search for: napi_complete_done

Since commit 39e6c8208d7b ("net: solve a NAPI race") napi has been able to be rescheduled within napi_complete_done() even in non-busypoll case, but virtnet_poll() always enabled interrupts before complete, and when napi was rescheduled within napi_complete_done() it did not disable interrupts. This caused more interrupts when event idx is disabled. According to commit cbdadbbf0c79 ("virtio_net: fix race i...

Since commit 39e6c8208d7b ("net: solve a NAPI race") napi has been able to be rescheduled within napi_complete_done() even in non-busypoll case, but virtnet_poll() always enabled interrupts before complete, and when napi was rescheduled within napi_complete_done() it did not disable interrupts. This caused more interrupts when event idx is disabled. According to commit cbdadbbf0c79 ("virtio_net: fix race i...

On 2017?12?07? 13:09, Michael S. Tsirkin wrote: > On Thu, Dec 07, 2017 at 01:15:15PM +0900, Toshiaki Makita wrote: >> Since commit 39e6c8208d7b ("net: solve a NAPI race") napi has been able >> to be rescheduled within napi_complete_done() even in non-busypoll case, >> but virtnet_poll() always enabled interrupts before complete, and when >> napi was rescheduled within napi_complete_done() it did not disable >> interrupts. >> This caused more interrupts when event idx is disabled. >> >> According...

On 2017?12?07? 13:09, Michael S. Tsirkin wrote: > On Thu, Dec 07, 2017 at 01:15:15PM +0900, Toshiaki Makita wrote: >> Since commit 39e6c8208d7b ("net: solve a NAPI race") napi has been able >> to be rescheduled within napi_complete_done() even in non-busypoll case, >> but virtnet_poll() always enabled interrupts before complete, and when >> napi was rescheduled within napi_complete_done() it did not disable >> interrupts. >> This caused more interrupts when event idx is disabled. >> >> According...

On Thu, Dec 07, 2017 at 01:15:15PM +0900, Toshiaki Makita wrote: > Since commit 39e6c8208d7b ("net: solve a NAPI race") napi has been able > to be rescheduled within napi_complete_done() even in non-busypoll case, > but virtnet_poll() always enabled interrupts before complete, and when > napi was rescheduled within napi_complete_done() it did not disable > interrupts. > This caused more interrupts when event idx is disabled. > > According to commit cbdadbbf0c79...

From: Eric Dumazet <edumazet at google.com> Straightforward patch to add GRO processing to virtio_net. napi_complete_done() usage allows more aggressive aggregation, opted-in by setting /sys/class/net/xxx/gro_flush_timeout Tested: Setting /sys/class/net/xxx/gro_flush_timeout to 1000 nsec, Rick Jones reported following results. One VM of each on a pair of OpenStack compute nodes with E5-2650Lv3 CPUs and Intel 82599E...

From: Eric Dumazet <edumazet at google.com> Straightforward patch to add GRO processing to virtio_net. napi_complete_done() usage allows more aggressive aggregation, opted-in by setting /sys/class/net/xxx/gro_flush_timeout Tested: Setting /sys/class/net/xxx/gro_flush_timeout to 1000 nsec, Rick Jones reported following results. One VM of each on a pair of OpenStack compute nodes with E5-2650Lv3 CPUs and Intel 82599E...

On Fri, Jul 31, 2015 at 06:25:17PM +0200, Eric Dumazet wrote: > From: Eric Dumazet <edumazet at google.com> > > Straightforward patch to add GRO processing to virtio_net. > > napi_complete_done() usage allows more aggressive aggregation, > opted-in by setting /sys/class/net/xxx/gro_flush_timeout > > Tested: > > Setting /sys/class/net/xxx/gro_flush_timeout to 1000 nsec, > Rick Jones reported following results. > > One VM of each on a pair of OpenStack compute nod...

...[ cut here ]------------ WARNING: CPU: 0 PID: 0 at net/ipv4/af_inet.c:1473 inet_gro_complete+0xbb/0xd0 Call Trace: <IRQ> dump_stack+0x63/0x8e __warn+0xd1/0xf0 warn_slowpath_null+0x1d/0x20 inet_gro_complete+0xbb/0xd0 napi_gro_complete+0x73/0xa0 napi_gro_flush+0x5f/0x80 napi_complete_done+0x6a/0xb0 igb_poll+0x38d/0x720 [igb] ? igb_msix_ring+0x2e/0x40 [igb] ? __handle_irq_event_percpu+0x4b/0x1a0 net_rx_action+0x158/0x360 __do_softirq+0xd1/0x283 irq_exit+0xe9/0x100 xen_evtchn_do_upcall+0x35/0x50 xen_do_hypervisor_callback+0x1e/0x40 <EOI> ? xen_hyper...

...[ 46.166817] ? print_irqtrace_events+0x280/0x280 [ 46.166830] ? rcu_process_callbacks+0xc5e/0x12d0 [ 46.166838] ? kvm_clock_read+0x1f/0x30 [ 46.166857] ? mark_held_locks+0xd5/0x170 [ 46.166867] ? net_rx_action+0x2aa/0x10e0 [ 46.166882] net_rx_action+0x4bc/0x10e0 [ 46.166906] ? napi_complete_done+0x480/0x480 [ 46.166925] ? print_irqtrace_events+0x280/0x280 [ 46.166935] ? sched_clock+0x5/0x10 [ 46.166952] ? __lock_is_held+0xcb/0x1a0 [ 46.166982] __do_softirq+0x2c4/0xdf4 [ 46.167010] do_softirq_own_stack+0x2a/0x40 [ 46.167031] </IRQ> [ 46.167038] ? virtnet_napi_enab...

...[ 46.166817] ? print_irqtrace_events+0x280/0x280 [ 46.166830] ? rcu_process_callbacks+0xc5e/0x12d0 [ 46.166838] ? kvm_clock_read+0x1f/0x30 [ 46.166857] ? mark_held_locks+0xd5/0x170 [ 46.166867] ? net_rx_action+0x2aa/0x10e0 [ 46.166882] net_rx_action+0x4bc/0x10e0 [ 46.166906] ? napi_complete_done+0x480/0x480 [ 46.166925] ? print_irqtrace_events+0x280/0x280 [ 46.166935] ? sched_clock+0x5/0x10 [ 46.166952] ? __lock_is_held+0xcb/0x1a0 [ 46.166982] __do_softirq+0x2c4/0xdf4 [ 46.167010] do_softirq_own_stack+0x2a/0x40 [ 46.167031] </IRQ> [ 46.167038] ? virtnet_napi_enab...

With the implementation of napi-tx in virtio driver, we clean tx descriptors from rx napi handler, for the purpose of reducing tx complete interrupts. But this introduces a race where tx complete interrupt has been raised, but the handler finds there is no work to do because we have done the work in the previous rx interrupt handler. A similar issue exists with polling from start_xmit, it is

With the implementation of napi-tx in virtio driver, we clean tx descriptors from rx napi handler, for the purpose of reducing tx complete interrupts. But this introduces a race where tx complete interrupt has been raised, but the handler finds there is no work to do because we have done the work in the previous rx interrupt handler. A similar issue exists with polling from start_xmit, it is

...ace_events+0x280/0x280 > [?? 46.166830]? ? rcu_process_callbacks+0xc5e/0x12d0 > [?? 46.166838]? ? kvm_clock_read+0x1f/0x30 > [?? 46.166857]? ? mark_held_locks+0xd5/0x170 > [?? 46.166867]? ? net_rx_action+0x2aa/0x10e0 > [?? 46.166882]? net_rx_action+0x4bc/0x10e0 > [?? 46.166906]? ? napi_complete_done+0x480/0x480 > [?? 46.166925]? ? print_irqtrace_events+0x280/0x280 > [?? 46.166935]? ? sched_clock+0x5/0x10 > [?? 46.166952]? ? __lock_is_held+0xcb/0x1a0 > [?? 46.166982]? __do_softirq+0x2c4/0xdf4 > [?? 46.167010]? do_softirq_own_stack+0x2a/0x40 > [?? 46.167031]? </IRQ> >...

On Mon, Jul 23, 2018 at 11:36:03PM +0900, Toshiaki Makita wrote: > From: Toshiaki Makita <makita.toshiaki at lab.ntt.co.jp> > > Add some ethtool stat items useful for performance analysis. > > Signed-off-by: Toshiaki Makita <makita.toshiaki at lab.ntt.co.jp> Series: Acked-by: Michael S. Tsirkin <mst at redhat.com> Patch 1 seems appropriate for stable, even

On Mon, Jul 23, 2018 at 11:36:03PM +0900, Toshiaki Makita wrote: > From: Toshiaki Makita <makita.toshiaki at lab.ntt.co.jp> > > Add some ethtool stat items useful for performance analysis. > > Signed-off-by: Toshiaki Makita <makita.toshiaki at lab.ntt.co.jp> Series: Acked-by: Michael S. Tsirkin <mst at redhat.com> Patch 1 seems appropriate for stable, even

...5920/0x5920 > [ 7.275689] ? do_raw_spin_unlock+0x54/0x220 > [ 7.275699] ? find_held_lock+0x32/0x1c0 > [ 7.275710] ? rcu_process_callbacks+0xa60/0xd20 > [ 7.275736] net_rx_action+0x2ee/0xad0 > [ 7.275748] ? rcu_note_context_switch+0x320/0x320 > [ 7.275754] ? napi_complete_done+0x300/0x300 > [ 7.275763] ? native_apic_msr_write+0x27/0x30 > [ 7.275768] ? lapic_next_event+0x5b/0x90 > [ 7.275775] ? clockevents_program_event+0x21d/0x2f0 > [ 7.275791] __do_softirq+0x19a/0x623 > [ 7.275807] do_softirq_own_stack+0x2a/0x40 > [ 7.275811] &l...

From: Willem de Bruijn <willemb at google.com> Add napi for virtio-net transmit completion processing. Based on previous patchsets by Jason Wang: [RFC V7 PATCH 0/7] enable tx interrupts for virtio-net http://lkml.iu.edu/hypermail/linux/kernel/1505.3/00245.html Changes: RFC -> v1: - dropped vhost interrupt moderation patch: not needed and likely expensive at light

From: Willem de Bruijn <willemb at google.com> Add napi for virtio-net transmit completion processing. Based on previous patchsets by Jason Wang: [RFC V7 PATCH 0/7] enable tx interrupts for virtio-net http://lkml.iu.edu/hypermail/linux/kernel/1505.3/00245.html Changes: RFC -> v1: - dropped vhost interrupt moderation patch: not needed and likely expensive at light

From: Willem de Bruijn <willemb at google.com> Add napi for virtio-net transmit completion processing. Changes: v1 -> v2: - disable by default - disable unless affinity_hint_set because cache misses add up to a third higher cycle cost, e.g., in TCP_RR tests. This is not limited to the patch that enables tx completion cleaning in rx napi. - use trylock to