Re: [patch] kwaitd, 2.5.31-A1

[Ingo Molnar]

On Tue, 13 Aug 2002, Andrew Morton wrote:

> Well what we can do in there is to just have a one-deep percpu list. So
> the exitting task frees the previous thread's stack (if any) and inserts
> its own stack.  And that stack can be used in fork, of course.  Which
> gives some per-cpu LIFO stack allocation.

yeah, this will work - and it's not a big chunk of RAM we are holding
onto, so we can keep it around indefinitely. I'll try it this way.

(the LIFO argument is not true once the proper per-CPU page caching
patches are integrated.)

(btw., the kernel has the same catch-22 problem as user-space's problem
with stack deallocation, with the difference that it's much easier and
cheaper to provide atomicity in kernel-space.)

	Ingo