[PATCH] xdp: Fix cleanup on map free for devmap_hash map type

[PATCH] xdp: Fix cleanup on map free for devmap_hash map type

[Toke Høiland-Jørgensen]

Tetsuo pointed out that it was not only the device unregister hook that was
broken for devmap_hash types, it was also cleanup on map free. So better
fix this as well.

While we're add it, there's no reason to allocate the netdev_map array for
DEVMAP_HASH, so skip that and adjust the cost accordingly.

Fixes: 6f9d451ab1a3 ("xdp: Add devmap_hash map type for looking up devices by hashed index")
Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
---
 kernel/bpf/devmap.c | 74 ++++++++++++++++++++++++++++-----------------
 1 file changed, 46 insertions(+), 28 deletions(-)

diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index 3867864cdc2f..3d3d61b5985b 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -74,7 +74,7 @@ struct bpf_dtab_netdev {
 
 struct bpf_dtab {
 	struct bpf_map map;
-	struct bpf_dtab_netdev **netdev_map;
+	struct bpf_dtab_netdev **netdev_map; /* DEVMAP type only */
 	struct list_head __percpu *flush_list;
 	struct list_head list;
 
@@ -101,6 +101,12 @@ static struct hlist_head *dev_map_create_hash(unsigned int entries)
 	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;
@@ -120,8 +126,7 @@ 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) dtab->map.max_entries * sizeof(struct bpf_dtab_netdev *);
-	cost += sizeof(struct list_head) * num_possible_cpus();
+	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);
@@ -129,6 +134,8 @@ static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
 		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 */
@@ -143,24 +150,22 @@ static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
 	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_percpu;
 
 		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_percpu;
 	}
 
 	return 0;
 
-free_map_area:
-	bpf_map_area_free(dtab->netdev_map);
 free_percpu:
 	free_percpu(dtab->flush_list);
 free_charge:
@@ -228,21 +233,40 @@ static void dev_map_free(struct bpf_map *map)
 			cond_resched();
 	}
 
-	for (i = 0; i < dtab->map.max_entries; i++) {
-		struct bpf_dtab_netdev *dev;
+	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;
 
-		dev = dtab->netdev_map[i];
-		if (!dev)
-			continue;
+			head = dev_map_index_hash(dtab, i);
 
-		free_percpu(dev->bulkq);
-		dev_put(dev->dev);
-		kfree(dev);
+			hlist_for_each_entry_safe(dev, next, head, index_hlist) {
+				hlist_del_rcu(&dev->index_hlist);
+				free_percpu(dev->bulkq);
+				dev_put(dev->dev);
+				kfree(dev);
+			}
+		}
+
+		kfree(dtab->dev_index_head);
+	} else {
+		for (i = 0; i < dtab->map.max_entries; i++) {
+			struct bpf_dtab_netdev *dev;
+
+			dev = dtab->netdev_map[i];
+			if (!dev)
+				continue;
+
+			free_percpu(dev->bulkq);
+			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);
 }
 
@@ -263,12 +287,6 @@ static int dev_map_get_next_key(struct bpf_map *map, void *key, void *next_key)
 	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);
-- 
2.24.0

RE: [PATCH] xdp: Fix cleanup on map free for devmap_hash map type

[John Fastabend]

Toke Høiland-Jørgensen wrote:
> Tetsuo pointed out that it was not only the device unregister hook that was
> broken for devmap_hash types, it was also cleanup on map free. So better
> fix this as well.
> 
> While we're add it, there's no reason to allocate the netdev_map array for
              ^^^
              at
> DEVMAP_HASH, so skip that and adjust the cost accordingly.

Beyond saving space without pulling these apart the free would have gotten
fairly ugly.

> 
> Fixes: 6f9d451ab1a3 ("xdp: Add devmap_hash map type for looking up devices by hashed index")
> Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
> Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
> ---
>  kernel/bpf/devmap.c | 74 ++++++++++++++++++++++++++++-----------------
>  1 file changed, 46 insertions(+), 28 deletions(-)

small typo in commit message otherwise

Acked-by: John Fastabend <john.fastabend@gmail.com>

Re: [PATCH] xdp: Fix cleanup on map free for devmap_hash map type

[Alexei Starovoitov]

On Thu, Nov 21, 2019 at 1:32 PM John Fastabend <john.fastabend@gmail.com> wrote:
>
> Toke Høiland-Jørgensen wrote:
> > Tetsuo pointed out that it was not only the device unregister hook that was
> > broken for devmap_hash types, it was also cleanup on map free. So better
> > fix this as well.
> >
> > While we're add it, there's no reason to allocate the netdev_map array for
>               ^^^
>               at
> > DEVMAP_HASH, so skip that and adjust the cost accordingly.
>
> Beyond saving space without pulling these apart the free would have gotten
> fairly ugly.
>
> >
> > Fixes: 6f9d451ab1a3 ("xdp: Add devmap_hash map type for looking up devices by hashed index")
> > Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
> > Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
> > ---
> >  kernel/bpf/devmap.c | 74 ++++++++++++++++++++++++++++-----------------
> >  1 file changed, 46 insertions(+), 28 deletions(-)
>
> small typo in commit message otherwise
>
> Acked-by: John Fastabend <john.fastabend@gmail.com>

Fixed the typo and applied to bpf-next. Thanks