ios – why my knowledge from buffer of AudioRenderCallback appears to be like unusual?

I am making an attempt to implement fft.
So I attempt to use audio buffer from audio render callback

static OSStatus renderCallback (AudioPlayer                 *   self,
                                    AudioUnitRenderActionFlags  *   ioActionFlags,
                                    const AudioTimeStamp        *   inTimeStamp,
                                    UInt32                          inBusNumber,
                                    UInt32                          inNumberFrames,
                                    AudioBufferList             *   ioData)
    {
        /*
        attempt to use ioData->mBuffers[0].mData
        */
        return OSStatus;
    }

nonetheless fft end result are crammed with nan.

I discovered pattern code’s audio buffer has values between -1 and 1 however my audio buffer values are like ‘-3.2342E-34’

possibly cheap worth shouldn’t be between -1 and 1 however ‘-3.2342E-34’ would not appear to be proper worth. (audio is taking part in usually, although)

beneath format is my output format of AudioGraph and I cannot change this now…

AudioStreamBasicDescription format =
         kAudioFormatFlagsNativeEndian ;

I would like to determine what that numbers are to step ahead however, frankly, I do not even know what info is important for somebody who look into this.

I can replace data right here as a lot as essential.

please give me some information.

thanks.