"Fossies" - the Fresh Open Source Software Archive

Member "pytorch-1.8.2/docs/source/multiprocessing.rst" (23 Jul 2021, 5806 Bytes) of package /linux/misc/pytorch-1.8.2.tar.gz:

As a special service "Fossies" has tried to format the requested source page into HTML format (assuming markdown format). Alternatively you can here view or download the uninterpreted source code file. A member file download can also be achieved by clicking within a package contents listing on the according byte size field.


Multiprocessing package - torch.multiprocessing




If the main process exits abruptly (e.g. because of an incoming signal), Python's multiprocessing sometimes fails to clean up its children. It's a known caveat, so if you're seeing any resource leaks after interrupting the interpreter, it probably means that this has just happened to you.

Strategy management




Sharing CUDA tensors

Sharing CUDA tensors between processes is supported only in Python 3, using a spawn or forkserver start methods.

Unlike CPU tensors, the sending process is required to keep the original tensor as long as the receiving process retains a copy of the tensor. The refcounting is implemented under the hood but requires users to follow the next best practices.


If the consumer process dies abnormally to a fatal signal, the shared tensor could be forever kept in memory as long as the sending process is running.

  1. Release memory ASAP in the consumer.
## Good
x = queue.get()
# do somethings with x
del x
## Bad
x = queue.get()
# do somethings with x
# do everything else (producer have to keep x in memory)

2. Keep producer process running until all consumers exits. This will prevent the situation when the producer process releasing memory which is still in use by the consumer.

## producer
# send tensors, do something
## consumer
# receive tensors and use them
  1. Don't pass received tensors.
# not going to work
x = queue.get()
# you need to create a process-local copy
x = queue.get()
x_clone = x.clone()
# putting and getting from the same queue in the same process will likely end up with segfault
x = queue.get()

Sharing strategies

This section provides a brief overview into how different sharing strategies work. Note that it applies only to CPU tensor - CUDA tensors will always use the CUDA API, as that's the only way they can be shared.

File descriptor - file_descriptor


This is the default strategy (except for macOS and OS X where it's not supported).

This strategy will use file descriptors as shared memory handles. Whenever a storage is moved to shared memory, a file descriptor obtained from shm_open is cached with the object, and when it's going to be sent to other processes, the file descriptor will be transferred (e.g. via UNIX sockets) to it. The receiver will also cache the file descriptor and mmap it, to obtain a shared view onto the storage data.

Note that if there will be a lot of tensors shared, this strategy will keep a large number of file descriptors open most of the time. If your system has low limits for the number of open file descriptors, and you can't raise them, you should use the file_system strategy.

File system - file_system

This strategy will use file names given to shm_open to identify the shared memory regions. This has a benefit of not requiring the implementation to cache the file descriptors obtained from it, but at the same time is prone to shared memory leaks. The file can't be deleted right after its creation, because other processes need to access it to open their views. If the processes fatally crash, or are killed, and don't call the storage destructors, the files will remain in the system. This is very serious, because they keep using up the memory until the system is restarted, or they're freed manually.

To counter the problem of shared memory file leaks, torch.multiprocessing will spawn a daemon named torch_shm_manager that will isolate itself from the current process group, and will keep track of all shared memory allocations. Once all processes connected to it exit, it will wait a moment to ensure there will be no new connections, and will iterate over all shared memory files allocated by the group. If it finds that any of them still exist, they will be deallocated. We've tested this method and it proved to be robust to various failures. Still, if your system has high enough limits, and file_descriptor is a supported strategy, we do not recommend switching to this one.

Spawning subprocesses


Available for Python >= 3.4.

This depends on the spawn start method in Python's multiprocessing package.

Spawning a number of subprocesses to perform some function can be done by creating Process instances and calling join to wait for their completion. This approach works fine when dealing with a single subprocess but presents potential issues when dealing with multiple processes.

Namely, joining processes sequentially implies they will terminate sequentially. If they don't, and the first process does not terminate, the process termination will go unnoticed. Also, there are no native facilities for error propagation.

The spawn function below addresses these concerns and takes care of error propagation, out of order termination, and will actively terminate processes upon detecting an error in one of them.


Returned by ~spawn when called with join=False.