Pierre Peiffer wrote: > But there can be a performance impact when several processes use > different futexes which have the same hash key. > In fact, the plist contains all waiters _of_all_futexes_ having the same > hash key, not only the waiters of a given futex. This can be more a > problem, s/can be/is/ There are systems with thousands of active futexes, maybe tens of thousands. Not only is hash collision likely, it's also a matter of using and administering the plist. We have to make futexes less connected, not more. Now I definitely want to see real world tests first. -- ➧ Ulrich Drepper ➧ Red Hat, Inc. ➧ 444 Castro St ➧ Mountain View, CA ❖
Attachment:
signature.asc
Description: OpenPGP digital signature
- References:
- [PATCH 2.6.20-rc4 1/4] futex priority based wakeup
- From: Pierre Peiffer <[email protected]>
- Re: [PATCH 2.6.20-rc4 1/4] futex priority based wakeup
- From: Ulrich Drepper <[email protected]>
- Re: [PATCH 2.6.20-rc4 1/4] futex priority based wakeup
- From: Pierre Peiffer <[email protected]>
- Re: [PATCH 2.6.20-rc4 1/4] futex priority based wakeup
- From: Jakub Jelinek <[email protected]>
- Re: [PATCH 2.6.20-rc4 1/4] futex priority based wakeup
- From: Pierre Peiffer <[email protected]>
- [PATCH 2.6.20-rc4 1/4] futex priority based wakeup
- Prev by Date: Re: Jumping into Kernel development: About -rc kernels...
- Next by Date: Re: SATA/IDE Dual Mode w/Intel 945 Chipset or HOW TO LIQUIFY a flash IDE chip under 2.6.18
- Previous by thread: Re: [PATCH 2.6.20-rc4 1/4] futex priority based wakeup
- Next by thread: Re: [PATCH 2.6.20-rc4 1/4] futex priority based wakeup
- Index(es):
 |