summaryrefslogtreecommitdiff
path: root/drivers/net/hyperv
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/net/hyperv')
-rw-r--r--drivers/net/hyperv/hyperv_net.h50
-rw-r--r--drivers/net/hyperv/netvsc.c267
-rw-r--r--drivers/net/hyperv/netvsc_drv.c448
-rw-r--r--drivers/net/hyperv/rndis_filter.c205
4 files changed, 617 insertions, 353 deletions
diff --git a/drivers/net/hyperv/hyperv_net.h b/drivers/net/hyperv/hyperv_net.h
index 8b3bd8ecd..591af71ea 100644
--- a/drivers/net/hyperv/hyperv_net.h
+++ b/drivers/net/hyperv/hyperv_net.h
@@ -158,7 +158,7 @@ enum rndis_device_state {
};
struct rndis_device {
- struct netvsc_device *net_dev;
+ struct net_device *ndev;
enum rndis_device_state state;
bool link_state;
@@ -173,6 +173,7 @@ struct rndis_device {
/* Interface */
struct rndis_message;
+struct netvsc_device;
int netvsc_device_add(struct hv_device *device, void *additional_info);
int netvsc_device_remove(struct hv_device *device);
int netvsc_send(struct hv_device *device,
@@ -189,8 +190,8 @@ int netvsc_recv_callback(struct hv_device *device_obj,
struct vmbus_channel *channel,
u16 vlan_tci);
void netvsc_channel_cb(void *context);
-int rndis_filter_open(struct hv_device *dev);
-int rndis_filter_close(struct hv_device *dev);
+int rndis_filter_open(struct netvsc_device *nvdev);
+int rndis_filter_close(struct netvsc_device *nvdev);
int rndis_filter_device_add(struct hv_device *dev,
void *additional_info);
void rndis_filter_device_remove(struct hv_device *dev);
@@ -200,7 +201,9 @@ int rndis_filter_receive(struct hv_device *dev,
struct vmbus_channel *channel);
int rndis_filter_set_packet_filter(struct rndis_device *dev, u32 new_filter);
-int rndis_filter_set_device_mac(struct hv_device *hdev, char *mac);
+int rndis_filter_set_device_mac(struct net_device *ndev, char *mac);
+
+void netvsc_switch_datapath(struct net_device *nv_dev, bool vf);
#define NVSP_INVALID_PROTOCOL_VERSION ((u32)0xFFFFFFFF)
@@ -645,6 +648,8 @@ struct netvsc_reconfig {
struct net_device_context {
/* point back to our device context */
struct hv_device *device_ctx;
+ /* netvsc_device */
+ struct netvsc_device *nvdev;
/* reconfigure work */
struct delayed_work dwork;
/* last reconfig time */
@@ -663,17 +668,26 @@ struct net_device_context {
/* Ethtool settings */
u8 duplex;
u32 speed;
+
+ /* the device is going away */
+ bool start_remove;
+
+ /* State to manage the associated VF interface. */
+ struct net_device *vf_netdev;
+ bool vf_inject;
+ atomic_t vf_use_cnt;
+ /* 1: allocated, serial number is valid. 0: not allocated */
+ u32 vf_alloc;
+ /* Serial number of the VF to team with */
+ u32 vf_serial;
};
/* Per netvsc device */
struct netvsc_device {
- struct hv_device *dev;
-
u32 nvsp_version;
atomic_t num_outstanding_sends;
wait_queue_head_t wait_drain;
- bool start_remove;
bool destroy;
/* Receive buffer allocated by us but manages by NetVSP */
@@ -699,8 +713,6 @@ struct netvsc_device {
struct nvsp_message revoke_packet;
/* unsigned char HwMacAddr[HW_MACADDR_LEN]; */
- struct net_device *ndev;
-
struct vmbus_channel *chn_table[VRSS_CHANNEL_MAX];
u32 send_table[VRSS_SEND_TAB_SIZE];
u32 max_chn;
@@ -723,15 +735,21 @@ struct netvsc_device {
u32 max_pkt; /* max number of pkt in one send, e.g. 8 */
u32 pkt_align; /* alignment bytes, e.g. 8 */
- /* The net device context */
- struct net_device_context *nd_ctx;
-
- /* 1: allocated, serial number is valid. 0: not allocated */
- u32 vf_alloc;
- /* Serial number of the VF to team with */
- u32 vf_serial;
+ atomic_t open_cnt;
};
+static inline struct netvsc_device *
+net_device_to_netvsc_device(struct net_device *ndev)
+{
+ return ((struct net_device_context *)netdev_priv(ndev))->nvdev;
+}
+
+static inline struct netvsc_device *
+hv_device_to_netvsc_device(struct hv_device *device)
+{
+ return net_device_to_netvsc_device(hv_get_drvdata(device));
+}
+
/* NdisInitialize message */
struct rndis_initialize_request {
u32 req_id;
diff --git a/drivers/net/hyperv/netvsc.c b/drivers/net/hyperv/netvsc.c
index ec313fc08..410fb8e81 100644
--- a/drivers/net/hyperv/netvsc.c
+++ b/drivers/net/hyperv/netvsc.c
@@ -33,11 +33,36 @@
#include "hyperv_net.h"
+/*
+ * Switch the data path from the synthetic interface to the VF
+ * interface.
+ */
+void netvsc_switch_datapath(struct net_device *ndev, bool vf)
+{
+ struct net_device_context *net_device_ctx = netdev_priv(ndev);
+ struct hv_device *dev = net_device_ctx->device_ctx;
+ struct netvsc_device *nv_dev = net_device_ctx->nvdev;
+ struct nvsp_message *init_pkt = &nv_dev->channel_init_pkt;
+
+ memset(init_pkt, 0, sizeof(struct nvsp_message));
+ init_pkt->hdr.msg_type = NVSP_MSG4_TYPE_SWITCH_DATA_PATH;
+ if (vf)
+ init_pkt->msg.v4_msg.active_dp.active_datapath =
+ NVSP_DATAPATH_VF;
+ else
+ init_pkt->msg.v4_msg.active_dp.active_datapath =
+ NVSP_DATAPATH_SYNTHETIC;
-static struct netvsc_device *alloc_net_device(struct hv_device *device)
+ vmbus_sendpacket(dev->channel, init_pkt,
+ sizeof(struct nvsp_message),
+ (unsigned long)init_pkt,
+ VM_PKT_DATA_INBAND, 0);
+}
+
+
+static struct netvsc_device *alloc_net_device(void)
{
struct netvsc_device *net_device;
- struct net_device *ndev = hv_get_drvdata(device);
net_device = kzalloc(sizeof(struct netvsc_device), GFP_KERNEL);
if (!net_device)
@@ -50,14 +75,11 @@ static struct netvsc_device *alloc_net_device(struct hv_device *device)
}
init_waitqueue_head(&net_device->wait_drain);
- net_device->start_remove = false;
net_device->destroy = false;
- net_device->dev = device;
- net_device->ndev = ndev;
+ atomic_set(&net_device->open_cnt, 0);
net_device->max_pkt = RNDIS_MAX_PKT_DEFAULT;
net_device->pkt_align = RNDIS_PKT_ALIGN_DEFAULT;
- hv_set_drvdata(device, net_device);
return net_device;
}
@@ -69,9 +91,8 @@ static void free_netvsc_device(struct netvsc_device *nvdev)
static struct netvsc_device *get_outbound_net_device(struct hv_device *device)
{
- struct netvsc_device *net_device;
+ struct netvsc_device *net_device = hv_device_to_netvsc_device(device);
- net_device = hv_get_drvdata(device);
if (net_device && net_device->destroy)
net_device = NULL;
@@ -80,9 +101,7 @@ static struct netvsc_device *get_outbound_net_device(struct hv_device *device)
static struct netvsc_device *get_inbound_net_device(struct hv_device *device)
{
- struct netvsc_device *net_device;
-
- net_device = hv_get_drvdata(device);
+ struct netvsc_device *net_device = hv_device_to_netvsc_device(device);
if (!net_device)
goto get_in_err;
@@ -96,11 +115,12 @@ get_in_err:
}
-static int netvsc_destroy_buf(struct netvsc_device *net_device)
+static int netvsc_destroy_buf(struct hv_device *device)
{
struct nvsp_message *revoke_packet;
int ret = 0;
- struct net_device *ndev = net_device->ndev;
+ struct net_device *ndev = hv_get_drvdata(device);
+ struct netvsc_device *net_device = net_device_to_netvsc_device(ndev);
/*
* If we got a section count, it means we received a
@@ -118,7 +138,7 @@ static int netvsc_destroy_buf(struct netvsc_device *net_device)
revoke_packet->msg.v1_msg.
