I have created one testing app for running deep counter loop. I run the loop function in background thread using performSelectorInBackground and also NSOperation subclass separately.
I am also using performSelectorOnMainThread to notify main thread within backgroundthread method and [NSNotificationCenter defaultCenter] postNotificationName within NSOperation subclass to notify main thread for updating UI.
Initially both the implementation giving me same result and i am able to update UI without having any problem. The only difference I found is the Thread count between two implementations.
The performSelectorInBackground implementation created one thread and got terminated after loop finished and my app thread count again goes to 1.
The NSOperation subclass implementation created two new threads and keep exists in the application and i can see 3 threads after loop got finished in main() function.
So, my question is why two threads created by NSOperation and why it didn't get terminated just like the first background thread implementation?
I am little bit confuse and unable to decide which implementation is best in-terms of performance and memory management.
It's likely the operation queue is keeping threads alive, waiting for new operations to appear.
You have to remember that the operation queue is designed to work efficiently with many operations, so creating and destroying threads for each operation is going to hurt performance. So what you are seeing is probably just the way the queue is designed to work by keeping a pool of threads alive.
Basically, as long as you are using the operation queue properly and according to the documentation I wouldn't worry about it.
Related
I have an iOS application using NSThreads for concurrency tasks. I will try to migrate it to be using the Grand Central Dispatch (GCD) for handling concurrency.
The problem is that the app needs information regarding how many threads has been created since a given time. And how many threads that was spawned since that given time is currently running.
At the moment this is done by creating a category that does a method swizzling on the -main method in NSThread. In the new swizzled method it simply increments the total number of threads running and then decrement the same variable before the new swizzled -main method returns.
The problem is that when I use GCD dispatch_async it does not create a NSThread, hence my category approach does not work. How can I achieve the same while using GCD to handle concurrency?
What I would like to detect is when a new block is added to GCD, and when that block has been executed.
Any suggestions on how to achieve the same is very welcome.
EDIT
Many thanks to #ipmcc and #RyanR for helping me out on this. :) I believe I need to tell some more about the background and what I am trying to accomplish.
What I am actually trying is to extend the iOS testing framework Frank. Frank embeds a small web-server within a given app which enables sending HTTP request to the iOS application and thereby simulating events, a swipe or a tap gesture as an example.
I would like to extend it in a way that enables it to wait until all work triggered by a specific simulated event has ended before returning upon a request.
However I found it hard to detect exactly what work was triggered by the received event. And thats how I came to the solution to just reset a thread counter and then increment this counter for all created threads after the event was simulated, and decrement it when the threads are finishing. And then block until threads count became zero again. I know this approach is not perfect either, and it wont work with GCP.
Is there any other way to achieve it? Another possible solution which I have thought of is to specify that everything must run synchronized except the thread handling the HTTP request. However I don't know if this possible.
Any suggestions on how to achieve blocking after each simulated event until work triggered by that event has completed?
The problem is that the app needs information regarding how many
threads has been created since a given time. And how many threads that
was spawned since that given time is currently running.
You will not be able to get this information from GCD. One of the points of GCD is that you do not manage the thread pool. It is opaque. You'll note that even pthreads, the underlying threading library on which NSThread and GCD are built, does not have a (public) means to enumerate all existing threads or get the number of running threads. This is not going to be doable without hard core low level hackery. If you need to control or know the number of threads, then you need to be the one to spawn and manage them, and GCD is the wrong abstraction for you.
At the moment this is done by creating a category that does a method
swizzling on the -main method in NSThread. In the new swizzled method
it simply increments the total number of threads running and then
decrement the same variable before the new swizzled -main method
returns.
Note that this only tells you the number of threads started using NSThread. As mentioned, NSThread is a fairly high level abstraction on top of pthreads. There is nothing to prevent library code from spawning its own threads using the pthreads API that will be invisible to your count.
The problem is that when I use GCD dispatch_async it does not create a
NSThread, hence my category approach does not work. How can I achieve
the same while using GCD to handle concurrency?
In short, you can't. If you want to go forth and patch functions all over the various frameworks, then you should look up a library called mach_override. (But please don't.)
What I would like to detect is when a new block is added to GCD, and
when that block has been executed.
