search for: rx_irq

...t only use one event channel to do tx / rx notification. This may cause unnecessary wake-up of process routines. This patch adds a new feature called feautre-split-event-channel to netback, enabling it to handle Tx and Rx event separately. Netback will use tx_irq to notify guest for tx completion, rx_irq for rx notification. If frontend doesn''t support this feature, tx_irq = rx_irq. Signed-off-by: Wei Liu <wei.liu2@citrix.com> --- drivers/net/xen-netback/common.h | 10 +++-- drivers/net/xen-netback/interface.c | 78 ++++++++++++++++++++++++++++------- drivers/net/xen-netbac...

...net/xen-netfront.c +++ b/drivers/net/xen-netfront.c @@ -85,7 +85,15 @@ struct netfront_info { struct napi_struct napi; - unsigned int evtchn; + /* Split event channels support, tx_* == rx_* when using + * single event channel. + */ + unsigned int tx_evtchn, rx_evtchn; + unsigned int tx_irq, rx_irq; + /* Only used when split event channels support is enabled */ + char tx_irq_name[IFNAMSIZ+4]; /* DEVNAME-tx */ + char rx_irq_name[IFNAMSIZ+4]; /* DEVNAME-rx */ + struct xenbus_device *xbdev; spinlock_t tx_lock; @@ -330,7 +338,7 @@ no_skb: push: RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&amp...

...affinity regardless of whether the >> optimization is used. > > > Yes, I noticed this in the past too. > >> Though this is not limited to napi-tx, it is more >> pronounced in that mode than without napi. >> >> 1x TCP_RR for affinity configuration {process, rx_irq, tx_irq}: >> >> upstream: >> >> 1,1,1: 28985 Mbps, 278 Gcyc >> 1,0,2: 30067 Mbps, 402 Gcyc >> >> napi tx: >> >> 1,1,1: 34492 Mbps, 269 Gcyc >> 1,0,2: 36527 Mbps, 537 Gcyc (!) >> 1,0,1: 36269 Mbps, 394 Gcyc >> 1,0,0: 34674 Mbps,...

...affinity regardless of whether the >> optimization is used. > > > Yes, I noticed this in the past too. > >> Though this is not limited to napi-tx, it is more >> pronounced in that mode than without napi. >> >> 1x TCP_RR for affinity configuration {process, rx_irq, tx_irq}: >> >> upstream: >> >> 1,1,1: 28985 Mbps, 278 Gcyc >> 1,0,2: 30067 Mbps, 402 Gcyc >> >> napi tx: >> >> 1,1,1: 34492 Mbps, 269 Gcyc >> 1,0,2: 36527 Mbps, 537 Gcyc (!) >> 1,0,1: 36269 Mbps, 394 Gcyc >> 1,0,0: 34674 Mbps,...

...s always a win over keeping it off, even without irq affinity. The cycle cost is significant without affinity regardless of whether the optimization is used. Though this is not limited to napi-tx, it is more pronounced in that mode than without napi. 1x TCP_RR for affinity configuration {process, rx_irq, tx_irq}: upstream: 1,1,1: 28985 Mbps, 278 Gcyc 1,0,2: 30067 Mbps, 402 Gcyc napi tx: 1,1,1: 34492 Mbps, 269 Gcyc 1,0,2: 36527 Mbps, 537 Gcyc (!) 1,0,1: 36269 Mbps, 394 Gcyc 1,0,0: 34674 Mbps, 402 Gcyc This is a particularly strong example. It is also representative of most RR tests. It is less...

...s always a win over keeping it off, even without irq affinity. The cycle cost is significant without affinity regardless of whether the optimization is used. Though this is not limited to napi-tx, it is more pronounced in that mode than without napi. 1x TCP_RR for affinity configuration {process, rx_irq, tx_irq}: upstream: 1,1,1: 28985 Mbps, 278 Gcyc 1,0,2: 30067 Mbps, 402 Gcyc napi tx: 1,1,1: 34492 Mbps, 269 Gcyc 1,0,2: 36527 Mbps, 537 Gcyc (!) 1,0,1: 36269 Mbps, 394 Gcyc 1,0,0: 34674 Mbps, 402 Gcyc This is a particularly strong example. It is also representative of most RR tests. It is less...

