zqz/platform-external-webrtc
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Prevent crash in NetEQ when decoder overflow.

Browse Source 
NetEQ can crash when decoder gives too many output samples than it can handle. A practical case this happens is when multiple opus packets are combined.

The best solution is to pass the max size to the ACM decode function and let it return a failure if the max size if too small.

BUG=4361
R=henrik.lundin@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/45619004

Cr-Commit-Position: refs/heads/master@{#8730}
git-svn-id: http://webrtc.googlecode.com/svn/trunk@8730 4adac7df-926f-26a2-2b94-8c16560cd09d
This commit is contained in:
minyue@webrtc.org
2015-03-16 12:30:37 +00:00
parent 4b89aa03bb
commit 7f7d7e3427
19 changed files with 453 additions and 188 deletions
Download Patch File Download Diff File Expand all files Collapse all files

40
webrtc/modules/audio_coding/codecs/audio_decoder.cc
View File

@ -16,12 +16,40 @@
namespace webrtc {
int AudioDecoder::DecodeRedundant(const uint8_t* encoded,
size_t encoded_len,
int sample_rate_hz,
int16_t* decoded,
SpeechType* speech_type) {
return Decode(encoded, encoded_len, sample_rate_hz, decoded, speech_type);
int AudioDecoder::Decode(const uint8_t* encoded, size_t encoded_len,
int sample_rate_hz, size_t max_decoded_bytes,
int16_t* decoded, SpeechType* speech_type) {
int duration = PacketDuration(encoded, encoded_len);
if (duration >= 0 && duration * sizeof(int16_t) > max_decoded_bytes) {
return -1;
}
return DecodeInternal(encoded, encoded_len, sample_rate_hz, decoded,
speech_type);
}
int AudioDecoder::DecodeRedundant(const uint8_t* encoded, size_t encoded_len,
int sample_rate_hz, size_t max_decoded_bytes,
int16_t* decoded, SpeechType* speech_type) {
int duration = PacketDurationRedundant(encoded, encoded_len);
if (duration >= 0 && duration * sizeof(int16_t) > max_decoded_bytes) {
return -1;
}
return DecodeRedundantInternal(encoded, encoded_len, sample_rate_hz, decoded,
speech_type);
}
int AudioDecoder::DecodeInternal(const uint8_t* encoded, size_t encoded_len,
int sample_rate_hz, int16_t* decoded,
SpeechType* speech_type) {
return kNotImplemented;
}
int AudioDecoder::DecodeRedundantInternal(const uint8_t* encoded,
size_t encoded_len,
int sample_rate_hz, int16_t* decoded,
SpeechType* speech_type) {
return DecodeInternal(encoded, encoded_len, sample_rate_hz, decoded,
speech_type);
}
bool AudioDecoder::HasDecodePlc() const { return false; }

21
webrtc/modules/audio_coding/codecs/audio_decoder.h
View File

@ -35,22 +35,25 @@ class AudioDecoder {
virtual ~AudioDecoder() {}
// Decodes |encode_len| bytes from |encoded| and writes the result in
// |decoded|. The number of samples from all channels produced is in
// the return value. If the decoder produced comfort noise, |speech_type|
// |decoded|. The maximum bytes allowed to be written into |decoded| is
// |max_decoded_bytes|. The number of samples from all channels produced is
// in the return value. If the decoder produced comfort noise, |speech_type|
// is set to kComfortNoise, otherwise it is kSpeech. The desired output
// sample rate is provided in |sample_rate_hz|, which must be valid for the
// codec at hand.
virtual int Decode(const uint8_t* encoded,
size_t encoded_len,
int sample_rate_hz,
size_t max_decoded_bytes,
int16_t* decoded,
SpeechType* speech_type) = 0;
SpeechType* speech_type);
// Same as Decode(), but interfaces to the decoders redundant decode function.
// The default implementation simply calls the regular Decode() method.
virtual int DecodeRedundant(const uint8_t* encoded,
size_t encoded_len,
int sample_rate_hz,
size_t max_decoded_bytes,
int16_t* decoded,
SpeechType* speech_type);
@ -99,6 +102,18 @@ class AudioDecoder {
protected:
static SpeechType ConvertSpeechType(int16_t type);
virtual int DecodeInternal(const uint8_t* encoded,
size_t encoded_len,
int sample_rate_hz,
int16_t* decoded,
SpeechType* speech_type);
virtual int DecodeRedundantInternal(const uint8_t* encoded,
size_t encoded_len,
int sample_rate_hz,
int16_t* decoded,
SpeechType* speech_type);
size_t channels_;
private:

12
webrtc/modules/audio_coding/codecs/isac/audio_encoder_isac_t.h
View File

@ -72,11 +72,6 @@ class AudioEncoderDecoderIsacT : public AudioEncoder, public AudioDecoder {
int Max10MsFramesInAPacket() const override;
// AudioDecoder methods.
int Decode(const uint8_t* encoded,
size_t encoded_len,
int sample_rate_hz,
int16_t* decoded,
SpeechType* speech_type) override;
bool HasDecodePlc() const override;
int DecodePlc(int num_frames, int16_t* decoded) override;
int Init() override;
@ -95,6 +90,13 @@ class AudioEncoderDecoderIsacT : public AudioEncoder, public AudioDecoder {
uint8_t* encoded,
EncodedInfo* info) override;
// AudioDecoder protected method.
int DecodeInternal(const uint8_t* encoded,
size_t encoded_len,
int sample_rate_hz,
int16_t* decoded,
SpeechType* speech_type) override;
private:
// This value is taken from STREAM_SIZE_MAX_60 for iSAC float (60 ms) and
// STREAM_MAXW16_60MS for iSAC fix (60 ms).

14
webrtc/modules/audio_coding/codecs/isac/audio_encoder_isac_t_impl.h
View File

@ -218,11 +218,11 @@ void AudioEncoderDecoderIsacT<T>::EncodeInternal(uint32_t rtp_timestamp,
}
template <typename T>
int AudioEncoderDecoderIsacT<T>::Decode(const uint8_t* encoded,
size_t encoded_len,
int sample_rate_hz,
int16_t* decoded,
SpeechType* speech_type) {
int AudioEncoderDecoderIsacT<T>::DecodeInternal(const uint8_t* encoded,
size_t encoded_len,
int sample_rate_hz,
int16_t* decoded,
SpeechType* speech_type) {
CriticalSectionScoped cs(state_lock_.get());
CHECK(sample_rate_hz == 16000 || sample_rate_hz == 32000)
<< "Unsupported sample rate " << sample_rate_hz;
@ -232,8 +232,8 @@ int AudioEncoderDecoderIsacT<T>::Decode(const uint8_t* encoded,
}
int16_t temp_type = 1; // Default is speech.
int16_t ret =
T::Decode(isac_state_, encoded, static_cast<int16_t>(encoded_len),
decoded, &temp_type);
T::DecodeInternal(isac_state_, encoded, static_cast<int16_t>(encoded_len),
decoded, &temp_type);
*speech_type = ConvertSpeechType(temp_type);
return ret;
}

10
webrtc/modules/audio_coding/codecs/isac/fix/interface/audio_encoder_isacfix.h
View File

@ -36,11 +36,11 @@ struct IsacFix {
static inline int16_t Create(instance_type** inst) {
return WebRtcIsacfix_Create(inst);
}
static inline int16_t Decode(instance_type* inst,
const uint8_t* encoded,
int16_t len,
int16_t* decoded,
int16_t* speech_type) {
static inline int16_t DecodeInternal(instance_type* inst,
const uint8_t* encoded,
int16_t len,
int16_t* decoded,
int16_t* speech_type) {
return WebRtcIsacfix_Decode(inst, encoded, len, decoded, speech_type);
}
static inline int16_t DecodePlc(instance_type* inst,

10
webrtc/modules/audio_coding/codecs/isac/main/interface/audio_encoder_isac.h
View File

@ -35,11 +35,11 @@ struct IsacFloat {
static inline int16_t Create(instance_type** inst) {
return WebRtcIsac_Create(inst);
}
static inline int16_t Decode(instance_type* inst,
const uint8_t* encoded,
int16_t len,
int16_t* decoded,
int16_t* speech_type) {
static inline int16_t DecodeInternal(instance_type* inst,
const uint8_t* encoded,
int16_t len,
int16_t* decoded,
int16_t* speech_type) {
return WebRtcIsac_Decode(inst, encoded, len, decoded, speech_type);
}
static inline int16_t DecodePlc(instance_type* inst,

48
webrtc/modules/audio_coding/codecs/opus/opus_unittest.cc
View File

@ -554,6 +554,54 @@ TEST_P(OpusTest, OpusDurationEstimation) {
EXPECT_EQ(0, WebRtcOpus_DecoderFree(opus_decoder_));
}
TEST_P(OpusTest, OpusDecodeRepacketized) {
const int kPackets = 6;
PrepareSpeechData(channels_, 20, 20 * kPackets);
// Create encoder memory.
ASSERT_EQ(0, WebRtcOpus_EncoderCreate(&opus_encoder_,
channels_,
application_));
ASSERT_EQ(0, WebRtcOpus_DecoderCreate(&opus_decoder_,
channels_));
// Set bitrate.
EXPECT_EQ(0, WebRtcOpus_SetBitRate(opus_encoder_,
channels_ == 1 ? 32000 : 64000));
// Check number of channels for decoder.
EXPECT_EQ(channels_, WebRtcOpus_DecoderChannels(opus_decoder_));
// Encode & decode.
int16_t audio_type;
rtc::scoped_ptr<int16_t[]> output_data_decode(
new int16_t[kPackets * kOpus20msFrameSamples * channels_]);
OpusRepacketizer* rp = opus_repacketizer_create();
for (int idx = 0; idx < kPackets; idx++) {
encoded_bytes_ = WebRtcOpus_Encode(opus_encoder_,
speech_data_.GetNextBlock(),
kOpus20msFrameSamples, kMaxBytes,
bitstream_);
EXPECT_EQ(OPUS_OK, opus_repacketizer_cat(rp, bitstream_, encoded_bytes_));
}
encoded_bytes_ = opus_repacketizer_out(rp, bitstream_, kMaxBytes);
EXPECT_EQ(kOpus20msFrameSamples * kPackets,
WebRtcOpus_DurationEst(opus_decoder_, bitstream_, encoded_bytes_));
EXPECT_EQ(kOpus20msFrameSamples * kPackets,
WebRtcOpus_Decode(opus_decoder_, bitstream_, encoded_bytes_,
output_data_decode.get(), &audio_type));
// Free memory.
opus_repacketizer_destroy(rp);
EXPECT_EQ(0, WebRtcOpus_EncoderFree(opus_encoder_));
EXPECT_EQ(0, WebRtcOpus_DecoderFree(opus_decoder_));
}
INSTANTIATE_TEST_CASE_P(VariousMode,
OpusTest,
Combine(Values(1, 2), Values(0, 1)));