Since GCD uses thread pools, the act of adding a block does not imply a new thread. (And that's sorta the whole point.)
If you have some limited resource whose consumption you need to manage, the traditional way to do that would be with a limiting semaphore, but that is just one option.
This whole question just reeks of a poor design. Like the number of pthreads, GCD's queue widths are opaque/non-public. Your previous solution was not particularly viable (as discussed), and further efforts are likely to yield similarly poor solutions. You should really rethink your architecture such that knowing how many threads are running isn't important.
EDIT: Thanks for the clarification. There's not really a generic way, from the outside, to tell when all the "work" is done. What if an action sets up a timer that won't call back for ten minutes? At the extreme, consider this: the main runloop continues to spin for the entire life of the app, and as long as the main runloop is spinning, "work" could be being done on it.
In order to detect "doneness" your app has to signal doneness. In order to signal doneness, the app has to have some way (internal to itself) to know it's done. Put differently, the app can't tell something else (i.e. Frank) something it doesn't know. One way to go about this would be to encapsulate all the work you do in your app in NSOperations. NSOperation/NSOperationQueue provide good ways of reporting "doneness." At the simplest level, you could wrap the code where you kickoff work in an NSBlockOperation, then add a completion block to that operation that signals something else when it's done, and enqueue it to an NSOperationQueue for execution. (You could also do this with dispatch_group and dispatch_group_notify if you prefer working in the GCD style.)
If you have specific questions about how to package up your app's work into NSOperations, I would suggest starting a new question.
You can hook into the dispatch introspection functions (introspection.h, methods all start with dispatch_introspection), but you have to link with that library which is supposed to be only for debugging. I don't think you can include that in a release build. Your best bet would be to encapsulate GCD into your own object, so all your code submits blocks to execute through that object and it submits them to GCD after tracking whatever you're interested in. You won't be able to track thread consumption though, because GCD intentionally abstracts that and reuses threads.
Is it safe to add an NSOperationQueue to an NSOperation, and then add this operation to another NSOperationQueue?
Here is some code to visualize what I am trying to do.
NSOperationQueue *mainQueue = [NSOperationQueue alloc] init];
// Here I declare some NSBlockOperation's, i.e. parseOperation1-2-3
// and also another operation called zipOperation, which includes
// an NSOperationQueue itself. This queue takes the processed (parsed) files
// and write them to a single zip file. Each operation's job is to write the data
// stream and add it to the zip file. After all operations are done,
// it closes the zip.
[zipOperation addDependency:parseOperation1];
[zipOperation addDependency:parseOperation2];
[zipOperation addDependency:parseOperation3];
[mainQueue addOperation:parseOperation1];
[mainQueue addOperation:parseOperation2];
[mainQueue addOperation:parseOperation3];
[mainQueue addOperation:zipOperation];
I have used this approach and have it running in live code deployed on the App Store. I haven't experienced any issues during development or in the last 2 months since the code has been live.
In my case I had a high level series of operations, some of which contained a set of sub operations. Rather than expose the detail of each sub operation into the high level code, I created NSOperations which themselves contained NSOperationQueues and enqueued their own sub operations. The code I ended up with was much cleaner and easier to maintain.
I read extensively into NSOperation and have not seen any commentary that warns against this approach. I reviewed a lot of information online, the Apple documentation, and WWDC videos.
The only possible "drawback" might be the added complexity of understanding and implementing a Concurrent operation. Embedding an NSOperationQueue in an NSOperation means that operation becomes Concurrent.
So that's a 'YES' from me.
Additional details about concurrent operations:
An NSOperationQueue calls the start method on a normal (non-concurrent) NSOperation and expects the operation to be finished by the time the start call returns. For instance some piece of code you supplied to NSBlockOperation is complete at the end of the block.
If the work will not be finished by the time the start call returns then you configure the NSOperation as a Concurrent operation, so the NSOperationQueue knows that it has to wait until you tell it that the operation is finished at some later point in time.
For example, concurrent operations are often used to run asynchronous network calls; the start method only starts the network call, which then runs in the background, and calls back to the operation when its finished. You then change the isFinished property of the NSOperation to flag that the work is now complete.
