I guess the question says it all.
I want to fork on windows. What is the most similar operation and how do I use it.
Cygwin has fully featured fork() on Windows. Thus if using Cygwin is acceptable for you, then the problem is solved in the case performance is not an issue.
Otherwise you can take a look at how Cygwin implements fork(). From a quite old Cygwin's architecture doc:
5.6. Process Creation The fork call in Cygwin is particularly interesting because it does not map well on top of the Win32 API. This makes it very difficult to implement correctly. Currently, the Cygwin fork is a non-copy-on-write implementation similar to what was present in early flavors of UNIX.
The first thing that happens when a parent process forks a child process is that the parent initializes a space in the Cygwin process table for the child. It then creates a suspended child process using the Win32 CreateProcess call. Next, the parent process calls setjmp to save its own context and sets a pointer to this in a Cygwin shared memory area (shared among all Cygwin tasks). It then fills in the child's .data and .bss sections by copying from its own address space into the suspended child's address space. After the child's address space is initialized, the child is run while the parent waits on a mutex. The child discovers it has been forked and longjumps using the saved jump buffer. The child then sets the mutex the parent is waiting on and blocks on another mutex. This is the signal for the parent to copy its stack and heap into the child, after which it releases the mutex the child is waiting on and returns from the fork call. Finally, the child wakes from blocking on the last mutex, recreates any memory-mapped areas passed to it via the shared area, and returns from fork itself.
While we have some ideas as to how to speed up our fork implementation by reducing the number of context switches between the parent and child process, fork will almost certainly always be inefficient under Win32. Fortunately, in most circumstances the spawn family of calls provided by Cygwin can be substituted for a fork/exec pair with only a little effort. These calls map cleanly on top of the Win32 API. As a result, they are much more efficient. Changing the compiler's driver program to call spawn instead of fork was a trivial change and increased compilation speeds by twenty to thirty percent in our tests.
However, spawn and exec present their own set of difficulties. Because there is no way to do an actual exec under Win32, Cygwin has to invent its own Process IDs (PIDs). As a result, when a process performs multiple exec calls, there will be multiple Windows PIDs associated with a single Cygwin PID. In some cases, stubs of each of these Win32 processes may linger, waiting for their exec'd Cygwin process to exit.
Sounds like a lot of work, doesn't it? And yes, it is slooooow.
EDIT: the doc is outdated, please see this excellent answer for an update
I certainly don't know the details on this because I've never done it it, but the native NT API has a capability to fork a process (the POSIX subsystem on Windows needs this capability - I'm not sure if the POSIX subsystem is even supported anymore).
A search for ZwCreateProcess() should get you some more details - for example this bit of information from Maxim Shatskih:
The most important parameter here is SectionHandle. If this parameter is NULL, the kernel will fork the current process. Otherwise, this parameter must be a handle of the SEC_IMAGE section object created on the EXE file before calling ZwCreateProcess().
Iker Arizmendi wrote:
> Because the Cygwin project relied solely on Win32 APIs its fork > implementation is non-COW and inefficient in those cases where a fork > is not followed by exec. It's also rather complex. See here (section > 5.6) for details: > > http://www.redhat.com/support/wpapers/cygnus/cygnus_cygwin/architecture.html
This document is rather old, 10 years or so. While we're still using Win32 calls to emulate fork, the method has changed noticably. Especially, we don't create the child process in the suspended state anymore, unless specific datastructes need a special handling in the parent before they get copied to the child. In the current 1.5.25 release the only case for a suspended child are open sockets in the parent. The upcoming 1.7.0 release will not suspend at all.
One reason not to use ZwCreateProcess was that up to the 1.5.25 release we're still supporting Windows 9x users. However, two attempts to use ZwCreateProcess on NT-based systems failed for one reason or another.
It would be really nice if this stuff would be better or at all documented, especially a couple of datastructures and how to connect a process to a subsystem. While fork is not a Win32 concept, I don't see that it would be a bad thing to make fork easier to implement.