search for: atomic_set_bit

...write side of a seqlock only provides write barriers. Access to map = vq->maps[VHOST_ADDR_USED]; Still needs a read side barrier, and then I think this will be no better than a normal spinlock. It also doesn't seem like this algorithm even needs a seqlock, as this is just a one bit flag atomic_set_bit(using map) smp_mb__after_atomic() .. maps [...] atomic_clear_bit(using map) map = NULL; smp_mb__before_atomic(); while (atomic_read_bit(using map)) relax() Again, not clear this could be faster than a spinlock when the barriers are correct... Jason

...write side of a seqlock only provides write barriers. Access to map = vq->maps[VHOST_ADDR_USED]; Still needs a read side barrier, and then I think this will be no better than a normal spinlock. It also doesn't seem like this algorithm even needs a seqlock, as this is just a one bit flag atomic_set_bit(using map) smp_mb__after_atomic() .. maps [...] atomic_clear_bit(using map) map = NULL; smp_mb__before_atomic(); while (atomic_read_bit(using map)) relax() Again, not clear this could be faster than a spinlock when the barriers are correct... Jason

...) ===== tools/ioemu/target-i386-dm/helper2.c 1.2 vs edited ===== --- 1.2/tools/ioemu/target-i386-dm/helper2.c 2005-05-20 10:01:22 -07:00 +++ edited/tools/ioemu/target-i386-dm/helper2.c 2005-05-23 16:35:40 -07:00 @@ -313,7 +313,8 @@ intr = &(((vcpu_iodata_t *) shared_page)->vp_intr[0]); atomic_set_bit(vector, intr); - fprintf(logfile, "injecting vector: %x\n", vector); + if (loglevel & CPU_LOG_INT) + fprintf(logfile, "injecting vector: %x\n", vector); env->send_event = 1; } _______________________________________________ Xen-devel mai...

...d then I think this will be no >> better than a normal spinlock. >> >> It also doesn't seem like this algorithm even needs a seqlock, as this >> is just a one bit flag > > > Right, so then I tend to use spinlock first for correctness. > > >> >> atomic_set_bit(using map) >> smp_mb__after_atomic() >> .. maps [...] >> atomic_clear_bit(using map) >> >> >> map = NULL; >> smp_mb__before_atomic(); >> while (atomic_read_bit(using map)) >> ??? relax() >> >> Again, not clear this could be faster th...

...> > Still needs a read side barrier, and then I think this will be no > better than a normal spinlock. > > It also doesn't seem like this algorithm even needs a seqlock, as this > is just a one bit flag Right, so then I tend to use spinlock first for correctness. > > atomic_set_bit(using map) > smp_mb__after_atomic() > .. maps [...] > atomic_clear_bit(using map) > > > map = NULL; > smp_mb__before_atomic(); > while (atomic_read_bit(using map)) > relax() > > Again, not clear this could be faster than a spinlock when the > barriers are cor...

Hi all: This series try to fix several issues introduced by meta data accelreation series. Please review. Changes from V3: - remove the unnecessary patch Changes from V2: - use seqlck helper to synchronize MMU notifier with vhost worker Changes from V1: - try not use RCU to syncrhonize MMU notifier with vhost worker - set dirty pages after no readers - return -EAGAIN only when we find the

Hi all: This series try to fix several issues introduced by meta data accelreation series. Please review. Changes from V3: - remove the unnecessary patch Changes from V2: - use seqlck helper to synchronize MMU notifier with vhost worker Changes from V1: - try not use RCU to syncrhonize MMU notifier with vhost worker - set dirty pages after no readers - return -EAGAIN only when we find the