Question

My Audio Unit analysis project is having some memory issues, whereby each time an Audio Unit is rendered (or somewhere around that) it is allocating a bunch of memory which isn’t being released, causing memory usage to swell and the app to eventually crash.

In instruments, I notice the following string of 32 byte mallocs occurring repeatedly, and they remain live:

BufferedAudioConverter::AllocateBuffers() x6
BufferedInputAudioConverter:BufferedInputAudioConverter(StreamDescPair const&) x 3

Any ideas where the problem might lie? When is that memory allocated in the process and how can it safely be released?

Many thanks.

The code was based on some non-Apple sample code, PitchDetector from sleepyleaf.com

Some code extracts where the problem might lie….. Please let me know if more code is needed.

renderErr = AudioUnitRender(rioUnit, ioActionFlags, 
                            inTimeStamp, bus1, inNumberFrames, THIS->bufferList); //128 inNumberFrames
if (renderErr < 0) {
    return renderErr;
}

// Fill the buffer with our sampled data. If we fill our buffer, run the
// fft.
int read = bufferCapacity - index;
if (read > inNumberFrames) {
    memcpy((SInt16 *)dataBuffer + index, THIS->bufferList->mBuffers[0].mData, inNumberFrames*sizeof(SInt16));
    THIS->index += inNumberFrames;
} else {  DO ANALYSIS

memset(outputBuffer, 0, n*sizeof(SInt16));

---

- (void)createAUProcessingGraph {
OSStatus err;
// Configure the search parameters to find the default playback output unit
// (called the kAudioUnitSubType_RemoteIO on iOS but
// kAudioUnitSubType_DefaultOutput on Mac OS X)
AudioComponentDescription ioUnitDescription;
ioUnitDescription.componentType = kAudioUnitType_Output;
ioUnitDescription.componentSubType = kAudioUnitSubType_RemoteIO;
ioUnitDescription.componentManufacturer = kAudioUnitManufacturer_Apple;
ioUnitDescription.componentFlags = 0;
enter code here

ioUnitDescription.componentFlagsMask = 0;

// Declare and instantiate an audio processing graph
NewAUGraph(&processingGraph);

// Add an audio unit node to the graph, then instantiate the audio unit.
/* 
 An AUNode is an opaque type that represents an audio unit in the context
 of an audio processing graph. You receive a reference to the new audio unit
 instance, in the ioUnit parameter, on output of the AUGraphNodeInfo 
 function call.
 */
AUNode ioNode;
AUGraphAddNode(processingGraph, &ioUnitDescription, &ioNode);

AUGraphOpen(processingGraph); // indirectly performs audio unit instantiation

// Obtain a reference to the newly-instantiated I/O unit. Each Audio Unit
// requires its own configuration.
AUGraphNodeInfo(processingGraph, ioNode, NULL, &ioUnit);

// Initialize below.
AURenderCallbackStruct callbackStruct = {0};
UInt32 enableInput;
UInt32 enableOutput;

// Enable input and disable output.
enableInput = 1; enableOutput = 0;
callbackStruct.inputProc = RenderFFTCallback;
callbackStruct.inputProcRefCon = (__bridge void*)self;

err = AudioUnitSetProperty(ioUnit, kAudioOutputUnitProperty_EnableIO, 
                           kAudioUnitScope_Input, 
                           kInputBus, &enableInput, sizeof(enableInput));

err = AudioUnitSetProperty(ioUnit, kAudioOutputUnitProperty_EnableIO, 
                           kAudioUnitScope_Output, 
                           kOutputBus, &enableOutput, sizeof(enableOutput));

err = AudioUnitSetProperty(ioUnit, kAudioOutputUnitProperty_SetInputCallback, 
                           kAudioUnitScope_Input, 
                           kOutputBus, &callbackStruct, sizeof(callbackStruct));


// Set the stream format.
size_t bytesPerSample = [self ASBDForSoundMode];

err = AudioUnitSetProperty(ioUnit, kAudioUnitProperty_StreamFormat, 
                           kAudioUnitScope_Output, 
                           kInputBus, &streamFormat, sizeof(streamFormat));

err = AudioUnitSetProperty(ioUnit, kAudioUnitProperty_StreamFormat, 
                           kAudioUnitScope_Input, 
                           kOutputBus, &streamFormat, sizeof(streamFormat));




// Disable system buffer allocation. We'll do it ourselves.
UInt32 flag = 0;
err = AudioUnitSetProperty(ioUnit, kAudioUnitProperty_ShouldAllocateBuffer,
                           kAudioUnitScope_Output, 
                           kInputBus, &flag, sizeof(flag));


// Allocate AudioBuffers for use when listening.
// TODO: Move into initialization...should only be required once.
    bufferList = (AudioBufferList *)malloc(sizeof(AudioBuffer));
    bufferList->mNumberBuffers = 1;
    bufferList->mBuffers[0].mNumberChannels = 1;

    bufferList->mBuffers[0].mDataByteSize = 512*bytesPerSample;
    bufferList->mBuffers[0].mData = calloc(512, bytesPerSample);

}

Answer 1

I managed to find and fix the issue, which was in an area of the code not posted above.

In a following step the output buffer was being converted into a different number format using an AudioConverter object. However, the converter object was not being disposed of, and remained live in the memory. I fixed it by using AudioConverterDispose as below:

err = AudioConverterNew(&inFormat, &outFormat, &converter);
err = AudioConverterConvertBuffer(converter, inSize, buf, &outSize, outputBuf);

err = AudioConverterDispose (converter);