Messages in this thread |  | | | From | Rusty Russell <> | | Subject | Re: [PATCH] Fast Userspace Mutexes III. | | Date | Wed, 06 Mar 2002 12:31:01 +1100 |
| |
In message <20020305151439.457E03FE06@smtp.linux.ibm.com> you write: > I agree to put it there if its not used as a means to define whether > user locks are permitted or not. If that is the intention, then the current > futex will need to check every access through find_vma(), which we > both know nobody wants to do. > > So it can only be used for architectural hints, agreed ?
Yes. It *might* work if you don't PROT_SEM the page the semaphore is on, but it's still a bug waiting to happen. OTOH, it'd be nice if PROT_SEM returns EINVAL was a reliably indicator of no futex support. This way you actually need to call the futex syscall once to see if it works.
Cheers! Rusty. -- Anyone who quotes me in their sig is an idiot. -- Rusty Russell. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
|  |