Subject: Re: Pthreads, libc, and the future
To: Michael Richardson <mcr@sandelman.ottawa.on.ca>
From: Michael Graff <explorer@flame.org>
List: tech-userlevel
Date: 12/13/1999 14:19:05
Michael Richardson <mcr@sandelman.ottawa.on.ca> writes:

>   No, I don't agree.  Jason is right here.
> 
>   The lock can exist in libc, but until pthreads registers itself with libc
> the locking primitives are no-ops.

The lock primitive (test and set) can exist in libc, but the rest of
the locking (queues, priority, etc) really is too thread-system
specific to be in libc.

Now, what we _are_ doing (right now in fact, and before I started
this) is to have locks with function calls that map to do-nothing
functions until the pthread_mutex_lock() or the like is present.

I have a hard time believing that we can make a generic lock in libc,
however.

--Michael