xdp: Make devmap flush_list common for all map instances

The devmap flush list is used to track entries that need to flushed
from via the xdp_do_flush_map() function. This list used to be
per-map, but there is really no reason for that. Instead make the
flush list global for all devmaps, which simplifies __dev_map_flush()
and dev_map_init_map().

Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/bpf/20191219061006.21980-6-bjorn.topel@gmail.com

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index d467983..3119180 100644 (file)
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -959,7 +959,7 @@ struct sk_buff;
 
 struct bpf_dtab_netdev *__dev_map_lookup_elem(struct bpf_map *map, u32 key);
 struct bpf_dtab_netdev *__dev_map_hash_lookup_elem(struct bpf_map *map, u32 key);
-void __dev_map_flush(struct bpf_map *map);
+void __dev_map_flush(void);
 int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
                    struct net_device *dev_rx);
 int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
@@ -1068,7 +1068,7 @@ static inline struct net_device  *__dev_map_hash_lookup_elem(struct bpf_map *map
        return NULL;
 }
 
-static inline void __dev_map_flush(struct bpf_map *map)
+static inline void __dev_map_flush(void)
 {
 }
 

diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index b7595de..da9c832 100644 (file)
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -75,7 +75,6 @@ struct bpf_dtab_netdev {
 struct bpf_dtab {
        struct bpf_map map;
        struct bpf_dtab_netdev **netdev_map; /* DEVMAP type only */
-       struct list_head __percpu *flush_list;
        struct list_head list;
 
        /* these are only used for DEVMAP_HASH type maps */
@@ -85,6 +84,7 @@ struct bpf_dtab {
        u32 n_buckets;
 };
 
+static DEFINE_PER_CPU(struct list_head, dev_map_flush_list);
 static DEFINE_SPINLOCK(dev_map_lock);
 static LIST_HEAD(dev_map_list);
 
@@ -109,8 +109,8 @@ static inline struct hlist_head *dev_map_index_hash(struct bpf_dtab *dtab,
 
 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 ||
@@ -125,9 +125,6 @@ static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
 
        bpf_map_init_from_attr(&dtab->map, attr);
 
-       /* make sure page count doesn't overflow */
-       cost = (u64) 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);
 
@@ -143,17 +140,10 @@ static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
        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));
-
        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_percpu;
+                       goto free_charge;
 
                spin_lock_init(&dtab->index_lock);
        } else {
@@ -161,13 +151,11 @@ static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
                                                      sizeof(struct bpf_dtab_netdev *),
                                                      dtab->map.numa_node);
                if (!dtab->netdev_map)
-                       goto free_percpu;
+                       goto free_charge;
        }
 
        return 0;
 
-free_percpu:
-       free_percpu(dtab->flush_list);
 free_charge:
        bpf_map_charge_finish(&dtab->map.memory);
        return -ENOMEM;
@@ -254,7 +242,6 @@ static void dev_map_free(struct bpf_map *map)
                bpf_map_area_free(dtab->netdev_map);
        }
 
-       free_percpu(dtab->flush_list);
        kfree(dtab);
 }
 
@@ -384,10 +371,9 @@ error:
  * 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.
  */
-void __dev_map_flush(struct bpf_map *map)
+void __dev_map_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 list_head *flush_list = this_cpu_ptr(&dev_map_flush_list);
        struct xdp_bulk_queue *bq, *tmp;
 
        rcu_read_lock();
@@ -419,7 +405,7 @@ static int bq_enqueue(struct bpf_dtab_netdev *obj, struct xdp_frame *xdpf,
                      struct net_device *dev_rx)
 
 {
-       struct list_head *flush_list = this_cpu_ptr(obj->dtab->flush_list);
+       struct list_head *flush_list = this_cpu_ptr(&dev_map_flush_list);
        struct xdp_bulk_queue *bq = this_cpu_ptr(obj->bulkq);
 
        if (unlikely(bq->count == DEV_MAP_BULK_SIZE))
@@ -777,10 +763,15 @@ static struct notifier_block dev_map_notifier = {
 
 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_map_flush_list, cpu));
        return 0;
 }
 

diff --git a/net/core/filter.c b/net/core/filter.c
index c51678c..b7570cb 100644 (file)
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -3555,7 +3555,7 @@ void xdp_do_flush_map(void)
                switch (map->map_type) {
                case BPF_MAP_TYPE_DEVMAP:
                case BPF_MAP_TYPE_DEVMAP_HASH:
-                       __dev_map_flush(map);
+                       __dev_map_flush();
                        break;
                case BPF_MAP_TYPE_CPUMAP:
                        __cpu_map_flush(map);