iOS Swift DispatchQueue: How do I run a process using multiple CPUs

So far I am trying to run the Grand Central Dispatch to process images on multiple cores.

The issue I am running into is that the processing only occurs on the main queue and stays on the first core.

I am using an iPhone 6 which has dual cores up to 200% cpu I can use and it caps out at 100% CPU and other threads are clearly not running the function.

CODE:

//value is an image inside an image array that I am using the .map function
//imageProcess processes an image on the device CPU.

let core1 = DispatchQueue(label: "core1", qos: .userInitiated)
let core2 = DispatchQueue(label: "core2", qos: .userInitiated)

core1.async

self.imageProcess(fromImage: value, completion: (error) in
if let error = error
print("Oops something failed!")

else
print("It has finished! Core1")

)

core2.async
self.imageProcess(fromImage: value, completion: (error) in
if let error = error
print("Oops something failed!")

else
print("It has finished! Core2")

)



What I am seeing is the DispatchQueue just drops everything into thread 11 on my CPU and all the images are processed that way. There is no multithreading and it stays on a single core.

Questions:

Thanks all!

Update

Looks like my function is not properly configured for DispatchQueue. Any suggestions?

Please note that faceDetector is calling a firebase function MLKit Firebase

private func imageProcess(fromImage image: UIImage, completion: @escaping (Error?) -> Void)
let visionImage = VisionImage(image: image)
faceDetector.detect(in: visionImage) [unowned self] (faces, error) in
guard error == nil, let detectedFaces = faces else
self.showAlert("An error occured", alertMessage: "The face detection failed.")
return


let faceStates = self.faceStates(forDetectedFaces: detectedFaces)
self.updateDetectedInfo(forFaceStates: faceStates)
completion(nil)




qos

let core1 = DispatchQueue(label: "core1", qos: .background)

imageProcess

detect

imageProcess

core1.async

core2.async

imageProcess returns immediately while

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.