revoke_recv_buf.id = NETVSC_RECEIVE_BUFFER_ID;
- ret = vmbus_sendpacket(net_device->dev->channel,
+ ret = vmbus_sendpacket(device->channel,
revoke_packet,
sizeof(struct nvsp_message),
(unsigned long)revoke_packet,
@@ -136,8 +156,8 @@ static int netvsc_destroy_buf(struct netvsc_device *net_device)
/* Teardown the gpadl on the vsp end */
if (net_device->recv_buf_gpadl_handle) {
- ret = vmbus_teardown_gpadl(net_device->dev->channel,
- net_device->recv_buf_gpadl_handle);
+ ret = vmbus_teardown_gpadl(device->channel,
+ net_device->recv_buf_gpadl_handle);
/* If we failed here, we might as well return and have a leak
* rather than continue and a bugchk
@@ -178,7 +198,7 @@ static int netvsc_destroy_buf(struct netvsc_device *net_device)
revoke_packet->msg.v1_msg.revoke_send_buf.id =
NETVSC_SEND_BUFFER_ID;
- ret = vmbus_sendpacket(net_device->dev->channel,
+ ret = vmbus_sendpacket(device->channel,
revoke_packet,
sizeof(struct nvsp_message),
(unsigned long)revoke_packet,
@@ -194,7 +214,7 @@ static int netvsc_destroy_buf(struct netvsc_device *net_device)
}
/* Teardown the gpadl on the vsp end */
if (net_device->send_buf_gpadl_handle) {
- ret = vmbus_teardown_gpadl(net_device->dev->channel,
+ ret = vmbus_teardown_gpadl(device->channel,
net_device->send_buf_gpadl_handle);
/* If we failed here, we might as well return and have a leak
@@ -220,7 +240,6 @@ static int netvsc_destroy_buf(struct netvsc_device *net_device)
static int netvsc_init_buf(struct hv_device *device)
{
int ret = 0;
- unsigned long t;
struct netvsc_device *net_device;
struct nvsp_message *init_packet;
struct net_device *ndev;
@@ -229,7 +248,7 @@ static int netvsc_init_buf(struct hv_device *device)
net_device = get_outbound_net_device(device);
if (!net_device)
return -ENODEV;
- ndev = net_device->ndev;
+ ndev = hv_get_drvdata(device);
node = cpu_to_node(device->channel->target_cpu);
net_device->recv_buf = vzalloc_node(net_device->recv_buf_size, node);
@@ -281,9 +300,7 @@ static int netvsc_init_buf(struct hv_device *device)
goto cleanup;
}
- t = wait_for_completion_timeout(&net_device->channel_init_wait, 5*HZ);
- BUG_ON(t == 0);
-
+ wait_for_completion(&net_device->channel_init_wait);
/* Check the response */
if (init_packet->msg.v1_msg.
@@ -366,8 +383,7 @@ static int netvsc_init_buf(struct hv_device *device)
goto cleanup;
}
- t = wait_for_completion_timeout(&net_device->channel_init_wait, 5*HZ);
- BUG_ON(t == 0);
+ wait_for_completion(&net_device->channel_init_wait);
/* Check the response */
if (init_packet->msg.v1_msg.
@@ -406,7 +422,7 @@ static int netvsc_init_buf(struct hv_device *device)
goto exit;
cleanup:
- netvsc_destroy_buf(net_device);
+ netvsc_destroy_buf(device);
exit:
return ret;
@@ -419,8 +435,8 @@ static int negotiate_nvsp_ver(struct hv_device *device,
struct nvsp_message *init_packet,
u32 nvsp_ver)
{
+ struct net_device *ndev = hv_get_drvdata(device);
int ret;
- unsigned long t;
memset(init_packet, 0, sizeof(struct nvsp_message));
init_packet->hdr.msg_type = NVSP_MSG_TYPE_INIT;
@@ -437,10 +453,7 @@ static int negotiate_nvsp_ver(struct hv_device *device,
if (ret != 0)
return ret;
- t = wait_for_completion_timeout(&net_device->channel_init_wait, 5*HZ);
-
- if (t == 0)
- return -ETIMEDOUT;
+ wait_for_completion(&net_device->channel_init_wait);
if (init_packet->msg.init_msg.init_complete.status !=
NVSP_STAT_SUCCESS)
@@ -452,8 +465,7 @@ static int negotiate_nvsp_ver(struct hv_device *device,
/* NVSPv2 or later: Send NDIS config */
memset(init_packet, 0, sizeof(struct nvsp_message));
init_packet->hdr.msg_type = NVSP_MSG2_TYPE_SEND_NDIS_CONFIG;
- init_packet->msg.v2_msg.send_ndis_config.mtu = net_device->ndev->mtu +
- ETH_HLEN;
+ init_packet->msg.v2_msg.send_ndis_config.mtu = ndev->mtu + ETH_HLEN;
init_packet->msg.v2_msg.send_ndis_config.capability.ieee8021q = 1;
if (nvsp_ver >= NVSP_PROTOCOL_VERSION_5)
@@ -473,7 +485,6 @@ static int netvsc_connect_vsp(struct hv_device *device)
struct netvsc_device *net_device;
struct nvsp_message *init_packet;
int ndis_version;
- struct net_device *ndev;
u32 ver_list[] = { NVSP_PROTOCOL_VERSION_1, NVSP_PROTOCOL_VERSION_2,
NVSP_PROTOCOL_VERSION_4, NVSP_PROTOCOL_VERSION_5 };
int i, num_ver = 4; /* number of different NVSP versions */
@@ -481,7 +492,6 @@ static int netvsc_connect_vsp(struct hv_device *device)
net_device = get_outbound_net_device(device);
if (!net_device)
return -ENODEV;
- ndev = net_device->ndev;
init_packet = &net_device->channel_init_pkt;
@@ -537,9 +547,9 @@ cleanup:
return ret;
}
-static void netvsc_disconnect_vsp(struct netvsc_device *net_device)
+static void netvsc_disconnect_vsp(struct hv_device *device)
{
- netvsc_destroy_buf(net_device);
+ netvsc_destroy_buf(device);
}
/*
@@ -547,24 +557,13 @@ static void netvsc_disconnect_vsp(struct netvsc_device *net_device)
*/
int netvsc_device_remove(struct hv_device *device)
{
- struct netvsc_device *net_device;
- unsigned long flags;
-
- net_device = hv_get_drvdata(device);
-
- netvsc_disconnect_vsp(net_device);
+ struct net_device *ndev = hv_get_drvdata(device);
+ struct net_device_context *net_device_ctx = netdev_priv(ndev);
+ struct netvsc_device *net_device = net_device_ctx->nvdev;
- /*
- * Since we have already drained, we don't need to busy wait
- * as was done in final_release_stor_device()
- * Note that we cannot set the ext pointer to NULL until
- * we have drained - to drain the outgoing packets, we need to
- * allow incoming packets.
- */
+ netvsc_disconnect_vsp(device);
- spin_lock_irqsave(&device->channel->inbound_lock, flags);
- hv_set_drvdata(device, NULL);
- spin_unlock_irqrestore(&device->channel->inbound_lock, flags);
+ net_device_ctx->nvdev = NULL;
/*
* At this point, no one should be accessing net_device
@@ -612,12 +611,11 @@ static void netvsc_send_completion(struct netvsc_device *net_device,
{
struct nvsp_message *nvsp_packet;
struct hv_netvsc_packet *nvsc_packet;
- struct net_device *ndev;
+ struct net_device *ndev = hv_get_drvdata(device);
+ struct net_device_context *net_device_ctx = netdev_priv(ndev);
u32 send_index;
struct sk_buff *skb;
- ndev = net_device->ndev;
-
nvsp_packet = (struct nvsp_message *)((unsigned long)packet +
(packet->offset8 << 3));
@@ -662,7 +660,7 @@ static void netvsc_send_completion(struct netvsc_device *net_device,
wake_up(&net_device->wait_drain);
if (netif_tx_queue_stopped(netdev_get_tx_queue(ndev, q_idx)) &&
- !net_device->start_remove &&
+ !net_device_ctx->start_remove &&
(hv_ringbuf_avail_percent(&channel->outbound) >
RING_AVAIL_PERCENT_HIWATER || queue_sends < 1))
netif_tx_wake_queue(netdev_get_tx_queue(
@@ -746,6 +744,7 @@ static u32 netvsc_copy_to_send_buf(struct netvsc_device *net_device,
}
static inline int netvsc_send_pkt(
+ struct hv_device *device,
struct hv_netvsc_packet *packet,
struct netvsc_device *net_device,
struct hv_page_buffer **pb,
@@ -754,7 +753,7 @@ static inline int netvsc_send_pkt(
struct nvsp_message nvmsg;
u16 q_idx = packet->q_idx;
struct vmbus_channel *out_channel = net_device->chn_table[q_idx];
- struct net_device *ndev = net_device->ndev;
+ struct net_device *ndev = hv_get_drvdata(device);
u64 req_id;
int ret;
struct hv_page_buffer *pgbuf;
@@ -949,7 +948,8 @@ int netvsc_send(struct hv_device *device,
}
if (msd_send) {
- m_ret = netvsc_send_pkt(msd_send, net_device, NULL, msd_skb);
+ m_ret = netvsc_send_pkt(device, msd_send, net_device,
+ NULL, msd_skb);
if (m_ret != 0) {
netvsc_free_send_slot(net_device,
@@ -960,7 +960,7 @@ int netvsc_send(struct hv_device *device,
send_now:
if (cur_send)
- ret = netvsc_send_pkt(cur_send, net_device, pb, skb);
+ ret = netvsc_send_pkt(device, cur_send, net_device, pb, skb);
if (ret != 0 && section_index != NETVSC_INVALID_INDEX)
netvsc_free_send_slot(net_device, section_index);
@@ -976,9 +976,7 @@ static void netvsc_send_recv_completion(struct hv_device *device,
struct nvsp_message recvcompMessage;
int retries = 0;
int ret;
- struct net_device *ndev;
-
- ndev = net_device->ndev;
+ struct net_device *ndev = hv_get_drvdata(device);
recvcompMessage.hdr.msg_type =
NVSP_MSG1_TYPE_SEND_RNDIS_PKT_COMPLETE;
@@ -1025,11 +1023,9 @@ static void netvsc_receive(struct netvsc_device *net_device,
u32 status = NVSP_STAT_SUCCESS;
int i;
int count = 0;
- struct net_device *ndev;
+ struct net_device *ndev = hv_get_drvdata(device);
void *data;
- ndev = net_device->ndev;
-
/*
* All inbound packets other than send completion should be xfer page
* packet
@@ -1085,14 +1081,13 @@ static void netvsc_send_table(struct hv_device *hdev,
struct nvsp_message *nvmsg)
{
struct netvsc_device *nvscdev;
- struct net_device *ndev;
+ struct net_device *ndev = hv_get_drvdata(hdev);
int i;
u32 count, *tab;
nvscdev = get_outbound_net_device(hdev);
if (!nvscdev)
return;
- ndev = nvscdev->ndev;
count = nvmsg->msg.v5_msg.send_table.count;
if (count != VRSS_SEND_TAB_SIZE) {
@@ -1107,16 +1102,16 @@ static void netvsc_send_table(struct hv_device *hdev,
nvscdev->send_table[i] = tab[i];
}
-static void netvsc_send_vf(struct netvsc_device *nvdev,
+static void netvsc_send_vf(struct net_device_context *net_device_ctx,
struct nvsp_message *nvmsg)
{
- nvdev->vf_alloc = nvmsg->msg.v4_msg.vf_assoc.allocated;
- nvdev->vf_serial = nvmsg->msg.v4_msg.vf_assoc.serial;
+ net_device_ctx->vf_alloc = nvmsg->msg.v4_msg.vf_assoc.allocated;
+ net_device_ctx->vf_serial = nvmsg->msg.v4_msg.vf_assoc.serial;
}
static inline void netvsc_receive_inband(struct hv_device *hdev,
- struct netvsc_device *nvdev,
- struct nvsp_message *nvmsg)
+ struct net_device_context *net_device_ctx,
+ struct nvsp_message *nvmsg)
{
switch (nvmsg->hdr.msg_type) {
case NVSP_MSG5_TYPE_SEND_INDIRECTION_TABLE:
@@ -1124,11 +1119,45 @@ static inline void netvsc_receive_inband(struct hv_device *hdev,
break;
case NVSP_MSG4_TYPE_SEND_VF_ASSOCIATION:
- netvsc_send_vf(nvdev, nvmsg);
+ netvsc_send_vf(net_device_ctx, nvmsg);
break;
}
}
+static void netvsc_process_raw_pkt(struct hv_device *device,
+ struct vmbus_channel *channel,
+ struct netvsc_device *net_device,
+ struct net_device *ndev,
+ u64 request_id,
+ struct vmpacket_descriptor *desc)
+{
+ struct nvsp_message *nvmsg;
+ struct net_device_context *net_device_ctx = netdev_priv(ndev);
+
+ nvmsg = (struct nvsp_message *)((unsigned long)
+ desc + (desc->offset8 << 3));
+
+ switch (desc->type) {
+ case VM_PKT_COMP:
+ netvsc_send_completion(net_device, channel, device, desc);
+ break;
+
+ case VM_PKT_DATA_USING_XFER_PAGES:
+ netvsc_receive(net_device, channel, device, desc);
+ break;
+
+ case VM_PKT_DATA_INBAND:
+ netvsc_receive_inband(device, net_device_ctx, nvmsg);
+ break;
+
+ default:
+ netdev_err(ndev, "unhandled packet type %d, tid %llx\n",
+ desc->type, request_id);
+ break;
+ }
+}
+
+
void netvsc_channel_cb(void *context)
{
int ret;
@@ -1141,7 +1170,7 @@ void netvsc_channel_cb(void *context)
unsigned char *buffer;
int bufferlen = NETVSC_PACKET_SIZE;
struct net_device *ndev;
- struct nvsp_message *nvmsg;
+ bool need_to_commit = false;
if (channel->primary_channel != NULL)
device = channel->primary_channel->device_obj;
@@ -1151,43 +1180,40 @@ void netvsc_channel_cb(void *context)
net_device = get_inbound_net_device(device);
if (!net_device)
return;
- ndev = net_device->ndev;
+ ndev = hv_get_drvdata(device);
buffer = get_per_channel_state(channel);
do {
+ desc = get_next_pkt_raw(channel);
+ if (desc != NULL) {
+ netvsc_process_raw_pkt(device,
+ channel,
+ net_device,
+ ndev,
+ desc->trans_id,
+ desc);
+
+ put_pkt_raw(channel, desc);
+ need_to_commit = true;
+ continue;
+ }
+ if (need_to_commit) {
+ need_to_commit = false;
+ commit_rd_index(channel);
+ }
+
ret = vmbus_recvpacket_raw(channel, buffer, bufferlen,
&bytes_recvd, &request_id);
if (ret == 0) {
if (bytes_recvd > 0) {
desc = (struct vmpacket_descriptor *)buffer;
- nvmsg = (struct nvsp_message *)((unsigned long)
- desc + (desc->offset8 << 3));
- switch (desc->type) {
- case VM_PKT_COMP:
- netvsc_send_completion(net_device,
- channel,
- device, desc);
- break;
-
- case VM_PKT_DATA_USING_XFER_PAGES:
- netvsc_receive(net_device, channel,
- device, desc);
- break;
-
- case VM_PKT_DATA_INBAND:
- netvsc_receive_inband(device,
- net_device,
- nvmsg);
- break;
-
- default:
- netdev_err(ndev,
- "unhandled packet type %d, "
- "tid %llx len %d\n",
- desc->type, request_id,
- bytes_recvd);
- break;
- }
+ netvsc_process_raw_pkt(device,
+ channel,
+ net_device,
+ ndev,
+ request_id,
+ desc);
+
} else {
/*
@@ -1224,30 +1250,19 @@ void netvsc_channel_cb(void *context)
*/
int netvsc_device_add(struct hv_device *device, void *additional_info)
{
- int ret = 0;
+ int i, ret = 0;
int ring_size =
((struct netvsc_device_info *)additional_info)->ring_size;
struct netvsc_device *net_device;
- struct net_device *ndev;
+ struct net_device *ndev = hv_get_drvdata(device);
+ struct net_device_context *net_device_ctx = netdev_priv(ndev);
- net_device = alloc_net_device(device);
+ net_device = alloc_net_device();
if (!net_device)
return -ENOMEM;
net_device->ring_size = ring_size;
- /*
- * Coming into this function, struct net_device * is
- * registered as the driver private data.
- * In alloc_net_device(), we register struct netvsc_device *
- * as the driver private data and stash away struct net_device *
- * in struct netvsc_device *.
- */
- ndev = net_device->ndev;
-
- /* Add netvsc_device context to netvsc_device */
- net_device->nd_ctx = netdev_priv(ndev);
-
/* Initialize the NetVSC channel extension */
init_completion(&net_device->channel_init_wait);
@@ -1266,7 +1281,19 @@ int netvsc_device_add(struct hv_device *device, void *additional_info)
/* Channel is opened */
pr_info("hv_netvsc channel opened successfully\n");
- net_device->chn_table[0] = device->channel;
+ /* If we're reopening the device we may have multiple queues, fill the
+ * chn_table with the default channel to use it before subchannels are
+ * opened.
+ */
+ for (i = 0; i < VRSS_CHANNEL_MAX; i++)
+ net_device->chn_table[i] = device->channel;
+
+ /* Writing nvdev pointer unlocks netvsc_send(), make sure chn_table is
+ * populated.
+ */
+ wmb();
+
+ net_device_ctx->nvdev = net_device;
/* Connect with the NetVsp */
ret = netvsc_connect_vsp(device);
diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index b8121eba3..c4d9653ca 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -67,18 +67,19 @@ static void do_set_multicast(struct work_struct *w)
{
struct net_device_context *ndevctx =
container_of(w, struct net_device_context, work);
- struct netvsc_device *nvdev;
+ struct hv_device *device_obj = ndevctx->device_ctx;
+ struct net_device *ndev = hv_get_drvdata(device_obj);
+ struct netvsc_device *nvdev = ndevctx->nvdev;
struct rndis_device *rdev;
- nvdev = hv_get_drvdata(ndevctx->device_ctx);
- if (nvdev == NULL || nvdev->ndev == NULL)
+ if (!nvdev)
return;
rdev = nvdev->extension;
if (rdev == NULL)
return;
- if (nvdev->ndev->flags & IFF_PROMISC)
+ if (ndev->flags & IFF_PROMISC)
rndis_filter_set_packet_filter(rdev,
NDIS_PACKET_TYPE_PROMISCUOUS);
else
@@ -97,16 +98,14 @@ static void netvsc_set_multicast_list(struct net_device *net)
static int netvsc_open(struct net_device *net)
{
- struct net_device_context *net_device_ctx = netdev_priv(net);
- struct hv_device *device_obj = net_device_ctx->device_ctx;
- struct netvsc_device *nvdev;
+ struct netvsc_device *nvdev = net_device_to_netvsc_device(net);
struct rndis_device *rdev;
int ret = 0;
netif_carrier_off(net);
/* Open up the device */
- ret = rndis_filter_open(device_obj);
+ ret = rndis_filter_open(nvdev);
if (ret != 0) {
netdev_err(net, "unable to open device (ret %d).\n", ret);
return ret;
@@ -114,7 +113,6 @@ static int netvsc_open(struct net_device *net)
netif_tx_wake_all_queues(net);
- nvdev = hv_get_drvdata(device_obj);
rdev = nvdev->extension;
if (!rdev->link_state)
netif_carrier_on(net);
@@ -125,8 +123,7 @@ static int netvsc_open(struct net_device *net)
static int netvsc_close(struct net_device *net)
{
struct net_device_context *net_device_ctx = netdev_priv(net);
- struct hv_device *device_obj = net_device_ctx->device_ctx;
- struct netvsc_device *nvdev = hv_get_drvdata(device_obj);
+ struct netvsc_device *nvdev = net_device_ctx->nvdev;
int ret;
u32 aread, awrite, i, msec = 10, retry = 0, retry_max = 20;
struct vmbus_channel *chn;
@@ -135,7 +132,7 @@ static int netvsc_close(struct net_device *net)
/* Make sure netvsc_set_multicast_list doesn't re-enable filter! */
cancel_work_sync(&net_device_ctx->work);
- ret = rndis_filter_close(device_obj);
+ ret = rndis_filter_close(nvdev);
if (ret != 0) {
netdev_err(net, "unable to close device (ret %d).\n", ret);
return ret;
@@ -205,8 +202,7 @@ static u16 netvsc_select_queue(struct net_device *ndev, struct sk_buff *skb,
void *accel_priv, select_queue_fallback_t fallback)
{
struct net_device_context *net_device_ctx = netdev_priv(ndev);
- struct hv_device *hdev = net_device_ctx->device_ctx;
- struct netvsc_device *nvsc_dev = hv_get_drvdata(hdev);
+ struct netvsc_device *nvsc_dev = net_device_ctx->nvdev;
u32 hash;
u16 q_idx = 0;
@@ -580,7 +576,6 @@ void netvsc_linkstatus_callback(struct hv_device *device_obj,
struct rndis_indicate_status *indicate = &resp->msg.indicate_status;
struct net_device *net;
struct net_device_context *ndev_ctx;
- struct netvsc_device *net_device;
struct netvsc_reconfig *event;
unsigned long flags;
@@ -590,8 +585,7 @@ void netvsc_linkstatus_callback(struct hv_device *device_obj,
indicate->status != RNDIS_STATUS_MEDIA_DISCONNECT)
return;
- net_device = hv_get_drvdata(device_obj);
- net = net_device->ndev;
+ net = hv_get_drvdata(device_obj);
if (!net || net->reg_state != NETREG_REGISTERED)
return;
@@ -610,6 +604,47 @@ void netvsc_linkstatus_callback(struct hv_device *device_obj,
schedule_delayed_work(&ndev_ctx->dwork, 0);
}
+
+static struct sk_buff *netvsc_alloc_recv_skb(struct net_device *net,
+ struct hv_netvsc_packet *packet,
+ struct ndis_tcp_ip_checksum_info *csum_info,
+ void *data, u16 vlan_tci)
+{
+ struct sk_buff *skb;
+
+ skb = netdev_alloc_skb_ip_align(net, packet->total_data_buflen);
+ if (!skb)
+ return skb;
+
+ /*
+ * Copy to skb. This copy is needed here since the memory pointed by
+ * hv_netvsc_packet cannot be deallocated
+ */
+ memcpy(skb_put(skb, packet->total_data_buflen), data,
+ packet->total_data_buflen);
+
+ skb->protocol = eth_type_trans(skb, net);
+
+ /* skb is already created with CHECKSUM_NONE */
+ skb_checksum_none_assert(skb);
+
+ /*
+ * In Linux, the IP checksum is always checked.
+ * Do L4 checksum offload if enabled and present.
+ */
+ if (csum_info && (net->features & NETIF_F_RXCSUM)) {
+ if (csum_info->receive.tcp_checksum_succeeded ||
+ csum_info->receive.udp_checksum_succeeded)
+ skb->ip_summed = CHECKSUM_UNNECESSARY;
+ }
+
+ if (vlan_tci & VLAN_TAG_PRESENT)
+ __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q),
+ vlan_tci);
+
+ return skb;
+}
+
/*
* netvsc_recv_callback - Callback when we receive a packet from the
* "wire" on the specified device.
@@ -621,48 +656,60 @@ int netvsc_recv_callback(struct hv_device *device_obj,
struct vmbus_channel *channel,
u16 vlan_tci)
{
- struct net_device *net;
- struct net_device_context *net_device_ctx;
+ struct net_device *net = hv_get_drvdata(device_obj);
+ struct net_device_context *net_device_ctx = netdev_priv(net);
struct sk_buff *skb;
+ struct sk_buff *vf_skb;
struct netvsc_stats *rx_stats;
+ u32 bytes_recvd = packet->total_data_buflen;
+ int ret = 0;
- net = ((struct netvsc_device *)hv_get_drvdata(device_obj))->ndev;
- if (!net || net->reg_state != NETREG_REGISTERED) {
+ if (!net || net->reg_state != NETREG_REGISTERED)
return NVSP_STAT_FAIL;
+
+ if (READ_ONCE(net_device_ctx->vf_inject)) {
+ atomic_inc(&net_device_ctx->vf_use_cnt);
+ if (!READ_ONCE(net_device_ctx->vf_inject)) {
+ /*
+ * We raced; just move on.
+ */
+ atomic_dec(&net_device_ctx->vf_use_cnt);
+ goto vf_injection_done;
+ }
+
+ /*
+ * Inject this packet into the VF inerface.
+ * On Hyper-V, multicast and brodcast packets
+ * are only delivered on the synthetic interface
+ * (after subjecting these to policy filters on
+ * the host). Deliver these via the VF interface
+ * in the guest.
+ */
+ vf_skb = netvsc_alloc_recv_skb(net_device_ctx->vf_netdev,
+ packet, csum_info, *data,
+ vlan_tci);
+ if (vf_skb != NULL) {
+ ++net_device_ctx->vf_netdev->stats.rx_packets;
+ net_device_ctx->vf_netdev->stats.rx_bytes +=
+ bytes_recvd;
+ netif_receive_skb(vf_skb);
+ } else {
+ ++net->stats.rx_dropped;
+ ret = NVSP_STAT_FAIL;
+ }
+ atomic_dec(&net_device_ctx->vf_use_cnt);
+ return ret;
}
- net_device_ctx = netdev_priv(net);
+
+vf_injection_done:
rx_stats = this_cpu_ptr(net_device_ctx->rx_stats);
/* Allocate a skb - TODO direct I/O to pages? */
- skb = netdev_alloc_skb_ip_align(net, packet->total_data_buflen);
+ skb = netvsc_alloc_recv_skb(net, packet, csum_info, *data, vlan_tci);
if (unlikely(!skb)) {
++net->stats.rx_dropped;
return NVSP_STAT_FAIL;
}
-
- /*
- * Copy to skb. This copy is needed here since the memory pointed by
- * hv_netvsc_packet cannot be deallocated
- */
- memcpy(skb_put(skb, packet->total_data_buflen), *data,
- packet->total_data_buflen);
-
- skb->protocol = eth_type_trans(skb, net);
- if (csum_info) {
- /* We only look at the IP checksum here.
- * Should we be dropping the packet if checksum
- * failed? How do we deal with other checksums - TCP/UDP?
- */
- if (csum_info->receive.ip_checksum_succeeded)
- skb->ip_summed = CHECKSUM_UNNECESSARY;
- else
- skb->ip_summed = CHECKSUM_NONE;
- }
-
- if (vlan_tci & VLAN_TAG_PRESENT)
- __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q),
- vlan_tci);
-
skb_record_rx_queue(skb, channel->
offermsg.offer.sub_channel_index);
@@ -692,8 +739,7 @@ static void netvsc_get_channels(struct net_device *net,
struct ethtool_channels *channel)
{
struct net_device_context *net_device_ctx = netdev_priv(net);
- struct hv_device *dev = net_device_ctx->device_ctx;
- struct netvsc_device *nvdev = hv_get_drvdata(dev);
+ struct netvsc_device *nvdev = net_device_ctx->nvdev;
if (nvdev) {
channel->max_combined = nvdev->max_chn;
@@ -706,14 +752,14 @@ static int netvsc_set_channels(struct net_device *net,
{
struct net_device_context *net_device_ctx = netdev_priv(net);
struct hv_device *dev = net_device_ctx->device_ctx;
- struct netvsc_device *nvdev = hv_get_drvdata(dev);
+ struct netvsc_device *nvdev = net_device_ctx->nvdev;
struct netvsc_device_info device_info;
u32 num_chn;
u32 max_chn;
int ret = 0;
bool recovering = false;
- if (!nvdev || nvdev->destroy)
+ if (net_device_ctx->start_remove || !nvdev || nvdev->destroy)
return -ENODEV;
num_chn = nvdev->num_chn;
@@ -742,14 +788,11 @@ static int netvsc_set_channels(struct net_device *net,
goto out;
do_set:
- nvdev->start_remove = true;
+ net_device_ctx->start_remove = true;
rndis_filter_device_remove(dev);
nvdev->num_chn = channels->combined_count;
- net_device_ctx->device_ctx = dev;
- hv_set_drvdata(dev, net);
-
memset(&device_info, 0, sizeof(device_info));
device_info.num_chn = nvdev->num_chn; /* passed to RNDIS */
device_info.ring_size = ring_size;
@@ -764,7 +807,7 @@ static int netvsc_set_channels(struct net_device *net,
goto recover;
}
- nvdev = hv_get_drvdata(dev);
+ nvdev = net_device_ctx->nvdev;
ret = netif_set_real_num_tx_queues(net, nvdev->num_chn);
if (ret) {
@@ -786,6 +829,9 @@ static int netvsc_set_channels(struct net_device *net,
out:
netvsc_open(net);
+ net_device_ctx->start_remove = false;
+ /* We may have missed link change notifications */
+ schedule_delayed_work(&net_device_ctx->dwork, 0);
return ret;
@@ -854,14 +900,14 @@ static int netvsc_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
static int netvsc_change_mtu(struct net_device *ndev, int mtu)
{
struct net_device_context *ndevctx = netdev_priv(ndev);
- struct hv_device *hdev = ndevctx->device_ctx;
- struct netvsc_device *nvdev = hv_get_drvdata(hdev);
+ struct netvsc_device *nvdev = ndevctx->nvdev;
+ struct hv_device *hdev = ndevctx->device_ctx;
struct netvsc_device_info device_info;
int limit = ETH_DATA_LEN;
u32 num_chn;
int ret = 0;
- if (nvdev == NULL || nvdev->destroy)
+ if (ndevctx->start_remove || !nvdev || nvdev->destroy)
return -ENODEV;
if (nvdev->nvsp_version >= NVSP_PROTOCOL_VERSION_2)
@@ -876,14 +922,11 @@ static int netvsc_change_mtu(struct net_device *ndev, int mtu)
num_chn = nvdev->num_chn;
- nvdev->start_remove = true;
+ ndevctx->start_remove = true;
rndis_filter_device_remove(hdev);
ndev->mtu = mtu;
- ndevctx->device_ctx = hdev;
- hv_set_drvdata(hdev, ndev);
-
memset(&device_info, 0, sizeof(device_info));
device_info.ring_size = ring_size;
device_info.num_chn = num_chn;
@@ -892,6 +935,10 @@ static int netvsc_change_mtu(struct net_device *ndev, int mtu)
out:
netvsc_open(ndev);
+ ndevctx->start_remove = false;
+
+ /* We may have missed link change notifications */
+ schedule_delayed_work(&ndevctx->dwork, 0);
return ret;
}
@@ -939,8 +986,6 @@ static struct rtnl_link_stats64 *netvsc_get_stats64(struct net_device *net,
static int netvsc_set_mac_addr(struct net_device *ndev, void *p)
{
- struct net_device_context *ndevctx = netdev_priv(ndev);
- struct hv_device *hdev = ndevctx->device_ctx;
struct sockaddr *addr = p;
char save_adr[ETH_ALEN];
unsigned char save_aatype;
@@ -953,7 +998,7 @@ static int netvsc_set_mac_addr(struct net_device *ndev, void *p)
if (err != 0)
return err;
- err = rndis_filter_set_device_mac(hdev, addr->sa_data);
+ err = rndis_filter_set_device_mac(ndev, addr->sa_data);
if (err != 0) {
/* roll back to saved MAC */
memcpy(ndev->dev_addr, save_adr, ETH_ALEN);
@@ -1004,18 +1049,22 @@ static const struct net_device_ops device_ops = {
*/
static void netvsc_link_change(struct work_struct *w)
{
- struct net_device_context *ndev_ctx;
- struct net_device *net;
+ struct net_device_context *ndev_ctx =
+ container_of(w, struct net_device_context, dwork.work);
+ struct hv_device *device_obj = ndev_ctx->device_ctx;
+ struct net_device *net = hv_get_drvdata(device_obj);
struct netvsc_device *net_device;
struct rndis_device *rdev;
struct netvsc_reconfig *event = NULL;
bool notify = false, reschedule = false;
unsigned long flags, next_reconfig, delay;
- ndev_ctx = container_of(w, struct net_device_context, dwork.work);
- net_device = hv_get_drvdata(ndev_ctx->device_ctx);
+ rtnl_lock();
+ if (ndev_ctx->start_remove)
+ goto out_unlock;
+
+ net_device = ndev_ctx->nvdev;
rdev = net_device->extension;
- net = net_device->ndev;
next_reconfig = ndev_ctx->last_reconfig + LINKCHANGE_INT;
if (time_is_after_jiffies(next_reconfig)) {
@@ -1026,7 +1075,7 @@ static void netvsc_link_change(struct work_struct *w)
delay = next_reconfig - jiffies;
delay = delay < LINKCHANGE_INT ? delay : LINKCHANGE_INT;
schedule_delayed_work(&ndev_ctx->dwork, delay);
- return;
+ goto out_unlock;
}
ndev_ctx->last_reconfig = jiffies;
@@ -1040,9 +1089,7 @@ static void netvsc_link_change(struct work_struct *w)
spin_unlock_irqrestore(&ndev_ctx->lock, flags);
if (!event)
- return;
-
- rtnl_lock();
+ goto out_unlock;
switch (event->event) {
/* Only the following events are possible due to the check in
@@ -1074,7 +1121,7 @@ static void netvsc_link_change(struct work_struct *w)
netif_tx_stop_all_queues(net);
event->event = RNDIS_STATUS_MEDIA_CONNECT;
spin_lock_irqsave(&ndev_ctx->lock, flags);
- list_add_tail(&event->list, &ndev_ctx->reconfig_events);
+ list_add(&event->list, &ndev_ctx->reconfig_events);
spin_unlock_irqrestore(&ndev_ctx->lock, flags);
reschedule = true;
}
@@ -1091,6 +1138,11 @@ static void netvsc_link_change(struct work_struct *w)
*/
if (reschedule)
schedule_delayed_work(&ndev_ctx->dwork, LINKCHANGE_INT);
+
+ return;
+
+out_unlock:
+ rtnl_unlock();
}
static void netvsc_free_netdev(struct net_device *netdev)
@@ -1102,6 +1154,176 @@ static void netvsc_free_netdev(struct net_device *netdev)
free_netdev(netdev);
}
+static struct net_device *get_netvsc_net_device(char *mac)
+{
+ struct net_device *dev, *found = NULL;
+ int rtnl_locked;
+
+ rtnl_locked = rtnl_trylock();
+
+ for_each_netdev(&init_net, dev) {
+ if (memcmp(dev->dev_addr, mac, ETH_ALEN) == 0) {
+ if (dev->netdev_ops != &device_ops)
+ continue;
+ found = dev;
+ break;
+ }
+ }
+ if (rtnl_locked)
+ rtnl_unlock();
+
+ return found;
+}
+
+static int netvsc_register_vf(struct net_device *vf_netdev)
+{
+ struct net_device *ndev;
+ struct net_device_context *net_device_ctx;
+ struct netvsc_device *netvsc_dev;
+ const struct ethtool_ops *eth_ops = vf_netdev->ethtool_ops;
+
+ if (eth_ops == NULL || eth_ops == &ethtool_ops)
+ return NOTIFY_DONE;
+
+ /*
+ * We will use the MAC address to locate the synthetic interface to
+ * associate with the VF interface. If we don't find a matching
+ * synthetic interface, move on.
+ */
+ ndev = get_netvsc_net_device(vf_netdev->dev_addr);
+ if (!ndev)
+ return NOTIFY_DONE;
+
+ net_device_ctx = netdev_priv(ndev);
+ netvsc_dev = net_device_ctx->nvdev;
+ if (!netvsc_dev || net_device_ctx->vf_netdev)
+ return NOTIFY_DONE;
+
+ netdev_info(ndev, "VF registering: %s\n", vf_netdev->name);
+ /*
+ * Take a reference on the module.
+ */
+ try_module_get(THIS_MODULE);
+ net_device_ctx->vf_netdev = vf_netdev;
+ return NOTIFY_OK;
+}
+
+static void netvsc_inject_enable(struct net_device_context *net_device_ctx)
+{
+ net_device_ctx->vf_inject = true;
+}
+
+static void netvsc_inject_disable(struct net_device_context *net_device_ctx)
+{
+ net_device_ctx->vf_inject = false;
+
+ /* Wait for currently active users to drain out. */
+ while (atomic_read(&net_device_ctx->vf_use_cnt) != 0)
+ udelay(50);
+}
+
+static int netvsc_vf_up(struct net_device *vf_netdev)
+{
+ struct net_device *ndev;
+ struct netvsc_device *netvsc_dev;
+ const struct ethtool_ops *eth_ops = vf_netdev->ethtool_ops;
+ struct net_device_context *net_device_ctx;
+
+ if (eth_ops == &ethtool_ops)
+ return NOTIFY_DONE;
+
+ ndev = get_netvsc_net_device(vf_netdev->dev_addr);
+ if (!ndev)
+ return NOTIFY_DONE;
+
+ net_device_ctx = netdev_priv(ndev);
+ netvsc_dev = net_device_ctx->nvdev;
+
+ if (!netvsc_dev || !net_device_ctx->vf_netdev)
+ return NOTIFY_DONE;
+
+ netdev_info(ndev, "VF up: %s\n", vf_netdev->name);
+ netvsc_inject_enable(net_device_ctx);
+
+ /*
+ * Open the device before switching data path.
+ */
+ rndis_filter_open(netvsc_dev);
+
+ /*
+ * notify the host to switch the data path.
+ */
+ netvsc_switch_datapath(ndev, true);
+ netdev_info(ndev, "Data path switched to VF: %s\n", vf_netdev->name);
+
+ netif_carrier_off(ndev);
+
+ /* Now notify peers through VF device. */
+ call_netdevice_notifiers(NETDEV_NOTIFY_PEERS, vf_netdev);
+
+ return NOTIFY_OK;
+}
+
+
+static int netvsc_vf_down(struct net_device *vf_netdev)
+{
+ struct net_device *ndev;
+ struct netvsc_device *netvsc_dev;
+ struct net_device_context *net_device_ctx;
+ const struct ethtool_ops *eth_ops = vf_netdev->ethtool_ops;
+
+ if (eth_ops == &ethtool_ops)
+ return NOTIFY_DONE;
+
+ ndev = get_netvsc_net_device(vf_netdev->dev_addr);
+ if (!ndev)
+ return NOTIFY_DONE;
+
+ net_device_ctx = netdev_priv(ndev);
+ netvsc_dev = net_device_ctx->nvdev;
+
+ if (!netvsc_dev || !net_device_ctx->vf_netdev)
+ return NOTIFY_DONE;
+
+ netdev_info(ndev, "VF down: %s\n", vf_netdev->name);
+ netvsc_inject_disable(net_device_ctx);
+ netvsc_switch_datapath(ndev, false);
+ netdev_info(ndev, "Data path switched from VF: %s\n", vf_netdev->name);
+ rndis_filter_close(netvsc_dev);
+ netif_carrier_on(ndev);
+
+ /* Now notify peers through netvsc device. */
+ call_netdevice_notifiers(NETDEV_NOTIFY_PEERS, ndev);
+
+ return NOTIFY_OK;
+}
+
+
+static int netvsc_unregister_vf(struct net_device *vf_netdev)
+{
+ struct net_device *ndev;
+ struct netvsc_device *netvsc_dev;
+ const struct ethtool_ops *eth_ops = vf_netdev->ethtool_ops;
+ struct net_device_context *net_device_ctx;
+
+ if (eth_ops == &ethtool_ops)
+ return NOTIFY_DONE;
+
+ ndev = get_netvsc_net_device(vf_netdev->dev_addr);
+ if (!ndev)
+ return NOTIFY_DONE;
+
+ net_device_ctx = netdev_priv(ndev);
+ netvsc_dev = net_device_ctx->nvdev;
+ if (!netvsc_dev || !net_device_ctx->vf_netdev)
+ return NOTIFY_DONE;
+ netdev_info(ndev, "VF unregistering: %s\n", vf_netdev->name);
+ netvsc_inject_disable(net_device_ctx);
+ net_device_ctx->vf_netdev = NULL;
+ module_put(THIS_MODULE);
+ return NOTIFY_OK;
+}
+
static int netvsc_probe(struct hv_device *dev,
const struct hv_vmbus_device_id *dev_id)
{
@@ -1138,12 +1360,19 @@ static int netvsc_probe(struct hv_device *dev,
}
hv_set_drvdata(dev, net);
+
+ net_device_ctx->start_remove = false;
+
INIT_DELAYED_WORK(&net_device_ctx->dwork, netvsc_link_change);
INIT_WORK(&net_device_ctx->work, do_set_multicast);
spin_lock_init(&net_device_ctx->lock);
INIT_LIST_HEAD(&net_device_ctx->reconfig_events);
+ atomic_set(&net_device_ctx->vf_use_cnt, 0);
+ net_device_ctx->vf_netdev = NULL;
+ net_device_ctx->vf_inject = false;
+
net->netdev_ops = &device_ops;
net->hw_features = NETVSC_HW_FEATURES;
@@ -1168,7 +1397,7 @@ static int netvsc_probe(struct hv_device *dev,
}
memcpy(net->dev_addr, device_info.mac_adr, ETH_ALEN);
- nvdev = hv_get_drvdata(dev);
+ nvdev = net_device_ctx->nvdev;
netif_set_real_num_tx_queues(net, nvdev->num_chn);
netif_set_real_num_rx_queues(net, nvdev->num_chn);
@@ -1190,17 +1419,24 @@ static int netvsc_remove(struct hv_device *dev)
struct net_device_context *ndev_ctx;
struct netvsc_device *net_device;
- net_device = hv_get_drvdata(dev);
- net = net_device->ndev;
+ net = hv_get_drvdata(dev);
if (net == NULL) {
dev_err(&dev->device, "No net device to remove\n");
return 0;
}
- net_device->start_remove = true;
ndev_ctx = netdev_priv(net);
+ net_device = ndev_ctx->nvdev;
+
+ /* Avoid racing with netvsc_change_mtu()/netvsc_set_channels()
+ * removing the device.
+ */
+ rtnl_lock();
+ ndev_ctx->start_remove = true;
+ rtnl_unlock();
+
cancel_delayed_work_sync(&ndev_ctx->dwork);
cancel_work_sync(&ndev_ctx->work);
@@ -1215,6 +1451,8 @@ static int netvsc_remove(struct hv_device *dev)
*/
rndis_filter_device_remove(dev);
+ hv_set_drvdata(dev, NULL);
+
netvsc_free_netdev(net);
return 0;
}
@@ -1235,19 +1473,67 @@ static struct hv_driver netvsc_drv = {
.remove = netvsc_remove,
};
+
+/*
+ * On Hyper-V, every VF interface is matched with a corresponding
+ * synthetic interface. The synthetic interface is presented first
+ * to the guest. When the corresponding VF instance is registered,
+ * we will take care of switching the data path.
+ */
+static int netvsc_netdev_event(struct notifier_block *this,
+ unsigned long event, void *ptr)
+{
+ struct net_device *event_dev = netdev_notifier_info_to_dev(ptr);
+
+ /* Avoid Vlan dev with same MAC registering as VF */
+ if (event_dev->priv_flags & IFF_802_1Q_VLAN)
+ return NOTIFY_DONE;
+
+ /* Avoid Bonding master dev with same MAC registering as VF */
+ if (event_dev->priv_flags & IFF_BONDING &&
+ event_dev->flags & IFF_MASTER)
+ return NOTIFY_DONE;
+
+ switch (event) {
+ case NETDEV_REGISTER:
+ return netvsc_register_vf(event_dev);
+ case NETDEV_UNREGISTER:
+ return netvsc_unregister_vf(event_dev);
+ case NETDEV_UP:
+ return netvsc_vf_up(event_dev);
+ case NETDEV_DOWN:
+ return netvsc_vf_down(event_dev);
+ default:
+ return NOTIFY_DONE;
+ }
+}
+
+static struct notifier_block netvsc_netdev_notifier = {
+ .notifier_call = netvsc_netdev_event,
+};
+
static void __exit netvsc_drv_exit(void)
{
+ unregister_netdevice_notifier(&netvsc_netdev_notifier);
vmbus_driver_unregister(&netvsc_drv);
}
static int __init netvsc_drv_init(void)
{
+ int ret;
+
if (ring_size < RING_SIZE_MIN) {
ring_size = RING_SIZE_MIN;
pr_info("Increased ring_size to %d (min allowed)\n",
ring_size);
}
- return vmbus_driver_register(&netvsc_drv);
+ ret = vmbus_driver_register(&netvsc_drv);
+
+ if (ret)
+ return ret;
+
+ register_netdevice_notifier(&netvsc_netdev_notifier);
+ return 0;
}
MODULE_LICENSE("GPL");
diff --git a/drivers/net/hyperv/rndis_filter.c b/drivers/net/hyperv/rndis_filter.c
index c4e1e0408..8e830f741 100644
--- a/drivers/net/hyperv/rndis_filter.c
+++ b/drivers/net/hyperv/rndis_filter.c
@@ -126,11 +126,7 @@ static void put_rndis_request(struct rndis_device *dev,
static void dump_rndis_message(struct hv_device *hv_dev,
struct rndis_message *rndis_msg)
{
- struct net_device *netdev;
- struct netvsc_device *net_device;
-
- net_device = hv_get_drvdata(hv_dev);
- netdev = net_device->ndev;
+ struct net_device *netdev = hv_get_drvdata(hv_dev);
switch (rndis_msg->ndis_msg_type) {
case RNDIS_MSG_PACKET:
@@ -211,6 +207,7 @@ static int rndis_filter_send_request(struct rndis_device *dev,
struct hv_netvsc_packet *packet;
struct hv_page_buffer page_buf[2];
struct hv_page_buffer *pb = page_buf;
+ struct net_device_context *net_device_ctx = netdev_priv(dev->ndev);
/* Setup the packet to send it */
packet = &req->pkt;
@@ -236,7 +233,7 @@ static int rndis_filter_send_request(struct rndis_device *dev,
pb[0].len;
}
- ret = netvsc_send(dev->net_dev->dev, packet, NULL, &pb, NULL);
+ ret = netvsc_send(net_device_ctx->device_ctx, packet, NULL, &pb, NULL);
return ret;
}
@@ -262,9 +259,7 @@ static void rndis_filter_receive_response(struct rndis_device *dev,
struct rndis_request *request = NULL;
bool found = false;
unsigned long flags;
- struct net_device *ndev;
-
- ndev = dev->net_dev->ndev;
+ struct net_device *ndev = dev->ndev;
spin_lock_irqsave(&dev->request_lock, flags);
list_for_each_entry(request, &dev->req_list, list_ent) {
@@ -355,6 +350,7 @@ static int rndis_filter_receive_data(struct rndis_device *dev,
struct ndis_pkt_8021q_info *vlan;
struct ndis_tcp_ip_checksum_info *csum_info;
u16 vlan_tci = 0;
+ struct net_device_context *net_device_ctx = netdev_priv(dev->ndev);
rndis_pkt = &msg->msg.pkt;
@@ -368,7 +364,7 @@ static int rndis_filter_receive_data(struct rndis_device *dev,
* should be the data packet size plus the trailer padding size
*/
if (pkt->total_data_buflen < rndis_pkt->data_len) {
- netdev_err(dev->net_dev->ndev, "rndis message buffer "
+ netdev_err(dev->ndev, "rndis message buffer "
"overflow detected (got %u, min %u)"
"...dropping this message!\n",
pkt->total_data_buflen, rndis_pkt->data_len);
@@ -390,7 +386,7 @@ static int rndis_filter_receive_data(struct rndis_device *dev,
}
csum_info = rndis_get_ppi(rndis_pkt, TCPIP_CHKSUM_PKTINFO);
- return netvsc_recv_callback(dev->net_dev->dev, pkt, data,
+ return netvsc_recv_callback(net_device_ctx->device_ctx, pkt, data,
csum_info, channel, vlan_tci);
}
@@ -399,10 +395,11 @@ int rndis_filter_receive(struct hv_device *dev,
void **data,
struct vmbus_channel *channel)
{
- struct netvsc_device *net_dev = hv_get_drvdata(dev);
+ struct net_device *ndev = hv_get_drvdata(dev);
+ struct net_device_context *net_device_ctx = netdev_priv(ndev);
+ struct netvsc_device *net_dev = net_device_ctx->nvdev;
struct rndis_device *rndis_dev;
struct rndis_message *rndis_msg;
- struct net_device *ndev;
int ret = 0;
if (!net_dev) {
@@ -410,8 +407,6 @@ int rndis_filter_receive(struct hv_device *dev,
goto exit;
}
- ndev = net_dev->ndev;
-
/* Make sure the rndis device state is initialized */
if (!net_dev->extension) {
netdev_err(ndev, "got rndis message but no rndis device - "
@@ -430,7 +425,7 @@ int rndis_filter_receive(struct hv_device *dev,
rndis_msg = *data;
- if (netif_msg_rx_err(net_dev->nd_ctx))
+ if (netif_msg_rx_err(net_device_ctx))
dump_rndis_message(dev, rndis_msg);
switch (rndis_msg->ndis_msg_type) {
@@ -471,7 +466,6 @@ static int rndis_filter_query_device(struct rndis_device *dev, u32 oid,
struct rndis_query_request *query;
struct rndis_query_complete *query_complete;
int ret = 0;
- unsigned long t;
if (!result)
return -EINVAL;
@@ -508,11 +502,7 @@ static int rndis_filter_query_device(struct rndis_device *dev, u32 oid,
if (ret != 0)
goto cleanup;
- t = wait_for_completion_timeout(&request->wait_event, 5*HZ);
- if (t == 0) {
- ret = -ETIMEDOUT;
- goto cleanup;
- }
+ wait_for_completion(&request->wait_event);
/* Copy the response back */
query_complete = &request->response_msg.msg.query_complete;
@@ -548,11 +538,10 @@ static int rndis_filter_query_device_mac(struct rndis_device *dev)
#define NWADR_STR "NetworkAddress"
#define NWADR_STRLEN 14
-int rndis_filter_set_device_mac(struct hv_device *hdev, char *mac)
+int rndis_filter_set_device_mac(struct net_device *ndev, char *mac)
{
- struct netvsc_device *nvdev = hv_get_drvdata(hdev);
+ struct netvsc_device *nvdev = net_device_to_netvsc_device(ndev);
struct rndis_device *rdev = nvdev->extension;
- struct net_device *ndev = nvdev->ndev;
struct rndis_request *request;
struct rndis_set_request *set;
struct rndis_config_parameter_info *cpi;
@@ -562,7 +551,6 @@ int rndis_filter_set_device_mac(struct hv_device *hdev, char *mac)
u32 extlen = sizeof(struct rndis_config_parameter_info) +
2*NWADR_STRLEN + 4*ETH_ALEN;
int ret;
- unsigned long t;
request = get_rndis_request(rdev, RNDIS_MSG_SET,
RNDIS_MESSAGE_SIZE(struct rndis_set_request) + extlen);
@@ -603,21 +591,13 @@ int rndis_filter_set_device_mac(struct hv_device *hdev, char *mac)
if (ret != 0)
goto cleanup;
- t = wait_for_completion_timeout(&request->wait_event, 5*HZ);
- if (t == 0) {
- netdev_err(ndev, "timeout before we got a set response...\n");
- /*
- * can't put_rndis_request, since we may still receive a
- * send-completion.
- */
- return -EBUSY;
- } else {
- set_complete = &request->response_msg.msg.set_complete;
- if (set_complete->status != RNDIS_STATUS_SUCCESS) {
- netdev_err(ndev, "Fail to set MAC on host side:0x%x\n",
- set_complete->status);
- ret = -EINVAL;
- }
+ wait_for_completion(&request->wait_event);
+
+ set_complete = &request->response_msg.msg.set_complete;
+ if (set_complete->status != RNDIS_STATUS_SUCCESS) {
+ netdev_err(ndev, "Fail to set MAC on host side:0x%x\n",
+ set_complete->status);
+ ret = -EINVAL;
}
cleanup:
@@ -626,19 +606,17 @@ cleanup:
}
static int
-rndis_filter_set_offload_params(struct hv_device *hdev,
+rndis_filter_set_offload_params(struct net_device *ndev,
struct ndis_offload_params *req_offloads)
{
- struct netvsc_device *nvdev = hv_get_drvdata(hdev);
+ struct netvsc_device *nvdev = net_device_to_netvsc_device(ndev);
struct rndis_device *rdev = nvdev->extension;
- struct net_device *ndev = nvdev->ndev;
struct rndis_request *request;
struct rndis_set_request *set;
struct ndis_offload_params *offload_params;
struct rndis_set_complete *set_complete;
u32 extlen = sizeof(struct ndis_offload_params);
int ret;
- unsigned long t;
u32 vsp_version = nvdev->nvsp_version;
if (vsp_version <= NVSP_PROTOCOL_VERSION_4) {
@@ -672,20 +650,12 @@ rndis_filter_set_offload_params(struct hv_device *hdev,
if (ret != 0)
goto cleanup;
- t = wait_for_completion_timeout(&request->wait_event, 5*HZ);
- if (t == 0) {
- netdev_err(ndev, "timeout before we got aOFFLOAD set response...\n");
- /* can't put_rndis_request, since we may still receive a
- * send-completion.
- */
- return -EBUSY;
- } else {
- set_complete = &request->response_msg.msg.set_complete;
- if (set_complete->status != RNDIS_STATUS_SUCCESS) {
- netdev_err(ndev, "Fail to set offload on host side:0x%x\n",
- set_complete->status);
- ret = -EINVAL;
- }
+ wait_for_completion(&request->wait_event);
+ set_complete = &request->response_msg.msg.set_complete;
+ if (set_complete->status != RNDIS_STATUS_SUCCESS) {
+ netdev_err(ndev, "Fail to set offload on host side:0x%x\n",
+ set_complete->status);
+ ret = -EINVAL;
}
cleanup:
@@ -703,7 +673,7 @@ u8 netvsc_hash_key[HASH_KEYLEN] = {
static int rndis_filter_set_rss_param(struct rndis_device *rdev, int num_queue)
{
- struct net_device *ndev = rdev->net_dev->ndev;
+ struct net_device *ndev = rdev->ndev;
struct rndis_request *request;
struct rndis_set_request *set;
struct rndis_set_complete *set_complete;
@@ -713,7 +683,6 @@ static int rndis_filter_set_rss_param(struct rndis_device *rdev, int num_queue)
u32 *itab;
u8 *keyp;
int i, ret;
- unsigned long t;
request = get_rndis_request(
rdev, RNDIS_MSG_SET,
@@ -756,20 +725,12 @@ static int rndis_filter_set_rss_param(struct rndis_device *rdev, int num_queue)
if (ret != 0)
goto cleanup;
- t = wait_for_completion_timeout(&request->wait_event, 5*HZ);
- if (t == 0) {
- netdev_err(ndev, "timeout before we got a set response...\n");
- /* can't put_rndis_request, since we may still receive a
- * send-completion.
- */
- return -ETIMEDOUT;
- } else {
- set_complete = &request->response_msg.msg.set_complete;
- if (set_complete->status != RNDIS_STATUS_SUCCESS) {
- netdev_err(ndev, "Fail to set RSS parameters:0x%x\n",
- set_complete->status);
- ret = -EINVAL;
- }
+ wait_for_completion(&request->wait_event);
+ set_complete = &request->response_msg.msg.set_complete;
+ if (set_complete->status != RNDIS_STATUS_SUCCESS) {
+ netdev_err(ndev, "Fail to set RSS parameters:0x%x\n",
+ set_complete->status);
+ ret = -EINVAL;
}
cleanup:
@@ -798,10 +759,6 @@ int rndis_filter_set_packet_filter(struct rndis_device *dev, u32 new_filter)
struct rndis_set_complete *set_complete;
u32 status;
int ret;
- unsigned long t;
- struct net_device *ndev;
-
- ndev = dev->net_dev->ndev;
request = get_rndis_request(dev, RNDIS_MSG_SET,
RNDIS_MESSAGE_SIZE(struct rndis_set_request) +
@@ -824,26 +781,14 @@ int rndis_filter_set_packet_filter(struct rndis_device *dev, u32 new_filter)
if (ret != 0)
goto cleanup;
- t = wait_for_completion_timeout(&request->wait_event, 5*HZ);
+ wait_for_completion(&request->wait_event);
- if (t == 0) {
- netdev_err(ndev,
- "timeout before we got a set response...\n");
- ret = -ETIMEDOUT;
- /*
- * We can't deallocate the request since we may still receive a
- * send completion for it.
- */
- goto exit;
- } else {
- set_complete = &request->response_msg.msg.set_complete;
- status = set_complete->status;
- }
+ set_complete = &request->response_msg.msg.set_complete;
+ status = set_complete->status;
cleanup:
if (request)
put_rndis_request(dev, request);
-exit:
return ret;
}
@@ -855,8 +800,7 @@ static int rndis_filter_init_device(struct rndis_device *dev)
struct rndis_initialize_complete *init_complete;
u32 status;
int ret;
- unsigned long t;
- struct netvsc_device *nvdev = dev->net_dev;
+ struct netvsc_device *nvdev = net_device_to_netvsc_device(dev->ndev);
request = get_rndis_request(dev, RNDIS_MSG_INIT,
RNDIS_MESSAGE_SIZE(struct rndis_initialize_request));
@@ -879,13 +823,7 @@ static int rndis_filter_init_device(struct rndis_device *dev)
goto cleanup;
}
-
- t = wait_for_completion_timeout(&request->wait_event, 5*HZ);
-
- if (t == 0) {
- ret = -ETIMEDOUT;
- goto cleanup;
- }
+ wait_for_completion(&request->wait_event);
init_complete = &request->response_msg.msg.init_complete;
status = init_complete->status;
@@ -910,8 +848,9 @@ static void rndis_filter_halt_device(struct rndis_device *dev)
{
struct rndis_request *request;
struct rndis_halt_request *halt;
- struct netvsc_device *nvdev = dev->net_dev;
- struct hv_device *hdev = nvdev->dev;
+ struct net_device_context *net_device_ctx = netdev_priv(dev->ndev);
+ struct netvsc_device *nvdev = net_device_ctx->nvdev;
+ struct hv_device *hdev = net_device_ctx->device_ctx;
ulong flags;
/* Attempt to do a rndis device halt */
@@ -979,13 +918,13 @@ static int rndis_filter_close_device(struct rndis_device *dev)
static void netvsc_sc_open(struct vmbus_channel *new_sc)
{
- struct netvsc_device *nvscdev;
+ struct net_device *ndev =
+ hv_get_drvdata(new_sc->primary_channel->device_obj);
+ struct netvsc_device *nvscdev = net_device_to_netvsc_device(ndev);
u16 chn_index = new_sc->offermsg.offer.sub_channel_index;
int ret;
unsigned long flags;
- nvscdev = hv_get_drvdata(new_sc->primary_channel->device_obj);
-
if (chn_index >= nvscdev->num_chn)
return;
@@ -1010,12 +949,13 @@ int rndis_filter_device_add(struct hv_device *dev,
void *additional_info)
{
int ret;
+ struct net_device *net = hv_get_drvdata(dev);
+ struct net_device_context *net_device_ctx = netdev_priv(net);
struct netvsc_device *net_device;
struct rndis_device *rndis_device;
struct netvsc_device_info *device_info = additional_info;
struct ndis_offload_params offloads;
struct nvsp_message *init_packet;
- unsigned long t;
struct ndis_recv_scale_cap rsscap;
u32 rsscap_size = sizeof(struct ndis_recv_scale_cap);
u32 mtu, size;
@@ -1040,16 +980,15 @@ int rndis_filter_device_add(struct hv_device *dev,
return ret;
}
-
/* Initialize the rndis device */
- net_device = hv_get_drvdata(dev);
+ net_device = net_device_ctx->nvdev;
net_device->max_chn = 1;
net_device->num_chn = 1;
spin_lock_init(&net_device->sc_lock);
net_device->extension = rndis_device;
- rndis_device->net_dev = net_device;
+ rndis_device->ndev = net;
/* Send the rndis initialization message */
ret = rndis_filter_init_device(rndis_device);
@@ -1063,8 +1002,8 @@ int rndis_filter_device_add(struct hv_device *dev,
ret = rndis_filter_query_device(rndis_device,
RNDIS_OID_GEN_MAXIMUM_FRAME_SIZE,
&mtu, &size);
- if (ret == 0 && size == sizeof(u32) && mtu < net_device->ndev->mtu)
- net_device->ndev->mtu = mtu;
+ if (ret == 0 && size == sizeof(u32) && mtu < net->mtu)
+ net->mtu = mtu;
/* Get the mac address */
ret = rndis_filter_query_device_mac(rndis_device);
@@ -1090,7 +1029,7 @@ int rndis_filter_device_add(struct hv_device *dev,
offloads.lso_v2_ipv4 = NDIS_OFFLOAD_PARAMETERS_LSOV2_ENABLED;
- ret = rndis_filter_set_offload_params(dev, &offloads);
+ ret = rndis_filter_set_offload_params(net, &offloads);
if (ret)
goto err_dev_remv;
@@ -1159,11 +1098,8 @@ int rndis_filter_device_add(struct hv_device *dev,
VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED);
if (ret)
goto out;
- t = wait_for_completion_timeout(&net_device->channel_init_wait, 5*HZ);
- if (t == 0) {
- ret = -ETIMEDOUT;
- goto out;
- }
+ wait_for_completion(&net_device->channel_init_wait);
+
if (init_packet->msg.v5_msg.subchn_comp.status !=
NVSP_STAT_SUCCESS) {
ret = -ENODEV;
@@ -1198,19 +1134,14 @@ err_dev_remv:
void rndis_filter_device_remove(struct hv_device *dev)
{
- struct netvsc_device *net_dev = hv_get_drvdata(dev);
+ struct netvsc_device *net_dev = hv_device_to_netvsc_device(dev);
struct rndis_device *rndis_dev = net_dev->extension;
- unsigned long t;
/* If not all subchannel offers are complete, wait for them until
* completion to avoid race.
*/
- while (net_dev->num_sc_offered > 0) {
- t = wait_for_completion_timeout(&net_dev->channel_init_wait,
- 10 * HZ);
- if (t == 0)
- WARN(1, "Netvsc: Waiting for sub-channel processing");
- }
+ if (net_dev->num_sc_offered > 0)
+ wait_for_completion(&net_dev->channel_init_wait);
/* Halt and release the rndis device */
rndis_filter_halt_device(rndis_dev);
@@ -1222,22 +1153,24 @@ void rndis_filter_device_remove(struct hv_device *dev)
}
-int rndis_filter_open(struct hv_device *dev)
+int rndis_filter_open(struct netvsc_device *nvdev)
{
- struct netvsc_device *net_device = hv_get_drvdata(dev);
-
- if (!net_device)
+ if (!nvdev)
return -EINVAL;
- return rndis_filter_open_device(net_device->extension);
+ if (atomic_inc_return(&nvdev->open_cnt) != 1)
+ return 0;
+
+ return rndis_filter_open_device(nvdev->extension);
}
-int rndis_filter_close(struct hv_device *dev)
+int rndis_filter_close(struct netvsc_device *nvdev)
{
- struct netvsc_device *nvdev = hv_get_drvdata(dev);
-
if (!nvdev)
return -EINVAL;
+ if (atomic_dec_return(&nvdev->open_cnt) != 0)
+ return 0;
+
return rndis_filter_close_device(nvdev->extension);
}
generated by cgit v1.2.3-54-g00ecf (git 2.45.2) at 2025-10-20 03:08:07 +0000