* [PATCH net-next 07/14] vsock: handle buffer_size sockopts in the core
From: Stefano Garzarella @ 2019-10-23 9:55 UTC (permalink / raw)
To: netdev
Cc: Michael S. Tsirkin, kvm, Greg Kroah-Hartman, Jason Wang,
David S. Miller, Dexuan Cui, Haiyang Zhang, Jorgen Hansen,
Sasha Levin, linux-kernel, Arnd Bergmann, Stefan Hajnoczi,
linux-hyperv, K. Y. Srinivasan, Stephen Hemminger, virtualization
In-Reply-To: <20191023095554.11340-1-sgarzare@redhat.com>
virtio_transport and vmci_transport handle the buffer_size
sockopts in a very similar way.
In order to support multiple transports, this patch moves this
handling in the core to allow the user to change the options
also if the socket is not yet assigned to any transport.
This patch also adds the '.notify_buffer_size' callback in the
'struct virtio_transport' in order to inform the transport,
when the buffer_size is changed by the user. It is also useful
to limit the 'buffer_size' requested (e.g. virtio transports).
Acked-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
---
RFC -> v1:
- changed .notify_buffer_size return to void (Stefan)
- documented that .notify_buffer_size is called with sk_lock held (Stefan)
---
drivers/vhost/vsock.c | 7 +-
include/linux/virtio_vsock.h | 15 +----
include/net/af_vsock.h | 15 ++---
net/vmw_vsock/af_vsock.c | 43 ++++++++++---
net/vmw_vsock/hyperv_transport.c | 36 -----------
net/vmw_vsock/virtio_transport.c | 8 +--
net/vmw_vsock/virtio_transport_common.c | 79 ++++-------------------
net/vmw_vsock/vmci_transport.c | 86 +++----------------------
net/vmw_vsock/vmci_transport.h | 3 -
9 files changed, 65 insertions(+), 227 deletions(-)
diff --git a/drivers/vhost/vsock.c b/drivers/vhost/vsock.c
index 92ab3852c954..6d7e4f022748 100644
--- a/drivers/vhost/vsock.c
+++ b/drivers/vhost/vsock.c
@@ -418,13 +418,8 @@ static struct virtio_transport vhost_transport = {
.notify_send_pre_block = virtio_transport_notify_send_pre_block,
.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,
- .set_buffer_size = virtio_transport_set_buffer_size,
- .set_min_buffer_size = virtio_transport_set_min_buffer_size,
- .set_max_buffer_size = virtio_transport_set_max_buffer_size,
- .get_buffer_size = virtio_transport_get_buffer_size,
- .get_min_buffer_size = virtio_transport_get_min_buffer_size,
- .get_max_buffer_size = virtio_transport_get_max_buffer_size,
},
.send_pkt = vhost_transport_send_pkt,
diff --git a/include/linux/virtio_vsock.h b/include/linux/virtio_vsock.h
index 96d8132acbd7..b79befd2a5a4 100644
--- a/include/linux/virtio_vsock.h
+++ b/include/linux/virtio_vsock.h
@@ -7,9 +7,6 @@
#include <net/sock.h>
#include <net/af_vsock.h>
-#define VIRTIO_VSOCK_DEFAULT_MIN_BUF_SIZE 128
-#define VIRTIO_VSOCK_DEFAULT_BUF_SIZE (1024 * 256)
-#define VIRTIO_VSOCK_DEFAULT_MAX_BUF_SIZE (1024 * 256)
#define VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE (1024 * 4)
#define VIRTIO_VSOCK_MAX_BUF_SIZE 0xFFFFFFFFUL
#define VIRTIO_VSOCK_MAX_PKT_BUF_SIZE (1024 * 64)
@@ -25,11 +22,6 @@ enum {
struct virtio_vsock_sock {
struct vsock_sock *vsk;
- /* Protected by lock_sock(sk_vsock(trans->vsk)) */
- u32 buf_size;
- u32 buf_size_min;
- u32 buf_size_max;
-
spinlock_t tx_lock;
spinlock_t rx_lock;
@@ -93,12 +85,6 @@ s64 virtio_transport_stream_has_space(struct vsock_sock *vsk);
int virtio_transport_do_socket_init(struct vsock_sock *vsk,
struct vsock_sock *psk);
-u64 virtio_transport_get_buffer_size(struct vsock_sock *vsk);
-u64 virtio_transport_get_min_buffer_size(struct vsock_sock *vsk);
-u64 virtio_transport_get_max_buffer_size(struct vsock_sock *vsk);
-void virtio_transport_set_buffer_size(struct vsock_sock *vsk, u64 val);
-void virtio_transport_set_min_buffer_size(struct vsock_sock *vsk, u64 val);
-void virtio_transport_set_max_buffer_size(struct vsock_sock *vs, u64 val);
int
virtio_transport_notify_poll_in(struct vsock_sock *vsk,
size_t target,
@@ -125,6 +111,7 @@ int virtio_transport_notify_send_pre_enqueue(struct vsock_sock *vsk,
struct vsock_transport_send_notify_data *data);
int virtio_transport_notify_send_post_enqueue(struct vsock_sock *vsk,
ssize_t written, struct vsock_transport_send_notify_data *data);
+void virtio_transport_notify_buffer_size(struct vsock_sock *vsk, u64 *val);
u64 virtio_transport_stream_rcvhiwat(struct vsock_sock *vsk);
bool virtio_transport_stream_is_active(struct vsock_sock *vsk);
diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h
index 2ca67d048de4..4b5d16840fd4 100644
--- a/include/net/af_vsock.h
+++ b/include/net/af_vsock.h
@@ -65,6 +65,11 @@ struct vsock_sock {
bool sent_request;
bool ignore_connecting_rst;
+ /* Protected by lock_sock(sk) */
+ u64 buffer_size;
+ u64 buffer_min_size;
+ u64 buffer_max_size;
+
/* Private to transport. */
void *trans;
};
@@ -140,18 +145,12 @@ struct vsock_transport {
struct vsock_transport_send_notify_data *);
int (*notify_send_post_enqueue)(struct vsock_sock *, ssize_t,
struct vsock_transport_send_notify_data *);
+ /* sk_lock held by the caller */
+ void (*notify_buffer_size)(struct vsock_sock *, u64 *);
/* Shutdown. */
int (*shutdown)(struct vsock_sock *, int);
- /* Buffer sizes. */
- void (*set_buffer_size)(struct vsock_sock *, u64);
- void (*set_min_buffer_size)(struct vsock_sock *, u64);
- void (*set_max_buffer_size)(struct vsock_sock *, u64);
- u64 (*get_buffer_size)(struct vsock_sock *);
- u64 (*get_min_buffer_size)(struct vsock_sock *);
- u64 (*get_max_buffer_size)(struct vsock_sock *);
-
/* Addressing. */
u32 (*get_local_cid)(void);
};
diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
index eaea159006c8..90ac46ea12ef 100644
--- a/net/vmw_vsock/af_vsock.c
+++ b/net/vmw_vsock/af_vsock.c
@@ -126,6 +126,10 @@ static struct proto vsock_proto = {
*/
#define VSOCK_DEFAULT_CONNECT_TIMEOUT (2 * HZ)
+#define VSOCK_DEFAULT_BUFFER_SIZE (1024 * 256)
+#define VSOCK_DEFAULT_BUFFER_MAX_SIZE (1024 * 256)
+#define VSOCK_DEFAULT_BUFFER_MIN_SIZE 128
+
static const struct vsock_transport *transport_single;
static DEFINE_MUTEX(vsock_register_mutex);
@@ -613,10 +617,16 @@ struct sock *__vsock_create(struct net *net,
vsk->trusted = psk->trusted;
vsk->owner = get_cred(psk->owner);
vsk->connect_timeout = psk->connect_timeout;
+ vsk->buffer_size = psk->buffer_size;
+ vsk->buffer_min_size = psk->buffer_min_size;
+ vsk->buffer_max_size = psk->buffer_max_size;
} else {
vsk->trusted = capable(CAP_NET_ADMIN);
vsk->owner = get_current_cred();
vsk->connect_timeout = VSOCK_DEFAULT_CONNECT_TIMEOUT;
+ vsk->buffer_size = VSOCK_DEFAULT_BUFFER_SIZE;
+ vsk->buffer_min_size = VSOCK_DEFAULT_BUFFER_MIN_SIZE;
+ vsk->buffer_max_size = VSOCK_DEFAULT_BUFFER_MAX_SIZE;
}
if (vsk->transport->init(vsk, psk) < 0) {
@@ -1368,6 +1378,23 @@ static int vsock_listen(struct socket *sock, int backlog)
return err;
}
+static void vsock_update_buffer_size(struct vsock_sock *vsk,
+ const struct vsock_transport *transport,
+ u64 val)
+{
+ if (val > vsk->buffer_max_size)
+ val = vsk->buffer_max_size;
+
+ if (val < vsk->buffer_min_size)
+ val = vsk->buffer_min_size;
+
+ if (val != vsk->buffer_size &&
+ transport && transport->notify_buffer_size)
+ transport->notify_buffer_size(vsk, &val);
+
+ vsk->buffer_size = val;
+}
+
static int vsock_stream_setsockopt(struct socket *sock,
int level,
int optname,
@@ -1405,17 +1432,19 @@ static int vsock_stream_setsockopt(struct socket *sock,
switch (optname) {
case SO_VM_SOCKETS_BUFFER_SIZE:
COPY_IN(val);
- transport->set_buffer_size(vsk, val);
+ vsock_update_buffer_size(vsk, transport, val);
break;
case SO_VM_SOCKETS_BUFFER_MAX_SIZE:
COPY_IN(val);
- transport->set_max_buffer_size(vsk, val);
+ vsk->buffer_max_size = val;
+ vsock_update_buffer_size(vsk, transport, vsk->buffer_size);
break;
case SO_VM_SOCKETS_BUFFER_MIN_SIZE:
COPY_IN(val);
- transport->set_min_buffer_size(vsk, val);
+ vsk->buffer_min_size = val;
+ vsock_update_buffer_size(vsk, transport, vsk->buffer_size);
break;
case SO_VM_SOCKETS_CONNECT_TIMEOUT: {
@@ -1456,7 +1485,6 @@ static int vsock_stream_getsockopt(struct socket *sock,
int len;
struct sock *sk;
struct vsock_sock *vsk;
- const struct vsock_transport *transport;
u64 val;
if (level != AF_VSOCK)
@@ -1480,21 +1508,20 @@ static int vsock_stream_getsockopt(struct socket *sock,
err = 0;
sk = sock->sk;
vsk = vsock_sk(sk);
- transport = vsk->transport;
switch (optname) {
case SO_VM_SOCKETS_BUFFER_SIZE:
- val = transport->get_buffer_size(vsk);
+ val = vsk->buffer_size;
COPY_OUT(val);
break;
case SO_VM_SOCKETS_BUFFER_MAX_SIZE:
- val = transport->get_max_buffer_size(vsk);
+ val = vsk->buffer_max_size;
COPY_OUT(val);
break;
case SO_VM_SOCKETS_BUFFER_MIN_SIZE:
- val = transport->get_min_buffer_size(vsk);
+ val = vsk->buffer_min_size;
COPY_OUT(val);
break;
diff --git a/net/vmw_vsock/hyperv_transport.c b/net/vmw_vsock/hyperv_transport.c
index bef8772116ec..d62297a62ca6 100644
--- a/net/vmw_vsock/hyperv_transport.c
+++ b/net/vmw_vsock/hyperv_transport.c
@@ -845,36 +845,6 @@ int hvs_notify_send_post_enqueue(struct vsock_sock *vsk, ssize_t written,
return 0;
}
-static void hvs_set_buffer_size(struct vsock_sock *vsk, u64 val)
-{
- /* Ignored. */
-}
-
-static void hvs_set_min_buffer_size(struct vsock_sock *vsk, u64 val)
-{
- /* Ignored. */
-}
-
-static void hvs_set_max_buffer_size(struct vsock_sock *vsk, u64 val)
-{
- /* Ignored. */
-}
-
-static u64 hvs_get_buffer_size(struct vsock_sock *vsk)
-{
- return -ENOPROTOOPT;
-}
-
-static u64 hvs_get_min_buffer_size(struct vsock_sock *vsk)
-{
- return -ENOPROTOOPT;
-}
-
-static u64 hvs_get_max_buffer_size(struct vsock_sock *vsk)
-{
- return -ENOPROTOOPT;
-}
-
static struct vsock_transport hvs_transport = {
.get_local_cid = hvs_get_local_cid,
@@ -908,12 +878,6 @@ static struct vsock_transport hvs_transport = {
.notify_send_pre_enqueue = hvs_notify_send_pre_enqueue,
.notify_send_post_enqueue = hvs_notify_send_post_enqueue,
- .set_buffer_size = hvs_set_buffer_size,
- .set_min_buffer_size = hvs_set_min_buffer_size,
- .set_max_buffer_size = hvs_set_max_buffer_size,
- .get_buffer_size = hvs_get_buffer_size,
- .get_min_buffer_size = hvs_get_min_buffer_size,
- .get_max_buffer_size = hvs_get_max_buffer_size,
};
static int hvs_probe(struct hv_device *hdev,
diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c
index 3756f0857946..fb1fc7760e8c 100644
--- a/net/vmw_vsock/virtio_transport.c
+++ b/net/vmw_vsock/virtio_transport.c
@@ -494,13 +494,7 @@ static struct virtio_transport virtio_transport = {
.notify_send_pre_block = virtio_transport_notify_send_pre_block,
.notify_send_pre_enqueue = virtio_transport_notify_send_pre_enqueue,
.notify_send_post_enqueue = virtio_transport_notify_send_post_enqueue,
-
- .set_buffer_size = virtio_transport_set_buffer_size,
- .set_min_buffer_size = virtio_transport_set_min_buffer_size,
- .set_max_buffer_size = virtio_transport_set_max_buffer_size,
- .get_buffer_size = virtio_transport_get_buffer_size,
- .get_min_buffer_size = virtio_transport_get_min_buffer_size,
- .get_max_buffer_size = virtio_transport_get_max_buffer_size,
+ .notify_buffer_size = virtio_transport_notify_buffer_size,
},
.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 37a1c7e7c7fe..b2a310dfa158 100644
--- a/net/vmw_vsock/virtio_transport_common.c
+++ b/net/vmw_vsock/virtio_transport_common.c
@@ -456,17 +456,13 @@ int virtio_transport_do_socket_init(struct vsock_sock *vsk,
if (psk) {
struct virtio_vsock_sock *ptrans = psk->trans;
- vvs->buf_size = ptrans->buf_size;
- vvs->buf_size_min = ptrans->buf_size_min;
- vvs->buf_size_max = ptrans->buf_size_max;
vvs->peer_buf_alloc = ptrans->peer_buf_alloc;
- } else {
- vvs->buf_size = VIRTIO_VSOCK_DEFAULT_BUF_SIZE;
- vvs->buf_size_min = VIRTIO_VSOCK_DEFAULT_MIN_BUF_SIZE;
- vvs->buf_size_max = VIRTIO_VSOCK_DEFAULT_MAX_BUF_SIZE;
}
- vvs->buf_alloc = vvs->buf_size;
+ if (vsk->buffer_size > VIRTIO_VSOCK_MAX_BUF_SIZE)
+ vsk->buffer_size = VIRTIO_VSOCK_MAX_BUF_SIZE;
+
+ vvs->buf_alloc = vsk->buffer_size;
spin_lock_init(&vvs->rx_lock);
spin_lock_init(&vvs->tx_lock);
@@ -476,71 +472,20 @@ int virtio_transport_do_socket_init(struct vsock_sock *vsk,
}
EXPORT_SYMBOL_GPL(virtio_transport_do_socket_init);
-u64 virtio_transport_get_buffer_size(struct vsock_sock *vsk)
-{
- struct virtio_vsock_sock *vvs = vsk->trans;
-
- return vvs->buf_size;
-}
-EXPORT_SYMBOL_GPL(virtio_transport_get_buffer_size);
-
-u64 virtio_transport_get_min_buffer_size(struct vsock_sock *vsk)
+/* sk_lock held by the caller */
+void virtio_transport_notify_buffer_size(struct vsock_sock *vsk, u64 *val)
{
struct virtio_vsock_sock *vvs = vsk->trans;
- return vvs->buf_size_min;
-}
-EXPORT_SYMBOL_GPL(virtio_transport_get_min_buffer_size);
-
-u64 virtio_transport_get_max_buffer_size(struct vsock_sock *vsk)
-{
- struct virtio_vsock_sock *vvs = vsk->trans;
-
- return vvs->buf_size_max;
-}
-EXPORT_SYMBOL_GPL(virtio_transport_get_max_buffer_size);
-
-void virtio_transport_set_buffer_size(struct vsock_sock *vsk, u64 val)
-{
- struct virtio_vsock_sock *vvs = vsk->trans;
+ if (*val > VIRTIO_VSOCK_MAX_BUF_SIZE)
+ *val = VIRTIO_VSOCK_MAX_BUF_SIZE;
- if (val > VIRTIO_VSOCK_MAX_BUF_SIZE)
- val = VIRTIO_VSOCK_MAX_BUF_SIZE;
- if (val < vvs->buf_size_min)
- vvs->buf_size_min = val;
- if (val > vvs->buf_size_max)
- vvs->buf_size_max = val;
- vvs->buf_size = val;
- vvs->buf_alloc = val;
+ vvs->buf_alloc = *val;
virtio_transport_send_credit_update(vsk, VIRTIO_VSOCK_TYPE_STREAM,
NULL);
}
-EXPORT_SYMBOL_GPL(virtio_transport_set_buffer_size);
-
-void virtio_transport_set_min_buffer_size(struct vsock_sock *vsk, u64 val)
-{
- struct virtio_vsock_sock *vvs = vsk->trans;
-
- if (val > VIRTIO_VSOCK_MAX_BUF_SIZE)
- val = VIRTIO_VSOCK_MAX_BUF_SIZE;
- if (val > vvs->buf_size)
- vvs->buf_size = val;
- vvs->buf_size_min = val;
-}
-EXPORT_SYMBOL_GPL(virtio_transport_set_min_buffer_size);
-
-void virtio_transport_set_max_buffer_size(struct vsock_sock *vsk, u64 val)
-{
- struct virtio_vsock_sock *vvs = vsk->trans;
-
- if (val > VIRTIO_VSOCK_MAX_BUF_SIZE)
- val = VIRTIO_VSOCK_MAX_BUF_SIZE;
- if (val < vvs->buf_size)
- vvs->buf_size = val;
- vvs->buf_size_max = val;
-}
-EXPORT_SYMBOL_GPL(virtio_transport_set_max_buffer_size);
+EXPORT_SYMBOL_GPL(virtio_transport_notify_buffer_size);
int
virtio_transport_notify_poll_in(struct vsock_sock *vsk,
@@ -632,9 +577,7 @@ EXPORT_SYMBOL_GPL(virtio_transport_notify_send_post_enqueue);
u64 virtio_transport_stream_rcvhiwat(struct vsock_sock *vsk)
{
- struct virtio_vsock_sock *vvs = vsk->trans;
-
- return vvs->buf_size;
+ return vsk->buffer_size;
}
EXPORT_SYMBOL_GPL(virtio_transport_stream_rcvhiwat);
diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c
index f8e3131ac480..8290d37b6587 100644
--- a/net/vmw_vsock/vmci_transport.c
+++ b/net/vmw_vsock/vmci_transport.c
@@ -74,10 +74,6 @@ static u32 vmci_transport_qp_resumed_sub_id = VMCI_INVALID_ID;
static int PROTOCOL_OVERRIDE = -1;
-#define VMCI_TRANSPORT_DEFAULT_QP_SIZE_MIN 128
-#define VMCI_TRANSPORT_DEFAULT_QP_SIZE 262144
-#define VMCI_TRANSPORT_DEFAULT_QP_SIZE_MAX 262144
-
/* Helper function to convert from a VMCI error code to a VSock error code. */
static s32 vmci_transport_error_to_vsock_error(s32 vmci_error)
@@ -1025,11 +1021,11 @@ static int vmci_transport_recv_listen(struct sock *sk,
/* If the proposed size fits within our min/max, accept it. Otherwise
* propose our own size.
*/
- if (pkt->u.size >= vmci_trans(vpending)->queue_pair_min_size &&
- pkt->u.size <= vmci_trans(vpending)->queue_pair_max_size) {
+ if (pkt->u.size >= vpending->buffer_min_size &&
+ pkt->u.size <= vpending->buffer_max_size) {
qp_size = pkt->u.size;
} else {
- qp_size = vmci_trans(vpending)->queue_pair_size;
+ qp_size = vpending->buffer_size;
}
/* Figure out if we are using old or new requests based on the
@@ -1098,7 +1094,7 @@ static int vmci_transport_recv_listen(struct sock *sk,
pending->sk_state = TCP_SYN_SENT;
vmci_trans(vpending)->produce_size =
vmci_trans(vpending)->consume_size = qp_size;
- vmci_trans(vpending)->queue_pair_size = qp_size;
+ vpending->buffer_size = qp_size;
vmci_trans(vpending)->notify_ops->process_request(pending);
@@ -1392,8 +1388,8 @@ static int vmci_transport_recv_connecting_client_negotiate(
vsk->ignore_connecting_rst = false;
/* Verify that we're OK with the proposed queue pair size */
- if (pkt->u.size < vmci_trans(vsk)->queue_pair_min_size ||
- pkt->u.size > vmci_trans(vsk)->queue_pair_max_size) {
+ if (pkt->u.size < vsk->buffer_min_size ||
+ pkt->u.size > vsk->buffer_max_size) {
err = -EINVAL;
goto destroy;
}
@@ -1498,8 +1494,7 @@ vmci_transport_recv_connecting_client_invalid(struct sock *sk,
vsk->sent_request = false;
vsk->ignore_connecting_rst = true;
- err = vmci_transport_send_conn_request(
- sk, vmci_trans(vsk)->queue_pair_size);
+ err = vmci_transport_send_conn_request(sk, vsk->buffer_size);
if (err < 0)
err = vmci_transport_error_to_vsock_error(err);
else
@@ -1583,21 +1578,6 @@ static int vmci_transport_socket_init(struct vsock_sock *vsk,
INIT_LIST_HEAD(&vmci_trans(vsk)->elem);
vmci_trans(vsk)->sk = &vsk->sk;
spin_lock_init(&vmci_trans(vsk)->lock);
- if (psk) {
- vmci_trans(vsk)->queue_pair_size =
- vmci_trans(psk)->queue_pair_size;
- vmci_trans(vsk)->queue_pair_min_size =
- vmci_trans(psk)->queue_pair_min_size;
- vmci_trans(vsk)->queue_pair_max_size =
- vmci_trans(psk)->queue_pair_max_size;
- } else {
- vmci_trans(vsk)->queue_pair_size =
- VMCI_TRANSPORT_DEFAULT_QP_SIZE;
- vmci_trans(vsk)->queue_pair_min_size =
- VMCI_TRANSPORT_DEFAULT_QP_SIZE_MIN;
- vmci_trans(vsk)->queue_pair_max_size =
- VMCI_TRANSPORT_DEFAULT_QP_SIZE_MAX;
- }
return 0;
}
@@ -1813,8 +1793,7 @@ static int vmci_transport_connect(struct vsock_sock *vsk)
if (vmci_transport_old_proto_override(&old_pkt_proto) &&
old_pkt_proto) {
- err = vmci_transport_send_conn_request(
- sk, vmci_trans(vsk)->queue_pair_size);
+ err = vmci_transport_send_conn_request(sk, vsk->buffer_size);
if (err < 0) {
sk->sk_state = TCP_CLOSE;
return err;
@@ -1822,8 +1801,7 @@ static int vmci_transport_connect(struct vsock_sock *vsk)
} else {
int supported_proto_versions =
vmci_transport_new_proto_supported_versions();
- err = vmci_transport_send_conn_request2(
- sk, vmci_trans(vsk)->queue_pair_size,
+ err = vmci_transport_send_conn_request2(sk, vsk->buffer_size,
supported_proto_versions);
if (err < 0) {
sk->sk_state = TCP_CLOSE;
@@ -1876,46 +1854,6 @@ static bool vmci_transport_stream_is_active(struct vsock_sock *vsk)
return !vmci_handle_is_invalid(vmci_trans(vsk)->qp_handle);
}
-static u64 vmci_transport_get_buffer_size(struct vsock_sock *vsk)
-{
- return vmci_trans(vsk)->queue_pair_size;
-}
-
-static u64 vmci_transport_get_min_buffer_size(struct vsock_sock *vsk)
-{
- return vmci_trans(vsk)->queue_pair_min_size;
-}
-
-static u64 vmci_transport_get_max_buffer_size(struct vsock_sock *vsk)
-{
- return vmci_trans(vsk)->queue_pair_max_size;
-}
-
-static void vmci_transport_set_buffer_size(struct vsock_sock *vsk, u64 val)
-{
- if (val < vmci_trans(vsk)->queue_pair_min_size)
- vmci_trans(vsk)->queue_pair_min_size = val;
- if (val > vmci_trans(vsk)->queue_pair_max_size)
- vmci_trans(vsk)->queue_pair_max_size = val;
- vmci_trans(vsk)->queue_pair_size = val;
-}
-
-static void vmci_transport_set_min_buffer_size(struct vsock_sock *vsk,
- u64 val)
-{
- if (val > vmci_trans(vsk)->queue_pair_size)
- vmci_trans(vsk)->queue_pair_size = val;
- vmci_trans(vsk)->queue_pair_min_size = val;
-}
-
-static void vmci_transport_set_max_buffer_size(struct vsock_sock *vsk,
- u64 val)
-{
- if (val < vmci_trans(vsk)->queue_pair_size)
- vmci_trans(vsk)->queue_pair_size = val;
- vmci_trans(vsk)->queue_pair_max_size = val;
-}
-
static int vmci_transport_notify_poll_in(
struct vsock_sock *vsk,
size_t target,
@@ -2098,12 +2036,6 @@ static const struct vsock_transport vmci_transport = {
.notify_send_pre_enqueue = vmci_transport_notify_send_pre_enqueue,
.notify_send_post_enqueue = vmci_transport_notify_send_post_enqueue,
.shutdown = vmci_transport_shutdown,
- .set_buffer_size = vmci_transport_set_buffer_size,
- .set_min_buffer_size = vmci_transport_set_min_buffer_size,
- .set_max_buffer_size = vmci_transport_set_max_buffer_size,
- .get_buffer_size = vmci_transport_get_buffer_size,
- .get_min_buffer_size = vmci_transport_get_min_buffer_size,
- .get_max_buffer_size = vmci_transport_get_max_buffer_size,
.get_local_cid = vmci_transport_get_local_cid,
};
diff --git a/net/vmw_vsock/vmci_transport.h b/net/vmw_vsock/vmci_transport.h
index 1ca1e8640b31..b7b072194282 100644
--- a/net/vmw_vsock/vmci_transport.h
+++ b/net/vmw_vsock/vmci_transport.h
@@ -108,9 +108,6 @@ struct vmci_transport {
struct vmci_qp *qpair;
u64 produce_size;
u64 consume_size;
- u64 queue_pair_size;
- u64 queue_pair_min_size;
- u64 queue_pair_max_size;
u32 detach_sub_id;
union vmci_transport_notify notify;
const struct vmci_transport_notify_ops *notify_ops;
--
2.21.0
^ permalink raw reply related
* [PATCH net-next 06/14] vsock: add 'struct vsock_sock *' param to vsock_core_get_transport()
From: Stefano Garzarella @ 2019-10-23 9:55 UTC (permalink / raw)
To: netdev
Cc: Michael S. Tsirkin, kvm, Greg Kroah-Hartman, Jason Wang,
David S. Miller, Dexuan Cui, Haiyang Zhang, Jorgen Hansen,
Sasha Levin, linux-kernel, Arnd Bergmann, Stefan Hajnoczi,
linux-hyperv, K. Y. Srinivasan, Stephen Hemminger, virtualization
In-Reply-To: <20191023095554.11340-1-sgarzare@redhat.com>
Since now the 'struct vsock_sock' object contains a pointer to
the transport, this patch adds a parameter to the
vsock_core_get_transport() to return the right transport
assigned to the socket.
This patch modifies also the virtio_transport_get_ops(), that
uses the vsock_core_get_transport(), adding the
'struct vsock_sock *' parameter.
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
---
RFC -> v1:
- Removed comment about protecting transport_single (Stefan)
---
include/net/af_vsock.h | 2 +-
net/vmw_vsock/af_vsock.c | 7 ++-----
net/vmw_vsock/virtio_transport_common.c | 9 +++++----
3 files changed, 8 insertions(+), 10 deletions(-)
diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h
index a5e1e134261d..2ca67d048de4 100644
--- a/include/net/af_vsock.h
+++ b/include/net/af_vsock.h
@@ -166,7 +166,7 @@ static inline int vsock_core_init(const struct vsock_transport *t)
void vsock_core_exit(void);
/* The transport may downcast this to access transport-specific functions */
-const struct vsock_transport *vsock_core_get_transport(void);
+const struct vsock_transport *vsock_core_get_transport(struct vsock_sock *vsk);
/**** UTILS ****/
diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
index c3a14f853eb0..eaea159006c8 100644
--- a/net/vmw_vsock/af_vsock.c
+++ b/net/vmw_vsock/af_vsock.c
@@ -2001,12 +2001,9 @@ void vsock_core_exit(void)
}
EXPORT_SYMBOL_GPL(vsock_core_exit);
-const struct vsock_transport *vsock_core_get_transport(void)
+const struct vsock_transport *vsock_core_get_transport(struct vsock_sock *vsk)
{
- /* vsock_register_mutex not taken since only the transport uses this
- * function and only while registered.
- */
- return transport_single;
+ return vsk->transport;
}
EXPORT_SYMBOL_GPL(vsock_core_get_transport);
diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c
index 9763394f7a61..37a1c7e7c7fe 100644
--- a/net/vmw_vsock/virtio_transport_common.c
+++ b/net/vmw_vsock/virtio_transport_common.c
@@ -29,9 +29,10 @@
/* Threshold for detecting small packets to copy */
#define GOOD_COPY_LEN 128
-static const struct virtio_transport *virtio_transport_get_ops(void)
+static const struct virtio_transport *
+virtio_transport_get_ops(struct vsock_sock *vsk)
{
- const struct vsock_transport *t = vsock_core_get_transport();
+ const struct vsock_transport *t = vsock_core_get_transport(vsk);
return container_of(t, struct virtio_transport, transport);
}
@@ -168,7 +169,7 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk,
struct virtio_vsock_pkt *pkt;
u32 pkt_len = info->pkt_len;
- src_cid = virtio_transport_get_ops()->transport.get_local_cid();
+ src_cid = virtio_transport_get_ops(vsk)->transport.get_local_cid();
src_port = vsk->local_addr.svm_port;
if (!info->remote_cid) {
dst_cid = vsk->remote_addr.svm_cid;
@@ -201,7 +202,7 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk,
virtio_transport_inc_tx_pkt(vvs, pkt);
- return virtio_transport_get_ops()->send_pkt(pkt);
+ return virtio_transport_get_ops(vsk)->send_pkt(pkt);
}
static bool virtio_transport_inc_rx_pkt(struct virtio_vsock_sock *vvs,
--
2.21.0
^ permalink raw reply related
* [PATCH net-next 05/14] vsock/virtio: add transport parameter to the virtio_transport_reset_no_sock()
From: Stefano Garzarella @ 2019-10-23 9:55 UTC (permalink / raw)
To: netdev
Cc: Michael S. Tsirkin, kvm, Greg Kroah-Hartman, Jason Wang,
David S. Miller, Dexuan Cui, Haiyang Zhang, Jorgen Hansen,
Sasha Levin, linux-kernel, Arnd Bergmann, Stefan Hajnoczi,
linux-hyperv, K. Y. Srinivasan, Stephen Hemminger, virtualization
In-Reply-To: <20191023095554.11340-1-sgarzare@redhat.com>
We are going to add 'struct vsock_sock *' parameter to
virtio_transport_get_ops().
In some cases, like in the virtio_transport_reset_no_sock(),
we don't have any socket assigned to the packet received,
so we can't use the virtio_transport_get_ops().
In order to allow virtio_transport_reset_no_sock() to use the
'.send_pkt' callback from the 'vhost_transport' or 'virtio_transport',
we add the 'struct virtio_transport *' to it and to its caller:
virtio_transport_recv_pkt().
We moved the 'vhost_transport' and 'virtio_transport' definition,
to pass their address to the virtio_transport_recv_pkt().
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
---
drivers/vhost/vsock.c | 94 +++++++-------
include/linux/virtio_vsock.h | 3 +-
net/vmw_vsock/virtio_transport.c | 160 ++++++++++++------------
net/vmw_vsock/virtio_transport_common.c | 12 +-
4 files changed, 135 insertions(+), 134 deletions(-)
diff --git a/drivers/vhost/vsock.c b/drivers/vhost/vsock.c
index 9f57736fe15e..92ab3852c954 100644
--- a/drivers/vhost/vsock.c
+++ b/drivers/vhost/vsock.c
@@ -384,6 +384,52 @@ static bool vhost_vsock_more_replies(struct vhost_vsock *vsock)
return val < vq->num;
}
+static struct virtio_transport vhost_transport = {
+ .transport = {
+ .get_local_cid = vhost_transport_get_local_cid,
+
+ .init = virtio_transport_do_socket_init,
+ .destruct = virtio_transport_destruct,
+ .release = virtio_transport_release,
+ .connect = virtio_transport_connect,
+ .shutdown = virtio_transport_shutdown,
+ .cancel_pkt = vhost_transport_cancel_pkt,
+
+ .dgram_enqueue = virtio_transport_dgram_enqueue,
+ .dgram_dequeue = virtio_transport_dgram_dequeue,
+ .dgram_bind = virtio_transport_dgram_bind,
+ .dgram_allow = virtio_transport_dgram_allow,
+
+ .stream_enqueue = virtio_transport_stream_enqueue,
+ .stream_dequeue = virtio_transport_stream_dequeue,
+ .stream_has_data = virtio_transport_stream_has_data,
+ .stream_has_space = virtio_transport_stream_has_space,
+ .stream_rcvhiwat = virtio_transport_stream_rcvhiwat,
+ .stream_is_active = virtio_transport_stream_is_active,
+ .stream_allow = virtio_transport_stream_allow,
+
+ .notify_poll_in = virtio_transport_notify_poll_in,
+ .notify_poll_out = virtio_transport_notify_poll_out,
+ .notify_recv_init = virtio_transport_notify_recv_init,
+ .notify_recv_pre_block = virtio_transport_notify_recv_pre_block,
+ .notify_recv_pre_dequeue = virtio_transport_notify_recv_pre_dequeue,
+ .notify_recv_post_dequeue = virtio_transport_notify_recv_post_dequeue,
+ .notify_send_init = virtio_transport_notify_send_init,
+ .notify_send_pre_block = virtio_transport_notify_send_pre_block,
+ .notify_send_pre_enqueue = virtio_transport_notify_send_pre_enqueue,
+ .notify_send_post_enqueue = virtio_transport_notify_send_post_enqueue,
+
+ .set_buffer_size = virtio_transport_set_buffer_size,
+ .set_min_buffer_size = virtio_transport_set_min_buffer_size,
+ .set_max_buffer_size = virtio_transport_set_max_buffer_size,
+ .get_buffer_size = virtio_transport_get_buffer_size,
+ .get_min_buffer_size = virtio_transport_get_min_buffer_size,
+ .get_max_buffer_size = virtio_transport_get_max_buffer_size,
+ },
+
+ .send_pkt = vhost_transport_send_pkt,
+};
+
static void vhost_vsock_handle_tx_kick(struct vhost_work *work)
{
struct vhost_virtqueue *vq = container_of(work, struct vhost_virtqueue,
@@ -438,7 +484,7 @@ static void vhost_vsock_handle_tx_kick(struct vhost_work *work)
/* Only accept correctly addressed packets */
if (le64_to_cpu(pkt->hdr.src_cid) == vsock->guest_cid)
- virtio_transport_recv_pkt(pkt);
+ virtio_transport_recv_pkt(&vhost_transport, pkt);
else
virtio_transport_free_pkt(pkt);
@@ -786,52 +832,6 @@ static struct miscdevice vhost_vsock_misc = {
.fops = &vhost_vsock_fops,
};
-static struct virtio_transport vhost_transport = {
- .transport = {
- .get_local_cid = vhost_transport_get_local_cid,
-
- .init = virtio_transport_do_socket_init,
- .destruct = virtio_transport_destruct,
- .release = virtio_transport_release,
- .connect = virtio_transport_connect,
- .shutdown = virtio_transport_shutdown,
- .cancel_pkt = vhost_transport_cancel_pkt,
-
- .dgram_enqueue = virtio_transport_dgram_enqueue,
- .dgram_dequeue = virtio_transport_dgram_dequeue,
- .dgram_bind = virtio_transport_dgram_bind,
- .dgram_allow = virtio_transport_dgram_allow,
-
- .stream_enqueue = virtio_transport_stream_enqueue,
- .stream_dequeue = virtio_transport_stream_dequeue,
- .stream_has_data = virtio_transport_stream_has_data,
- .stream_has_space = virtio_transport_stream_has_space,
- .stream_rcvhiwat = virtio_transport_stream_rcvhiwat,
- .stream_is_active = virtio_transport_stream_is_active,
- .stream_allow = virtio_transport_stream_allow,
-
- .notify_poll_in = virtio_transport_notify_poll_in,
- .notify_poll_out = virtio_transport_notify_poll_out,
- .notify_recv_init = virtio_transport_notify_recv_init,
- .notify_recv_pre_block = virtio_transport_notify_recv_pre_block,
- .notify_recv_pre_dequeue = virtio_transport_notify_recv_pre_dequeue,
- .notify_recv_post_dequeue = virtio_transport_notify_recv_post_dequeue,
- .notify_send_init = virtio_transport_notify_send_init,
- .notify_send_pre_block = virtio_transport_notify_send_pre_block,
- .notify_send_pre_enqueue = virtio_transport_notify_send_pre_enqueue,
- .notify_send_post_enqueue = virtio_transport_notify_send_post_enqueue,
-
- .set_buffer_size = virtio_transport_set_buffer_size,
- .set_min_buffer_size = virtio_transport_set_min_buffer_size,
- .set_max_buffer_size = virtio_transport_set_max_buffer_size,
- .get_buffer_size = virtio_transport_get_buffer_size,
- .get_min_buffer_size = virtio_transport_get_min_buffer_size,
- .get_max_buffer_size = virtio_transport_get_max_buffer_size,
- },
-
- .send_pkt = vhost_transport_send_pkt,
-};
-
static int __init vhost_vsock_init(void)
{
int ret;
diff --git a/include/linux/virtio_vsock.h b/include/linux/virtio_vsock.h
index 4c7781f4b29b..96d8132acbd7 100644
--- a/include/linux/virtio_vsock.h
+++ b/include/linux/virtio_vsock.h
@@ -151,7 +151,8 @@ virtio_transport_dgram_enqueue(struct vsock_sock *vsk,
void virtio_transport_destruct(struct vsock_sock *vsk);
-void virtio_transport_recv_pkt(struct virtio_vsock_pkt *pkt);
+void virtio_transport_recv_pkt(struct virtio_transport *t,
+ struct virtio_vsock_pkt *pkt);
void virtio_transport_free_pkt(struct virtio_vsock_pkt *pkt);
void virtio_transport_inc_tx_pkt(struct virtio_vsock_sock *vvs, struct virtio_vsock_pkt *pkt);
u32 virtio_transport_get_credit(struct virtio_vsock_sock *vvs, u32 wanted);
diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c
index 082a30936690..3756f0857946 100644
--- a/net/vmw_vsock/virtio_transport.c
+++ b/net/vmw_vsock/virtio_transport.c
@@ -86,33 +86,6 @@ static u32 virtio_transport_get_local_cid(void)
return ret;
}
-static void virtio_transport_loopback_work(struct work_struct *work)
-{
- struct virtio_vsock *vsock =
- container_of(work, struct virtio_vsock, loopback_work);
- LIST_HEAD(pkts);
-
- spin_lock_bh(&vsock->loopback_list_lock);
- list_splice_init(&vsock->loopback_list, &pkts);
- spin_unlock_bh(&vsock->loopback_list_lock);
-
- mutex_lock(&vsock->rx_lock);
-
- if (!vsock->rx_run)
- goto out;
-
- while (!list_empty(&pkts)) {
- struct virtio_vsock_pkt *pkt;
-
- pkt = list_first_entry(&pkts, struct virtio_vsock_pkt, list);
- list_del_init(&pkt->list);
-
- virtio_transport_recv_pkt(pkt);
- }
-out:
- mutex_unlock(&vsock->rx_lock);
-}
-
static int virtio_transport_send_pkt_loopback(struct virtio_vsock *vsock,
struct virtio_vsock_pkt *pkt)
{
@@ -370,59 +343,6 @@ static bool virtio_transport_more_replies(struct virtio_vsock *vsock)
return val < virtqueue_get_vring_size(vq);
}
-static void virtio_transport_rx_work(struct work_struct *work)
-{
- struct virtio_vsock *vsock =
- container_of(work, struct virtio_vsock, rx_work);
- struct virtqueue *vq;
-
- vq = vsock->vqs[VSOCK_VQ_RX];
-
- mutex_lock(&vsock->rx_lock);
-
- if (!vsock->rx_run)
- goto out;
-
- do {
- virtqueue_disable_cb(vq);
- for (;;) {
- struct virtio_vsock_pkt *pkt;
- unsigned int len;
-
- if (!virtio_transport_more_replies(vsock)) {
- /* Stop rx until the device processes already
- * pending replies. Leave rx virtqueue
- * callbacks disabled.
- */
- goto out;
- }
-
- pkt = virtqueue_get_buf(vq, &len);
- if (!pkt) {
- break;
- }
-
- vsock->rx_buf_nr--;
-
- /* Drop short/long packets */
- if (unlikely(len < sizeof(pkt->hdr) ||
- len > sizeof(pkt->hdr) + pkt->len)) {
- virtio_transport_free_pkt(pkt);
- continue;
- }
-
- pkt->len = len - sizeof(pkt->hdr);
- virtio_transport_deliver_tap_pkt(pkt);
- virtio_transport_recv_pkt(pkt);
- }
- } while (!virtqueue_enable_cb(vq));
-
-out:
- if (vsock->rx_buf_nr < vsock->rx_buf_max_nr / 2)
- virtio_vsock_rx_fill(vsock);
- mutex_unlock(&vsock->rx_lock);
-}
-
/* event_lock must be held */
static int virtio_vsock_event_fill_one(struct virtio_vsock *vsock,
struct virtio_vsock_event *event)
@@ -586,6 +506,86 @@ static struct virtio_transport virtio_transport = {
.send_pkt = virtio_transport_send_pkt,
};
+static void virtio_transport_loopback_work(struct work_struct *work)
+{
+ struct virtio_vsock *vsock =
+ container_of(work, struct virtio_vsock, loopback_work);
+ LIST_HEAD(pkts);
+
+ spin_lock_bh(&vsock->loopback_list_lock);
+ list_splice_init(&vsock->loopback_list, &pkts);
+ spin_unlock_bh(&vsock->loopback_list_lock);
+
+ mutex_lock(&vsock->rx_lock);
+
+ if (!vsock->rx_run)
+ goto out;
+
+ while (!list_empty(&pkts)) {
+ struct virtio_vsock_pkt *pkt;
+
+ pkt = list_first_entry(&pkts, struct virtio_vsock_pkt, list);
+ list_del_init(&pkt->list);
+
+ virtio_transport_recv_pkt(&virtio_transport, pkt);
+ }
+out:
+ mutex_unlock(&vsock->rx_lock);
+}
+
+static void virtio_transport_rx_work(struct work_struct *work)
+{
+ struct virtio_vsock *vsock =
+ container_of(work, struct virtio_vsock, rx_work);
+ struct virtqueue *vq;
+
+ vq = vsock->vqs[VSOCK_VQ_RX];
+
+ mutex_lock(&vsock->rx_lock);
+
+ if (!vsock->rx_run)
+ goto out;
+
+ do {
+ virtqueue_disable_cb(vq);
+ for (;;) {
+ struct virtio_vsock_pkt *pkt;
+ unsigned int len;
+
+ if (!virtio_transport_more_replies(vsock)) {
+ /* Stop rx until the device processes already
+ * pending replies. Leave rx virtqueue
+ * callbacks disabled.
+ */
+ goto out;
+ }
+
+ pkt = virtqueue_get_buf(vq, &len);
+ if (!pkt) {
+ break;
+ }
+
+ vsock->rx_buf_nr--;
+
+ /* Drop short/long packets */
+ if (unlikely(len < sizeof(pkt->hdr) ||
+ len > sizeof(pkt->hdr) + pkt->len)) {
+ virtio_transport_free_pkt(pkt);
+ continue;
+ }
+
+ pkt->len = len - sizeof(pkt->hdr);
+ virtio_transport_deliver_tap_pkt(pkt);
+ virtio_transport_recv_pkt(&virtio_transport, pkt);
+ }
+ } while (!virtqueue_enable_cb(vq));
+
+out:
+ if (vsock->rx_buf_nr < vsock->rx_buf_max_nr / 2)
+ virtio_vsock_rx_fill(vsock);
+ mutex_unlock(&vsock->rx_lock);
+}
+
static int virtio_vsock_probe(struct virtio_device *vdev)
{
vq_callback_t *callbacks[] = {
diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c
index b1cd16ed66ea..9763394f7a61 100644
--- a/net/vmw_vsock/virtio_transport_common.c
+++ b/net/vmw_vsock/virtio_transport_common.c
@@ -745,9 +745,9 @@ static int virtio_transport_reset(struct vsock_sock *vsk,
/* Normally packets are associated with a socket. There may be no socket if an
* attempt was made to connect to a socket that does not exist.
*/
-static int virtio_transport_reset_no_sock(struct virtio_vsock_pkt *pkt)
+static int virtio_transport_reset_no_sock(const struct virtio_transport *t,
+ struct virtio_vsock_pkt *pkt)
{
- const struct virtio_transport *t;
struct virtio_vsock_pkt *reply;
struct virtio_vsock_pkt_info info = {
.op = VIRTIO_VSOCK_OP_RST,
@@ -767,7 +767,6 @@ static int virtio_transport_reset_no_sock(struct virtio_vsock_pkt *pkt)
if (!reply)
return -ENOMEM;
- t = virtio_transport_get_ops();
if (!t) {
virtio_transport_free_pkt(reply);
return -ENOTCONN;
@@ -1107,7 +1106,8 @@ static bool virtio_transport_space_update(struct sock *sk,
/* We are under the virtio-vsock's vsock->rx_lock or vhost-vsock's vq->mutex
* lock.
*/
-void virtio_transport_recv_pkt(struct virtio_vsock_pkt *pkt)
+void virtio_transport_recv_pkt(struct virtio_transport *t,
+ struct virtio_vsock_pkt *pkt)
{
struct sockaddr_vm src, dst;
struct vsock_sock *vsk;
@@ -1129,7 +1129,7 @@ void virtio_transport_recv_pkt(struct virtio_vsock_pkt *pkt)
le32_to_cpu(pkt->hdr.fwd_cnt));
if (le16_to_cpu(pkt->hdr.type) != VIRTIO_VSOCK_TYPE_STREAM) {
- (void)virtio_transport_reset_no_sock(pkt);
+ (void)virtio_transport_reset_no_sock(t, pkt);
goto free_pkt;
}
@@ -1140,7 +1140,7 @@ void virtio_transport_recv_pkt(struct virtio_vsock_pkt *pkt)
if (!sk) {
sk = vsock_find_bound_socket(&dst);
if (!sk) {
- (void)virtio_transport_reset_no_sock(pkt);
+ (void)virtio_transport_reset_no_sock(t, pkt);
goto free_pkt;
}
}
--
2.21.0
^ permalink raw reply related
* [PATCH net-next 04/14] vsock: add 'transport' member in the struct vsock_sock
From: Stefano Garzarella @ 2019-10-23 9:55 UTC (permalink / raw)
To: netdev
Cc: Michael S. Tsirkin, kvm, Greg Kroah-Hartman, Jason Wang,
David S. Miller, Dexuan Cui, Haiyang Zhang, Jorgen Hansen,
Sasha Levin, linux-kernel, Arnd Bergmann, Stefan Hajnoczi,
linux-hyperv, K. Y. Srinivasan, Stephen Hemminger, virtualization
In-Reply-To: <20191023095554.11340-1-sgarzare@redhat.com>
As a preparation to support multiple transports, this patch adds
the 'transport' member at the 'struct vsock_sock'.
This new field is initialized during the creation in the
__vsock_create() function.
This patch also renames the global 'transport' pointer to
'transport_single', since for now we're only supporting a single
transport registered at run-time.
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
---
include/net/af_vsock.h | 1 +
net/vmw_vsock/af_vsock.c | 56 +++++++++++++++++++++++++++-------------
2 files changed, 39 insertions(+), 18 deletions(-)
diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h
index c660402b10f2..a5e1e134261d 100644
--- a/include/net/af_vsock.h
+++ b/include/net/af_vsock.h
@@ -27,6 +27,7 @@ extern spinlock_t vsock_table_lock;
struct vsock_sock {
/* sk must be the first member. */
struct sock sk;
+ const struct vsock_transport *transport;
struct sockaddr_vm local_addr;
struct sockaddr_vm remote_addr;
/* Links for the global tables of bound and connected sockets. */
diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
index 2f2582fb7fdd..c3a14f853eb0 100644
--- a/net/vmw_vsock/af_vsock.c
+++ b/net/vmw_vsock/af_vsock.c
@@ -126,7 +126,7 @@ static struct proto vsock_proto = {
*/
#define VSOCK_DEFAULT_CONNECT_TIMEOUT (2 * HZ)
-static const struct vsock_transport *transport;
+static const struct vsock_transport *transport_single;
static DEFINE_MUTEX(vsock_register_mutex);
/**** UTILS ****/
@@ -408,7 +408,9 @@ static bool vsock_is_pending(struct sock *sk)
static int vsock_send_shutdown(struct sock *sk, int mode)
{
- return transport->shutdown(vsock_sk(sk), mode);
+ struct vsock_sock *vsk = vsock_sk(sk);
+
+ return vsk->transport->shutdown(vsk, mode);
}
static void vsock_pending_work(struct work_struct *work)
@@ -518,7 +520,7 @@ static int __vsock_bind_stream(struct vsock_sock *vsk,
static int __vsock_bind_dgram(struct vsock_sock *vsk,
struct sockaddr_vm *addr)
{
- return transport->dgram_bind(vsk, addr);
+ return vsk->transport->dgram_bind(vsk, addr);
}
static int __vsock_bind(struct sock *sk, struct sockaddr_vm *addr)
@@ -536,7 +538,7 @@ static int __vsock_bind(struct sock *sk, struct sockaddr_vm *addr)
* like AF_INET prevents binding to a non-local IP address (in most
* cases), we only allow binding to the local CID.
*/
- cid = transport->get_local_cid();
+ cid = vsk->transport->get_local_cid();
if (addr->svm_cid != cid && addr->svm_cid != VMADDR_CID_ANY)
return -EADDRNOTAVAIL;
@@ -586,6 +588,7 @@ struct sock *__vsock_create(struct net *net,
sk->sk_type = type;
vsk = vsock_sk(sk);
+ vsk->transport = transport_single;
vsock_addr_init(&vsk->local_addr, VMADDR_CID_ANY, VMADDR_PORT_ANY);
vsock_addr_init(&vsk->remote_addr, VMADDR_CID_ANY, VMADDR_PORT_ANY);
@@ -616,7 +619,7 @@ struct sock *__vsock_create(struct net *net,
vsk->connect_timeout = VSOCK_DEFAULT_CONNECT_TIMEOUT;
}
- if (transport->init(vsk, psk) < 0) {
+ if (vsk->transport->init(vsk, psk) < 0) {
sk_free(sk);
return NULL;
}
@@ -641,7 +644,7 @@ static void __vsock_release(struct sock *sk, int level)
/* The release call is supposed to use lock_sock_nested()
* rather than lock_sock(), if a sock lock should be acquired.
*/
- transport->release(vsk);
+ vsk->transport->release(vsk);
/* When "level" is SINGLE_DEPTH_NESTING, use the nested
* version to avoid the warning "possible recursive locking
@@ -670,7 +673,7 @@ static void vsock_sk_destruct(struct sock *sk)
{
struct vsock_sock *vsk = vsock_sk(sk);
- transport->destruct(vsk);
+ vsk->transport->destruct(vsk);
/* When clearing these addresses, there's no need to set the family and
* possibly register the address family with the kernel.
@@ -694,13 +697,13 @@ static int vsock_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
s64 vsock_stream_has_data(struct vsock_sock *vsk)
{
- return transport->stream_has_data(vsk);
+ return vsk->transport->stream_has_data(vsk);
}
EXPORT_SYMBOL_GPL(vsock_stream_has_data);
s64 vsock_stream_has_space(struct vsock_sock *vsk)
{
- return transport->stream_has_space(vsk);
+ return vsk->transport->stream_has_space(vsk);
}
EXPORT_SYMBOL_GPL(vsock_stream_has_space);
@@ -869,6 +872,7 @@ static __poll_t vsock_poll(struct file *file, struct socket *sock,
mask |= EPOLLOUT | EPOLLWRNORM | EPOLLWRBAND;
} else if (sock->type == SOCK_STREAM) {
+ const struct vsock_transport *transport = vsk->transport;
lock_sock(sk);
/* Listening sockets that have connections in their accept
@@ -944,6 +948,7 @@ static int vsock_dgram_sendmsg(struct socket *sock, struct msghdr *msg,
struct sock *sk;
struct vsock_sock *vsk;
struct sockaddr_vm *remote_addr;
+ const struct vsock_transport *transport;
if (msg->msg_flags & MSG_OOB)
return -EOPNOTSUPP;
@@ -952,6 +957,7 @@ static int vsock_dgram_sendmsg(struct socket *sock, struct msghdr *msg,
err = 0;
sk = sock->sk;
vsk = vsock_sk(sk);
+ transport = vsk->transport;
lock_sock(sk);
@@ -1036,8 +1042,8 @@ static int vsock_dgram_connect(struct socket *sock,
if (err)
goto out;
- if (!transport->dgram_allow(remote_addr->svm_cid,
- remote_addr->svm_port)) {
+ if (!vsk->transport->dgram_allow(remote_addr->svm_cid,
+ remote_addr->svm_port)) {
err = -EINVAL;
goto out;
}
@@ -1053,7 +1059,9 @@ static int vsock_dgram_connect(struct socket *sock,
static int vsock_dgram_recvmsg(struct 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 +1125,7 @@ static int vsock_stream_connect(struct socket *sock, struct sockaddr *addr,
int err;
struct sock *sk;
struct vsock_sock *vsk;
+ const struct vsock_transport *transport;
struct sockaddr_vm *remote_addr;
long timeout;
DEFINE_WAIT(wait);
@@ -1122,6 +1133,7 @@ static int vsock_stream_connect(struct socket *sock, struct sockaddr *addr,
err = 0;
sk = sock->sk;
vsk = vsock_sk(sk);
+ transport = vsk->transport;
lock_sock(sk);
@@ -1365,6 +1377,7 @@ static int vsock_stream_setsockopt(struct socket *sock,
int err;
struct sock *sk;
struct vsock_sock *vsk;
+ const struct vsock_transport *transport;
u64 val;
if (level != AF_VSOCK)
@@ -1385,6 +1398,7 @@ static int vsock_stream_setsockopt(struct socket *sock,
err = 0;
sk = sock->sk;
vsk = vsock_sk(sk);
+ transport = vsk->transport;
lock_sock(sk);
@@ -1442,6 +1456,7 @@ static int vsock_stream_getsockopt(struct socket *sock,
int len;
struct sock *sk;
struct vsock_sock *vsk;
+ const struct vsock_transport *transport;
u64 val;
if (level != AF_VSOCK)
@@ -1465,6 +1480,7 @@ static int vsock_stream_getsockopt(struct socket *sock,
err = 0;
sk = sock->sk;
vsk = vsock_sk(sk);
+ transport = vsk->transport;
switch (optname) {
case SO_VM_SOCKETS_BUFFER_SIZE:
@@ -1509,6 +1525,7 @@ static int vsock_stream_sendmsg(struct socket *sock, struct msghdr *msg,
{
struct sock *sk;
struct vsock_sock *vsk;
+ const struct vsock_transport *transport;
ssize_t total_written;
long timeout;
int err;
@@ -1517,6 +1534,7 @@ static int vsock_stream_sendmsg(struct socket *sock, struct msghdr *msg,
sk = sock->sk;
vsk = vsock_sk(sk);
+ transport = vsk->transport;
total_written = 0;
err = 0;
@@ -1648,6 +1666,7 @@ vsock_stream_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
{
struct sock *sk;
struct vsock_sock *vsk;
+ const struct vsock_transport *transport;
int err;
size_t target;
ssize_t copied;
@@ -1658,6 +1677,7 @@ vsock_stream_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
sk = sock->sk;
vsk = vsock_sk(sk);
+ transport = vsk->transport;
err = 0;
lock_sock(sk);
@@ -1872,7 +1892,7 @@ static long vsock_dev_do_ioctl(struct file *filp,
switch (cmd) {
case IOCTL_VM_SOCKETS_GET_LOCAL_CID:
- if (put_user(transport->get_local_cid(), p) != 0)
+ if (put_user(transport_single->get_local_cid(), p) != 0)
retval = -EFAULT;
break;
@@ -1919,7 +1939,7 @@ int __vsock_core_init(const struct vsock_transport *t, struct module *owner)
if (err)
return err;
- if (transport) {
+ if (transport_single) {
err = -EBUSY;
goto err_busy;
}
@@ -1928,7 +1948,7 @@ int __vsock_core_init(const struct vsock_transport *t, struct module *owner)
* unload while there are open sockets.
