Messages in this thread | | | Date | Tue, 28 Mar 2000 20:36:11 -0800 (PST) | From | Jeremy Fitzhardinge <> | Subject | Re: Slow pthread_create() under high load |
| |
On 29-Mar-00 Mike Castle wrote: > On Tue, Mar 28, 2000 at 03:42:30PM -0800, Jeremy Fitzhardinge wrote: >> Yes, but it needn't be as complex as putting locking everywhere. In >> the >> case where there was no shared credential change, the fast path is >> still >> pretty fast. > > Don't you still need to lock/make use of credential/unlock? Otherwise > you > have a race condition on SMP boxes.
If you use a generation counter, you wouldn't even need to take the lock unless the creds had actually changed; you'd only get contention if someone were actually updating the credentials at that instant.
J
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |