On Tue, 23 Mar 1999, Alan Cox wrote:

> > Also, isn't it overly restrictive to true SMP efforts to manage all
> > semaphore wait queues under the control of a single, global lock?
> 
> They don't take much time to work out, but yes it needs to be using smaller
> locks in 2.3.

i already have something for this, i have implemented quite finegrained
locking for sem.c. Will post it once it's ready. (it's not all that
trivial btw., the semaphore code is not structured in a way that can
easily be threaded)

-- mingo

-
Linux SMP list: FIRST see FAQ at http://www.irisa.fr/prive/mentre/smp-faq/
To Unsubscribe: send "unsubscribe linux-smp" to [EMAIL PROTECTED]

Reply via email to