summaryrefslogtreecommitdiff
path: root/net/openvswitch
diff options
context:
space:
mode:
Diffstat (limited to 'net/openvswitch')
-rw-r--r--net/openvswitch/actions.c29
-rw-r--r--net/openvswitch/datapath.c50
-rw-r--r--net/openvswitch/datapath.h6
-rw-r--r--net/openvswitch/vport.c1
4 files changed, 62 insertions, 24 deletions
diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c
index e7269a3eec79..2832e0794197 100644
--- a/net/openvswitch/actions.c
+++ b/net/openvswitch/actions.c
@@ -39,16 +39,14 @@
#include "flow_netlink.h"
#include "openvswitch_trace.h"
-DEFINE_PER_CPU(struct ovs_pcpu_storage, ovs_pcpu_storage) = {
- .bh_lock = INIT_LOCAL_LOCK(bh_lock),
-};
+struct ovs_pcpu_storage __percpu *ovs_pcpu_storage;
/* Make a clone of the 'key', using the pre-allocated percpu 'flow_keys'
* space. Return NULL if out of key spaces.
*/
static struct sw_flow_key *clone_key(const struct sw_flow_key *key_)
{
- struct ovs_pcpu_storage *ovs_pcpu = this_cpu_ptr(&ovs_pcpu_storage);
+ struct ovs_pcpu_storage *ovs_pcpu = this_cpu_ptr(ovs_pcpu_storage);
struct action_flow_keys *keys = &ovs_pcpu->flow_keys;
int level = ovs_pcpu->exec_level;
struct sw_flow_key *key = NULL;
@@ -94,7 +92,7 @@ static struct deferred_action *add_deferred_actions(struct sk_buff *skb,
const struct nlattr *actions,
const int actions_len)
{
- struct action_fifo *fifo = this_cpu_ptr(&ovs_pcpu_storage.action_fifos);
+ struct action_fifo *fifo = this_cpu_ptr(&ovs_pcpu_storage->action_fifos);
struct deferred_action *da;
da = action_fifo_put(fifo);
@@ -755,7 +753,7 @@ static int set_sctp(struct sk_buff *skb, struct sw_flow_key *flow_key,
static int ovs_vport_output(struct net *net, struct sock *sk,
struct sk_buff *skb)
{
- struct ovs_frag_data *data = this_cpu_ptr(&ovs_pcpu_storage.frag_data);
+ struct ovs_frag_data *data = this_cpu_ptr(&ovs_pcpu_storage->frag_data);
struct vport *vport = data->vport;
if (skb_cow_head(skb, data->l2_len) < 0) {
@@ -807,7 +805,7 @@ static void prepare_frag(struct vport *vport, struct sk_buff *skb,
unsigned int hlen = skb_network_offset(skb);
struct ovs_frag_data *data;
- data = this_cpu_ptr(&ovs_pcpu_storage.frag_data);
+ data = this_cpu_ptr(&ovs_pcpu_storage->frag_data);
data->dst = skb->_skb_refdst;
data->vport = vport;
data->cb = *OVS_CB(skb);
@@ -943,8 +941,10 @@ static int output_userspace(struct datapath *dp, struct sk_buff *skb,
break;
case OVS_USERSPACE_ATTR_PID:
- if (dp->user_features &
- OVS_DP_F_DISPATCH_UPCALL_PER_CPU)
+ if (OVS_CB(skb)->upcall_pid)
+ upcall.portid = OVS_CB(skb)->upcall_pid;
+ else if (dp->user_features &
+ OVS_DP_F_DISPATCH_UPCALL_PER_CPU)
upcall.portid =
ovs_dp_get_upcall_portid(dp,
smp_processor_id());
@@ -1566,16 +1566,15 @@ static int clone_execute(struct datapath *dp, struct sk_buff *skb,
clone = clone_flow_key ? clone_key(key) : key;
if (clone) {
int err = 0;
-
if (actions) { /* Sample action */
if (clone_flow_key)
- __this_cpu_inc(ovs_pcpu_storage.exec_level);
+ __this_cpu_inc(ovs_pcpu_storage->exec_level);
err = do_execute_actions(dp, skb, clone,
actions, len);
if (clone_flow_key)
- __this_cpu_dec(ovs_pcpu_storage.exec_level);
+ __this_cpu_dec(ovs_pcpu_storage->exec_level);
} else { /* Recirc action */
clone->recirc_id = recirc_id;
ovs_dp_process_packet(skb, clone);
@@ -1611,7 +1610,7 @@ static int clone_execute(struct datapath *dp, struct sk_buff *skb,
