Messages in this thread | | | Date | Tue, 21 Apr 2009 13:22:23 +0400 | From | Evgeniy Polyakov <> | Subject | Re: [PATCH] netfilter: use per-cpu recursive lock (v11) |
| |
On Tue, Apr 21, 2009 at 10:55:59AM +0200, Eric Dumazet (dada1@cosmosbay.com) wrote: > Maybe just dont care about calling several time local_bh_disable() > (since we were doing this in previous kernels anyway, we used to call read_lock_bh()) > > This shortens fastpath, is faster than local_irq_save()/local_irq_restore(), > and looks better.
Yeah, given that non-nested locking is more likely condition, it will be even faster than preemption case.
> void xt_info_rdlock_bh(void) > { > struct xt_info_lock *lock; > > local_bh_disable(); > lock = &__get_cpu_var(xt_info_locks); > if (likely(++lock->depth == 0)) > spin_lock(&lock->lock); > } > > void xt_info_rdunlock_bh(void) > { > struct xt_info_lock *lock = &__get_cpu_var(xt_info_locks); > > BUG_ON(lock->depth < 0); > if (likely(--lock->depth < 0)) > spin_unlock(&lock->lock); > local_bh_enable(); > } > >
-- Evgeniy Polyakov
| |