* Re: [PATCH net-next v3 1/3] vsock: support sockmap [not found] ` <20230227-vsock-sockmap-upstream-v3-1-7e7f4ce623ee@bytedance.com> @ 2023-02-28 21:36 ` Michael S. Tsirkin [not found] ` <Y/B9ddkfQw6Ae/lY@bullseye> 2023-03-02 9:42 ` Stefano Garzarella 1 sibling, 1 reply; 6+ messages in thread From: Michael S. Tsirkin @ 2023-02-28 21:36 UTC (permalink / raw) To: Bobby Eshleman Cc: Cong Wang, kvm, Alexei Starovoitov, virtualization, Song Liu, Eric Dumazet, Stanislav Fomichev, linux-kselftest, Shuah Khan, Mykola Lysenko, Daniel Borkmann, John Fastabend, Andrii Nakryiko, Yonghong Song, Paolo Abeni, KP Singh, Stefan Hajnoczi, Jakub Kicinski, Hao Luo, netdev, linux-kernel, Jiri Olsa, bpf, Martin KaFai Lau, David S. Miller On Tue, Feb 28, 2023 at 07:04:34PM +0000, Bobby Eshleman wrote: > @@ -1241,19 +1252,34 @@ static int vsock_dgram_connect(struct socket *sock, > > memcpy(&vsk->remote_addr, remote_addr, sizeof(vsk->remote_addr)); > sock->state = SS_CONNECTED; > + sk->sk_state = TCP_ESTABLISHED; > > out: > release_sock(sk); > return err; > } How is this related? Maybe add a comment to explain? Does TCP_ESTABLISHED make sense for all types of sockets? -- MST _______________________________________________ Virtualization mailing list Virtualization@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/virtualization ^ permalink raw reply [flat|nested] 6+ messages in thread
[parent not found: <Y/B9ddkfQw6Ae/lY@bullseye>]
* Re: [PATCH net-next v3 1/3] vsock: support sockmap [not found] ` <Y/B9ddkfQw6Ae/lY@bullseye> @ 2023-03-02 9:20 ` Stefano Garzarella 0 siblings, 0 replies; 6+ messages in thread From: Stefano Garzarella @ 2023-03-02 9:20 UTC (permalink / raw) To: Bobby Eshleman Cc: Cong Wang, Bobby Eshleman, kvm, Michael S. Tsirkin, KP Singh, virtualization, Song Liu, Eric Dumazet, Stanislav Fomichev, linux-kselftest, Shuah Khan, Mykola Lysenko, Daniel Borkmann, John Fastabend, Andrii Nakryiko, Jakub Kicinski, Paolo Abeni, Alexei Starovoitov, Stefan Hajnoczi, Yonghong Song, Hao Luo, netdev, linux-kernel, Jiri Olsa, bpf, Martin KaFai Lau, David S. Miller On Sat, Feb 18, 2023 at 07:25:41AM +0000, Bobby Eshleman wrote: >On Tue, Feb 28, 2023 at 04:36:22PM -0500, Michael S. Tsirkin wrote: >> On Tue, Feb 28, 2023 at 07:04:34PM +0000, Bobby Eshleman wrote: >> > @@ -1241,19 +1252,34 @@ static int vsock_dgram_connect(struct socket *sock, >> > >> > memcpy(&vsk->remote_addr, remote_addr, sizeof(vsk->remote_addr)); >> > sock->state = SS_CONNECTED; >> > + sk->sk_state = TCP_ESTABLISHED; >> > >> > out: >> > release_sock(sk); >> > return err; >> > } >> >> >> How is this related? Maybe add a comment to explain? Does >> TCP_ESTABLISHED make sense for all types of sockets? >> > >Hey Michael, definitely, I can leave a comment. I agree, since I had the same doubt in previous versions, I think it's worth putting a comment in the code to explain why. Since there may be a v4, I'll leave some small comments in a separate email. Thanks, Stefano > >The real reason is due to this piece of logic in sockmap: >https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/net/core/sock_map.c?h=v6.2#n531 > >And because of it, you see the same thing in (for example) >unix_dgram_connect(): >https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/net/unix/af_unix.c?h=v6.2#n1394 > >I believe it makes sense for these other socket types. > _______________________________________________ Virtualization mailing list Virtualization@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/virtualization ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH net-next v3 1/3] vsock: support sockmap [not found] ` <20230227-vsock-sockmap-upstream-v3-1-7e7f4ce623ee@bytedance.com> 2023-02-28 21:36 ` [PATCH net-next v3 1/3] vsock: support sockmap Michael S. Tsirkin @ 2023-03-02 9:42 ` Stefano Garzarella 1 sibling, 0 replies; 6+ messages in thread From: Stefano Garzarella @ 2023-03-02 9:42 UTC (permalink / raw) To: Bobby Eshleman Cc: Cong Wang, kvm, Michael S. Tsirkin, Alexei Starovoitov, virtualization, Song Liu, Eric Dumazet, Stanislav Fomichev, linux-kselftest, Shuah Khan, Mykola Lysenko, Daniel Borkmann, John Fastabend, Andrii Nakryiko, Yonghong Song, Paolo Abeni, KP Singh, Stefan Hajnoczi, Jakub Kicinski, Hao Luo, netdev, linux-kernel, Jiri Olsa, bpf, Martin KaFai Lau, David S. Miller On Tue, Feb 28, 2023 at 07:04:34PM +0000, Bobby Eshleman wrote: >This patch adds sockmap support for vsock sockets. It is intended to be >usable by all transports, but only the virtio transport is implemented. > >Signed-off-by: Bobby Eshleman <bobby.eshleman@bytedance.com> >--- > drivers/vhost/vsock.c | 1 + > include/linux/virtio_vsock.h | 1 + > include/net/af_vsock.h | 17 ++++ > net/vmw_vsock/Makefile | 1 + > net/vmw_vsock/af_vsock.c | 55 ++++++++-- > net/vmw_vsock/virtio_transport.c | 2 + > net/vmw_vsock/virtio_transport_common.c | 25 +++++ > net/vmw_vsock/vsock_bpf.c | 174 ++++++++++++++++++++++++++++++++ > net/vmw_vsock/vsock_loopback.c | 2 + > 9 files changed, 272 insertions(+), 6 deletions(-) > >diff --git a/drivers/vhost/vsock.c b/drivers/vhost/vsock.c >index 1f3b89c885cc..3c6dc036b904 100644 >--- a/drivers/vhost/vsock.c >+++ b/drivers/vhost/vsock.c >@@ -439,6 +439,7 @@ static struct virtio_transport vhost_transport = { > .notify_send_post_enqueue = virtio_transport_notify_send_post_enqueue, > .notify_buffer_size = virtio_transport_notify_buffer_size, > >+ .read_skb = virtio_transport_read_skb, > }, > > .send_pkt = vhost_transport_send_pkt, >diff --git a/include/linux/virtio_vsock.h b/include/linux/virtio_vsock.h >index 3f9c16611306..c58453699ee9 100644 >--- a/include/linux/virtio_vsock.h >+++ b/include/linux/virtio_vsock.h >@@ -245,4 +245,5 @@ u32 virtio_transport_get_credit(struct virtio_vsock_sock *vvs, u32 wanted); > void virtio_transport_put_credit(struct virtio_vsock_sock *vvs, u32 credit); > void virtio_transport_deliver_tap_pkt(struct sk_buff *skb); > int virtio_transport_purge_skbs(void *vsk, struct sk_buff_head *list); >+int virtio_transport_read_skb(struct vsock_sock *vsk, skb_read_actor_t read_actor); > #endif /* _LINUX_VIRTIO_VSOCK_H */ >diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h >index 568a87c5e0d0..a73f5fbd296a 100644 >--- a/include/net/af_vsock.h >+++ b/include/net/af_vsock.h >@@ -75,6 +75,7 @@ struct vsock_sock { > void *trans; > }; > >+s64 vsock_connectible_has_data(struct vsock_sock *vsk); > s64 vsock_stream_has_data(struct vsock_sock *vsk); > s64 vsock_stream_has_space(struct vsock_sock *vsk); > struct sock *vsock_create_connected(struct sock *parent); >@@ -173,6 +174,9 @@ struct vsock_transport { > > /* Addressing. */ > u32 (*get_local_cid)(void); >+ >+ /* Read a single skb */ >+ int (*read_skb)(struct vsock_sock *, skb_read_actor_t); > }; > > /**** CORE ****/ >@@ -225,5 +229,18 @@ int vsock_init_tap(void); > int vsock_add_tap(struct vsock_tap *vt); > int vsock_remove_tap(struct vsock_tap *vt); > void vsock_deliver_tap(struct sk_buff *build_skb(void *opaque), void *opaque); >+int vsock_connectible_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, >+ int flags); >+int vsock_dgram_recvmsg(struct socket *sock, struct msghdr *msg, >+ size_t len, int flags); Just because you have to resend a v4, here we can align as `checkpatch --strict` suggests: CHECK: Alignment should match open parenthesis #63: FILE: include/net/af_vsock.h:235: +int vsock_dgram_recvmsg(struct socket *sock, struct msghdr *msg, + size_t len, int flags); >+ >+#ifdef CONFIG_BPF_SYSCALL >+extern struct proto vsock_proto; >+int vsock_bpf_update_proto(struct sock *sk, struct sk_psock *psock, bool restore); >+void __init vsock_bpf_build_proto(void); >+#else >+static inline void __init vsock_bpf_build_proto(void) >+{} >+#endif > > #endif /* __AF_VSOCK_H__ */ >diff --git a/net/vmw_vsock/Makefile b/net/vmw_vsock/Makefile >index 6a943ec95c4a..5da74c4a9f1d 100644 >--- a/net/vmw_vsock/Makefile >+++ b/net/vmw_vsock/Makefile >@@ -8,6 +8,7 @@ obj-$(CONFIG_HYPERV_VSOCKETS) += hv_sock.o > obj-$(CONFIG_VSOCKETS_LOOPBACK) += vsock_loopback.o > > vsock-y += af_vsock.o af_vsock_tap.o vsock_addr.o >+vsock-$(CONFIG_BPF_SYSCALL) += vsock_bpf.o > > vsock_diag-y += diag.o > >diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c >index 19aea7cba26e..f2cc04fb8b13 100644 >--- a/net/vmw_vsock/af_vsock.c >+++ b/net/vmw_vsock/af_vsock.c >@@ -116,10 +116,13 @@ static void vsock_sk_destruct(struct sock *sk); > static int vsock_queue_rcv_skb(struct sock *sk, struct sk_buff *skb); > > /* Protocol family. */ >-static struct proto vsock_proto = { >+struct proto vsock_proto = { > .name = "AF_VSOCK", > .owner = THIS_MODULE, > .obj_size = sizeof(struct vsock_sock), >+#ifdef CONFIG_BPF_SYSCALL >+ .psock_update_sk_prot = vsock_bpf_update_proto, >+#endif > }; > > /* The default peer timeout indicates how long we will wait for a peer response >@@ -865,7 +868,7 @@ s64 vsock_stream_has_data(struct vsock_sock *vsk) > } > EXPORT_SYMBOL_GPL(vsock_stream_has_data); > >-static s64 vsock_connectible_has_data(struct vsock_sock *vsk) >+s64 vsock_connectible_has_data(struct vsock_sock *vsk) > { > struct sock *sk = sk_vsock(vsk); > >@@ -874,6 +877,7 @@ static s64 vsock_connectible_has_data(struct vsock_sock *vsk) > else > return vsock_stream_has_data(vsk); > } >+EXPORT_SYMBOL_GPL(vsock_connectible_has_data); > > s64 vsock_stream_has_space(struct vsock_sock *vsk) > { >@@ -1131,6 +1135,13 @@ static __poll_t vsock_poll(struct file *file, struct socket *sock, > return mask; > } > >+static int vsock_read_skb(struct sock *sk, skb_read_actor_t read_actor) >+{ >+ struct vsock_sock *vsk = vsock_sk(sk); >+ >+ return vsk->transport->read_skb(vsk, read_actor); >+} >+ > static int vsock_dgram_sendmsg(struct socket *sock, struct msghdr *msg, > size_t len) > { >@@ -1241,19 +1252,34 @@ static int vsock_dgram_connect(struct socket *sock, > > memcpy(&vsk->remote_addr, remote_addr, sizeof(vsk->remote_addr)); > sock->state = SS_CONNECTED; >+ sk->sk_state = TCP_ESTABLISHED; > > out: > release_sock(sk); > return err; > } > >-static int vsock_dgram_recvmsg(struct socket *sock, struct msghdr *msg, >- size_t len, int flags) >+int vsock_dgram_recvmsg(struct socket *sock, struct msghdr *msg, >+ size_t len, int flags) > { >- struct vsock_sock *vsk = vsock_sk(sock->sk); >+#ifdef CONFIG_BPF_SYSCALL >+ const struct proto *prot; >+#endif >+ struct vsock_sock *vsk; >+ struct sock *sk; >+ >+ sk = sock->sk; >+ vsk = vsock_sk(sk); >+ >+#ifdef CONFIG_BPF_SYSCALL >+ prot = READ_ONCE(sk->sk_prot); >+ if (prot != &vsock_proto) >+ return prot->recvmsg(sk, msg, len, flags, NULL); >+#endif > > return vsk->transport->dgram_dequeue(vsk, msg, len, flags); > } >+EXPORT_SYMBOL_GPL(vsock_dgram_recvmsg); > > static const struct proto_ops vsock_dgram_ops = { > .family = PF_VSOCK, >@@ -1272,6 +1298,7 @@ static const struct proto_ops vsock_dgram_ops = { > .recvmsg = vsock_dgram_recvmsg, > .mmap = sock_no_mmap, > .sendpage = sock_no_sendpage, >+ .read_skb = vsock_read_skb, > }; > > static int vsock_transport_cancel_pkt(struct vsock_sock *vsk) >@@ -2086,13 +2113,16 @@ static int __vsock_seqpacket_recvmsg(struct sock *sk, struct msghdr *msg, > return err; > } > >-static int >+int > vsock_connectible_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, > int flags) > { > struct sock *sk; > struct vsock_sock *vsk; > const struct vsock_transport *transport; >+#ifdef CONFIG_BPF_SYSCALL >+ const struct proto *prot; >+#endif > int err; > > sk = sock->sk; >@@ -2139,6 +2169,14 @@ vsock_connectible_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, > goto out; > } > >+#ifdef CONFIG_BPF_SYSCALL >+ prot = READ_ONCE(sk->sk_prot); >+ if (prot != &vsock_proto) { >+ release_sock(sk); >+ return prot->recvmsg(sk, msg, len, flags, NULL); >+ } >+#endif >+ > if (sk->sk_type == SOCK_STREAM) > err = __vsock_stream_recvmsg(sk, msg, len, flags); > else >@@ -2148,6 +2186,7 @@ vsock_connectible_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, > release_sock(sk); > return err; > } >+EXPORT_SYMBOL_GPL(vsock_connectible_recvmsg); > > static int vsock_set_rcvlowat(struct sock *sk, int val) > { >@@ -2188,6 +2227,7 @@ static const struct proto_ops vsock_stream_ops = { > .mmap = sock_no_mmap, > .sendpage = sock_no_sendpage, > .set_rcvlowat = vsock_set_rcvlowat, >+ .read_skb = vsock_read_skb, > }; > > static const struct proto_ops vsock_seqpacket_ops = { >@@ -2209,6 +2249,7 @@ static const struct proto_ops vsock_seqpacket_ops = { > .recvmsg = vsock_connectible_recvmsg, > .mmap = sock_no_mmap, > .sendpage = sock_no_sendpage, >+ .read_skb = vsock_read_skb, > }; > > static int vsock_create(struct net *net, struct socket *sock, >@@ -2348,6 +2389,8 @@ static int __init vsock_init(void) > goto err_unregister_proto; > } > >+ vsock_bpf_build_proto(); >+ > return 0; > > err_unregister_proto: >diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c >index 28b5a8e8e094..e95df847176b 100644 >--- a/net/vmw_vsock/virtio_transport.c >+++ b/net/vmw_vsock/virtio_transport.c >@@ -457,6 +457,8 @@ static struct virtio_transport virtio_transport = { > .notify_send_pre_enqueue = virtio_transport_notify_send_pre_enqueue, > .notify_send_post_enqueue = virtio_transport_notify_send_post_enqueue, > .notify_buffer_size = virtio_transport_notify_buffer_size, >+ >+ .read_skb = virtio_transport_read_skb, > }, > > .send_pkt = virtio_transport_send_pkt, >diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c >index a1581c77cf84..f64527f32385 100644 >--- a/net/vmw_vsock/virtio_transport_common.c >+++ b/net/vmw_vsock/virtio_transport_common.c >@@ -1388,6 +1388,31 @@ int virtio_transport_purge_skbs(void *vsk, struct sk_buff_head *queue) > } > EXPORT_SYMBOL_GPL(virtio_transport_purge_skbs); > >+int virtio_transport_read_skb(struct vsock_sock *vsk, skb_read_actor_t recv_actor) >+{ >+ struct virtio_vsock_sock *vvs = vsk->trans; >+ struct sock *sk = sk_vsock(vsk); >+ struct sk_buff *skb; >+ int off = 0; >+ int copied; >+ int err; >+ >+ spin_lock_bh(&vvs->rx_lock); >+ /* Use __skb_recv_datagram() for race-free handling of the receive. It >+ * works for types other than dgrams. >+ */ >+ skb = __skb_recv_datagram(sk, &vvs->rx_queue, MSG_DONTWAIT, &off, &err); >+ spin_unlock_bh(&vvs->rx_lock); >+ >+ if (!skb) >+ return err; >+ >+ copied = recv_actor(sk, skb); >+ kfree_skb(skb); >+ return copied; >+} >+EXPORT_SYMBOL_GPL(virtio_transport_read_skb); >+ > MODULE_LICENSE("GPL v2"); > MODULE_AUTHOR("Asias He"); > MODULE_DESCRIPTION("common code for virtio vsock"); >diff --git a/net/vmw_vsock/vsock_bpf.c b/net/vmw_vsock/vsock_bpf.c >new file mode 100644 >index 000000000000..9c71c163c684 >--- /dev/null >+++ b/net/vmw_vsock/vsock_bpf.c >@@ -0,0 +1,174 @@ >+// SPDX-License-Identifier: GPL-2.0 >+/* Copyright (c) 2022 Bobby Eshleman <bobby.eshleman@bytedance.com> >+ * >+ * Based off of net/unix/unix_bpf.c >+ */ >+ >+#include <linux/bpf.h> >+#include <linux/module.h> >+#include <linux/skmsg.h> >+#include <linux/socket.h> >+#include <linux/wait.h> >+#include <net/af_vsock.h> >+#include <net/sock.h> >+ >+#define vsock_sk_has_data(__sk, __psock) \ >+ ({ !skb_queue_empty(&(__sk)->sk_receive_queue) || \ >+ !skb_queue_empty(&(__psock)->ingress_skb) || \ >+ !list_empty(&(__psock)->ingress_msg); \ >+ }) >+ >+static struct proto *vsock_prot_saved __read_mostly; >+static DEFINE_SPINLOCK(vsock_prot_lock); >+static struct proto vsock_bpf_prot; >+ >+static bool vsock_has_data(struct sock *sk, struct sk_psock *psock) >+{ >+ struct vsock_sock *vsk = vsock_sk(sk); >+ s64 ret; >+ >+ ret = vsock_connectible_has_data(vsk); >+ if (ret > 0) >+ return true; >+ >+ return vsock_sk_has_data(sk, psock); >+} >+ >+static bool vsock_msg_wait_data(struct sock *sk, struct sk_psock *psock, long timeo) >+{ >+ int ret; `ret` can be bool now. The rest LGTM. Thanks, Stefano _______________________________________________ Virtualization mailing list Virtualization@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/virtualization ^ permalink raw reply [flat|nested] 6+ messages in thread
[parent not found: <20230227-vsock-sockmap-upstream-v3-2-7e7f4ce623ee@bytedance.com>]
* Re: [PATCH net-next v3 2/3] selftests/bpf: add vsock to vmtest.sh [not found] ` <20230227-vsock-sockmap-upstream-v3-2-7e7f4ce623ee@bytedance.com> @ 2023-03-02 9:53 ` Stefano Garzarella 0 siblings, 0 replies; 6+ messages in thread From: Stefano Garzarella @ 2023-03-02 9:53 UTC (permalink / raw) To: Bobby Eshleman Cc: Cong Wang, kvm, Michael S. Tsirkin, Alexei Starovoitov, virtualization, Song Liu, Eric Dumazet, Stanislav Fomichev, linux-kselftest, Shuah Khan, Mykola Lysenko, Daniel Borkmann, John Fastabend, Andrii Nakryiko, Yonghong Song, Paolo Abeni, KP Singh, Stefan Hajnoczi, Jakub Kicinski, Hao Luo, netdev, linux-kernel, Jiri Olsa, bpf, Martin KaFai Lau, David S. Miller On Tue, Feb 28, 2023 at 07:04:35PM +0000, Bobby Eshleman wrote: >Add vsock loopback to the test kernel. > >This allows sockmap for vsock to be tested. > >Signed-off-by: Bobby Eshleman <bobby.eshleman@bytedance.com> >--- > tools/testing/selftests/bpf/config.aarch64 | 2 ++ > tools/testing/selftests/bpf/config.s390x | 3 +++ > tools/testing/selftests/bpf/config.x86_64 | 3 +++ > 3 files changed, 8 insertions(+) I already acked this in the v2, please bring them between versions if there are no changes. Acked-by: Stefano Garzarella <sgarzare@redhat.com> _______________________________________________ Virtualization mailing list Virtualization@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/virtualization ^ permalink raw reply [flat|nested] 6+ messages in thread
[parent not found: <20230227-vsock-sockmap-upstream-v3-3-7e7f4ce623ee@bytedance.com>]
* Re: [PATCH net-next v3 3/3] selftests/bpf: add a test case for vsock sockmap [not found] ` <20230227-vsock-sockmap-upstream-v3-3-7e7f4ce623ee@bytedance.com> @ 2023-03-02 9:54 ` Stefano Garzarella 0 siblings, 0 replies; 6+ messages in thread From: Stefano Garzarella @ 2023-03-02 9:54 UTC (permalink / raw) To: Bobby Eshleman Cc: Cong Wang, kvm, Michael S. Tsirkin, Alexei Starovoitov, virtualization, Song Liu, Eric Dumazet, Stanislav Fomichev, linux-kselftest, Shuah Khan, Mykola Lysenko, Daniel Borkmann, John Fastabend, Andrii Nakryiko, Yonghong Song, Paolo Abeni, KP Singh, Stefan Hajnoczi, Jakub Kicinski, Hao Luo, netdev, linux-kernel, Jiri Olsa, bpf, Martin KaFai Lau, David S. Miller On Tue, Feb 28, 2023 at 07:04:36PM +0000, Bobby Eshleman wrote: >Add a test case testing the redirection from connectible AF_VSOCK >sockets to connectible AF_UNIX sockets. > >Signed-off-by: Bobby Eshleman <bobby.eshleman@bytedance.com> >--- > .../selftests/bpf/prog_tests/sockmap_listen.c | 163 +++++++++++++++++++++ > 1 file changed, 163 insertions(+) Ditto. For the vsock part: Acked-by: Stefano Garzarella <sgarzare@redhat.com> _______________________________________________ Virtualization mailing list Virtualization@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/virtualization ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH net-next v3 0/3] vsock: add support for sockmap [not found] <20230227-vsock-sockmap-upstream-v3-0-7e7f4ce623ee@bytedance.com> ` (2 preceding siblings ...) [not found] ` <20230227-vsock-sockmap-upstream-v3-3-7e7f4ce623ee@bytedance.com> @ 2023-03-10 8:53 ` Michael S. Tsirkin 3 siblings, 0 replies; 6+ messages in thread From: Michael S. Tsirkin @ 2023-03-10 8:53 UTC (permalink / raw) To: Bobby Eshleman Cc: Cong Wang, kvm, Alexei Starovoitov, virtualization, Song Liu, Eric Dumazet, Stanislav Fomichev, linux-kselftest, Shuah Khan, Mykola Lysenko, Daniel Borkmann, John Fastabend, Andrii Nakryiko, Yonghong Song, Paolo Abeni, KP Singh, Stefan Hajnoczi, Jakub Kicinski, Hao Luo, netdev, linux-kernel, Jiri Olsa, bpf, Martin KaFai Lau, David S. Miller On Tue, Feb 28, 2023 at 07:04:33PM +0000, Bobby Eshleman wrote: > Add support for sockmap to vsock. > > We're testing usage of vsock as a way to redirect guest-local UDS > requests to the host and this patch series greatly improves the > performance of such a setup. > > Compared to copying packets via userspace, this improves throughput by > 121% in basic testing. besides the small comment, looks ok. Feel free to include my ack in v4: Acked-by: Michael S. Tsirkin <mst@redhat.com> > Tested as follows. > > Setup: guest unix dgram sender -> guest vsock redirector -> host vsock > server > Threads: 1 > Payload: 64k > No sockmap: > - 76.3 MB/s > - The guest vsock redirector was > "socat VSOCK-CONNECT:2:1234 UNIX-RECV:/path/to/sock" > Using sockmap (this patch): > - 168.8 MB/s (+121%) > - The guest redirector was a simple sockmap echo server, > redirecting unix ingress to vsock 2:1234 egress. > - Same sender and server programs > > *Note: these numbers are from RFC v1 > > Only the virtio transport has been tested. The loopback transport was > used in writing bpf/selftests, but not thoroughly tested otherwise. > > This series requires the skb patch. > > Changes in v3: > - vsock/bpf: Refactor wait logic in vsock_bpf_recvmsg() to avoid > backwards goto > - vsock/bpf: Check psock before acquiring slock > - vsock/bpf: Return bool instead of int of 0 or 1 > - vsock/bpf: Wrap macro args __sk/__psock in parens > - vsock/bpf: Place comment trailer */ on separate line > > Changes in v2: > - vsock/bpf: rename vsock_dgram_* -> vsock_* > - vsock/bpf: change sk_psock_{get,put} and {lock,release}_sock() order > to minimize slock hold time > - vsock/bpf: use "new style" wait > - vsock/bpf: fix bug in wait log > - vsock/bpf: add check that recvmsg sk_type is one dgram, seqpacket, or > stream. Return error if not one of the three. > - virtio/vsock: comment __skb_recv_datagram() usage > - virtio/vsock: do not init copied in read_skb() > - vsock/bpf: add ifdef guard around struct proto in dgram_recvmsg() > - selftests/bpf: add vsock loopback config for aarch64 > - selftests/bpf: add vsock loopback config for s390x > - selftests/bpf: remove vsock device from vmtest.sh qemu machine > - selftests/bpf: remove CONFIG_VIRTIO_VSOCKETS=y from config.x86_64 > - vsock/bpf: move transport-related (e.g., if (!vsk->transport)) checks > out of fast path > > Signed-off-by: Bobby Eshleman <bobby.eshleman@bytedance.com> > --- > Bobby Eshleman (3): > vsock: support sockmap > selftests/bpf: add vsock to vmtest.sh > selftests/bpf: Add a test case for vsock sockmap > > drivers/vhost/vsock.c | 1 + > include/linux/virtio_vsock.h | 1 + > include/net/af_vsock.h | 17 ++ > net/vmw_vsock/Makefile | 1 + > net/vmw_vsock/af_vsock.c | 55 ++++++- > net/vmw_vsock/virtio_transport.c | 2 + > net/vmw_vsock/virtio_transport_common.c | 24 +++ > net/vmw_vsock/vsock_bpf.c | 175 +++++++++++++++++++++ > net/vmw_vsock/vsock_loopback.c | 2 + > tools/testing/selftests/bpf/config.aarch64 | 2 + > tools/testing/selftests/bpf/config.s390x | 3 + > tools/testing/selftests/bpf/config.x86_64 | 3 + > .../selftests/bpf/prog_tests/sockmap_listen.c | 163 +++++++++++++++++++ > 13 files changed, 443 insertions(+), 6 deletions(-) > --- > base-commit: d83115ce337a632f996e44c9f9e18cadfcf5a094 > change-id: 20230118-support-vsock-sockmap-connectible-2e1297d2111a > > Best regards, > -- > Bobby Eshleman <bobby.eshleman@bytedance.com> > > --- > Bobby Eshleman (3): > vsock: support sockmap > selftests/bpf: add vsock to vmtest.sh > selftests/bpf: add a test case for vsock sockmap > > drivers/vhost/vsock.c | 1 + > include/linux/virtio_vsock.h | 1 + > include/net/af_vsock.h | 17 ++ > net/vmw_vsock/Makefile | 1 + > net/vmw_vsock/af_vsock.c | 55 ++++++- > net/vmw_vsock/virtio_transport.c | 2 + > net/vmw_vsock/virtio_transport_common.c | 25 +++ > net/vmw_vsock/vsock_bpf.c | 174 +++++++++++++++++++++ > net/vmw_vsock/vsock_loopback.c | 2 + > tools/testing/selftests/bpf/config.aarch64 | 2 + > tools/testing/selftests/bpf/config.s390x | 3 + > tools/testing/selftests/bpf/config.x86_64 | 3 + > .../selftests/bpf/prog_tests/sockmap_listen.c | 163 +++++++++++++++++++ > 13 files changed, 443 insertions(+), 6 deletions(-) > --- > base-commit: c2ea552065e43d05bce240f53c3185fd3a066204 > change-id: 20230227-vsock-sockmap-upstream-9d65c84174a2 > > Best regards, > -- > Bobby Eshleman <bobby.eshleman@bytedance.com> _______________________________________________ Virtualization mailing list Virtualization@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/virtualization ^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2023-03-10 8:53 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <20230227-vsock-sockmap-upstream-v3-0-7e7f4ce623ee@bytedance.com>
[not found] ` <20230227-vsock-sockmap-upstream-v3-1-7e7f4ce623ee@bytedance.com>
2023-02-28 21:36 ` [PATCH net-next v3 1/3] vsock: support sockmap Michael S. Tsirkin
[not found] ` <Y/B9ddkfQw6Ae/lY@bullseye>
2023-03-02 9:20 ` Stefano Garzarella
2023-03-02 9:42 ` Stefano Garzarella
[not found] ` <20230227-vsock-sockmap-upstream-v3-2-7e7f4ce623ee@bytedance.com>
2023-03-02 9:53 ` [PATCH net-next v3 2/3] selftests/bpf: add vsock to vmtest.sh Stefano Garzarella
[not found] ` <20230227-vsock-sockmap-upstream-v3-3-7e7f4ce623ee@bytedance.com>
2023-03-02 9:54 ` [PATCH net-next v3 3/3] selftests/bpf: add a test case for vsock sockmap Stefano Garzarella
2023-03-10 8:53 ` [PATCH net-next v3 0/3] vsock: add support for sockmap Michael S. Tsirkin
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).