static void process_deferred_actions(struct datapath *dp)
{
- struct action_fifo *fifo = this_cpu_ptr(&ovs_pcpu_storage.action_fifos);
+ struct action_fifo *fifo = this_cpu_ptr(&ovs_pcpu_storage->action_fifos);
/* Do not touch the FIFO in case there is no deferred actions. */
if (action_fifo_is_empty(fifo))
@@ -1642,7 +1641,7 @@ int ovs_execute_actions(struct datapath *dp, struct sk_buff *skb,
{
int err, level;
- level = __this_cpu_inc_return(ovs_pcpu_storage.exec_level);
+ level = __this_cpu_inc_return(ovs_pcpu_storage->exec_level);
if (unlikely(level > OVS_RECURSION_LIMIT)) {
net_crit_ratelimited("ovs: recursion limit reached on datapath %s, probable configuration error\n",
ovs_dp_name(dp));
@@ -1659,6 +1658,6 @@ int ovs_execute_actions(struct datapath *dp, struct sk_buff *skb,
process_deferred_actions(dp);
out:
- __this_cpu_dec(ovs_pcpu_storage.exec_level);
+ __this_cpu_dec(ovs_pcpu_storage->exec_level);
return err;
}
diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index 6a304ae2d959..d5b6e2002bc1 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -244,7 +244,7 @@ void ovs_dp_detach_port(struct vport *p)
/* Must be called with rcu_read_lock. */
void ovs_dp_process_packet(struct sk_buff *skb, struct sw_flow_key *key)
{
- struct ovs_pcpu_storage *ovs_pcpu = this_cpu_ptr(&ovs_pcpu_storage);
+ struct ovs_pcpu_storage *ovs_pcpu = this_cpu_ptr(ovs_pcpu_storage);
const struct vport *p = OVS_CB(skb)->input_vport;
struct datapath *dp = p->dp;
struct sw_flow *flow;
@@ -267,7 +267,9 @@ void ovs_dp_process_packet(struct sk_buff *skb, struct sw_flow_key *key)
memset(&upcall, 0, sizeof(upcall));
upcall.cmd = OVS_PACKET_CMD_MISS;
- if (dp->user_features & OVS_DP_F_DISPATCH_UPCALL_PER_CPU)
+ if (OVS_CB(skb)->upcall_pid)
+ upcall.portid = OVS_CB(skb)->upcall_pid;
+ else if (dp->user_features & OVS_DP_F_DISPATCH_UPCALL_PER_CPU)
upcall.portid =
ovs_dp_get_upcall_portid(dp, smp_processor_id());
else
@@ -299,7 +301,7 @@ void ovs_dp_process_packet(struct sk_buff *skb, struct sw_flow_key *key)
* avoided.
*/
if (IS_ENABLED(CONFIG_PREEMPT_RT) && ovs_pcpu->owner != current) {
- local_lock_nested_bh(&ovs_pcpu_storage.bh_lock);
+ local_lock_nested_bh(&ovs_pcpu_storage->bh_lock);
ovs_pcpu->owner = current;
ovs_pcpu_locked = true;
}
@@ -310,7 +312,7 @@ void ovs_dp_process_packet(struct sk_buff *skb, struct sw_flow_key *key)
ovs_dp_name(dp), error);
if (ovs_pcpu_locked) {
ovs_pcpu->owner = NULL;
- local_unlock_nested_bh(&ovs_pcpu_storage.bh_lock);
+ local_unlock_nested_bh(&ovs_pcpu_storage->bh_lock);
}
stats_counter = &stats->n_hit;
@@ -651,6 +653,9 @@ static int ovs_packet_cmd_execute(struct sk_buff *skb, struct genl_info *info)
!!(hash & OVS_PACKET_HASH_L4_BIT));
}
+ OVS_CB(packet)->upcall_pid =
+ nla_get_u32_default(a[OVS_PACKET_ATTR_UPCALL_PID], 0);
+
/* Build an sw_flow for sending this packet. */
flow = ovs_flow_alloc();
err = PTR_ERR(flow);
@@ -689,13 +694,13 @@ static int ovs_packet_cmd_execute(struct sk_buff *skb, struct genl_info *info)
sf_acts = rcu_dereference(flow->sf_acts);
local_bh_disable();
- local_lock_nested_bh(&ovs_pcpu_storage.bh_lock);
+ local_lock_nested_bh(&ovs_pcpu_storage->bh_lock);
if (IS_ENABLED(CONFIG_PREEMPT_RT))
- this_cpu_write(ovs_pcpu_storage.owner, current);
+ this_cpu_write(ovs_pcpu_storage->owner, current);
err = ovs_execute_actions(dp, packet, sf_acts, &flow->key);
