similar to: [PATCH net-next 0/8] vhost-net rx batching

On Tue, Mar 21, 2017 at 12:04:46PM +0800, Jason Wang wrote: > We used to dequeue one skb during recvmsg() from skb_array, this could > be inefficient because of the bad cache utilization and spinlock > touching for each packet. This patch tries to batch them by calling > batch dequeuing helpers explicitly on the exported skb array and pass > the skb back through msg_control for

On Tue, Mar 21, 2017 at 12:04:46PM +0800, Jason Wang wrote: > We used to dequeue one skb during recvmsg() from skb_array, this could > be inefficient because of the bad cache utilization and spinlock > touching for each packet. This patch tries to batch them by calling > batch dequeuing helpers explicitly on the exported skb array and pass > the skb back through msg_control for

On Tue, Mar 21, 2017 at 12:04:47PM +0800, Jason Wang wrote: > For the socket that exports its skb array, we can use lockless polling > to avoid touching spinlock during busy polling. > > Signed-off-by: Jason Wang <jasowang at redhat.com> > --- > drivers/vhost/net.c | 7 +++++-- > 1 file changed, 5 insertions(+), 2 deletions(-) > > diff --git

On Tue, Mar 21, 2017 at 12:04:47PM +0800, Jason Wang wrote: > For the socket that exports its skb array, we can use lockless polling > to avoid touching spinlock during busy polling. > > Signed-off-by: Jason Wang <jasowang at redhat.com> > --- > drivers/vhost/net.c | 7 +++++-- > 1 file changed, 5 insertions(+), 2 deletions(-) > > diff --git

On Thu, Mar 30, 2017 at 10:16:15AM +0800, Jason Wang wrote: > > > On 2017?03?29? 20:07, Michael S. Tsirkin wrote: > > On Tue, Mar 21, 2017 at 12:04:47PM +0800, Jason Wang wrote: > > > For the socket that exports its skb array, we can use lockless polling > > > to avoid touching spinlock during busy polling. > > > > > > Signed-off-by: Jason Wang

On Thu, Mar 30, 2017 at 10:16:15AM +0800, Jason Wang wrote: > > > On 2017?03?29? 20:07, Michael S. Tsirkin wrote: > > On Tue, Mar 21, 2017 at 12:04:47PM +0800, Jason Wang wrote: > > > For the socket that exports its skb array, we can use lockless polling > > > to avoid touching spinlock during busy polling. > > > > > > Signed-off-by: Jason Wang

Hi all: This series tries to switch to use skb array in tun. This is used to eliminate the spinlock contention between producer and consumer. The conversion was straightforward: just introdce a tx skb array and use it instead of sk_receive_queue. A minor issue is to keep the tx_queue_len behaviour, since tun used to use it for the length of sk_receive_queue. This is done through: - add the

Hi all: This series tries to switch to use skb array in tun. This is used to eliminate the spinlock contention between producer and consumer. The conversion was straightforward: just introdce a tx skb array and use it instead of sk_receive_queue. A minor issue is to keep the tx_queue_len behaviour, since tun used to use it for the length of sk_receive_queue. This is done through: - add the

Hi all: This series tries to switch to use skb array in tun. This is used to eliminate the spinlock contention between producer and consumer. The conversion was straightforward: just introdce a tx skb array and use it instead of sk_receive_queue. A minor issue is to keep the tx_queue_len behaviour, since tun used to use it for the length of sk_receive_queue. This is done through: - add the

Hi all: This series tries to switch to use skb array in tun. This is used to eliminate the spinlock contention between producer and consumer. The conversion was straightforward: just introdce a tx skb array and use it instead of sk_receive_queue. A minor issue is to keep the tx_queue_len behaviour, since tun used to use it for the length of sk_receive_queue. This is done through: - add the

We used to queue tx packets in sk_receive_queue, this is less efficient since it requires spinlocks to synchronize between producer and consumer. This patch tries to address this by: - introduce a new mode which will be only enabled with IFF_TX_ARRAY set and switch from sk_receive_queue to a fixed size of skb array with 256 entries in this mode. - introduce a new proto_ops peek_len which was

We used to queue tx packets in sk_receive_queue, this is less efficient since it requires spinlocks to synchronize between producer and consumer. This patch tries to address this by: - introduce a new mode which will be only enabled with IFF_TX_ARRAY set and switch from sk_receive_queue to a fixed size of skb array with 256 entries in this mode. - introduce a new proto_ops peek_len which was

On Tue, Mar 21, 2017 at 12:04:40PM +0800, Jason Wang wrote: > Signed-off-by: Jason Wang <jasowang at redhat.com> > --- > include/linux/ptr_ring.h | 65 ++++++++++++++++++++++++++++++++++++++++++++++++ > 1 file changed, 65 insertions(+) > > diff --git a/include/linux/ptr_ring.h b/include/linux/ptr_ring.h > index 6c70444..4771ded 100644 > ---

On Tue, Mar 21, 2017 at 12:04:40PM +0800, Jason Wang wrote: > Signed-off-by: Jason Wang <jasowang at redhat.com> > --- > include/linux/ptr_ring.h | 65 ++++++++++++++++++++++++++++++++++++++++++++++++ > 1 file changed, 65 insertions(+) > > diff --git a/include/linux/ptr_ring.h b/include/linux/ptr_ring.h > index 6c70444..4771ded 100644 > ---

On 2018?07?02? 20:57, xiangxia.m.yue at gmail.com wrote: > From: Tonghao Zhang <xiangxia.m.yue at gmail.com> > > Factor out generic busy polling logic and will be > used for in tx path in the next patch. And with the patch, > qemu can set differently the busyloop_timeout for rx queue. > > Signed-off-by: Tonghao Zhang <zhangtonghao at didichuxing.com> > --- >

On 2018?07?02? 20:57, xiangxia.m.yue at gmail.com wrote: > From: Tonghao Zhang <xiangxia.m.yue at gmail.com> > > Factor out generic busy polling logic and will be > used for in tx path in the next patch. And with the patch, > qemu can set differently the busyloop_timeout for rx queue. > > Signed-off-by: Tonghao Zhang <zhangtonghao at didichuxing.com> > --- >

From: Tonghao Zhang <xiangxia.m.yue at gmail.com> This patches improve the guest receive and transmit performance. On the handle_tx side, we poll the sock receive queue at the same time. handle_rx do that in the same way. For more performance report, see patch 4. v3 -> v4: fix some issues v2 -> v3: This patches are splited from previous big patch:

We used to dequeue one skb during recvmsg() from skb_array, this could be inefficient because of the bad cache utilization and spinlock touching for each packet. This patch tries to batch them by calling batch dequeuing helpers explicitly on the exported skb array and pass the skb back through msg_control for underlayer socket to finish the userspace copying. Tests were done by XDP1: - small

On 2018?06?30? 14:33, xiangxia.m.yue at gmail.com wrote: > From: Tonghao Zhang <xiangxia.m.yue at gmail.com> > > Factor out generic busy polling logic and will be > used for tx path in the next patch. And with the patch, > qemu can set differently the busyloop_timeout for rx queue. > > Signed-off-by: Tonghao Zhang <zhangtonghao at didichuxing.com> > --- >

From: Tonghao Zhang <xiangxia.m.yue at gmail.com> This patches improve the guest receive performance. On the handle_tx side, we poll the sock receive queue at the same time. handle_rx do that in the same way. For more performance report, see patch 4. v6->v7: fix issue and rebase codes: 1. on tx, busypoll will vhost_net_disable/enable_vq rx vq. [This is suggested by Toshiaki Makita