So.... Normally when you add operations to an NSOperationQueue that queue runs those operations in the background. So if you put an NSOperationQueue inside an NSOperation then that operations work will be done in the background. Therefore the operation is concurrent and you need to flag when the internal NSOperationQueue has finished processing all it's operations.
Alternatively there are some methods on NSOperationQueue such as waitUntilAllOperationsAreFinished which could be used to ensure all the work was done before the start call returns, however these involve blocking threads and I avoided them, you may feel more comfortable with that approach, and making sure you don't have any side effects from blocking threads.
In my case I was already familiar with Concurrent operations so it was straightforward just to set it up as a Concurrent operation.
Some documentation about concurrent operations:
Concurrency Programming Guide: Configuring Operations for Concurrent Execution
In this example they are detaching a thread to perform work in the background, in our case we would be starting the NSOperationQueue here.
I am using GCD to start a long-running background process ('run_loop') that creates an NSManagedObjectContext ('MOC'), monitors CoreData objects, and sometimes (when they're ready) uploads a serialization of them to a webserver and then deletes them.
I am using AFNetworking for the HTTP calls. The problem is in the request completion handler blocks, as the blocks run in a different thread to the owner of the MOC, which isn't supported by CoreData.
I have tried storing the NSThread from the start of the GCD run_loop block, and using performSelector:onThread:run_thread but this just doesn't seem to actually call the selector at all.
I have tried using dispatch_sync(run_queue) but this doesn't guarantee the thread is the same, only the GCD queue. A different MOC save in the main thread later hangs.
Eventually the only thing that worked was to set a boolean in the completion callback handler, and to introduce extra logic to detect the boolean switch and to perform the MOC work from the main run_loop.
Would anyone be able to suggest a more elegant fix? Or is CoreData simply not compatible with an AFNetworking request started from a GCD queue, and I should look at a lower-level thread control from the start?
Hmm .. the recommended way to deal with MOC and threads is to always make a new MOC that is a sub-moc of your main thread's MOC. Let the main thread do all the saving, but your GCD threads can basically merge changes to the main MOC.
I've had pretty good success working with https://github.com/magicalpanda/MagicalRecord/ to facilitate this in a simpler fashion.
How can I kill a thread created by performSelectorInBackground:withObject: from the main thread? I need to force termination of freezing threads.
You cannot kill background threads from the main thread, the method that is executing in a background thread has to return for the thread to end.
Your actual problem seems to be that your background thread is freezing, you should solve that instead of trying to work around it.
I'm not sure if this may help but here goes:
Assuming you're calling that performSelector call from class A. And assuming that class A is about to be released from memory in class B (which is where if the selector hasn't been performed yet, you might be getting a crash - Hence you're posting this question on SO):
Wherever you're releasing A from B, do this:
[NSObject cancelPreviousPerformRequestsWithTarget:A];
Apple documentation says
The recommended way to exit a thread is to let it exit its entry point
routine normally. Although Cocoa, POSIX, and Multiprocessing Services
offer routines for killing threads directly, the use of such routines
is strongly discouraged. Killing a thread prevents that thread from
cleaning up after itself. Memory allocated by the thread could
potentially be leaked and any other resources currently in use by the
thread might not be cleaned up properly, creating potential problems
later.
I'm not very 'up' on multi-threading, but I've been using detachNewThreadSelector in a couple of places to access data in a database. Unfortunately, on slower devices the first request may still be happening when I call the second... Which naturally causes a crash because both are calling the same method (which is obviously not thread-safe).
What is the best way to go about sorting a bug like this? Can I somehow queue them so that the second thread doesn't start until the first one has finished?
Thanks!
You may want to have a look at NSOperation and NSOperationQueue which is an abstraction for a queue of tasks that can be run asynchronously from the main thread. If you want the NSOperationQueue to run just a NSOperation at a time (so to wait after the current task is compleate before firing the next one) you can just set the maxConcurrentOperationCount property of the queue to 1
To extend ranos answer a bit, be sure to add operations from the same thread because if you add several NSOperations from several threads, they will run concurrently, even though maxConcurrentOperationCount on NSOperationQueue is set to 1.