Re: [patch] user-vm-unlock-2.5.31-A2

[Ingo Molnar]

On Thu, 15 Aug 2002, Linus Torvalds wrote:

> The problem spot is the _fork_ from process X. Which gives a address in
> process' _X_ virtual address space - used for SETTID.
> 
> See? Process _X_ is not threaded, and is not maintaining any thread data
> structures.

okay, this is the misunderstanding then. If it fork()s and then uses some
threading (which uses clone()) then in all cases i know about it must be
linked against some threading library. Otherwise Y couldnt do a clone()  
call and expect threading to work. In theory Y could 'become' a
threading-capable process, but right now no threading library i'm aware of
allows this - lots of standard C calls must be threading-aware and
threading-safe. So right now 'threading' is a property that comes with the
process image at exec()  time. But this must not be so from a conceptual
angle, so i agree with you.

(but the question is mostly academic anyway, because it makes perfect
sense to use a pure SETTID for a completely unthreaded application, to get
the fork() PID return value in the child's context as well.)

	Ingo