if (IS_ENABLED(CONFIG_PREEMPT_RT))
- this_cpu_write(ovs_pcpu_storage.owner, NULL);
- local_unlock_nested_bh(&ovs_pcpu_storage.bh_lock);
+ this_cpu_write(ovs_pcpu_storage->owner, NULL);
+ local_unlock_nested_bh(&ovs_pcpu_storage->bh_lock);
local_bh_enable();
rcu_read_unlock();
@@ -719,6 +724,7 @@ static const struct nla_policy packet_policy[OVS_PACKET_ATTR_MAX + 1] = {
[OVS_PACKET_ATTR_PROBE] = { .type = NLA_FLAG },
[OVS_PACKET_ATTR_MRU] = { .type = NLA_U16 },
[OVS_PACKET_ATTR_HASH] = { .type = NLA_U64 },
+ [OVS_PACKET_ATTR_UPCALL_PID] = { .type = NLA_U32 },
};
static const struct genl_small_ops dp_packet_genl_ops[] = {
@@ -2744,6 +2750,28 @@ static struct drop_reason_list drop_reason_list_ovs = {
.n_reasons = ARRAY_SIZE(ovs_drop_reasons),
};
+static int __init ovs_alloc_percpu_storage(void)
+{
+ unsigned int cpu;
+
+ ovs_pcpu_storage = alloc_percpu(*ovs_pcpu_storage);
+ if (!ovs_pcpu_storage)
+ return -ENOMEM;
+
+ for_each_possible_cpu(cpu) {
+ struct ovs_pcpu_storage *ovs_pcpu;
+
+ ovs_pcpu = per_cpu_ptr(ovs_pcpu_storage, cpu);
+ local_lock_init(&ovs_pcpu->bh_lock);
+ }
+ return 0;
+}
+
+static void ovs_free_percpu_storage(void)
+{
+ free_percpu(ovs_pcpu_storage);
+}
+
static int __init dp_init(void)
{
int err;
@@ -2753,6 +2781,10 @@ static int __init dp_init(void)
pr_info("Open vSwitch switching datapath\n");
+ err = ovs_alloc_percpu_storage();
+ if (err)
+ goto error;
+
err = ovs_internal_dev_rtnl_link_register();
if (err)
goto error;
@@ -2799,6 +2831,7 @@ error_flow_exit:
error_unreg_rtnl_link:
ovs_internal_dev_rtnl_link_unregister();
error:
+ ovs_free_percpu_storage();
return err;
}
@@ -2813,6 +2846,7 @@ static void dp_cleanup(void)
ovs_vport_exit();
ovs_flow_exit();
ovs_internal_dev_rtnl_link_unregister();
+ ovs_free_percpu_storage();
}
module_init(dp_init);
diff --git a/net/openvswitch/datapath.h b/net/openvswitch/datapath.h
index 1b5348b0f559..db0c3e69d66c 100644
--- a/net/openvswitch/datapath.h
+++ b/net/openvswitch/datapath.h
@@ -121,6 +121,8 @@ struct datapath {
* @cutlen: The number of bytes from the packet end to be removed.
* @probability: The sampling probability that was applied to this skb; 0 means
* no sampling has occurred; U32_MAX means 100% probability.
+ * @upcall_pid: Netlink socket PID to use for sending this packet to userspace;
+ * 0 means "not set" and default per-CPU or per-vport dispatch should be used.
*/
struct ovs_skb_cb {
struct vport *input_vport;
@@ -128,6 +130,7 @@ struct ovs_skb_cb {
u16 acts_origlen;
u32 cutlen;
u32 probability;
+ u32 upcall_pid;
};
#define OVS_CB(skb) ((struct ovs_skb_cb *)(skb)->cb)
@@ -220,7 +223,8 @@ struct ovs_pcpu_storage {
struct task_struct *owner;
local_lock_t bh_lock;
};
-DECLARE_PER_CPU(struct ovs_pcpu_storage, ovs_pcpu_storage);
+
+extern struct ovs_pcpu_storage __percpu *ovs_pcpu_storage;
/**
* enum ovs_pkt_hash_types - hash info to include with a packet
diff --git a/net/openvswitch/vport.c b/net/openvswitch/vport.c
index 8732f6e51ae5..6bbbc16ab778 100644
--- a/net/openvswitch/vport.c
+++ b/net/openvswitch/vport.c
@@ -501,6 +501,7 @@ int ovs_vport_receive(struct vport *vport, struct sk_buff *skb,
OVS_CB(skb)->mru = 0;
OVS_CB(skb)->cutlen = 0;
OVS_CB(skb)->probability = 0;
+ OVS_CB(skb)->upcall_pid = 0;
if (unlikely(dev_net(skb->dev) != ovs_dp_get_net(vport->dp))) {
u32 mark;
generated by cgit (git 2.34.1) at 2025-07-31 17:45:14 +0000