*/
vsock_proto.owner = owner;
- transport = t;
+ transport_single = t;
vsock_device.minor = MISC_DYNAMIC_MINOR;
err = misc_register(&vsock_device);
@@ -1958,7 +1978,7 @@ int __vsock_core_init(const struct vsock_transport *t, struct module *owner)
err_deregister_misc:
misc_deregister(&vsock_device);
err_reset_transport:
- transport = NULL;
+ transport_single = NULL;
err_busy:
mutex_unlock(&vsock_register_mutex);
return err;
@@ -1975,7 +1995,7 @@ void vsock_core_exit(void)
/* We do not want the assignment below re-ordered. */
mb();
- transport = NULL;
+ transport_single = NULL;
mutex_unlock(&vsock_register_mutex);
}
@@ -1986,7 +2006,7 @@ const struct vsock_transport *vsock_core_get_transport(void)
/* vsock_register_mutex not taken since only the transport uses this
* function and only while registered.
*/
- return transport;
+ return transport_single;
}
EXPORT_SYMBOL_GPL(vsock_core_get_transport);
--
2.21.0
^ permalink raw reply related
* [PATCH net-next 03/14] vsock: remove include/linux/vm_sockets.h file
From: Stefano Garzarella @ 2019-10-23 9:55 UTC (permalink / raw)
To: netdev
Cc: Michael S. Tsirkin, kvm, Greg Kroah-Hartman, Jason Wang,
David S. Miller, Dexuan Cui, Haiyang Zhang, Jorgen Hansen,
Sasha Levin, linux-kernel, Arnd Bergmann, Stefan Hajnoczi,
linux-hyperv, K. Y. Srinivasan, Stephen Hemminger, virtualization
In-Reply-To: <20191023095554.11340-1-sgarzare@redhat.com>
This header file now only includes the "uapi/linux/vm_sockets.h".
We can include directly it when needed.
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
---
include/linux/vm_sockets.h | 13 -------------
include/net/af_vsock.h | 2 +-
include/net/vsock_addr.h | 2 +-
net/vmw_vsock/vmci_transport_notify.h | 1 -
4 files changed, 2 insertions(+), 16 deletions(-)
delete mode 100644 include/linux/vm_sockets.h
diff --git a/include/linux/vm_sockets.h b/include/linux/vm_sockets.h
deleted file mode 100644
index 7dd899ccb920..000000000000
--- a/include/linux/vm_sockets.h
+++ /dev/null
@@ -1,13 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0-only */
-/*
- * VMware vSockets Driver
- *
- * Copyright (C) 2007-2013 VMware, Inc. All rights reserved.
- */
-
-#ifndef _VM_SOCKETS_H
-#define _VM_SOCKETS_H
-
-#include <uapi/linux/vm_sockets.h>
-
-#endif /* _VM_SOCKETS_H */
diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h
index 80ea0f93d3f7..c660402b10f2 100644
--- a/include/net/af_vsock.h
+++ b/include/net/af_vsock.h
@@ -10,7 +10,7 @@
#include <linux/kernel.h>
#include <linux/workqueue.h>
-#include <linux/vm_sockets.h>
+#include <uapi/linux/vm_sockets.h>
#include "vsock_addr.h"
diff --git a/include/net/vsock_addr.h b/include/net/vsock_addr.h
index 57d2db5c4bdf..cf8cc140d68d 100644
--- a/include/net/vsock_addr.h
+++ b/include/net/vsock_addr.h
@@ -8,7 +8,7 @@
#ifndef _VSOCK_ADDR_H_
#define _VSOCK_ADDR_H_
-#include <linux/vm_sockets.h>
+#include <uapi/linux/vm_sockets.h>
void vsock_addr_init(struct sockaddr_vm *addr, u32 cid, u32 port);
int vsock_addr_validate(const struct sockaddr_vm *addr);
diff --git a/net/vmw_vsock/vmci_transport_notify.h b/net/vmw_vsock/vmci_transport_notify.h
index 7843f08d4290..a1aa5a998c0e 100644
--- a/net/vmw_vsock/vmci_transport_notify.h
+++ b/net/vmw_vsock/vmci_transport_notify.h
@@ -11,7 +11,6 @@
#include <linux/types.h>
#include <linux/vmw_vmci_defs.h>
#include <linux/vmw_vmci_api.h>
-#include <linux/vm_sockets.h>
#include "vmci_transport.h"
--
2.21.0
^ permalink raw reply related
* [PATCH net-next 02/14] vsock: remove vm_sockets_get_local_cid()
From: Stefano Garzarella @ 2019-10-23 9:55 UTC (permalink / raw)
To: netdev
Cc: Michael S. Tsirkin, kvm, Greg Kroah-Hartman, Jason Wang,
David S. Miller, Dexuan Cui, Haiyang Zhang, Jorgen Hansen,
Sasha Levin, linux-kernel, Arnd Bergmann, Stefan Hajnoczi,
linux-hyperv, K. Y. Srinivasan, Stephen Hemminger, virtualization
In-Reply-To: <20191023095554.11340-1-sgarzare@redhat.com>
vm_sockets_get_local_cid() is only used in virtio_transport_common.c.
We can replace it calling the virtio_transport_get_ops() and
using the get_local_cid() callback registered by the transport.
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
---
include/linux/vm_sockets.h | 2 --
net/vmw_vsock/af_vsock.c | 10 ----------
net/vmw_vsock/virtio_transport_common.c | 2 +-
3 files changed, 1 insertion(+), 13 deletions(-)
diff --git a/include/linux/vm_sockets.h b/include/linux/vm_sockets.h
index 33f1a2ecd905..7dd899ccb920 100644
--- a/include/linux/vm_sockets.h
+++ b/include/linux/vm_sockets.h
@@ -10,6 +10,4 @@
#include <uapi/linux/vm_sockets.h>
-int vm_sockets_get_local_cid(void);
-
#endif /* _VM_SOCKETS_H */
diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
index 2ab43b2bba31..2f2582fb7fdd 100644
--- a/net/vmw_vsock/af_vsock.c
+++ b/net/vmw_vsock/af_vsock.c
@@ -129,16 +129,6 @@ static struct proto vsock_proto = {
static const struct vsock_transport *transport;
static DEFINE_MUTEX(vsock_register_mutex);
-/**** EXPORTS ****/
-
-/* Get the ID of the local context. This is transport dependent. */
-
-int vm_sockets_get_local_cid(void)
-{
- return transport->get_local_cid();
-}
-EXPORT_SYMBOL_GPL(vm_sockets_get_local_cid);
-
/**** UTILS ****/
/* Each bound VSocket is stored in the bind hash table and each connected
diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c
index d02c9b41a768..b1cd16ed66ea 100644
--- a/net/vmw_vsock/virtio_transport_common.c
+++ b/net/vmw_vsock/virtio_transport_common.c
@@ -168,7 +168,7 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk,
struct virtio_vsock_pkt *pkt;
u32 pkt_len = info->pkt_len;
- src_cid = vm_sockets_get_local_cid();
+ src_cid = virtio_transport_get_ops()->transport.get_local_cid();
src_port = vsk->local_addr.svm_port;
if (!info->remote_cid) {
dst_cid = vsk->remote_addr.svm_cid;
--
2.21.0
^ permalink raw reply related
* [PATCH net-next 01/14] vsock/vmci: remove unused VSOCK_DEFAULT_CONNECT_TIMEOUT
From: Stefano Garzarella @ 2019-10-23 9:55 UTC (permalink / raw)
To: netdev
Cc: Michael S. Tsirkin, kvm, Greg Kroah-Hartman, Jason Wang,
David S. Miller, Dexuan Cui, Haiyang Zhang, Jorgen Hansen,
Sasha Levin, linux-kernel, Arnd Bergmann, Stefan Hajnoczi,
linux-hyperv, K. Y. Srinivasan, Stephen Hemminger, virtualization
In-Reply-To: <20191023095554.11340-1-sgarzare@redhat.com>
The VSOCK_DEFAULT_CONNECT_TIMEOUT definition was introduced with
commit d021c344051af ("VSOCK: Introduce VM Sockets"), but it is
never used in the net/vmw_vsock/vmci_transport.c.
VSOCK_DEFAULT_CONNECT_TIMEOUT is used and defined in
net/vmw_vsock/af_vsock.c
Cc: Jorgen Hansen <jhansen@vmware.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
---
net/vmw_vsock/vmci_transport.c | 5 -----
1 file changed, 5 deletions(-)
diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c
index 8c9c4ed90fa7..f8e3131ac480 100644
--- a/net/vmw_vsock/vmci_transport.c
+++ b/net/vmw_vsock/vmci_transport.c
@@ -78,11 +78,6 @@ static int PROTOCOL_OVERRIDE = -1;
#define VMCI_TRANSPORT_DEFAULT_QP_SIZE 262144
#define VMCI_TRANSPORT_DEFAULT_QP_SIZE_MAX 262144
-/* The default peer timeout indicates how long we will wait for a peer response
- * to a control message.
- */
-#define VSOCK_DEFAULT_CONNECT_TIMEOUT (2 * HZ)
-
/* Helper function to convert from a VMCI error code to a VSock error code. */
static s32 vmci_transport_error_to_vsock_error(s32 vmci_error)
--
2.21.0
^ permalink raw reply related
* [PATCH net-next 00/14] vsock: add multi-transports support
From: Stefano Garzarella @ 2019-10-23 9:55 UTC (permalink / raw)
To: netdev
Cc: Michael S. Tsirkin, kvm, Greg Kroah-Hartman, Jason Wang,
David S. Miller, Dexuan Cui, Haiyang Zhang, Jorgen Hansen,
Sasha Levin, linux-kernel, Arnd Bergmann, Stefan Hajnoczi,
linux-hyperv, K. Y. Srinivasan, Stephen Hemminger, virtualization
This series adds the multi-transports support to vsock, following
this proposal: https://www.spinics.net/lists/netdev/msg575792.html
With the multi-transports support, we can use VSOCK with nested VMs
(using also different hypervisors) loading both guest->host and
host->guest transports at the same time.
Before this series, vmci-transport supported this behavior but only
using VMware hypervisor on L0, L1, etc.
RFC: https://patchwork.ozlabs.org/cover/1168442/
RFC -> v1:
- Added R-b/A-b from Dexuan and Stefan
- Fixed comments and typos in several patches (Stefan)
- Patch 7: changed .notify_buffer_size return to void (Stefan)
- Added patch 8 to simplify the API exposed to the transports (Stefan)
- Patch 11:
+ documented VSOCK_TRANSPORT_F_* flags (Stefan)
+ fixed vsock_assign_transport() when the socket is already assigned
+ moved features outside of struct vsock_transport, and used as
parameter of vsock_core_register() as a preparation of Patch 12
- Removed "vsock: add 'transport_hg' to handle g2h\h2g transports" patch
- Added patch 12 to register vmci_transport only when VMCI guest/host
are active
The first 9 patches are cleanups and preparations, maybe some of
these can go regardless of this series.
Patch 10 changes the hvs_remote_addr_init(). setting the
VMADDR_CID_HOST as remote CID instead of VMADDR_CID_ANY to make
the choice of transport to be used work properly.
Patch 11 adds multi-transports support.
Patch 12 touch a little bit the vmci_transport and the vmci driver
to register the vmci_transport only when there are active host/guest.
Patch 13 prevents the transport modules unloading while sockets are
assigned to them.
Patch 14 fixes an issue in the bind() logic discoverable only with
the new multi-transport support.
I've tested this series with nested KVM (vsock-transport [L0,L1],
virtio-transport[L1,L2]) and with VMware (L0) + KVM (L1)
(vmci-transport [L0,L1], vhost-transport [L1], virtio-transport[L2]).
Dexuan successfully tested the RFC series on HyperV with a Linux guest.
Stefano Garzarella (14):
vsock/vmci: remove unused VSOCK_DEFAULT_CONNECT_TIMEOUT
vsock: remove vm_sockets_get_local_cid()
vsock: remove include/linux/vm_sockets.h file
vsock: add 'transport' member in the struct vsock_sock
vsock/virtio: add transport parameter to the
virtio_transport_reset_no_sock()
vsock: add 'struct vsock_sock *' param to vsock_core_get_transport()
vsock: handle buffer_size sockopts in the core
vsock: add vsock_create_connected() called by transports
vsock: move vsock_insert_unbound() in the vsock_create()
hv_sock: set VMADDR_CID_HOST in the hvs_remote_addr_init()
vsock: add multi-transports support
vsock/vmci: register vmci_transport only when VMCI guest/host are
active
vsock: prevent transport modules unloading
vsock: fix bind() behaviour taking care of CID
drivers/misc/vmw_vmci/vmci_driver.c | 50 ++++
drivers/misc/vmw_vmci/vmci_driver.h | 2 +
drivers/misc/vmw_vmci/vmci_guest.c | 2 +
drivers/misc/vmw_vmci/vmci_host.c | 7 +
drivers/vhost/vsock.c | 96 +++---
include/linux/virtio_vsock.h | 18 +-
include/linux/vm_sockets.h | 15 -
include/linux/vmw_vmci_api.h | 2 +
include/net/af_vsock.h | 44 +--
include/net/vsock_addr.h | 2 +-
net/vmw_vsock/af_vsock.c | 376 ++++++++++++++++++------
net/vmw_vsock/hyperv_transport.c | 70 ++---
net/vmw_vsock/virtio_transport.c | 177 ++++++-----
net/vmw_vsock/virtio_transport_common.c | 131 +++------
net/vmw_vsock/vmci_transport.c | 137 +++------
net/vmw_vsock/vmci_transport.h | 3 -
net/vmw_vsock/vmci_transport_notify.h | 1 -
17 files changed, 627 insertions(+), 506 deletions(-)
delete mode 100644 include/linux/vm_sockets.h
--
2.21.0
^ permalink raw reply
* Re: [PATCH] video: hyperv: hyperv_fb: Use physical memory for fb on HyperV Gen 1 VMs.
From: hch @ 2019-10-23 9:10 UTC (permalink / raw)
To: Wei Hu
Cc: b.zolnierkie@samsung.com, KY Srinivasan, Haiyang Zhang,
Stephen Hemminger, sashal@kernel.org, hch@lst.de,
m.szyprowski@samsung.com, robin.murphy@arm.com,
mchehab+samsung@kernel.org, sam@ravnborg.org,
gregkh@linuxfoundation.org, alexandre.belloni@bootlin.com,
info@metux.net, arnd@arndb.de, dri-devel@lists.freedesktop.org,
linux-fbdev@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-hyperv@vger.kernel.org, iommu@lists.linux-foundation.org,
dcui@microsoft.com, Michael Kelley
In-Reply-To: <20191022110905.4032-1-weh@microsoft.com>
> + select DMA_CMA
Thіs needs to be
select DMA_CMA if HAVE_DMA_CONTIGUOUS
> +#include <linux/dma-contiguous.h>
> + /* Allocate from CMA */
> + // request_pages = (request_size >> PAGE_SHIFT) + 1;
> + request_pages = (round_up(request_size, PAGE_SIZE) >> PAGE_SHIFT);
> + page = dma_alloc_from_contiguous(NULL, request_pages, 0, false);
dma_alloc_from_contiguous is an internal helper, you must use it
through dma_alloc_coherent and pass a struct device to that function.
> + if (!gen2vm) {
> + pdev = pci_get_device(PCI_VENDOR_ID_MICROSOFT,
> + PCI_DEVICE_ID_HYPERV_VIDEO, NULL);
> + if (!pdev) {
> + pr_err("Unable to find PCI Hyper-V video\n");
> + return -ENODEV;
> + }
> + }
Please actually implement a pci_driver instead of hacks like this.
> + par->need_docopy = false;
> + goto getmem1;
> + } else {
No need for an else after a goto.
^ permalink raw reply
* Re: [PATCH v7 3/5] x86/kvm: Add "nopvspin" parameter to disable PV spinlocks
From: Zhenzhong Duan @ 2019-10-23 1:36 UTC (permalink / raw)
To: Sean Christopherson
Cc: Vitaly Kuznetsov, linux-kernel, tglx, mingo, bp, x86, pbonzini,
rkrcmar, wanpengli, jmattson, joro, boris.ostrovsky, jgross,
peterz, will, linux-hyperv, kvm, mikelley, kys, haiyangz,
sthemmin, sashal, Jonathan Corbet, H. Peter Anvin
In-Reply-To: <20191022210355.GR2343@linux.intel.com>
On 2019/10/23 5:03, Sean Christopherson wrote:
> On Tue, Oct 22, 2019 at 08:46:46PM +0800, Zhenzhong Duan wrote:
>> Hi Vitaly,
>>
>> On 2019/10/22 19:36, Vitaly Kuznetsov wrote:
>>
>>> Zhenzhong Duan<zhenzhong.duan@oracle.com> writes:
>>>
>> ...snip
>>
>>>> diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
>>>> index 249f14a..3945aa5 100644
>>>> --- a/arch/x86/kernel/kvm.c
>>>> +++ b/arch/x86/kernel/kvm.c
>>>> @@ -825,18 +825,36 @@ __visible bool __kvm_vcpu_is_preempted(long cpu)
>>>> */
>>>> void __init kvm_spinlock_init(void)
>>>> {
>>>> - /* Does host kernel support KVM_FEATURE_PV_UNHALT? */
>>>> - if (!kvm_para_has_feature(KVM_FEATURE_PV_UNHALT))
>>>> + /*
>>>> + * In case host doesn't support KVM_FEATURE_PV_UNHALT there is still an
>>>> + * advantage of keeping virt_spin_lock_key enabled: virt_spin_lock() is
>>>> + * preferred over native qspinlock when vCPU is preempted.
>>>> + */
>>>> + if (!kvm_para_has_feature(KVM_FEATURE_PV_UNHALT)) {
>>>> + pr_info("PV spinlocks disabled, no host support.\n");
>>>> return;
>>>> + }
>>>> + /*
>>>> + * Disable PV qspinlock and use native qspinlock when dedicated pCPUs
>>>> + * are available.
>>>> + */
>>>> if (kvm_para_has_hint(KVM_HINTS_REALTIME)) {
>>>> - static_branch_disable(&virt_spin_lock_key);
>>>> - return;
>>>> + pr_info("PV spinlocks disabled with KVM_HINTS_REALTIME hints.\n");
>>>> + goto out;
>>>> }
>>>> - /* Don't use the pvqspinlock code if there is only 1 vCPU. */
>>>> - if (num_possible_cpus() == 1)
>>>> - return;
>>>> + if (num_possible_cpus() == 1) {
>>>> + pr_info("PV spinlocks disabled, single CPU.\n");
>>>> + goto out;
>>>> + }
>>>> +
>>>> + if (nopvspin) {
>>>> + pr_info("PV spinlocks disabled, forced by \"nopvspin\" parameter.\n");
>>>> + goto out;
>>>> + }
>>>> +
>>>> + pr_info("PV spinlocks enabled\n");
>>>> __pv_init_lock_hash();
>>>> pv_ops.lock.queued_spin_lock_slowpath = __pv_queued_spin_lock_slowpath;
>>>> @@ -849,6 +867,8 @@ void __init kvm_spinlock_init(void)
>>>> pv_ops.lock.vcpu_is_preempted =
>>>> PV_CALLEE_SAVE(__kvm_vcpu_is_preempted);
>>>> }
>>>> +out:
>>>> + static_branch_disable(&virt_spin_lock_key);
>>> You probably need to add 'return' before 'out:' as it seems you're
>>> disabling virt_spin_lock_key in all cases now).
>> virt_spin_lock_key is kept enabled in !kvm_para_has_feature(KVM_FEATURE_PV_UNHALT)
>> case which is the only case virt_spin_lock() optimization is used.
>>
>> When PV qspinlock is enabled, virt_spin_lock() isn't called in
>> __pv_queued_spin_lock_slowpath() in which case we don't care
>> virt_spin_lock_key's value.
>>
>> So adding 'return' or not are both ok, I chosed to save a line,
>> let me know if you prefer to add a 'return' and I'll change it.
> It'd be worth adding a comment here if you end up spinning another version
> to change the logging prefix. The logic is sound and I like the end
> result, but I had the same knee jerk "this can't be right!?!?" reaction as
> Vitaly.
Sure, will do in next version.
Thanks
Zhenzhong
^ permalink raw reply
* Re: [PATCH v7 2/5] x86/kvm: Change print code to use pr_*() format
From: Zhenzhong Duan @ 2019-10-23 1:29 UTC (permalink / raw)
To: Sean Christopherson
Cc: linux-kernel, tglx, mingo, bp, x86, pbonzini, rkrcmar, vkuznets,
wanpengli, jmattson, joro, boris.ostrovsky, jgross, peterz, will,
linux-hyperv, kvm, mikelley, kys, haiyangz, sthemmin, sashal,
H. Peter Anvin
In-Reply-To: <20191022210120.GQ2343@linux.intel.com>
On 2019/10/23 5:01, Sean Christopherson wrote:
> On Mon, Oct 21, 2019 at 05:11:13PM +0800, Zhenzhong Duan wrote:
>> pr_*() is preferred than printk(KERN_* ...), after change all the print
>> in arch/x86/kernel/kvm.c will have "kvm_guest: xxx" style.
>>
>> No functional change.
>>
>> Signed-off-by: Zhenzhong Duan <zhenzhong.duan@oracle.com>
>> Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
>> Cc: Paolo Bonzini <pbonzini@redhat.com>
>> Cc: Radim Krcmar <rkrcmar@redhat.com>
>> Cc: Sean Christopherson <sean.j.christopherson@intel.com>
>> Cc: Vitaly Kuznetsov <vkuznets@redhat.com>
>> Cc: Wanpeng Li <wanpengli@tencent.com>
>> Cc: Jim Mattson <jmattson@google.com>
>> Cc: Joerg Roedel <joro@8bytes.org>
>> Cc: Thomas Gleixner <tglx@linutronix.de>
>> Cc: Ingo Molnar <mingo@redhat.com>
>> Cc: Borislav Petkov <bp@alien8.de>
>> Cc: "H. Peter Anvin" <hpa@zytor.com>
>> ---
>> arch/x86/kernel/kvm.c | 30 ++++++++++++++++--------------
>> 1 file changed, 16 insertions(+), 14 deletions(-)
>>
>> diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
>> index 3bc6a266..249f14a 100644
>> --- a/arch/x86/kernel/kvm.c
>> +++ b/arch/x86/kernel/kvm.c
>> @@ -7,6 +7,8 @@
>> * Authors: Anthony Liguori <aliguori@us.ibm.com>
>> */
>>
>> +#define pr_fmt(fmt) "kvm_guest: " fmt
> Sort of a silly nit, especially since I suggested kvm_guest...
>
> What about using kvm-guest instead of kvm_guest to be consistent with
> kvm-clock, the other prolific logger in a KVM guest.
>
> E.g.
>
> kvm-clock: cpu 1, msr 551e041, secondary cpu clock
> kvm-guest: setup async PF for cpu 1
> kvm-guest: stealtime: cpu 1, msr 277695f40
> kvm-clock: cpu 2, msr 551e081, secondary cpu clock
> kvm-guest: setup async PF for cpu 2
> kvm-guest: stealtime: cpu 2, msr 277715f40
> kvm-clock: cpu 3, msr 551e0c1, secondary cpu clock
> kvm-guest: setup async PF for cpu 3
> kvm-guest: stealtime: cpu 3, msr 277795f40
> kvm-clock: cpu 4, msr 551e101, secondary cpu clock
>
> instead of
>
> kvm-clock: cpu 1, msr 551e041, secondary cpu clock
> kvm_guest: setup async PF for cpu 1
> kvm_guest: stealtime: cpu 1, msr 277695f40
> kvm-clock: cpu 2, msr 551e081, secondary cpu clock
> kvm_guest: setup async PF for cpu 2
> kvm_guest: stealtime: cpu 2, msr 277715f40
> kvm-clock: cpu 3, msr 551e0c1, secondary cpu clock
> kvm_guest: setup async PF for cpu 3
> kvm_guest: stealtime: cpu 3, msr 277795f40
> kvm-clock: cpu 4, msr 551e101, secondary cpu clock
Good suggestion, will do, thanks for point out.
Zhenzhong
^ permalink raw reply
* Re: [PATCH v7 3/5] x86/kvm: Add "nopvspin" parameter to disable PV spinlocks
From: Sean Christopherson @ 2019-10-22 21:03 UTC (permalink / raw)
To: Zhenzhong Duan
Cc: Vitaly Kuznetsov, linux-kernel, tglx, mingo, bp, x86, pbonzini,
rkrcmar, wanpengli, jmattson, joro, boris.ostrovsky, jgross,
peterz, will, linux-hyperv, kvm, mikelley, kys, haiyangz,
sthemmin, sashal, Jonathan Corbet, H. Peter Anvin
In-Reply-To: <dbc50272-a4f5-ce7c-ba71-75031521f420@oracle.com>
On Tue, Oct 22, 2019 at 08:46:46PM +0800, Zhenzhong Duan wrote:
> Hi Vitaly,
>
> On 2019/10/22 19:36, Vitaly Kuznetsov wrote:
>
> >Zhenzhong Duan<zhenzhong.duan@oracle.com> writes:
> >
> ...snip
>
> >>diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
> >>index 249f14a..3945aa5 100644
> >>--- a/arch/x86/kernel/kvm.c
> >>+++ b/arch/x86/kernel/kvm.c
> >>@@ -825,18 +825,36 @@ __visible bool __kvm_vcpu_is_preempted(long cpu)
> >> */
> >> void __init kvm_spinlock_init(void)
> >> {
> >>- /* Does host kernel support KVM_FEATURE_PV_UNHALT? */
> >>- if (!kvm_para_has_feature(KVM_FEATURE_PV_UNHALT))
> >>+ /*
> >>+ * In case host doesn't support KVM_FEATURE_PV_UNHALT there is still an
> >>+ * advantage of keeping virt_spin_lock_key enabled: virt_spin_lock() is
> >>+ * preferred over native qspinlock when vCPU is preempted.
> >>+ */
> >>+ if (!kvm_para_has_feature(KVM_FEATURE_PV_UNHALT)) {
> >>+ pr_info("PV spinlocks disabled, no host support.\n");
> >> return;
> >>+ }
> >>+ /*
> >>+ * Disable PV qspinlock and use native qspinlock when dedicated pCPUs
> >>+ * are available.
> >>+ */
> >> if (kvm_para_has_hint(KVM_HINTS_REALTIME)) {
> >>- static_branch_disable(&virt_spin_lock_key);
> >>- return;
> >>+ pr_info("PV spinlocks disabled with KVM_HINTS_REALTIME hints.\n");
> >>+ goto out;
> >> }
> >>- /* Don't use the pvqspinlock code if there is only 1 vCPU. */
> >>- if (num_possible_cpus() == 1)
> >>- return;
> >>+ if (num_possible_cpus() == 1) {
> >>+ pr_info("PV spinlocks disabled, single CPU.\n");
> >>+ goto out;
> >>+ }
> >>+
> >>+ if (nopvspin) {
> >>+ pr_info("PV spinlocks disabled, forced by \"nopvspin\" parameter.\n");
> >>+ goto out;
> >>+ }
> >>+
> >>+ pr_info("PV spinlocks enabled\n");
> >> __pv_init_lock_hash();
> >> pv_ops.lock.queued_spin_lock_slowpath = __pv_queued_spin_lock_slowpath;
> >>@@ -849,6 +867,8 @@ void __init kvm_spinlock_init(void)
> >> pv_ops.lock.vcpu_is_preempted =
> >> PV_CALLEE_SAVE(__kvm_vcpu_is_preempted);
> >> }
> >>+out:
> >>+ static_branch_disable(&virt_spin_lock_key);
> >You probably need to add 'return' before 'out:' as it seems you're
> >disabling virt_spin_lock_key in all cases now).
>
> virt_spin_lock_key is kept enabled in !kvm_para_has_feature(KVM_FEATURE_PV_UNHALT)
> case which is the only case virt_spin_lock() optimization is used.
>
> When PV qspinlock is enabled, virt_spin_lock() isn't called in
> __pv_queued_spin_lock_slowpath() in which case we don't care
> virt_spin_lock_key's value.
>
> So adding 'return' or not are both ok, I chosed to save a line,
> let me know if you prefer to add a 'return' and I'll change it.
It'd be worth adding a comment here if you end up spinning another version
to change the logging prefix. The logic is sound and I like the end
result, but I had the same knee jerk "this can't be right!?!?" reaction as
Vitaly.
^ permalink raw reply
* Re: [PATCH v7 2/5] x86/kvm: Change print code to use pr_*() format
From: Sean Christopherson @ 2019-10-22 21:01 UTC (permalink / raw)
To: Zhenzhong Duan
Cc: linux-kernel, tglx, mingo, bp, x86, pbonzini, rkrcmar, vkuznets,
wanpengli, jmattson, joro, boris.ostrovsky, jgross, peterz, will,
linux-hyperv, kvm, mikelley, kys, haiyangz, sthemmin, sashal,
H. Peter Anvin
In-Reply-To: <1571649076-2421-3-git-send-email-zhenzhong.duan@oracle.com>
On Mon, Oct 21, 2019 at 05:11:13PM +0800, Zhenzhong Duan wrote:
> pr_*() is preferred than printk(KERN_* ...), after change all the print
> in arch/x86/kernel/kvm.c will have "kvm_guest: xxx" style.
>
> No functional change.
>
> Signed-off-by: Zhenzhong Duan <zhenzhong.duan@oracle.com>
> Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
> Cc: Paolo Bonzini <pbonzini@redhat.com>
> Cc: Radim Krcmar <rkrcmar@redhat.com>
> Cc: Sean Christopherson <sean.j.christopherson@intel.com>
> Cc: Vitaly Kuznetsov <vkuznets@redhat.com>
> Cc: Wanpeng Li <wanpengli@tencent.com>
> Cc: Jim Mattson <jmattson@google.com>
> Cc: Joerg Roedel <joro@8bytes.org>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: Ingo Molnar <mingo@redhat.com>
> Cc: Borislav Petkov <bp@alien8.de>
> Cc: "H. Peter Anvin" <hpa@zytor.com>
> ---
> arch/x86/kernel/kvm.c | 30 ++++++++++++++++--------------
> 1 file changed, 16 insertions(+), 14 deletions(-)
>
> diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
> index 3bc6a266..249f14a 100644
> --- a/arch/x86/kernel/kvm.c
> +++ b/arch/x86/kernel/kvm.c
> @@ -7,6 +7,8 @@
> * Authors: Anthony Liguori <aliguori@us.ibm.com>
> */
>
> +#define pr_fmt(fmt) "kvm_guest: " fmt
Sort of a silly nit, especially since I suggested kvm_guest...
What about using kvm-guest instead of kvm_guest to be consistent with
kvm-clock, the other prolific logger in a KVM guest.
E.g.
kvm-clock: cpu 1, msr 551e041, secondary cpu clock
kvm-guest: setup async PF for cpu 1
kvm-guest: stealtime: cpu 1, msr 277695f40
kvm-clock: cpu 2, msr 551e081, secondary cpu clock
kvm-guest: setup async PF for cpu 2
kvm-guest: stealtime: cpu 2, msr 277715f40
kvm-clock: cpu 3, msr 551e0c1, secondary cpu clock
kvm-guest: setup async PF for cpu 3
kvm-guest: stealtime: cpu 3, msr 277795f40
kvm-clock: cpu 4, msr 551e101, secondary cpu clock
instead of
kvm-clock: cpu 1, msr 551e041, secondary cpu clock
kvm_guest: setup async PF for cpu 1
kvm_guest: stealtime: cpu 1, msr 277695f40
kvm-clock: cpu 2, msr 551e081, secondary cpu clock
kvm_guest: setup async PF for cpu 2
kvm_guest: stealtime: cpu 2, msr 277715f40
kvm-clock: cpu 3, msr 551e0c1, secondary cpu clock
kvm_guest: setup async PF for cpu 3
kvm_guest: stealtime: cpu 3, msr 277795f40
kvm-clock: cpu 4, msr 551e101, secondary cpu clock
^ permalink raw reply
* Re: [PATCH v7 3/5] x86/kvm: Add "nopvspin" parameter to disable PV spinlocks
From: Vitaly Kuznetsov @ 2019-10-22 13:11 UTC (permalink / raw)
To: Zhenzhong Duan, linux-kernel
Cc: tglx, mingo, bp, x86, pbonzini, rkrcmar, sean.j.christopherson,
wanpengli, jmattson, joro, boris.ostrovsky, jgross, peterz, will,
linux-hyperv, kvm, mikelley, kys, haiyangz, sthemmin, sashal,
Jonathan Corbet, H. Peter Anvin
In-Reply-To: <dbc50272-a4f5-ce7c-ba71-75031521f420@oracle.com>
Zhenzhong Duan <zhenzhong.duan@oracle.com> writes:
> Hi Vitaly,
>
> On 2019/10/22 19:36, Vitaly Kuznetsov wrote:
>
>> Zhenzhong Duan<zhenzhong.duan@oracle.com> writes:
>>
> ...snip
>
>>> diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
>>> index 249f14a..3945aa5 100644
>>> --- a/arch/x86/kernel/kvm.c
>>> +++ b/arch/x86/kernel/kvm.c
>>> @@ -825,18 +825,36 @@ __visible bool __kvm_vcpu_is_preempted(long cpu)
>>> */
>>> void __init kvm_spinlock_init(void)
>>> {
>>> - /* Does host kernel support KVM_FEATURE_PV_UNHALT? */
>>> - if (!kvm_para_has_feature(KVM_FEATURE_PV_UNHALT))
>>> + /*
>>> + * In case host doesn't support KVM_FEATURE_PV_UNHALT there is still an
>>> + * advantage of keeping virt_spin_lock_key enabled: virt_spin_lock() is
>>> + * preferred over native qspinlock when vCPU is preempted.
>>> + */
>>> + if (!kvm_para_has_feature(KVM_FEATURE_PV_UNHALT)) {
>>> + pr_info("PV spinlocks disabled, no host support.\n");
>>> return;
>>> + }
>>>
>>> + /*
>>> + * Disable PV qspinlock and use native qspinlock when dedicated pCPUs
>>> + * are available.
>>> + */
>>> if (kvm_para_has_hint(KVM_HINTS_REALTIME)) {
>>> - static_branch_disable(&virt_spin_lock_key);
>>> - return;
>>> + pr_info("PV spinlocks disabled with KVM_HINTS_REALTIME hints.\n");
>>> + goto out;
>>> }
>>>
>>> - /* Don't use the pvqspinlock code if there is only 1 vCPU. */
>>> - if (num_possible_cpus() == 1)
>>> - return;
>>> + if (num_possible_cpus() == 1) {
>>> + pr_info("PV spinlocks disabled, single CPU.\n");
>>> + goto out;
>>> + }
>>> +
>>> + if (nopvspin) {
>>> + pr_info("PV spinlocks disabled, forced by \"nopvspin\" parameter.\n");
>>> + goto out;
>>> + }
>>> +
>>> + pr_info("PV spinlocks enabled\n");
>>>
>>> __pv_init_lock_hash();
>>> pv_ops.lock.queued_spin_lock_slowpath = __pv_queued_spin_lock_slowpath;
>>> @@ -849,6 +867,8 @@ void __init kvm_spinlock_init(void)
>>> pv_ops.lock.vcpu_is_preempted =
>>> PV_CALLEE_SAVE(__kvm_vcpu_is_preempted);
>>> }
>>> +out:
>>> + static_branch_disable(&virt_spin_lock_key);
>> You probably need to add 'return' before 'out:' as it seems you're
>> disabling virt_spin_lock_key in all cases now).
>
> virt_spin_lock_key is kept enabled in !kvm_para_has_feature(KVM_FEATURE_PV_UNHALT)
> case which is the only case virt_spin_lock() optimization is used.
>
> When PV qspinlock is enabled, virt_spin_lock() isn't called in
> __pv_queued_spin_lock_slowpath() in which case we don't care
> virt_spin_lock_key's value.
>
True, my bad: I though we still need it enabled for something.
> So adding 'return' or not are both ok, I chosed to save a line,
> let me know if you prefer to add a 'return' and I'll change it.
No, please ignore.
>
> btw: __pv_queued_spin_lock_slowpath() is alias of queued_spin_lock_slowpath()
>
> Thanks
> Zhenzhong
>
--
Vitaly
^ permalink raw reply
* Re: [PATCH v7 3/5] x86/kvm: Add "nopvspin" parameter to disable PV spinlocks
From: Zhenzhong Duan @ 2019-10-22 12:46 UTC (permalink / raw)
To: Vitaly Kuznetsov, linux-kernel
Cc: tglx, mingo, bp, x86, pbonzini, rkrcmar, sean.j.christopherson,
wanpengli, jmattson, joro, boris.ostrovsky, jgross, peterz, will,
linux-hyperv, kvm, mikelley, kys, haiyangz, sthemmin, sashal,
Jonathan Corbet, H. Peter Anvin
In-Reply-To: <8736fl1071.fsf@vitty.brq.redhat.com>
Hi Vitaly,
On 2019/10/22 19:36, Vitaly Kuznetsov wrote:
> Zhenzhong Duan<zhenzhong.duan@oracle.com> writes:
>
...snip
>> diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
>> index 249f14a..3945aa5 100644
>> --- a/arch/x86/kernel/kvm.c
>> +++ b/arch/x86/kernel/kvm.c
>> @@ -825,18 +825,36 @@ __visible bool __kvm_vcpu_is_preempted(long cpu)
>> */
>> void __init kvm_spinlock_init(void)
>> {
>> - /* Does host kernel support KVM_FEATURE_PV_UNHALT? */
>> - if (!kvm_para_has_feature(KVM_FEATURE_PV_UNHALT))
>> + /*
>> + * In case host doesn't support KVM_FEATURE_PV_UNHALT there is still an
>> + * advantage of keeping virt_spin_lock_key enabled: virt_spin_lock() is
>> + * preferred over native qspinlock when vCPU is preempted.
>> + */
>> + if (!kvm_para_has_feature(KVM_FEATURE_PV_UNHALT)) {
>> + pr_info("PV spinlocks disabled, no host support.\n");
>> return;
>> + }
>>
>> + /*
>> + * Disable PV qspinlock and use native qspinlock when dedicated pCPUs
>> + * are available.
