Question

Really basic question here:

I’ve seen a bunch of CoreAudio code that uses memset when dealing with structs and I haven’t been able to figure out why. This is from a .m ObjC file.

Here’s the line from the code block below:

memset(&clientFormat, 0, sizeof(clientFormat));

//—————

    AudioStreamBasicDescription clientFormat;
    if ( sourceFormat.mFormatID == kAudioFormatLinearPCM ) {
    clientFormat = sourceFormat;
    } else {
    memset(&clientFormat, 0, sizeof(clientFormat));
    int sampleSize = sizeof(AudioSampleType);
    clientFormat.mFormatID = kAudioFormatLinearPCM;
    clientFormat.mFormatFlags = kAudioFormatFlagsCanonical;
    clientFormat.mBitsPerChannel = 8 * sampleSize;
    clientFormat.mChannelsPerFrame = sourceFormat.mChannelsPerFrame;
    clientFormat.mFramesPerPacket = 1;
    clientFormat.mBytesPerPacket = clientFormat.mBytesPerFrame = sourceFormat.mChannelsPerFrame * sampleSize;
    clientFormat.mSampleRate = sourceFormat.mSampleRate;
}

Answer 1

As called, it sets the memory at &clientFormat, with a length of sizeof(clientFormat), to 0. This is required because memory in C is not cleared to 0 after allocation with malloc() and most other allocation functions.