3SuppressorTuningOverrideAllParams) {
webrtc::test::ScopedFieldTrials field_trials(
"WebRTC-Aec3SuppressorTuningOverride/"
"nearend_tuning_mask_lf_enr_transparent:0.1,nearend_tuning_mask_lf_enr_"
"suppress:0.2,nearend_tuning_mask_hf_enr_transparent:0.3,nearend_tuning_"
"mask_hf_enr_suppress:0.4,nearend_tuning_max_inc_factor:0.5,nearend_"
"tuning_max_dec_factor_lf:0.6,normal_tuning_mask_lf_enr_transparent:0.7,"
"normal_tuning_mask_lf_enr_suppress:0.8,normal_tuning_mask_hf_enr_"
"transparent:0.9,normal_tuning_mask_hf_enr_suppress:1.0,normal_tuning_"
"max_inc_factor:1.1,normal_tuning_max_dec_factor_lf:1.2,dominant_nearend_"
"detection_enr_threshold:1.3,dominant_nearend_detection_enr_exit_"
"threshold:1.4,dominant_nearend_detection_snr_threshold:1.5,dominant_"
"nearend_detection_hold_duration:10,dominant_nearend_detection_trigger_"
"threshold:11,ep_strength_default_len:1.6/");
EchoCanceller3Config default_config;
EchoCanceller3Config adjusted_config = AdjustConfig(default_config);
ASSERT_NE(adjusted_config.suppressor.nearend_tuning.mask_lf.enr_transparent,
default_config.suppressor.nearend_tuning.mask_lf.enr_transparent);
ASSERT_NE(adjusted_config.suppressor.nearend_tuning.mask_lf.enr_suppress,
default_config.suppressor.nearend_tuning.mask_lf.enr_suppress);
ASSERT_NE(adjusted_config.suppressor.nearend_tuning.mask_hf.enr_transparent,
default_config.suppressor.nearend_tuning.mask_hf.enr_transparent);
ASSERT_NE(adjusted_config.suppressor.nearend_tuning.mask_hf.enr_suppress,
default_config.suppressor.nearend_tuning.mask_hf.enr_suppress);
ASSERT_NE(adjusted_config.suppressor.nearend_tuning.max_inc_factor,
default_config.suppressor.nearend_tuning.max_inc_factor);
ASSERT_NE(adjusted_config.suppressor.nearend_tuning.max_dec_factor_lf,
default_config.suppressor.nearend_tuning.max_dec_factor_lf);
ASSERT_NE(adjusted_config.suppressor.normal_tuning.mask_lf.enr_transparent,
default_config.suppressor.normal_tuning.mask_lf.enr_transparent);
ASSERT_NE(adjusted_config.suppressor.normal_tuning.mask_lf.enr_suppress,
default_config.suppressor.normal_tuning.mask_lf.enr_suppress);
ASSERT_NE(adjusted_config.suppressor.normal_tuning.mask_hf.enr_transparent,
default_config.suppressor.normal_tuning.mask_hf.enr_transparent);
ASSERT_NE(adjusted_config.suppressor.normal_tuning.mask_hf.enr_suppress,
default_config.suppressor.normal_tuning.mask_hf.enr_suppress);
ASSERT_NE(adjusted_config.suppressor.normal_tuning.max_inc_factor,
default_config.suppressor.normal_tuning.max_inc_factor);
ASSERT_NE(adjusted_config.suppressor.normal_tuning.max_dec_factor_lf,
default_config.suppressor.normal_tuning.max_dec_factor_lf);
ASSERT_NE(adjusted_config.suppressor.dominant_nearend_detection.enr_threshold,
default_config.suppressor.dominant_nearend_detection.enr_threshold);
ASSERT_NE(
adjusted_config.suppressor.dominant_nearend_detection.enr_exit_threshold,
default_config.suppressor.dominant_nearend_detection.enr_exit_threshold);
ASSERT_NE(adjusted_config.suppressor.dominant_nearend_detection.snr_threshold,
default_config.suppressor.dominant_nearend_detection.snr_threshold);
ASSERT_NE(adjusted_config.suppressor.dominant_nearend_detection.hold_duration,
default_config.suppressor.dominant_nearend_detection.hold_duration);
ASSERT_NE(
adjusted_config.suppressor.dominant_nearend_detection.trigger_threshold,
default_config.suppressor.dominant_nearend_detection.trigger_threshold);
ASSERT_NE(adjusted_config.ep_strength.default_len,
default_config.ep_strength.default_len);
EXPECT_FLOAT_EQ(
adjusted_config.suppressor.nearend_tuning.mask_lf.enr_transparent, 0.1);
EXPECT_FLOAT_EQ(
adjusted_config.suppressor.nearend_tuning.mask_lf.enr_suppress, 0.2);
EXPECT_FLOAT_EQ(
adjusted_config.suppressor.nearend_tuning.mask_hf.enr_transparent, 0.3);
EXPECT_FLOAT_EQ(
adjusted_config.suppressor.nearend_tuning.mask_hf.enr_suppress, 0.4);
EXPECT_FLOAT_EQ(adjusted_config.suppressor.nearend_tuning.max_inc_factor,
0.5);
EXPECT_FLOAT_EQ(adjusted_config.suppressor.nearend_tuning.max_dec_factor_lf,
0.6);
EXPECT_FLOAT_EQ(
adjusted_config.suppressor.normal_tuning.mask_lf.enr_transparent, 0.7);
EXPECT_FLOAT_EQ(adjusted_config.suppressor.normal_tuning.mask_lf.enr_suppress,
0.8);
EXPECT_FLOAT_EQ(
adjusted_config.suppressor.normal_tuning.mask_hf.enr_transparent, 0.9);