...gt; > >> > >> > Yes, I noticed this in the past too. >> > >> >> Though this is not limited to napi-tx, it is more >> >> pronounced in that mode than without napi. >> >> >> >> 1x TCP_RR for affinity configuration {process, rx_irq, tx_irq}: >> >> >> >> upstream: >> >> >> >> 1,1,1: 28985 Mbps, 278 Gcyc >> >> 1,0,2: 30067 Mbps, 402 Gcyc >> >> >> >> napi tx: >> >> >> >> 1,1,1: 34492 Mbps, 269 Gcyc >> >> 1,0,2:...

...gt; > >> > >> > Yes, I noticed this in the past too. >> > >> >> Though this is not limited to napi-tx, it is more >> >> pronounced in that mode than without napi. >> >> >> >> 1x TCP_RR for affinity configuration {process, rx_irq, tx_irq}: >> >> >> >> upstream: >> >> >> >> 1,1,1: 28985 Mbps, 278 Gcyc >> >> 1,0,2: 30067 Mbps, 402 Gcyc >> >> >> >> napi tx: >> >> >> >> 1,1,1: 34492 Mbps, 269 Gcyc >> >> 1,0,2:...

...The cycle cost is significant without affinity regardless of whether the > optimization is used. Yes, I noticed this in the past too. > Though this is not limited to napi-tx, it is more > pronounced in that mode than without napi. > > 1x TCP_RR for affinity configuration {process, rx_irq, tx_irq}: > > upstream: > > 1,1,1: 28985 Mbps, 278 Gcyc > 1,0,2: 30067 Mbps, 402 Gcyc > > napi tx: > > 1,1,1: 34492 Mbps, 269 Gcyc > 1,0,2: 36527 Mbps, 537 Gcyc (!) > 1,0,1: 36269 Mbps, 394 Gcyc > 1,0,0: 34674 Mbps, 402 Gcyc > > This is a particularly stron...

...> optimization is used. > > > > > > Yes, I noticed this in the past too. > > > >> Though this is not limited to napi-tx, it is more > >> pronounced in that mode than without napi. > >> > >> 1x TCP_RR for affinity configuration {process, rx_irq, tx_irq}: > >> > >> upstream: > >> > >> 1,1,1: 28985 Mbps, 278 Gcyc > >> 1,0,2: 30067 Mbps, 402 Gcyc > >> > >> napi tx: > >> > >> 1,1,1: 34492 Mbps, 269 Gcyc > >> 1,0,2: 36527 Mbps, 537 Gcyc (!) > >>...

...t;> > Yes, I noticed this in the past too. > >> > > >> >> Though this is not limited to napi-tx, it is more > >> >> pronounced in that mode than without napi. > >> >> > >> >> 1x TCP_RR for affinity configuration {process, rx_irq, tx_irq}: > >> >> > >> >> upstream: > >> >> > >> >> 1,1,1: 28985 Mbps, 278 Gcyc > >> >> 1,0,2: 30067 Mbps, 402 Gcyc > >> >> > >> >> napi tx: > >> >> > >> >> 1,1,1:...

From: Willem de Bruijn <willemb at google.com> Convert virtio-net to a standard napi tx completion path. This enables better TCP pacing using TCP small queues and increases single stream throughput. The virtio-net driver currently cleans tx descriptors on transmission of new packets in ndo_start_xmit. Latency depends on new traffic, so is unbounded. To avoid deadlock when a socket reaches

...ntel/ixgb/ixgb_main.c b/drivers/net/ethernet/intel/ixgb/ixgb_main.c index 0940a0da16f2..28465fc76dda 100644 --- a/drivers/net/ethernet/intel/ixgb/ixgb_main.c +++ b/drivers/net/ethernet/intel/ixgb/ixgb_main.c @@ -70,7 +70,7 @@ static int ixgb_clean(struct napi_struct *, int); static bool ixgb_clean_rx_irq(struct ixgb_adapter *, int *, int); static void ixgb_alloc_rx_buffers(struct ixgb_adapter *, int); -static void ixgb_tx_timeout(struct net_device *dev); +static void ixgb_tx_timeout(struct net_device *dev, int txqueue); static void ixgb_tx_timeout_task(struct work_struct *work); static void...

...rivers/net/ethernet/intel/ixgb/ixgb_main.c > index 0940a0da16f2..28465fc76dda 100644 > --- a/drivers/net/ethernet/intel/ixgb/ixgb_main.c > +++ b/drivers/net/ethernet/intel/ixgb/ixgb_main.c > @@ -70,7 +70,7 @@ static int ixgb_clean(struct napi_struct *, int); > static bool ixgb_clean_rx_irq(struct ixgb_adapter *, int *, int); > static void ixgb_alloc_rx_buffers(struct ixgb_adapter *, int); > > -static void ixgb_tx_timeout(struct net_device *dev); > +static void ixgb_tx_timeout(struct net_device *dev, int txqueue); > static void ixgb_tx_timeout_task(struct work_struct...

Hi Michael, Em sex., 22 de nov. de 2019 ?s 07:31, Michael S. Tsirkin <mst at redhat.com> escreveu: > > On Thu, Nov 21, 2019 at 10:36:36PM -0300, Julio Faracco wrote: > > Driver virtio_net is not handling error events for TX provided by > > dev_watchdog. This event is reached when transmission queue is having > > problems to transmit packets. This could happen for any

Hi Michael, Em sex., 22 de nov. de 2019 ?s 07:31, Michael S. Tsirkin <mst at redhat.com> escreveu: > > On Thu, Nov 21, 2019 at 10:36:36PM -0300, Julio Faracco wrote: > > Driver virtio_net is not handling error events for TX provided by > > dev_watchdog. This event is reached when transmission queue is having > > problems to transmit packets. This could happen for any

...igb/igb_main.c index ed7e667d7eb2..0eb8470050c2 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -146,7 +146,7 @@ static int igb_poll(struct napi_struct *, int); static bool igb_clean_tx_irq(struct igb_q_vector *, int); static int igb_clean_rx_irq(struct igb_q_vector *, int); static int igb_ioctl(struct net_device *, struct ifreq *, int cmd); -static void igb_tx_timeout(struct net_device *); +static void igb_tx_timeout(struct net_device *, int txqueue); static void igb_reset_task(struct work_struct *); static void igb_vlan_mode(struct net...

...igb/igb_main.c index ed7e667d7eb2..0eb8470050c2 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -146,7 +146,7 @@ static int igb_poll(struct napi_struct *, int); static bool igb_clean_tx_irq(struct igb_q_vector *, int); static int igb_clean_rx_irq(struct igb_q_vector *, int); static int igb_ioctl(struct net_device *, struct ifreq *, int cmd); -static void igb_tx_timeout(struct net_device *); +static void igb_tx_timeout(struct net_device *, int txqueue); static void igb_reset_task(struct work_struct *); static void igb_vlan_mode(struct net...

...igb/igb_main.c index ed7e667d7eb2..0eb8470050c2 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -146,7 +146,7 @@ static int igb_poll(struct napi_struct *, int); static bool igb_clean_tx_irq(struct igb_q_vector *, int); static int igb_clean_rx_irq(struct igb_q_vector *, int); static int igb_ioctl(struct net_device *, struct ifreq *, int cmd); -static void igb_tx_timeout(struct net_device *); +static void igb_tx_timeout(struct net_device *, int txqueue); static void igb_reset_task(struct work_struct *); static void igb_vlan_mode(struct net...

...igb/igb_main.c index ed7e667d7eb2..0eb8470050c2 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -146,7 +146,7 @@ static int igb_poll(struct napi_struct *, int); static bool igb_clean_tx_irq(struct igb_q_vector *, int); static int igb_clean_rx_irq(struct igb_q_vector *, int); static int igb_ioctl(struct net_device *, struct ifreq *, int cmd); -static void igb_tx_timeout(struct net_device *); +static void igb_tx_timeout(struct net_device *, unsigned int txqueue); static void igb_reset_task(struct work_struct *); static void igb_vlan_mode(s...