(BPF_F_NUMA_NODE | BPF_F_RDONLY | BPF_F_WRONLY)
#define DEV_MAP_BULK_SIZE 16
-struct bpf_dtab_netdev;
-
-struct xdp_bulk_queue {
+struct xdp_dev_bulk_queue {
struct xdp_frame *q[DEV_MAP_BULK_SIZE];
struct list_head flush_node;
+ struct net_device *dev;
struct net_device *dev_rx;
- struct bpf_dtab_netdev *obj;
unsigned int count;
};
struct net_device *dev; /* must be first member, due to tracepoint */
struct hlist_node index_hlist;
struct bpf_dtab *dtab;
- struct xdp_bulk_queue __percpu *bulkq;
struct rcu_head rcu;
- unsigned int idx; /* keep track of map index for tracepoint */
+ unsigned int idx;
};
struct bpf_dtab {
struct bpf_map map;
- struct bpf_dtab_netdev **netdev_map;
- struct list_head __percpu *flush_list;
+ struct bpf_dtab_netdev **netdev_map; /* DEVMAP type only */
struct list_head list;
/* these are only used for DEVMAP_HASH type maps */
u32 n_buckets;
};
+static DEFINE_PER_CPU(struct list_head, dev_flush_list);
static DEFINE_SPINLOCK(dev_map_lock);
static LIST_HEAD(dev_map_list);
return hash;
}
+static inline struct hlist_head *dev_map_index_hash(struct bpf_dtab *dtab,
+ int idx)
+{
+ return &dtab->dev_index_head[idx & (dtab->n_buckets - 1)];
+}
+
static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
{
- int err, cpu;
- u64 cost;
+ u64 cost = 0;
+ int err;
/* check sanity of attributes */
if (attr->max_entries == 0 || attr->key_size != 4 ||
bpf_map_init_from_attr(&dtab->map, attr);
- /* make sure page count doesn't overflow */
- cost = (u64) dtab->map.max_entries * sizeof(struct bpf_dtab_netdev *);
- cost += sizeof(struct list_head) * num_possible_cpus();
-
if (attr->map_type == BPF_MAP_TYPE_DEVMAP_HASH) {
dtab->n_buckets = roundup_pow_of_two(dtab->map.max_entries);
if (!dtab->n_buckets) /* Overflow check */
return -EINVAL;
cost += (u64) sizeof(struct hlist_head) * dtab->n_buckets;
+ } else {
+ cost += (u64) dtab->map.max_entries * sizeof(struct bpf_dtab_netdev *);
}
/* if map size is larger than memlock limit, reject it */
if (err)
return -EINVAL;
- dtab->flush_list = alloc_percpu(struct list_head);
- if (!dtab->flush_list)
- goto free_charge;
-
- for_each_possible_cpu(cpu)
- INIT_LIST_HEAD(per_cpu_ptr(dtab->flush_list, cpu));
-
- dtab->netdev_map = bpf_map_area_alloc(dtab->map.max_entries *
- sizeof(struct bpf_dtab_netdev *),
- dtab->map.numa_node);
- if (!dtab->netdev_map)
- goto free_percpu;
-
if (attr->map_type == BPF_MAP_TYPE_DEVMAP_HASH) {
dtab->dev_index_head = dev_map_create_hash(dtab->n_buckets);
if (!dtab->dev_index_head)
- goto free_map_area;
+ goto free_charge;
spin_lock_init(&dtab->index_lock);
+ } else {
+ dtab->netdev_map = bpf_map_area_alloc(dtab->map.max_entries *
+ sizeof(struct bpf_dtab_netdev *),
+ dtab->map.numa_node);
+ if (!dtab->netdev_map)
+ goto free_charge;
}
return 0;
-free_map_area:
- bpf_map_area_free(dtab->netdev_map);
-free_percpu:
- free_percpu(dtab->flush_list);
free_charge:
bpf_map_charge_finish(&dtab->map.memory);
return -ENOMEM;
static void dev_map_free(struct bpf_map *map)
{
struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map);
- int i, cpu;
+ int i;
/* At this point bpf_prog->aux->refcnt == 0 and this map->refcnt == 0,
* so the programs (can be more than one that used this map) were
- * disconnected from events. Wait for outstanding critical sections in
- * these programs to complete. The rcu critical section only guarantees
- * no further reads against netdev_map. It does __not__ ensure pending
- * flush operations (if any) are complete.
+ * disconnected from events. The following synchronize_rcu() guarantees
+ * both rcu read critical sections complete and waits for
+ * preempt-disable regions (NAPI being the relevant context here) so we
+ * are certain there will be no further reads against the netdev_map and
+ * all flush operations are complete. Flush operations can only be done
+ * from NAPI context for this reason.
*/
spin_lock(&dev_map_lock);
/* Make sure prior __dev_map_entry_free() have completed. */
rcu_barrier();
- /* To ensure all pending flush operations have completed wait for flush
- * list to empty on _all_ cpus.
- * Because the above synchronize_rcu() ensures the map is disconnected
- * from the program we can assume no new items will be added.
- */
- for_each_online_cpu(cpu) {
- struct list_head *flush_list = per_cpu_ptr(dtab->flush_list, cpu);
+ if (dtab->map.map_type == BPF_MAP_TYPE_DEVMAP_HASH) {
+ for (i = 0; i < dtab->n_buckets; i++) {
+ struct bpf_dtab_netdev *dev;
+ struct hlist_head *head;
+ struct hlist_node *next;
- while (!list_empty(flush_list))
- cond_resched();
- }
+ head = dev_map_index_hash(dtab, i);
- for (i = 0; i < dtab->map.max_entries; i++) {
- struct bpf_dtab_netdev *dev;
+ hlist_for_each_entry_safe(dev, next, head, index_hlist) {
+ hlist_del_rcu(&dev->index_hlist);
+ dev_put(dev->dev);
+ kfree(dev);
+ }
+ }
- dev = dtab->netdev_map[i];
- if (!dev)
- continue;
+ kfree(dtab->dev_index_head);
+ } else {
+ for (i = 0; i < dtab->map.max_entries; i++) {
+ struct bpf_dtab_netdev *dev;
- free_percpu(dev->bulkq);
- dev_put(dev->dev);
- kfree(dev);
+ dev = dtab->netdev_map[i];
+ if (!dev)
+ continue;
+
+ dev_put(dev->dev);
+ kfree(dev);
+ }
+
+ bpf_map_area_free(dtab->netdev_map);
}
- free_percpu(dtab->flush_list);
- bpf_map_area_free(dtab->netdev_map);
- kfree(dtab->dev_index_head);
kfree(dtab);
}
return 0;
}
-static inline struct hlist_head *dev_map_index_hash(struct bpf_dtab *dtab,
- int idx)
-{
- return &dtab->dev_index_head[idx & (dtab->n_buckets - 1)];
-}
-
struct bpf_dtab_netdev *__dev_map_hash_lookup_elem(struct bpf_map *map, u32 key)
{
struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map);
struct hlist_head *head = dev_map_index_hash(dtab, key);
struct bpf_dtab_netdev *dev;
- hlist_for_each_entry_rcu(dev, head, index_hlist)
+ hlist_for_each_entry_rcu(dev, head, index_hlist,
+ lockdep_is_held(&dtab->index_lock))
if (dev->idx == key)
return dev;
return -ENOENT;
}
-static int bq_xmit_all(struct xdp_bulk_queue *bq, u32 flags,
- bool in_napi_ctx)
+static int bq_xmit_all(struct xdp_dev_bulk_queue *bq, u32 flags)
{
- struct bpf_dtab_netdev *obj = bq->obj;
- struct net_device *dev = obj->dev;
+ struct net_device *dev = bq->dev;
int sent = 0, drops = 0, err = 0;
int i;
out:
bq->count = 0;
- trace_xdp_devmap_xmit(&obj->dtab->map, obj->idx,
- sent, drops, bq->dev_rx, dev, err);
+ trace_xdp_devmap_xmit(bq->dev_rx, dev, sent, drops, err);
bq->dev_rx = NULL;
__list_del_clearprev(&bq->flush_node);
return 0;
for (i = 0; i < bq->count; i++) {
struct xdp_frame *xdpf = bq->q[i];
- /* RX path under NAPI protection, can return frames faster */
- if (likely(in_napi_ctx))
- xdp_return_frame_rx_napi(xdpf);
- else
- xdp_return_frame(xdpf);
+ xdp_return_frame_rx_napi(xdpf);
drops++;
}
goto out;
}
-/* __dev_map_flush is called from xdp_do_flush_map() which _must_ be signaled
+/* __dev_flush is called from xdp_do_flush() which _must_ be signaled
* from the driver before returning from its napi->poll() routine. The poll()
* routine is called either from busy_poll context or net_rx_action signaled
* from NET_RX_SOFTIRQ. Either way the poll routine must complete before the
* net device can be torn down. On devmap tear down we ensure the flush list
* is empty before completing to ensure all flush operations have completed.
+ * When drivers update the bpf program they may need to ensure any flush ops
+ * are also complete. Using synchronize_rcu or call_rcu will suffice for this
+ * because both wait for napi context to exit.
*/
-void __dev_map_flush(struct bpf_map *map)
+void __dev_flush(void)
{
- struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map);
- struct list_head *flush_list = this_cpu_ptr(dtab->flush_list);
- struct xdp_bulk_queue *bq, *tmp;
+ struct list_head *flush_list = this_cpu_ptr(&dev_flush_list);
+ struct xdp_dev_bulk_queue *bq, *tmp;
- rcu_read_lock();
list_for_each_entry_safe(bq, tmp, flush_list, flush_node)
- bq_xmit_all(bq, XDP_XMIT_FLUSH, true);
- rcu_read_unlock();
+ bq_xmit_all(bq, XDP_XMIT_FLUSH);
}
/* rcu_read_lock (from syscall and BPF contexts) ensures that if a delete and/or
/* Runs under RCU-read-side, plus in softirq under NAPI protection.
* Thus, safe percpu variable access.
*/
-static int bq_enqueue(struct bpf_dtab_netdev *obj, struct xdp_frame *xdpf,
+static int bq_enqueue(struct net_device *dev, struct xdp_frame *xdpf,
struct net_device *dev_rx)
-
{
- struct list_head *flush_list = this_cpu_ptr(obj->dtab->flush_list);
- struct xdp_bulk_queue *bq = this_cpu_ptr(obj->bulkq);
+ struct list_head *flush_list = this_cpu_ptr(&dev_flush_list);
+ struct xdp_dev_bulk_queue *bq = this_cpu_ptr(dev->xdp_bulkq);
if (unlikely(bq->count == DEV_MAP_BULK_SIZE))
- bq_xmit_all(bq, 0, true);
+ bq_xmit_all(bq, 0);
/* Ingress dev_rx will be the same for all xdp_frame's in
* bulk_queue, because bq stored per-CPU and must be flushed
return 0;
}
-int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
- struct net_device *dev_rx)
+static inline int __xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
+ struct net_device *dev_rx)
{
- struct net_device *dev = dst->dev;
struct xdp_frame *xdpf;
int err;
if (unlikely(!xdpf))
return -EOVERFLOW;
- return bq_enqueue(dst, xdpf, dev_rx);
+ return bq_enqueue(dev, xdpf, dev_rx);
+}
+
+int dev_xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
+ struct net_device *dev_rx)
+{
+ return __xdp_enqueue(dev, xdp, dev_rx);
+}
+
+int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
+ struct net_device *dev_rx)
+{
+ struct net_device *dev = dst->dev;
+
+ return __xdp_enqueue(dev, xdp, dev_rx);
}
int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
return dev ? &dev->ifindex : NULL;
}
-static void dev_map_flush_old(struct bpf_dtab_netdev *dev)
-{
- if (dev->dev->netdev_ops->ndo_xdp_xmit) {
- struct xdp_bulk_queue *bq;
- int cpu;
-
- rcu_read_lock();
- for_each_online_cpu(cpu) {
- bq = per_cpu_ptr(dev->bulkq, cpu);
- bq_xmit_all(bq, XDP_XMIT_FLUSH, false);
- }
- rcu_read_unlock();
- }
-}
-
static void __dev_map_entry_free(struct rcu_head *rcu)
{
struct bpf_dtab_netdev *dev;
dev = container_of(rcu, struct bpf_dtab_netdev, rcu);
- dev_map_flush_old(dev);
- free_percpu(dev->bulkq);
dev_put(dev->dev);
kfree(dev);
}
return -EINVAL;
/* Use call_rcu() here to ensure any rcu critical sections have
- * completed, but this does not guarantee a flush has happened
- * yet. Because driver side rcu_read_lock/unlock only protects the
- * running XDP program. However, for pending flush operations the
- * dev and ctx are stored in another per cpu map. And additionally,
- * the driver tear down ensures all soft irqs are complete before
- * removing the net device in the case of dev_put equals zero.
+ * completed as well as any flush operations because call_rcu
+ * will wait for preempt-disable region to complete, NAPI in this
+ * context. And additionally, the driver tear down ensures all
+ * soft irqs are complete before removing the net device in the
+ * case of dev_put equals zero.
*/
old_dev = xchg(&dtab->netdev_map[k], NULL);
if (old_dev)
u32 ifindex,
unsigned int idx)
{
- gfp_t gfp = GFP_ATOMIC | __GFP_NOWARN;
struct bpf_dtab_netdev *dev;
- struct xdp_bulk_queue *bq;
- int cpu;
- dev = kmalloc_node(sizeof(*dev), gfp, dtab->map.numa_node);
+ dev = kmalloc_node(sizeof(*dev), GFP_ATOMIC | __GFP_NOWARN,
+ dtab->map.numa_node);
if (!dev)
return ERR_PTR(-ENOMEM);
- dev->bulkq = __alloc_percpu_gfp(sizeof(*dev->bulkq),
- sizeof(void *), gfp);
- if (!dev->bulkq) {
- kfree(dev);
- return ERR_PTR(-ENOMEM);
- }
-
- for_each_possible_cpu(cpu) {
- bq = per_cpu_ptr(dev->bulkq, cpu);
- bq->obj = dev;
- }
-
dev->dev = dev_get_by_index(net, ifindex);
if (!dev->dev) {
- free_percpu(dev->bulkq);
kfree(dev);
return ERR_PTR(-EINVAL);
}
{
struct net_device *netdev = netdev_notifier_info_to_dev(ptr);
struct bpf_dtab *dtab;
- int i;
+ int i, cpu;
switch (event) {
+ case NETDEV_REGISTER:
+ if (!netdev->netdev_ops->ndo_xdp_xmit || netdev->xdp_bulkq)
+ break;
+
+ /* will be freed in free_netdev() */
+ netdev->xdp_bulkq =
+ __alloc_percpu_gfp(sizeof(struct xdp_dev_bulk_queue),
+ sizeof(void *), GFP_ATOMIC);
+ if (!netdev->xdp_bulkq)
+ return NOTIFY_BAD;
+
+ for_each_possible_cpu(cpu)
+ per_cpu_ptr(netdev->xdp_bulkq, cpu)->dev = netdev;
+ break;
case NETDEV_UNREGISTER:
/* This rcu_read_lock/unlock pair is needed because
* dev_map_list is an RCU list AND to ensure a delete
static int __init dev_map_init(void)
{
+ int cpu;
+
/* Assure tracepoint shadow struct _bpf_dtab_netdev is in sync */
BUILD_BUG_ON(offsetof(struct bpf_dtab_netdev, dev) !=
offsetof(struct _bpf_dtab_netdev, dev));
register_netdevice_notifier(&dev_map_notifier);
+
+ for_each_possible_cpu(cpu)
+ INIT_LIST_HEAD(&per_cpu(dev_flush_list, cpu));
return 0;
}