>> + */
>> if (kvm_para_has_hint(KVM_HINTS_REALTIME)) {
>> - static_branch_disable(&virt_spin_lock_key);
>> - return;
>> + pr_info("PV spinlocks disabled with KVM_HINTS_REALTIME hints.\n");
>> + goto out;
>> }
>>
>> - /* Don't use the pvqspinlock code if there is only 1 vCPU. */
>> - if (num_possible_cpus() == 1)
>> - return;
>> + if (num_possible_cpus() == 1) {
>> + pr_info("PV spinlocks disabled, single CPU.\n");
>> + goto out;
>> + }
>> +
>> + if (nopvspin) {
>> + pr_info("PV spinlocks disabled, forced by \"nopvspin\" parameter.\n");
>> + goto out;
>> + }
>> +
>> + pr_info("PV spinlocks enabled\n");
>>
>> __pv_init_lock_hash();
>> pv_ops.lock.queued_spin_lock_slowpath = __pv_queued_spin_lock_slowpath;
>> @@ -849,6 +867,8 @@ void __init kvm_spinlock_init(void)
>> pv_ops.lock.vcpu_is_preempted =
>> PV_CALLEE_SAVE(__kvm_vcpu_is_preempted);
>> }
>> +out:
>> + static_branch_disable(&virt_spin_lock_key);
> You probably need to add 'return' before 'out:' as it seems you're
> disabling virt_spin_lock_key in all cases now).
virt_spin_lock_key is kept enabled in !kvm_para_has_feature(KVM_FEATURE_PV_UNHALT)
case which is the only case virt_spin_lock() optimization is used.
When PV qspinlock is enabled, virt_spin_lock() isn't called in
__pv_queued_spin_lock_slowpath() in which case we don't care
virt_spin_lock_key's value.
So adding 'return' or not are both ok, I chosed to save a line,
let me know if you prefer to add a 'return' and I'll change it.
btw: __pv_queued_spin_lock_slowpath() is alias of queued_spin_lock_slowpath()
Thanks
Zhenzhong
^ permalink raw reply
* Re: [PATCH v7 3/5] x86/kvm: Add "nopvspin" parameter to disable PV spinlocks
From: Vitaly Kuznetsov @ 2019-10-22 11:36 UTC (permalink / raw)
To: Zhenzhong Duan, linux-kernel
Cc: tglx, mingo, bp, x86, pbonzini, rkrcmar, sean.j.christopherson,
wanpengli, jmattson, joro, boris.ostrovsky, jgross, peterz, will,
linux-hyperv, kvm, mikelley, kys, haiyangz, sthemmin, sashal,
Zhenzhong Duan, Jonathan Corbet, H. Peter Anvin
In-Reply-To: <1571649076-2421-4-git-send-email-zhenzhong.duan@oracle.com>
Zhenzhong Duan <zhenzhong.duan@oracle.com> writes:
> There are cases where a guest tries to switch spinlocks to bare metal
> behavior (e.g. by setting "xen_nopvspin" on XEN platform and
> "hv_nopvspin" on HYPER_V).
>
> That feature is missed on KVM, add a new parameter "nopvspin" to disable
> PV spinlocks for KVM guest.
>
> The new 'nopvspin' parameter will also replace Xen and Hyper-V specific
> parameters in future patches.
>
> Define variable nopvsin as global because it will be used in future
> patches as above.
>
> Signed-off-by: Zhenzhong Duan <zhenzhong.duan@oracle.com>
> Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
> Cc: Jonathan Corbet <corbet@lwn.net>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: Ingo Molnar <mingo@redhat.com>
> Cc: Borislav Petkov <bp@alien8.de>
> Cc: "H. Peter Anvin" <hpa@zytor.com>
> Cc: Paolo Bonzini <pbonzini@redhat.com>
> Cc: Radim Krcmar <rkrcmar@redhat.com>
> Cc: Sean Christopherson <sean.j.christopherson@intel.com>
> Cc: Vitaly Kuznetsov <vkuznets@redhat.com>
> Cc: Wanpeng Li <wanpengli@tencent.com>
> Cc: Jim Mattson <jmattson@google.com>
> Cc: Joerg Roedel <joro@8bytes.org>
> Cc: Peter Zijlstra <peterz@infradead.org>
> Cc: Will Deacon <will@kernel.org>
> ---
> Documentation/admin-guide/kernel-parameters.txt | 5 ++++
> arch/x86/include/asm/qspinlock.h | 1 +
> arch/x86/kernel/kvm.c | 34 ++++++++++++++++++++-----
> kernel/locking/qspinlock.c | 7 +++++
> 4 files changed, 40 insertions(+), 7 deletions(-)
>
> diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
> index a84a83f..bd49ed2 100644
> --- a/Documentation/admin-guide/kernel-parameters.txt
> +++ b/Documentation/admin-guide/kernel-parameters.txt
> @@ -5334,6 +5334,11 @@
> as generic guest with no PV drivers. Currently support
> XEN HVM, KVM, HYPER_V and VMWARE guest.
>
> + nopvspin [X86,KVM]
> + Disables the qspinlock slow path using PV optimizations
> + which allow the hypervisor to 'idle' the guest on lock
> + contention.
> +
> xirc2ps_cs= [NET,PCMCIA]
> Format:
> <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
> diff --git a/arch/x86/include/asm/qspinlock.h b/arch/x86/include/asm/qspinlock.h
> index 444d6fd..d86ab94 100644
> --- a/arch/x86/include/asm/qspinlock.h
> +++ b/arch/x86/include/asm/qspinlock.h
> @@ -32,6 +32,7 @@ static __always_inline u32 queued_fetch_set_pending_acquire(struct qspinlock *lo
> extern void __pv_init_lock_hash(void);
> extern void __pv_queued_spin_lock_slowpath(struct qspinlock *lock, u32 val);
> extern void __raw_callee_save___pv_queued_spin_unlock(struct qspinlock *lock);
> +extern bool nopvspin;
>
> #define queued_spin_unlock queued_spin_unlock
> /**
> diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
> index 249f14a..3945aa5 100644
> --- a/arch/x86/kernel/kvm.c
> +++ b/arch/x86/kernel/kvm.c
> @@ -825,18 +825,36 @@ __visible bool __kvm_vcpu_is_preempted(long cpu)
> */
> void __init kvm_spinlock_init(void)
> {
> - /* Does host kernel support KVM_FEATURE_PV_UNHALT? */
> - if (!kvm_para_has_feature(KVM_FEATURE_PV_UNHALT))
> + /*
> + * In case host doesn't support KVM_FEATURE_PV_UNHALT there is still an
> + * advantage of keeping virt_spin_lock_key enabled: virt_spin_lock() is
> + * preferred over native qspinlock when vCPU is preempted.
> + */
> + if (!kvm_para_has_feature(KVM_FEATURE_PV_UNHALT)) {
> + pr_info("PV spinlocks disabled, no host support.\n");
> return;
> + }
>
> + /*
> + * Disable PV qspinlock and use native qspinlock when dedicated pCPUs
> + * are available.
> + */
> if (kvm_para_has_hint(KVM_HINTS_REALTIME)) {
> - static_branch_disable(&virt_spin_lock_key);
> - return;
> + pr_info("PV spinlocks disabled with KVM_HINTS_REALTIME hints.\n");
> + goto out;
> }
>
> - /* Don't use the pvqspinlock code if there is only 1 vCPU. */
> - if (num_possible_cpus() == 1)
> - return;
> + if (num_possible_cpus() == 1) {
> + pr_info("PV spinlocks disabled, single CPU.\n");
> + goto out;
> + }
> +
> + if (nopvspin) {
> + pr_info("PV spinlocks disabled, forced by \"nopvspin\" parameter.\n");
> + goto out;
> + }
> +
> + pr_info("PV spinlocks enabled\n");
>
> __pv_init_lock_hash();
> pv_ops.lock.queued_spin_lock_slowpath = __pv_queued_spin_lock_slowpath;
> @@ -849,6 +867,8 @@ void __init kvm_spinlock_init(void)
> pv_ops.lock.vcpu_is_preempted =
> PV_CALLEE_SAVE(__kvm_vcpu_is_preempted);
> }
> +out:
> + static_branch_disable(&virt_spin_lock_key);
You probably need to add 'return' before 'out:' as it seems you're
disabling virt_spin_lock_key in all cases now).
> }
>
> #endif /* CONFIG_PARAVIRT_SPINLOCKS */
> diff --git a/kernel/locking/qspinlock.c b/kernel/locking/qspinlock.c
> index 2473f10..75193d6 100644
> --- a/kernel/locking/qspinlock.c
> +++ b/kernel/locking/qspinlock.c
> @@ -580,4 +580,11 @@ void queued_spin_lock_slowpath(struct qspinlock *lock, u32 val)
> #include "qspinlock_paravirt.h"
> #include "qspinlock.c"
>
> +bool nopvspin __initdata;
> +static __init int parse_nopvspin(char *arg)
> +{
> + nopvspin = true;
> + return 0;
> +}
> +early_param("nopvspin", parse_nopvspin);
> #endif
--
Vitaly
^ permalink raw reply
* [PATCH] video: hyperv: hyperv_fb: Use physical memory for fb on HyperV Gen 1 VMs.
From: Wei Hu @ 2019-10-22 11:10 UTC (permalink / raw)
To: b.zolnierkie@samsung.com, KY Srinivasan, Haiyang Zhang,
Stephen Hemminger, sashal@kernel.org, hch@lst.de,
m.szyprowski@samsung.com, robin.murphy@arm.com,
mchehab+samsung@kernel.org, sam@ravnborg.org,
gregkh@linuxfoundation.org, alexandre.belloni@bootlin.com,
info@metux.net, arnd@arndb.de, dri-devel@lists.freedesktop.org,
linux-fbdev@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-hyperv@vger.kernel.org, iommu@lists.linux-foundation.org,
dcui@microsoft.com, Michael Kelley
Cc: Wei Hu
On Hyper-V, Generation 1 VMs can directly use VM's physical memory for
their framebuffers. This can improve the efficiency of framebuffer and
overall performence for VM. The physical memory assigned to framebuffer
must be contiguous. We use CMA allocator to get contiguouse physicial
memory when the framebuffer size is greater than 4MB. For size under
4MB, we use alloc_pages to achieve this.
To enable framebuffer memory allocation from CMA, supply a kernel
parameter to give enough space to CMA allocator at boot time. For
example:
cma=130m
This gives 130MB memory to CAM allocator that can be allocated to
framebuffer. If this fails, we fall back to the old way of using
mmio for framebuffer.
Signed-off-by: Wei Hu <weh@microsoft.com>
---
drivers/video/fbdev/Kconfig | 1 +
drivers/video/fbdev/hyperv_fb.c | 179 +++++++++++++++++++++++++-------
kernel/dma/contiguous.c | 2 +
3 files changed, 147 insertions(+), 35 deletions(-)
diff --git a/drivers/video/fbdev/Kconfig b/drivers/video/fbdev/Kconfig
index aa9541bf964b..f534059461ee 100644
--- a/drivers/video/fbdev/Kconfig
+++ b/drivers/video/fbdev/Kconfig
@@ -2215,6 +2215,7 @@ config FB_HYPERV
select FB_CFB_COPYAREA
select FB_CFB_IMAGEBLIT
select FB_DEFERRED_IO
+ select DMA_CMA
help
This framebuffer driver supports Microsoft Hyper-V Synthetic Video.
diff --git a/drivers/video/fbdev/hyperv_fb.c b/drivers/video/fbdev/hyperv_fb.c
index 3f60b7bc8589..ea2fd3481225 100644
--- a/drivers/video/fbdev/hyperv_fb.c
+++ b/drivers/video/fbdev/hyperv_fb.c
@@ -31,6 +31,16 @@
* "set-vmvideo" command. For example
* set-vmvideo -vmname name -horizontalresolution:1920 \
* -verticalresolution:1200 -resolutiontype single
+ *
+ * Gen 1 VMs also support directly using VM's phyiscal memory for framebuffer.
+ * It could improve the efficiency and performance for framebuffer and VM.
+ * This requires to allocate contiguous physical memory from Linux kernel's
+ * CMA memory allocator. To enable this, supply a kernel parameter to give
+ * enough memory space to CMA allocator for framebuffer. For example:
+ * cma=130m
+ * This gives 130MB memory to CMA allocator that can be allocated to
+ * framebuffer. For reference, 8K resolution (7680x4320) takes about
+ * 127MB memory.
*/
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
@@ -42,6 +52,7 @@
#include <linux/fb.h>
#include <linux/pci.h>
#include <linux/efi.h>
+#include <linux/dma-contiguous.h>
#include <linux/hyperv.h>
@@ -227,7 +238,6 @@ struct synthvid_msg {
} __packed;
-
/* FB driver definitions and structures */
#define HVFB_WIDTH 1152 /* default screen width */
#define HVFB_HEIGHT 864 /* default screen height */
@@ -256,6 +266,9 @@ struct hvfb_par {
/* If true, the VSC notifies the VSP on every framebuffer change */
bool synchronous_fb;
+ /* If true, need to copy from deferred IO mem to framebuffer mem */
+ bool need_docopy;
+
struct notifier_block hvfb_panic_nb;
/* Memory for deferred IO and frame buffer itself */
@@ -432,7 +445,7 @@ static void synthvid_deferred_io(struct fb_info *p,
maxy = max_t(int, maxy, y2);
/* Copy from dio space to mmio address */
- if (par->fb_ready)
+ if (par->fb_ready && par->need_docopy)
hvfb_docopy(par, start, PAGE_SIZE);
}
@@ -749,12 +762,12 @@ static void hvfb_update_work(struct work_struct *w)
return;
/* Copy the dirty rectangle to frame buffer memory */
- for (j = y1; j < y2; j++) {
- hvfb_docopy(par,
- j * info->fix.line_length +
- (x1 * screen_depth / 8),
- (x2 - x1) * screen_depth / 8);
- }
+ if (par->need_docopy)
+ for (j = y1; j < y2; j++)
+ hvfb_docopy(par,
+ j * info->fix.line_length +
+ (x1 * screen_depth / 8),
+ (x2 - x1) * screen_depth / 8);
/* Refresh */
if (par->fb_ready && par->update)
@@ -799,7 +812,8 @@ static int hvfb_on_panic(struct notifier_block *nb,
par = container_of(nb, struct hvfb_par, hvfb_panic_nb);
par->synchronous_fb = true;
info = par->info;
- hvfb_docopy(par, 0, dio_fb_size);
+ if (par->need_docopy)
+ hvfb_docopy(par, 0, dio_fb_size);
synthvid_update(info, 0, 0, INT_MAX, INT_MAX);
return NOTIFY_DONE;
@@ -938,6 +952,62 @@ static void hvfb_get_option(struct fb_info *info)
return;
}
+/*
+ * Allocate enough contiguous physical memory.
+ * Return physical address if succeeded or -1 if failed.
+ */
+static unsigned long hvfb_get_phymem(unsigned int request_size)
+{
+ struct page *page = NULL;
+ unsigned int request_pages;
+ unsigned long paddr = 0;
+ unsigned int order = get_order(request_size);
+
+ if (request_size == 0)
+ return -1;
+
+ /* Try call alloc_pages if the size is less than 2^MAX_ORDER */
+ if (order < MAX_ORDER) {
+ page = alloc_pages(GFP_KERNEL | __GFP_ZERO, order);
+ if (!page)
+ return -1;
+
+ request_pages = (1 << order);
+ goto get_phymem1;
+ }
+
+ /* Allocate from CMA */
+ // request_pages = (request_size >> PAGE_SHIFT) + 1;
+ request_pages = (round_up(request_size, PAGE_SIZE) >> PAGE_SHIFT);
+ page = dma_alloc_from_contiguous(NULL, request_pages, 0, false);
+
+ if (page == NULL)
+ return -1;
+
+get_phymem1:
+ paddr = (page_to_pfn(page) << PAGE_SHIFT);
+
+ pr_info("Allocated %d pages starts at physical addr 0x%lx\n",
+ request_pages, paddr);
+
+ return paddr;
+}
+
+/* Release contiguous physical memory */
+static void hvfb_release_phymem(unsigned long paddr, unsigned int size)
+{
+ unsigned int order = get_order(size);
+
+ if (order < MAX_ORDER)
+ __free_pages(pfn_to_page(paddr >> PAGE_SHIFT), order);
+ else
+ dma_release_from_contiguous(NULL,
+ pfn_to_page(paddr >> PAGE_SHIFT),
+ (round_up(size, PAGE_SIZE) >>
+ PAGE_SHIFT));
+ // (size >> PAGE_SHIFT) + 1);
+}
+
/* Get framebuffer memory from Hyper-V video pci space */
static int hvfb_getmem(struct hv_device *hdev, struct fb_info *info)
@@ -947,8 +1017,58 @@ static int hvfb_getmem(struct hv_device *hdev, struct fb_info *info)
void __iomem *fb_virt;
int gen2vm = efi_enabled(EFI_BOOT);
resource_size_t pot_start, pot_end;
+ unsigned long paddr;
int ret;
+ if (!gen2vm) {
+ pdev = pci_get_device(PCI_VENDOR_ID_MICROSOFT,
+ PCI_DEVICE_ID_HYPERV_VIDEO, NULL);
+ if (!pdev) {
+ pr_err("Unable to find PCI Hyper-V video\n");
+ return -ENODEV;
+ }
+ }
+
+ info->apertures = alloc_apertures(1);
+ if (!info->apertures)
+ goto err1;
+
+ if (gen2vm) {
+ info->apertures->ranges[0].base = screen_info.lfb_base;
+ info->apertures->ranges[0].size = screen_info.lfb_size;
+ } else {
+ info->apertures->ranges[0].base = pci_resource_start(pdev, 0);
+ info->apertures->ranges[0].size = pci_resource_len(pdev, 0);
+ }
+
+ /*
+ * For Gen 1 VM, we can directly use the contiguous memory
+ * from VM. If we success, deferred IO happens directly
+ * on this allocated framebuffer memory, avoiding extra
+ * memory copy.
+ */
+ if (!gen2vm) {
+ paddr = hvfb_get_phymem(screen_fb_size);
+ if (paddr != (unsigned long) -1) {
+ par->mmio_pp = paddr;
+ par->mmio_vp = par->dio_vp = __va(paddr);
+
+ info->fix.smem_start = paddr;
+ info->fix.smem_len = screen_fb_size;
+ info->screen_base = par->mmio_vp;
+ info->screen_size = screen_fb_size;
+
+ par->need_docopy = false;
+ goto getmem1;
+ } else {
+ pr_info("Unable to allocate enough contiguous physical memory on Gen 1 VM. Use MMIO instead.\n");
+ }
+ }
+
+ /*
+ * Cannot use the contiguous physical memory.
+ * Allocate mmio space for framebuffer.