EXPECT_FLOAT_EQ(adjusted_config.suppressor.normal_tuning.mask_hf.enr_suppress,
1.0);
EXPECT_FLOAT_EQ(adjusted_config.suppressor.normal_tuning.max_inc_factor, 1.1);
EXPECT_FLOAT_EQ(adjusted_config.suppressor.normal_tuning.max_dec_factor_lf,
1.2);
EXPECT_FLOAT_EQ(
adjusted_config.suppressor.dominant_nearend_detection.enr_threshold, 1.3);
EXPECT_FLOAT_EQ(
adjusted_config.suppressor.dominant_nearend_detection.enr_exit_threshold,
1.4);
EXPECT_FLOAT_EQ(
adjusted_config.suppressor.dominant_nearend_detection.snr_threshold, 1.5);
EXPECT_EQ(adjusted_config.suppressor.dominant_nearend_detection.hold_duration,
10);
EXPECT_EQ(
adjusted_config.suppressor.dominant_nearend_detection.trigger_threshold,
11);
EXPECT_FLOAT_EQ(adjusted_config.ep_strength.default_len, 1.6);
}
// Testing the field trial-based override of the suppressor parameters for
// passing one parameter.
TEST(EchoCanceller3FieldTrials, Aec3SuppressorTuningOverrideOneParam) {
webrtc::test::ScopedFieldTrials field_trials(
"WebRTC-Aec3SuppressorTuningOverride/nearend_tuning_max_inc_factor:0.5/");
EchoCanceller3Config default_config;
EchoCanceller3Config adjusted_config = AdjustConfig(default_config);
ASSERT_EQ(adjusted_config.suppressor.nearend_tuning.mask_lf.enr_transparent,
default_config.suppressor.nearend_tuning.mask_lf.enr_transparent);
ASSERT_EQ(adjusted_config.suppressor.nearend_tuning.mask_lf.enr_suppress,
default_config.suppressor.nearend_tuning.mask_lf.enr_suppress);
ASSERT_EQ(adjusted_config.suppressor.nearend_tuning.mask_hf.enr_transparent,
default_config.suppressor.nearend_tuning.mask_hf.enr_transparent);
ASSERT_EQ(adjusted_config.suppressor.nearend_tuning.mask_hf.enr_suppress,
default_config.suppressor.nearend_tuning.mask_hf.enr_suppress);
ASSERT_EQ(adjusted_config.suppressor.nearend_tuning.max_dec_factor_lf,
default_config.suppressor.nearend_tuning.max_dec_factor_lf);
ASSERT_EQ(adjusted_config.suppressor.normal_tuning.mask_lf.enr_transparent,
default_config.suppressor.normal_tuning.mask_lf.enr_transparent);
ASSERT_EQ(adjusted_config.suppressor.normal_tuning.mask_lf.enr_suppress,
default_config.suppressor.normal_tuning.mask_lf.enr_suppress);
ASSERT_EQ(adjusted_config.suppressor.normal_tuning.mask_hf.enr_transparent,
default_config.suppressor.normal_tuning.mask_hf.enr_transparent);
ASSERT_EQ(adjusted_config.suppressor.normal_tuning.mask_hf.enr_suppress,
default_config.suppressor.normal_tuning.mask_hf.enr_suppress);
ASSERT_EQ(adjusted_config.suppressor.normal_tuning.max_inc_factor,
default_config.suppressor.normal_tuning.max_inc_factor);
ASSERT_EQ(adjusted_config.suppressor.normal_tuning.max_dec_factor_lf,
default_config.suppressor.normal_tuning.max_dec_factor_lf);
ASSERT_EQ(adjusted_config.suppressor.dominant_nearend_detection.enr_threshold,
default_config.suppressor.dominant_nearend_detection.enr_threshold);
ASSERT_EQ(
adjusted_config.suppressor.dominant_nearend_detection.enr_exit_threshold,
default_config.suppressor.dominant_nearend_detection.enr_exit_threshold);
ASSERT_EQ(adjusted_config.suppressor.dominant_nearend_detection.snr_threshold,
default_config.suppressor.dominant_nearend_detection.snr_threshold);
ASSERT_EQ(adjusted_config.suppressor.dominant_nearend_detection.hold_duration,
default_config.suppressor.dominant_nearend_detection.hold_duration);
ASSERT_EQ(
adjusted_config.suppressor.dominant_nearend_detection.trigger_threshold,
default_config.suppressor.dominant_nearend_detection.trigger_threshold);
ASSERT_NE(adjusted_config.suppressor.nearend_tuning.max_inc_factor,
default_config.suppressor.nearend_tuning.max_inc_factor);
EXPECT_FLOAT_EQ(adjusted_config.suppressor.nearend_tuning.max_inc_factor,
0.5);
}
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
#if RTC_DCHECK_IS_ON && GTEST_HAS_DEATH_TEST && !defined(WEBRTC_ANDROID)
TEST(EchoCanceller3InputCheckDeathTest, WrongCaptureNumBandsCheckVerification) {
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
for (auto rate : {16000, 32000, 48000}) {
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
SCOPED_TRACE(ProduceDebugText(rate));
EchoCanceller3Tester(rate).RunProcessCaptureNumBandsCheckVerification();
}
}
// Verifiers that the verification for null input to the capture processing api
// call works.
TEST(EchoCanceller3InputCheckDeathTest, NullCaptureProcessingParameter) {
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
EXPECT_DEATH(EchoCanceller3(EchoCanceller3Config(), 16000, 1, 1)
.ProcessCapture(nullptr, false),
"");
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
}
// Verifies the check for correct sample rate.
// TODO(peah): Re-enable the test once the issue with memory leaks during DEATH
// tests on test bots has been fixed.
TEST(EchoCanceller3InputCheckDeathTest, DISABLED_WrongSampleRate) {
ApmDataDumper data_dumper(0);
Reland "Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."" This is a reland of a66395e72f9fc86873bf443579ec73c3d78af240 Original change's description: > Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3." > > This is a reland of f3a197e55323aee974a932c52dd19fa88e5d4e38 > > Original change's description: > > Add core multi-channel pipeline in AEC3 > > This CL adds basic the basic pipeline to support multi-channel > > processing in AEC3. > > > > Apart from that, it removes the 8 kHz processing support in several > > places of the AEC3 code. > > > > Bug: webrtc:10913 > > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137 > > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332 > > Commit-Queue: Per Åhgren <peah@webrtc.org> > > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > > Cr-Commit-Position: refs/heads/master@{#29017} > > Bug: webrtc:10913 > Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124 > Reviewed-by: Sam Zackrisson <saza@webrtc.org> > Commit-Queue: Per Åhgren <peah@webrtc.org> > Cr-Commit-Position: refs/heads/master@{#29034} Bug: webrtc:10913 Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127 Commit-Queue: Sam Zackrisson <saza@webrtc.org> Reviewed-by: Sam Zackrisson <saza@webrtc.org> Cr-Commit-Position: refs/heads/master@{#29042}
2019-09-02 17:01:19 +02:00
EXPECT_DEATH(EchoCanceller3(EchoCanceller3Config(), 8001, 1, 1), "");
}
Reland of Added first layer of the echo canceller 3 functionality. Reason for reland: Added disabling of the BlockProcessor DEATH tests due to false alarms on the trybots. These errors cannot be reproduced locally and they occur only in the DEATH test code, which means that the memory leaks do not occur under normal conditions. Original issue's description: > Reason for revert: > Memcheck buildbot detected memory leaks in the death tests. The code looks fine > to me, but please investigate what causes the error and reland. > > Original issue's description: > > Added first layer of the echo canceller 3 functionality. > > > > This CL adds the first layer of the echo canceller 3. > > All of the code is as it should, apart from > > block_processor.* which only contains placeholder > > functionality. (Upcoming CLs will add proper > > functionality into those files.) > > > > > > > > BUG=webrtc:6018 > > > > Review-Url: https://codereview.webrtc.org/2584493002 > > Cr-Commit-Position: refs/heads/master@{#15861} > > Committed: > https://chromium.googlesource.com/external/webrtc/+/38fd1758e90bcdc7690a552e7ef0ec0d143d2f30 > > TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org,peah@webrtc.org > # Skipping CQ checks because original CL landed less than 1 days ago. > NOPRESUBMIT=true > NOTREECHECKS=true > NOTRY=true > BUG=webrtc:6018 BUG=webrtc:6018 TBR=ivoc@webrtc.org,aleloi@webrtc.org,henrik.lundin@webrtc.org Review-Url: https://codereview.webrtc.org/2608233002 Cr-Commit-Position: refs/heads/master@{#15884}
2017-01-03 04:20:34 -08:00
#endif
} // namespace webrtc