search for: proto_ops

...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 used for peeking the skb length. - implement a tun version of peek_len for vhost_net to use and convert vhost_net to use peek_len if possible. Pktgen test shows about 18% improvement on guest receiving pps for small buffers: Before: ~1220000pps After : ~1440000pps The reas...

...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 used for peeking the skb length. - implement a tun version of peek_len for vhost_net to use and convert vhost_net to use peek_len if possible. Pktgen test shows about 18% improvement on guest receiving pps for small buffers: Before: ~1220000pps After : ~1440000pps The reas...

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: - switch from sk_receive_queue to a skb_array, and resize it when tx_queue_len was changed. - introduce a new proto_ops peek_len which was used for peeking the skb length. - implement a tun version of peek_len for vhost_net to use and convert vhost_net to use peek_len if possible. Pktgen test shows about 15.3% improvement on guest receiving pps for small buffers: Before: ~1300000pps After : ~1500000pps Signed...

...gmail.com> --- net/vmw_vsock/af_vsock.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 8a398b3..c73b03a 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -1101,10 +1101,19 @@ static const struct proto_ops vsock_dgram_ops = { .sendpage = sock_no_sendpage, }; +static int vsock_transport_cancel_pkt(struct vsock_sock *vsk) +{ + if (!transport->cancel_pkt) + return -EOPNOTSUPP; + + return transport->cancel_pkt(vsk); +} + static void vsock_connect_timeout(struct work_struct *work) { struct...

...gmail.com> --- net/vmw_vsock/af_vsock.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 8a398b3..c73b03a 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -1101,10 +1101,19 @@ static const struct proto_ops vsock_dgram_ops = { .sendpage = sock_no_sendpage, }; +static int vsock_transport_cancel_pkt(struct vsock_sock *vsk) +{ + if (!transport->cancel_pkt) + return -EOPNOTSUPP; + + return transport->cancel_pkt(vsk); +} + static void vsock_connect_timeout(struct work_struct *work) { struct...

...nplug of vsock virtio transport device in a proper way. Maybe move the vsock_core_init()/vsock_core_exit() functions in the module_init and module_exit of vsock_virtio_transport module can't be the best way, but the architecture of vsock_core forces us to this approach for now. The vsock_core proto_ops expect a valid pointer to the transport device, so we can't call vsock_core_exit() until there are open sockets. v2 -> v3: - Rebased on master v1 -> v2: - Fixed commit message of patch 1. - Added Reviewed-by, Acked-by tags by Stefan Stefano Garzarella (2): vsock/virtio: fix kernel...

...nplug of vsock virtio transport device in a proper way. Maybe move the vsock_core_init()/vsock_core_exit() functions in the module_init and module_exit of vsock_virtio_transport module can't be the best way, but the architecture of vsock_core forces us to this approach for now. The vsock_core proto_ops expect a valid pointer to the transport device, so we can't call vsock_core_exit() until there are open sockets. v2 -> v3: - Rebased on master v1 -> v2: - Fixed commit message of patch 1. - Added Reviewed-by, Acked-by tags by Stefan Stefano Garzarella (2): vsock/virtio: fix kernel...

...ze 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 used for peeking the > skb length. > - implement a tun version of peek_len for vhost_net to use and convert > vhost_net to use peek_len if possible. > > Pktgen test shows about 18% improvement on guest receiving pps for small > buffers: > > Before: ~...

...file changed, 14 insertions(+), 6 deletions(-) > > diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c > index 880cc09..75febd4 100644 > --- a/drivers/net/macvtap.c > +++ b/drivers/net/macvtap.c > @@ -66,7 +66,7 @@ static struct cdev macvtap_cdev; > static const struct proto_ops macvtap_socket_ops; > > #define TUN_OFFLOADS (NETIF_F_HW_CSUM | NETIF_F_TSO_ECN | NETIF_F_TSO | \ > - NETIF_F_TSO6) > + NETIF_F_TSO6 | NETIF_F_UFO) > #define RX_OFFLOADS (NETIF_F_GRO | NETIF_F_LRO) > #define TAP_FEATURES (NETIF_F_GSO | NETIF_F_SG) > > @@...

...file changed, 14 insertions(+), 6 deletions(-) > > diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c > index 880cc09..75febd4 100644 > --- a/drivers/net/macvtap.c > +++ b/drivers/net/macvtap.c > @@ -66,7 +66,7 @@ static struct cdev macvtap_cdev; > static const struct proto_ops macvtap_socket_ops; > > #define TUN_OFFLOADS (NETIF_F_HW_CSUM | NETIF_F_TSO_ECN | NETIF_F_TSO | \ > - NETIF_F_TSO6) > + NETIF_F_TSO6 | NETIF_F_UFO) > #define RX_OFFLOADS (NETIF_F_GRO | NETIF_F_LRO) > #define TAP_FEATURES (NETIF_F_GSO | NETIF_F_SG) > > @@...

...gt; >> diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c > >> index 880cc09..75febd4 100644 > >> --- a/drivers/net/macvtap.c > >> +++ b/drivers/net/macvtap.c > >> @@ -66,7 +66,7 @@ static struct cdev macvtap_cdev; > >> static const struct proto_ops macvtap_socket_ops; > >> > >> #define TUN_OFFLOADS (NETIF_F_HW_CSUM | NETIF_F_TSO_ECN | NETIF_F_TSO | \ > >> - NETIF_F_TSO6) > >> + NETIF_F_TSO6 | NETIF_F_UFO) > >> #define RX_OFFLOADS (NETIF_F_GRO | NETIF_F_LRO) > >> #define...

...gt; >> diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c > >> index 880cc09..75febd4 100644 > >> --- a/drivers/net/macvtap.c > >> +++ b/drivers/net/macvtap.c > >> @@ -66,7 +66,7 @@ static struct cdev macvtap_cdev; > >> static const struct proto_ops macvtap_socket_ops; > >> > >> #define TUN_OFFLOADS (NETIF_F_HW_CSUM | NETIF_F_TSO_ECN | NETIF_F_TSO | \ > >> - NETIF_F_TSO6) > >> + NETIF_F_TSO6 | NETIF_F_UFO) > >> #define RX_OFFLOADS (NETIF_F_GRO | NETIF_F_LRO) > >> #define...

On Tue, Nov 11, 2014 at 12:17:26PM +0000, Ben Hutchings wrote: > On Tue, 2014-11-11 at 10:58 +0000, Stefan Hajnoczi wrote: > > Commit 3d0ad09412ffe00c9afa201d01effdb6023d09b4 ("drivers/net: Disable > > UFO through virtio") breaks live migration of KVM guests from older to > > newer host kernels: > > > >

On Tue, Nov 11, 2014 at 12:17:26PM +0000, Ben Hutchings wrote: > On Tue, 2014-11-11 at 10:58 +0000, Stefan Hajnoczi wrote: > > Commit 3d0ad09412ffe00c9afa201d01effdb6023d09b4 ("drivers/net: Disable > > UFO through virtio") breaks live migration of KVM guests from older to > > newer host kernels: > > > >

...ruct socket *sock, struct msghdr *msg, size_t len, int flags) { - return transport->dgram_dequeue(vsock_sk(sock->sk), msg, len, flags); + struct vsock_sock *vsk = vsock_sk(sock->sk); + + return vsk->transport->dgram_dequeue(vsk, msg, len, flags); } static const struct proto_ops vsock_dgram_ops = { @@ -1071,6 +1079,8 @@ static const struct proto_ops vsock_dgram_ops = { static int vsock_transport_cancel_pkt(struct vsock_sock *vsk) { + const struct vsock_transport *transport = vsk->transport; + if (!transport->cancel_pkt) return -EOPNOTSUPP; @@ -1107,6 +1117...

...ruct socket *sock, struct msghdr *msg, size_t len, int flags) { - return transport->dgram_dequeue(vsock_sk(sock->sk), msg, len, flags); + struct vsock_sock *vsk = vsock_sk(sock->sk); + + return vsk->transport->dgram_dequeue(vsk, msg, len, flags); } static const struct proto_ops vsock_dgram_ops = { @@ -1079,6 +1087,8 @@ static const struct proto_ops vsock_dgram_ops = { static int vsock_transport_cancel_pkt(struct vsock_sock *vsk) { + const struct vsock_transport *transport = vsk->transport; + if (!transport->cancel_pkt) return -EOPNOTSUPP; @@ -1115,6 +1125...

...e. If we will use the virtio-net datapath, it will become simpler. Yes, unix domain socket is one example that uses general skb and socket structure. And we probably need some kind of socket pair on host. Using socket can also simplify the unification with vhost-net which depends on the socket proto_ops to work. I admit it's a huge change probably, we can do it gradually. >> >>> Signed-off-by: Stefano Garzarella <sgarzare at redhat.com> >>> --- >>> drivers/vhost/vsock.c | 2 + >>> include/linux/virtio_vsock.h...

...e. If we will use the virtio-net datapath, it will become simpler. Yes, unix domain socket is one example that uses general skb and socket structure. And we probably need some kind of socket pair on host. Using socket can also simplify the unification with vhost-net which depends on the socket proto_ops to work. I admit it's a huge change probably, we can do it gradually. >> >>> Signed-off-by: Stefano Garzarella <sgarzare at redhat.com> >>> --- >>> drivers/vhost/vsock.c | 2 + >>> include/linux/virtio_vsock.h...

...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 used for peeking the > skb length. > - implement a tun version of peek_len for vhost_net to use and convert > vhost_net to use peek_len if possible. > > Pktgen test shows about 18% improvement on guest receiving pps for small > buffers: > > Before:...

...ice in > a proper way. > > Maybe move the vsock_core_init()/vsock_core_exit() functions in the module_init > and module_exit of vsock_virtio_transport module can't be the best way, but the > architecture of vsock_core forces us to this approach for now. > > The vsock_core proto_ops expect a valid pointer to the transport device, so we > can't call vsock_core_exit() until there are open sockets. > > v2 -> v3: > - Rebased on master > > v1 -> v2: > - Fixed commit message of patch 1. > - Added Reviewed-by, Acked-by tags by Stefan Series app...