+ */
dio_fb_size =
screen_width * screen_height * screen_depth / 8;
@@ -956,13 +1076,6 @@ static int hvfb_getmem(struct hv_device *hdev, struct fb_info *info)
pot_start = 0;
pot_end = -1;
} else {
- pdev = pci_get_device(PCI_VENDOR_ID_MICROSOFT,
- PCI_DEVICE_ID_HYPERV_VIDEO, NULL);
- if (!pdev) {
- pr_err("Unable to find PCI Hyper-V video\n");
- return -ENODEV;
- }
-
if (!(pci_resource_flags(pdev, 0) & IORESOURCE_MEM) ||
pci_resource_len(pdev, 0) < screen_fb_size) {
pr_err("Resource not available or (0x%lx < 0x%lx)\n",
@@ -991,20 +1104,6 @@ static int hvfb_getmem(struct hv_device *hdev, struct fb_info *info)
if (par->dio_vp == NULL)
goto err3;
- info->apertures = alloc_apertures(1);
- if (!info->apertures)
- goto err4;
-
- if (gen2vm) {
- info->apertures->ranges[0].base = screen_info.lfb_base;
- info->apertures->ranges[0].size = screen_info.lfb_size;
- remove_conflicting_framebuffers(info->apertures,
- KBUILD_MODNAME, false);
- } else {
- info->apertures->ranges[0].base = pci_resource_start(pdev, 0);
- info->apertures->ranges[0].size = pci_resource_len(pdev, 0);
- }
-
/* Physical address of FB device */
par->mmio_pp = par->mem->start;
/* Virtual address of FB device */
@@ -1015,13 +1114,17 @@ static int hvfb_getmem(struct hv_device *hdev, struct fb_info *info)
info->screen_base = par->dio_vp;
info->screen_size = dio_fb_size;
+ par->need_docopy = true;
+
+getmem1:
+ remove_conflicting_framebuffers(info->apertures,
+ KBUILD_MODNAME, false);
+
if (!gen2vm)
pci_dev_put(pdev);
return 0;
-err4:
- vfree(par->dio_vp);
err3:
iounmap(fb_virt);
err2:
@@ -1039,9 +1142,14 @@ static void hvfb_putmem(struct fb_info *info)
{
struct hvfb_par *par = info->par;
- vfree(par->dio_vp);
- iounmap(info->screen_base);
- vmbus_free_mmio(par->mem->start, screen_fb_size);
+ if (par->need_docopy) {
+ vfree(par->dio_vp);
+ iounmap(info->screen_base);
+ vmbus_free_mmio(par->mem->start, screen_fb_size);
+ } else {
+ hvfb_release_phymem(info->fix.smem_start, screen_fb_size);
+ }
+
par->mem = NULL;
}
@@ -1060,6 +1168,7 @@ static int hvfb_probe(struct hv_device *hdev,
par = info->par;
par->info = info;
par->fb_ready = false;
+ par->need_docopy = false;
init_completion(&par->wait);
INIT_DELAYED_WORK(&par->dwork, hvfb_update_work);
diff --git a/kernel/dma/contiguous.c b/kernel/dma/contiguous.c
index 69cfb4345388..4553f4cca80e 100644
--- a/kernel/dma/contiguous.c
+++ b/kernel/dma/contiguous.c
@@ -197,6 +197,7 @@ struct page *dma_alloc_from_contiguous(struct device *dev, size_t count,
return cma_alloc(dev_get_cma_area(dev), count, align, no_warn);
}
+EXPORT_SYMBOL(dma_alloc_from_contiguous);
/**
* dma_release_from_contiguous() - release allocated pages
@@ -213,6 +214,7 @@ bool dma_release_from_contiguous(struct device *dev, struct page *pages,
{
return cma_release(dev_get_cma_area(dev), pages, count);
}
+EXPORT_SYMBOL(dma_release_from_contiguous);
/**
* dma_alloc_contiguous() - allocate contiguous pages
--
2.20.1
^ permalink raw reply related
* [PATCH v7 3/5] x86/kvm: Add "nopvspin" parameter to disable PV spinlocks
From: Zhenzhong Duan @ 2019-10-21 9:11 UTC (permalink / raw)
To: linux-kernel
Cc: tglx, mingo, bp, x86, pbonzini, rkrcmar, sean.j.christopherson,
vkuznets, wanpengli, jmattson, joro, boris.ostrovsky, jgross,
peterz, will, linux-hyperv, kvm, mikelley, kys, haiyangz,
sthemmin, sashal, Zhenzhong Duan, Jonathan Corbet, H. Peter Anvin
In-Reply-To: <1571649076-2421-1-git-send-email-zhenzhong.duan@oracle.com>
There are cases where a guest tries to switch spinlocks to bare metal
behavior (e.g. by setting "xen_nopvspin" on XEN platform and
"hv_nopvspin" on HYPER_V).
That feature is missed on KVM, add a new parameter "nopvspin" to disable
PV spinlocks for KVM guest.
The new 'nopvspin' parameter will also replace Xen and Hyper-V specific
parameters in future patches.
Define variable nopvsin as global because it will be used in future
patches as above.
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@oracle.com>
Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krcmar <rkrcmar@redhat.com>
Cc: Sean Christopherson <sean.j.christopherson@intel.com>
Cc: Vitaly Kuznetsov <vkuznets@redhat.com>
Cc: Wanpeng Li <wanpengli@tencent.com>
Cc: Jim Mattson <jmattson@google.com>
Cc: Joerg Roedel <joro@8bytes.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will@kernel.org>
---
Documentation/admin-guide/kernel-parameters.txt | 5 ++++
arch/x86/include/asm/qspinlock.h | 1 +
arch/x86/kernel/kvm.c | 34 ++++++++++++++++++++-----
kernel/locking/qspinlock.c | 7 +++++
4 files changed, 40 insertions(+), 7 deletions(-)
diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index a84a83f..bd49ed2 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -5334,6 +5334,11 @@
as generic guest with no PV drivers. Currently support
XEN HVM, KVM, HYPER_V and VMWARE guest.
+ nopvspin [X86,KVM]
+ Disables the qspinlock slow path using PV optimizations
+ which allow the hypervisor to 'idle' the guest on lock
+ contention.
+
xirc2ps_cs= [NET,PCMCIA]
Format:
<irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
diff --git a/arch/x86/include/asm/qspinlock.h b/arch/x86/include/asm/qspinlock.h
index 444d6fd..d86ab94 100644
--- a/arch/x86/include/asm/qspinlock.h
+++ b/arch/x86/include/asm/qspinlock.h
@@ -32,6 +32,7 @@ static __always_inline u32 queued_fetch_set_pending_acquire(struct qspinlock *lo
extern void __pv_init_lock_hash(void);
extern void __pv_queued_spin_lock_slowpath(struct qspinlock *lock, u32 val);
extern void __raw_callee_save___pv_queued_spin_unlock(struct qspinlock *lock);
+extern bool nopvspin;
#define queued_spin_unlock queued_spin_unlock
/**
diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
index 249f14a..3945aa5 100644
--- a/arch/x86/kernel/kvm.c
+++ b/arch/x86/kernel/kvm.c
@@ -825,18 +825,36 @@ __visible bool __kvm_vcpu_is_preempted(long cpu)
*/
void __init kvm_spinlock_init(void)
{
- /* Does host kernel support KVM_FEATURE_PV_UNHALT? */
- if (!kvm_para_has_feature(KVM_FEATURE_PV_UNHALT))
+ /*
+ * In case host doesn't support KVM_FEATURE_PV_UNHALT there is still an
+ * advantage of keeping virt_spin_lock_key enabled: virt_spin_lock() is
+ * preferred over native qspinlock when vCPU is preempted.
+ */
+ if (!kvm_para_has_feature(KVM_FEATURE_PV_UNHALT)) {
+ pr_info("PV spinlocks disabled, no host support.\n");
return;
+ }
+ /*
+ * Disable PV qspinlock and use native qspinlock when dedicated pCPUs
+ * are available.
+ */
if (kvm_para_has_hint(KVM_HINTS_REALTIME)) {
- static_branch_disable(&virt_spin_lock_key);
- return;
+ pr_info("PV spinlocks disabled with KVM_HINTS_REALTIME hints.\n");
+ goto out;
}
- /* Don't use the pvqspinlock code if there is only 1 vCPU. */
- if (num_possible_cpus() == 1)
- return;
+ if (num_possible_cpus() == 1) {
+ pr_info("PV spinlocks disabled, single CPU.\n");
+ goto out;
+ }
+
+ if (nopvspin) {
+ pr_info("PV spinlocks disabled, forced by \"nopvspin\" parameter.\n");
+ goto out;
+ }
+
+ pr_info("PV spinlocks enabled\n");
__pv_init_lock_hash();
pv_ops.lock.queued_spin_lock_slowpath = __pv_queued_spin_lock_slowpath;
@@ -849,6 +867,8 @@ void __init kvm_spinlock_init(void)
pv_ops.lock.vcpu_is_preempted =
PV_CALLEE_SAVE(__kvm_vcpu_is_preempted);
}
+out:
+ static_branch_disable(&virt_spin_lock_key);
}
#endif /* CONFIG_PARAVIRT_SPINLOCKS */
diff --git a/kernel/locking/qspinlock.c b/kernel/locking/qspinlock.c
index 2473f10..75193d6 100644
--- a/kernel/locking/qspinlock.c
+++ b/kernel/locking/qspinlock.c
@@ -580,4 +580,11 @@ void queued_spin_lock_slowpath(struct qspinlock *lock, u32 val)
#include "qspinlock_paravirt.h"
#include "qspinlock.c"
+bool nopvspin __initdata;
+static __init int parse_nopvspin(char *arg)
+{
+ nopvspin = true;
+ return 0;
+}
+early_param("nopvspin", parse_nopvspin);
#endif
--
1.8.3.1
^ permalink raw reply related
* [PATCH v7 5/5] x86/hyperv: Mark "hv_nopvspin" parameter obsolete
From: Zhenzhong Duan @ 2019-10-21 9:11 UTC (permalink / raw)
To: linux-kernel
Cc: tglx, mingo, bp, x86, pbonzini, rkrcmar, sean.j.christopherson,
vkuznets, wanpengli, jmattson, joro, boris.ostrovsky, jgross,
peterz, will, linux-hyperv, kvm, mikelley, kys, haiyangz,
sthemmin, sashal, Zhenzhong Duan, Jonathan Corbet, H. Peter Anvin
In-Reply-To: <1571649076-2421-1-git-send-email-zhenzhong.duan@oracle.com>
Map "hv_nopvspin" to "nopvspin".
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@oracle.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: "K. Y. Srinivasan" <kys@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Cc: Stephen Hemminger <sthemmin@microsoft.com>
Cc: Sasha Levin <sashal@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
---
Documentation/admin-guide/kernel-parameters.txt | 6 +++++-
arch/x86/hyperv/hv_spinlock.c | 4 ++++
2 files changed, 9 insertions(+), 1 deletion(-)
diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index 85059dd..78648bb 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -1436,6 +1436,10 @@
hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations
which allow the hypervisor to 'idle' the
guest on lock contention.
+ This parameter is obsoleted by "nopvspin"
+ parameter, which has equivalent effect for
+ HYPER_V platform.
+
keep_bootcon [KNL]
Do not unregister boot console at start. This is only
@@ -5335,7 +5339,7 @@
as generic guest with no PV drivers. Currently support
XEN HVM, KVM, HYPER_V and VMWARE guest.
- nopvspin [X86,XEN,KVM]
+ nopvspin [X86,XEN,KVM,HYPER_V]
Disables the qspinlock slow path using PV optimizations
which allow the hypervisor to 'idle' the guest on lock
contention.
diff --git a/arch/x86/hyperv/hv_spinlock.c b/arch/x86/hyperv/hv_spinlock.c
index 07f21a0..47c7d6c 100644
--- a/arch/x86/hyperv/hv_spinlock.c
+++ b/arch/x86/hyperv/hv_spinlock.c
@@ -64,6 +64,9 @@ __visible bool hv_vcpu_is_preempted(int vcpu)
void __init hv_init_spinlocks(void)
{
+ if (nopvspin)
+ hv_pvspin = false;
+
if (!hv_pvspin || !apic ||
!(ms_hyperv.hints & HV_X64_CLUSTER_IPI_RECOMMENDED) ||
!(ms_hyperv.features & HV_X64_MSR_GUEST_IDLE_AVAILABLE)) {
@@ -82,6 +85,7 @@ void __init hv_init_spinlocks(void)
static __init int hv_parse_nopvspin(char *arg)
{
+ pr_notice("\"hv_nopvspin\" is deprecated, please use \"nopvspin\" instead\n");
hv_pvspin = false;
return 0;
}
--
1.8.3.1
^ permalink raw reply related
* [PATCH v7 4/5] xen: Mark "xen_nopvspin" parameter obsolete
From: Zhenzhong Duan @ 2019-10-21 9:11 UTC (permalink / raw)
To: linux-kernel
Cc: tglx, mingo, bp, x86, pbonzini, rkrcmar, sean.j.christopherson,
vkuznets, wanpengli, jmattson, joro, boris.ostrovsky, jgross,
peterz, will, linux-hyperv, kvm, mikelley, kys, haiyangz,
sthemmin, sashal, Zhenzhong Duan, Jonathan Corbet,
Stefano Stabellini, H. Peter Anvin
In-Reply-To: <1571649076-2421-1-git-send-email-zhenzhong.duan@oracle.com>
Map "xen_nopvspin" to "nopvspin", fix stale description of "xen_nopvspin"
as we use qspinlock now.
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@oracle.com>
Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: Juergen Gross <jgross@suse.com>
Cc: Stefano Stabellini <sstabellini@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
---
Documentation/admin-guide/kernel-parameters.txt | 7 ++++---
arch/x86/xen/spinlock.c | 4 ++--
2 files changed, 6 insertions(+), 5 deletions(-)
diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index bd49ed2..85059dd 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -5307,8 +5307,9 @@
panic() code such as dumping handler.
xen_nopvspin [X86,XEN]
- Disables the ticketlock slowpath using Xen PV
- optimizations.
+ Disables the qspinlock slowpath using Xen PV optimizations.
+ This parameter is obsoleted by "nopvspin" parameter, which
+ has equivalent effect for XEN platform.
xen_nopv [X86]
Disables the PV optimizations forcing the HVM guest to
@@ -5334,7 +5335,7 @@
as generic guest with no PV drivers. Currently support
XEN HVM, KVM, HYPER_V and VMWARE guest.
- nopvspin [X86,KVM]
+ nopvspin [X86,XEN,KVM]
Disables the qspinlock slow path using PV optimizations
which allow the hypervisor to 'idle' the guest on lock
contention.
diff --git a/arch/x86/xen/spinlock.c b/arch/x86/xen/spinlock.c
index 6deb490..799f4eb 100644
--- a/arch/x86/xen/spinlock.c
+++ b/arch/x86/xen/spinlock.c
@@ -114,9 +114,8 @@ void xen_uninit_lock_cpu(int cpu)
*/
void __init xen_init_spinlocks(void)
{
-
/* Don't need to use pvqspinlock code if there is only 1 vCPU. */
- if (num_possible_cpus() == 1)
+ if (num_possible_cpus() == 1 || nopvspin)
xen_pvspin = false;
if (!xen_pvspin) {
@@ -137,6 +136,7 @@ void __init xen_init_spinlocks(void)
static __init int xen_parse_nopvspin(char *arg)
{
+ pr_notice("\"xen_nopvspin\" is deprecated, please use \"nopvspin\" instead\n");
xen_pvspin = false;
return 0;
}
--
1.8.3.1
^ permalink raw reply related
* [PATCH v7 0/5] Add a unified parameter "nopvspin"
From: Zhenzhong Duan @ 2019-10-21 9:11 UTC (permalink / raw)
To: linux-kernel
Cc: tglx, mingo, bp, x86, pbonzini, rkrcmar, sean.j.christopherson,
vkuznets, wanpengli, jmattson, joro, boris.ostrovsky, jgross,
peterz, will, linux-hyperv, kvm, mikelley, kys, haiyangz,
sthemmin, sashal, Zhenzhong Duan
All the patches have Reviewed-by now, I think v7 could be the final
version.
There are cases folks want to disable spinlock optimization for
debug/test purpose. Xen and hyperv already have parameters "xen_nopvspin"
and "hv_nopvspin" to support that, but kvm doesn't.
The first patch adds that feature to KVM guest with "nopvspin".
For compatibility reason original parameters "xen_nopvspin" and
"hv_nopvspin" are retained and marked obsolete.
v7:
PATCH3: update comment and use goto, add RB [Vitaly Kuznetsov]
v6:
PATCH1: add Reviewed-by [Vitaly Kuznetsov]
PATCH2: change 'pv' to 'PV', add Reviewed-by [Vitaly Kuznetsov]
PATCH3: refactor 'if' branch in kvm_spinlock_init() [Vitaly Kuznetsov]
v5:
PATCH1: new patch to revert a currently unnecessory commit,
code is simpler a bit after that change. [Boris Ostrovsky]
PATCH3: fold 'if' statement,add comments on virt_spin_lock_key,
reorder with PATCH2 to better reflect dependency
PATCH4: fold 'if' statement, add Reviewed-by [Boris Ostrovsky]
PATCH5: add Reviewed-by [Michael Kelley]
v4:
PATCH1: use variable name nopvspin instead of pvspin and
defined it as __initdata, changed print message,
updated patch description [Sean Christopherson]
PATCH2: remove Suggested-by, use "kvm-guest:" prefix [Sean Christopherson]
PATCH3: make variable nopvsin and xen_pvspin coexist
remove Reviewed-by due to code change [Sean Christopherson]
PATCH4: make variable nopvsin and hv_pvspin coexist [Sean Christopherson]
v3:
PATCH2: Fix indentation
v2:
PATCH1: pick the print code change into separate PATCH2,
updated patch description [Vitaly Kuznetsov]
PATCH2: new patch with print code change [Vitaly Kuznetsov]
PATCH3: add Reviewed-by [Juergen Gross]
Zhenzhong Duan (5):
Revert "KVM: X86: Fix setup the virt_spin_lock_key before static key
get initialized"
x86/kvm: Change print code to use pr_*() format
x86/kvm: Add "nopvspin" parameter to disable PV spinlocks
xen: Mark "xen_nopvspin" parameter obsolete
x86/hyperv: Mark "hv_nopvspin" parameter obsolete
Documentation/admin-guide/kernel-parameters.txt | 14 ++++-
arch/x86/hyperv/hv_spinlock.c | 4 ++
arch/x86/include/asm/qspinlock.h | 1 +
arch/x86/kernel/kvm.c | 74 +++++++++++++++----------
arch/x86/xen/spinlock.c | 4 +-
kernel/locking/qspinlock.c | 7 +++
6 files changed, 71 insertions(+), 33 deletions(-)
--
1.8.3.1
^ permalink raw reply
* [PATCH v7 1/5] Revert "KVM: X86: Fix setup the virt_spin_lock_key before static key get initialized"
From: Zhenzhong Duan @ 2019-10-21 9:11 UTC (permalink / raw)
To: linux-kernel
Cc: tglx, mingo, bp, x86, pbonzini, rkrcmar, sean.j.christopherson,
vkuznets, wanpengli, jmattson, joro, boris.ostrovsky, jgross,
peterz, will, linux-hyperv, kvm, mikelley, kys, haiyangz,
sthemmin, sashal, Zhenzhong Duan, H. Peter Anvin
In-Reply-To: <1571649076-2421-1-git-send-email-zhenzhong.duan@oracle.com>
This reverts commit 34226b6b70980a8f81fff3c09a2c889f77edeeff.
Commit 8990cac6e5ea ("x86/jump_label: Initialize static branching
early") adds jump_label_init() call in setup_arch() to make static
keys initialized early, so we could use the original simpler code
again.
The similar change for XEN is in commit 090d54bcbc54 ("Revert
"x86/paravirt: Set up the virt_spin_lock_key after static keys get
initialized"")
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@oracle.com>
Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krcmar <rkrcmar@redhat.com>
Cc: Sean Christopherson <sean.j.christopherson@intel.com>
Cc: Vitaly Kuznetsov <vkuznets@redhat.com>
Cc: Wanpeng Li <wanpengli@tencent.com>
Cc: Jim Mattson <jmattson@google.com>
Cc: Joerg Roedel <joro@8bytes.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
---
arch/x86/kernel/kvm.c | 12 +++---------
1 file changed, 3 insertions(+), 9 deletions(-)
diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
index e820568..3bc6a266 100644
--- a/arch/x86/kernel/kvm.c
+++ b/arch/x86/kernel/kvm.c
@@ -527,13 +527,6 @@ static void kvm_smp_send_call_func_ipi(const struct cpumask *mask)
}
}
-static void __init kvm_smp_prepare_cpus(unsigned int max_cpus)
-{
- native_smp_prepare_cpus(max_cpus);
- if (kvm_para_has_hint(KVM_HINTS_REALTIME))
- static_branch_disable(&virt_spin_lock_key);
-}
-
static void __init kvm_smp_prepare_boot_cpu(void)
{
/*
@@ -633,7 +626,6 @@ static void __init kvm_guest_init(void)
apic_set_eoi_write(kvm_guest_apic_eoi_write);
#ifdef CONFIG_SMP
- smp_ops.smp_prepare_cpus = kvm_smp_prepare_cpus;
smp_ops.smp_prepare_boot_cpu = kvm_smp_prepare_boot_cpu;
if (kvm_para_has_feature(KVM_FEATURE_PV_SCHED_YIELD) &&
!kvm_para_has_hint(KVM_HINTS_REALTIME) &&
@@ -835,8 +827,10 @@ void __init kvm_spinlock_init(void)
if (!kvm_para_has_feature(KVM_FEATURE_PV_UNHALT))
return;
- if (kvm_para_has_hint(KVM_HINTS_REALTIME))
+ if (kvm_para_has_hint(KVM_HINTS_REALTIME)) {
+ static_branch_disable(&virt_spin_lock_key);
return;
+ }
/* Don't use the pvqspinlock code if there is only 1 vCPU. */
if (num_possible_cpus() == 1)
--
1.8.3.1
^ permalink raw reply related
* [PATCH v7 2/5] x86/kvm: Change print code to use pr_*() format
From: Zhenzhong Duan @ 2019-10-21 9:11 UTC (permalink / raw)
To: linux-kernel
Cc: tglx, mingo, bp, x86, pbonzini, rkrcmar, sean.j.christopherson,
vkuznets, wanpengli, jmattson, joro, boris.ostrovsky, jgross,
peterz, will, linux-hyperv, kvm, mikelley, kys, haiyangz,
sthemmin, sashal, Zhenzhong Duan, H. Peter Anvin
In-Reply-To: <1571649076-2421-1-git-send-email-zhenzhong.duan@oracle.com>
pr_*() is preferred than printk(KERN_* ...), after change all the print
in arch/x86/kernel/kvm.c will have "kvm_guest: xxx" style.
No functional change.
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@oracle.com>
Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krcmar <rkrcmar@redhat.com>
Cc: Sean Christopherson <sean.j.christopherson@intel.com>
Cc: Vitaly Kuznetsov <vkuznets@redhat.com>
Cc: Wanpeng Li <wanpengli@tencent.com>
Cc: Jim Mattson <jmattson@google.com>
Cc: Joerg Roedel <joro@8bytes.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
---
arch/x86/kernel/kvm.c | 30 ++++++++++++++++--------------
1 file changed, 16 insertions(+), 14 deletions(-)
diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
index 3bc6a266..249f14a 100644
--- a/arch/x86/kernel/kvm.c
+++ b/arch/x86/kernel/kvm.c
@@ -7,6 +7,8 @@
* Authors: Anthony Liguori <aliguori@us.ibm.com>
*/
+#define pr_fmt(fmt) "kvm_guest: " fmt
+
#include <linux/context_tracking.h>
#include <linux/init.h>
#include <linux/kernel.h>
@@ -286,8 +288,8 @@ static void kvm_register_steal_time(void)
return;
wrmsrl(MSR_KVM_STEAL_TIME, (slow_virt_to_phys(st) | KVM_MSR_ENABLED));
- pr_info("kvm-stealtime: cpu %d, msr %llx\n",
- cpu, (unsigned long long) slow_virt_to_phys(st));
+ pr_info("stealtime: cpu %d, msr %llx\n", cpu,
+ (unsigned long long) slow_virt_to_phys(st));
}
static DEFINE_PER_CPU_DECRYPTED(unsigned long, kvm_apic_eoi) = KVM_PV_EOI_DISABLED;
@@ -321,8 +323,7 @@ static void kvm_guest_cpu_init(void)
wrmsrl(MSR_KVM_ASYNC_PF_EN, pa);
__this_cpu_write(apf_reason.enabled, 1);
- printk(KERN_INFO"KVM setup async PF for cpu %d\n",
- smp_processor_id());
+ pr_info("setup async PF for cpu %d\n", smp_processor_id());
}
if (kvm_para_has_feature(KVM_FEATURE_PV_EOI)) {
@@ -347,8 +348,7 @@ static void kvm_pv_disable_apf(void)
wrmsrl(MSR_KVM_ASYNC_PF_EN, 0);
__this_cpu_write(apf_reason.enabled, 0);
- printk(KERN_INFO"Unregister pv shared memory for cpu %d\n",
- smp_processor_id());
+ pr_info("Unregister PV shared memory for cpu %d\n", smp_processor_id());
}
static void kvm_pv_guest_cpu_reboot(void *unused)
@@ -469,7 +469,8 @@ static void __send_ipi_mask(const struct cpumask *mask, int vector)
} else {
ret = kvm_hypercall4(KVM_HC_SEND_IPI, (unsigned long)ipi_bitmap,
(unsigned long)(ipi_bitmap >> BITS_PER_LONG), min, icr);
- WARN_ONCE(ret < 0, "KVM: failed to send PV IPI: %ld", ret);
+ WARN_ONCE(ret < 0, "kvm_guest: failed to send PV IPI: %ld",
+ ret);
min = max = apic_id;
ipi_bitmap = 0;
}
@@ -479,7 +480,8 @@ static void __send_ipi_mask(const struct cpumask *mask, int vector)
if (ipi_bitmap) {
ret = kvm_hypercall4(KVM_HC_SEND_IPI, (unsigned long)ipi_bitmap,
(unsigned long)(ipi_bitmap >> BITS_PER_LONG), min, icr);
- WARN_ONCE(ret < 0, "KVM: failed to send PV IPI: %ld", ret);
+ WARN_ONCE(ret < 0, "kvm_guest: failed to send PV IPI: %ld",
+ ret);
}
local_irq_restore(flags);
@@ -509,7 +511,7 @@ static void kvm_setup_pv_ipi(void)
{
apic->send_IPI_mask = kvm_send_ipi_mask;
apic->send_IPI_mask_allbutself = kvm_send_ipi_mask_allbutself;
- pr_info("KVM setup pv IPIs\n");
+ pr_info("setup PV IPIs\n");
}
static void kvm_smp_send_call_func_ipi(const struct cpumask *mask)
@@ -631,11 +633,11 @@ static void __init kvm_guest_init(void)
!kvm_para_has_hint(KVM_HINTS_REALTIME) &&
kvm_para_has_feature(KVM_FEATURE_STEAL_TIME)) {
smp_ops.send_call_func_ipi = kvm_smp_send_call_func_ipi;
- pr_info("KVM setup pv sched yield\n");
+ pr_info("setup PV sched yield\n");
}
if (cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN, "x86/kvm:online",
kvm_cpu_online, kvm_cpu_down_prepare) < 0)
- pr_err("kvm_guest: Failed to install cpu hotplug callbacks\n");
+ pr_err("failed to install cpu hotplug callbacks\n");
#else
sev_map_percpu_data();
kvm_guest_cpu_init();
@@ -738,7 +740,7 @@ static __init int kvm_setup_pv_tlb_flush(void)
zalloc_cpumask_var_node(per_cpu_ptr(&__pv_tlb_mask, cpu),
GFP_KERNEL, cpu_to_node(cpu));
}
- pr_info("KVM setup pv remote TLB flush\n");
+ pr_info("setup PV remote TLB flush\n");
}
return 0;
@@ -866,8 +868,8 @@ static void kvm_enable_host_haltpoll(void *i)
void arch_haltpoll_enable(unsigned int cpu)
{
if (!kvm_para_has_feature(KVM_FEATURE_POLL_CONTROL)) {
- pr_err_once("kvm: host does not support poll control\n");
- pr_err_once("kvm: host upgrade recommended\n");
+ pr_err_once("host does not support poll control\n");
+ pr_err_once("host upgrade recommended\n");
return;
}
--
1.8.3.1
^ permalink raw reply related
* Re: [PATCH v6 3/5] x86/kvm: Add "nopvspin" parameter to disable PV spinlocks
From: Zhenzhong Duan @ 2019-10-22 2:45 UTC (permalink / raw)
To: Vitaly Kuznetsov, linux-kernel
Cc: linux-hyperv, kvm, kys, haiyangz, sthemmin, sashal, tglx, mingo,
bp, pbonzini, rkrcmar, sean.j.christopherson, wanpengli, jmattson,
joro, boris.ostrovsky, jgross, sstabellini, peterz,
Jonathan Corbet, H. Peter Anvin, Will Deacon
In-Reply-To: <87k18y1hc1.fsf@vitty.brq.redhat.com>
On 2019/10/21 19:14, Vitaly Kuznetsov wrote:
>> index 249f14a..e9c76d8 100644
>> --- a/arch/x86/kernel/kvm.c
>> +++ b/arch/x86/kernel/kvm.c
>> @@ -825,18 +825,44 @@ __visible bool __kvm_vcpu_is_preempted(long cpu)
>> */
>> void __init kvm_spinlock_init(void)
>> {
>> - /* Does host kernel support KVM_FEATURE_PV_UNHALT? */
>> - if (!kvm_para_has_feature(KVM_FEATURE_PV_UNHALT))
>> + /*
>> + * PV spinlocks is disabled if no host side support, then native
>> + * qspinlock will be used. As native qspinlock is a fair lock, there is
>> + * lock holder preemption issue using it in a guest, imaging one pCPU
>> + * running 10 vCPUs of same guest contending same lock.
>> + *
>> + * virt_spin_lock() is introduced as an optimization for that scenario
>> + * which is enabled by virt_spin_lock_key key. To use that optimization,
>> + * virt_spin_lock_key isn't disabled here.
>> + */
> My take (if I properly understood what you say) would be:
>
> "In case host doesn't support KVM_FEATURE_PV_UNHALT there is still an
> advantage of keeping virt_spin_lock_key enabled: virt_spin_lock() is
> preferred over native qspinlock when vCPU is preempted."
Yes, that's what I mean, maybe I didn't explain clearly due to my pool
english,
I'll use your explanation instead.
>
>> + if (!kvm_para_has_feature(KVM_FEATURE_PV_UNHALT)) {
>> + pr_info("PV spinlocks disabled, no host support.\n");
>> return;
>> + }
>>
>> + /*
>> + * Disable PV qspinlock and use native qspinlock when dedicated pCPUs
>> + * are available.
>> + */
>> if (kvm_para_has_hint(KVM_HINTS_REALTIME)) {
>> + pr_info("PV spinlocks disabled with KVM_HINTS_REALTIME hints.\n");
>> + static_branch_disable(&virt_spin_lock_key);
>> + return;
>> + }
>> +
>> + if (num_possible_cpus() == 1) {
>> + pr_info("PV spinlocks disabled, single CPU.\n");
>> static_branch_disable(&virt_spin_lock_key);
>> return;
>> }
>>
>> - /* Don't use the pvqspinlock code if there is only 1 vCPU. */
>> - if (num_possible_cpus() == 1)
>> + if (nopvspin) {
>> + pr_info("PV spinlocks disabled, forced by \"nopvspin\" parameter.\n");
>> + static_branch_disable(&virt_spin_lock_key);
>> return;
> You could've replaced this 'static_branch_disable(); return;' pattern
> with a goto to the end of the function to save a few lines but this
> looks good anyways.
>
> Reviewed-by: Vitaly Kuznetsov<vkuznets@redhat.com>
Ok, will do, thanks for review.
Zhenzhong
^ permalink raw reply
* Re: [PATCH v6 3/5] x86/kvm: Add "nopvspin" parameter to disable PV spinlocks
From: Vitaly Kuznetsov @ 2019-10-21 11:14 UTC (permalink / raw)
To: Zhenzhong Duan, linux-kernel
Cc: linux-hyperv, kvm, kys, haiyangz, sthemmin, sashal, tglx, mingo,
bp, pbonzini, rkrcmar, sean.j.christopherson, wanpengli, jmattson,
joro, boris.ostrovsky, jgross, sstabellini, peterz,
Zhenzhong Duan, Jonathan Corbet, H. Peter Anvin, Will Deacon
In-Reply-To: <1571102367-31595-4-git-send-email-zhenzhong.duan@oracle.com>
Zhenzhong Duan <zhenzhong.duan@oracle.com> writes:
> There are cases where a guest tries to switch spinlocks to bare metal
> behavior (e.g. by setting "xen_nopvspin" on XEN platform and
> "hv_nopvspin" on HYPER_V).
>
> That feature is missed on KVM, add a new parameter "nopvspin" to disable
> PV spinlocks for KVM guest.
>
> The new 'nopvspin' parameter will also replace Xen and Hyper-V specific
> parameters in future patches.
>
> Define variable nopvsin as global because it will be used in future
> patches as above.
>
> Signed-off-by: Zhenzhong Duan <zhenzhong.duan@oracle.com>
> Cc: Jonathan Corbet <corbet@lwn.net>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: Ingo Molnar <mingo@redhat.com>
> Cc: Borislav Petkov <bp@alien8.de>
> Cc: "H. Peter Anvin" <hpa@zytor.com>
> Cc: Paolo Bonzini <pbonzini@redhat.com>
> Cc: Radim Krcmar <rkrcmar@redhat.com>
> Cc: Sean Christopherson <sean.j.christopherson@intel.com>
> Cc: Vitaly Kuznetsov <vkuznets@redhat.com>
> Cc: Wanpeng Li <wanpengli@tencent.com>
> Cc: Jim Mattson <jmattson@google.com>
> Cc: Joerg Roedel <joro@8bytes.org>
> Cc: Peter Zijlstra <peterz@infradead.org>
> Cc: Will Deacon <will@kernel.org>
> ---
> Documentation/admin-guide/kernel-parameters.txt | 5 ++++
> arch/x86/include/asm/qspinlock.h | 1 +
> arch/x86/kernel/kvm.c | 34 ++++++++++++++++++++++---
> kernel/locking/qspinlock.c | 7 +++++
> 4 files changed, 43 insertions(+), 4 deletions(-)
>
> diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
> index a84a83f..bd49ed2 100644
> --- a/Documentation/admin-guide/kernel-parameters.txt
> +++ b/Documentation/admin-guide/kernel-parameters.txt
> @@ -5334,6 +5334,11 @@
> as generic guest with no PV drivers. Currently support
> XEN HVM, KVM, HYPER_V and VMWARE guest.
>
> + nopvspin [X86,KVM]
> + Disables the qspinlock slow path using PV optimizations
> + which allow the hypervisor to 'idle' the guest on lock
> + contention.
> +
> xirc2ps_cs= [NET,PCMCIA]
> Format:
> <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
> diff --git a/arch/x86/include/asm/qspinlock.h b/arch/x86/include/asm/qspinlock.h
> index 444d6fd..d86ab94 100644
> --- a/arch/x86/include/asm/qspinlock.h
> +++ b/arch/x86/include/asm/qspinlock.h
> @@ -32,6 +32,7 @@ static __always_inline u32 queued_fetch_set_pending_acquire(struct qspinlock *lo
> extern void __pv_init_lock_hash(void);
> extern void __pv_queued_spin_lock_slowpath(struct qspinlock *lock, u32 val);
> extern void __raw_callee_save___pv_queued_spin_unlock(struct qspinlock *lock);
> +extern bool nopvspin;
>
> #define queued_spin_unlock queued_spin_unlock
> /**
> diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
> index 249f14a..e9c76d8 100644
> --- a/arch/x86/kernel/kvm.c
> +++ b/arch/x86/kernel/kvm.c
> @@ -825,18 +825,44 @@ __visible bool __kvm_vcpu_is_preempted(long cpu)
> */
> void __init kvm_spinlock_init(void)
> {
> - /* Does host kernel support KVM_FEATURE_PV_UNHALT? */
> - if (!kvm_para_has_feature(KVM_FEATURE_PV_UNHALT))
> + /*
> + * PV spinlocks is disabled if no host side support, then native
> + * qspinlock will be used. As native qspinlock is a fair lock, there is
> + * lock holder preemption issue using it in a guest, imaging one pCPU
> + * running 10 vCPUs of same guest contending same lock.
> + *
> + * virt_spin_lock() is introduced as an optimization for that scenario
> + * which is enabled by virt_spin_lock_key key. To use that optimization,
> + * virt_spin_lock_key isn't disabled here.
> + */
My take (if I properly understood what you say) would be:
"In case host doesn't support KVM_FEATURE_PV_UNHALT there is still an
advantage of keeping virt_spin_lock_key enabled: virt_spin_lock() is
preferred over native qspinlock when vCPU is preempted."
> + if (!kvm_para_has_feature(KVM_FEATURE_PV_UNHALT)) {
> + pr_info("PV spinlocks disabled, no host support.\n");
> return;
> + }
>
> + /*
> + * Disable PV qspinlock and use native qspinlock when dedicated pCPUs
> + * are available.
> + */
> if (kvm_para_has_hint(KVM_HINTS_REALTIME)) {
> + pr_info("PV spinlocks disabled with KVM_HINTS_REALTIME hints.\n");
> + static_branch_disable(&virt_spin_lock_key);
> + return;
> + }
> +
> + if (num_possible_cpus() == 1) {
> + pr_info("PV spinlocks disabled, single CPU.\n");
> static_branch_disable(&virt_spin_lock_key);
> return;
> }
>
> - /* Don't use the pvqspinlock code if there is only 1 vCPU. */
> - if (num_possible_cpus() == 1)
> + if (nopvspin) {
> + pr_info("PV spinlocks disabled, forced by \"nopvspin\" parameter.\n");
> + static_branch_disable(&virt_spin_lock_key);
> return;
You could've replaced this 'static_branch_disable(); return;' pattern
with a goto to the end of the function to save a few lines but this
looks good anyways.
Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
> + }
> +
> + pr_info("PV spinlocks enabled\n");
>
> __pv_init_lock_hash();
> pv_ops.lock.queued_spin_lock_slowpath = __pv_queued_spin_lock_slowpath;
> diff --git a/kernel/locking/qspinlock.c b/kernel/locking/qspinlock.c
> index 2473f10..75193d6 100644
> --- a/kernel/locking/qspinlock.c
> +++ b/kernel/locking/qspinlock.c
> @@ -580,4 +580,11 @@ void queued_spin_lock_slowpath(struct qspinlock *lock, u32 val)
> #include "qspinlock_paravirt.h"
> #include "qspinlock.c"
>
> +bool nopvspin __initdata;
> +static __init int parse_nopvspin(char *arg)
> +{
> + nopvspin = true;
> + return 0;
> +}
> +early_param("nopvspin", parse_nopvspin);
> #endif
--
Vitaly
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox