summaryrefslogtreecommitdiffstats
path: root/kernel
diff options
context:
space:
mode:
authorDavid S. Miller <davem@davemloft.net>2018-08-13 10:07:23 -0700
committerDavid S. Miller <davem@davemloft.net>2018-08-13 10:07:23 -0700
commitc1617fb4c5eea2ad38b5ffb937a732dbb137117f (patch)
tree7331bdc1fd66799533d2339816c688ded60e0475 /kernel
parent961d9735357ec59be3e3e2c37c0ca6494f04461b (diff)
parent2ce3206b9eb3943de09f3bf4ec9134568420d8b9 (diff)
downloadlinux-c1617fb4c5eea2ad38b5ffb937a732dbb137117f.tar.bz2
Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next
Daniel Borkmann says: ==================== pull-request: bpf-next 2018-08-13 The following pull-request contains BPF updates for your *net-next* tree. The main changes are: 1) Add driver XDP support for veth. This can be used in conjunction with redirect of another XDP program e.g. sitting on NIC so the xdp_frame can be forwarded to the peer veth directly without modification, from Toshiaki. 2) Add a new BPF map type REUSEPORT_SOCKARRAY and prog type SK_REUSEPORT in order to provide more control and visibility on where a SO_REUSEPORT sk should be located, and the latter enables to directly select a sk from the bpf map. This also enables map-in-map for application migration use cases, from Martin. 3) Add a new BPF helper bpf_skb_ancestor_cgroup_id() that returns the id of cgroup v2 that is the ancestor of the cgroup associated with the skb at the ancestor_level, from Andrey. 4) Implement BPF fs map pretty-print support based on BTF data for regular hash table and LRU map, from Yonghong. 5) Decouple the ability to attach BTF for a map from the key and value pretty-printer in BPF fs, and enable further support of BTF for maps for percpu and LPM trie, from Daniel. 6) Implement a better BPF sample of using XDP's CPU redirect feature for load balancing SKB processing to remote CPU. The sample implements the same XDP load balancing as Suricata does which is symmetric hash based on IP and L4 protocol, from Jesper. 7) Revert adding NULL pointer check with WARN_ON_ONCE() in __xdp_return()'s critical path as it is ensured that the allocator is present, from Björn. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'kernel')
-rw-r--r--kernel/bpf/Makefile3
-rw-r--r--kernel/bpf/arraymap.c28
-rw-r--r--kernel/bpf/cpumap.c1
-rw-r--r--kernel/bpf/devmap.c1
-rw-r--r--kernel/bpf/hashtab.c26
-rw-r--r--kernel/bpf/inode.c11
-rw-r--r--kernel/bpf/local_storage.c1
-rw-r--r--kernel/bpf/lpm_trie.c12
-rw-r--r--kernel/bpf/reuseport_array.c363
-rw-r--r--kernel/bpf/sockmap.c2
-rw-r--r--kernel/bpf/stackmap.c1
-rw-r--r--kernel/bpf/syscall.c42
-rw-r--r--kernel/bpf/verifier.c9
-rw-r--r--kernel/bpf/xskmap.c3
14 files changed, 478 insertions, 25 deletions
diff --git a/kernel/bpf/Makefile b/kernel/bpf/Makefile
index e8906cbad81f..0488b8258321 100644
--- a/kernel/bpf/Makefile
+++ b/kernel/bpf/Makefile
@@ -23,3 +23,6 @@ ifeq ($(CONFIG_PERF_EVENTS),y)
obj-$(CONFIG_BPF_SYSCALL) += stackmap.o
endif
obj-$(CONFIG_CGROUP_BPF) += cgroup.o
+ifeq ($(CONFIG_INET),y)
+obj-$(CONFIG_BPF_SYSCALL) += reuseport_array.o
+endif
diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c
index 2aa55d030c77..0c17aab3ce5f 100644
--- a/kernel/bpf/arraymap.c
+++ b/kernel/bpf/arraymap.c
@@ -54,7 +54,7 @@ static int bpf_array_alloc_percpu(struct bpf_array *array)
}
/* Called from syscall */
-static int array_map_alloc_check(union bpf_attr *attr)
+int array_map_alloc_check(union bpf_attr *attr)
{
bool percpu = attr->map_type == BPF_MAP_TYPE_PERCPU_ARRAY;
int numa_node = bpf_map_attr_numa_node(attr);
@@ -358,27 +358,20 @@ static void array_map_seq_show_elem(struct bpf_map *map, void *key,
rcu_read_unlock();
}
-static int array_map_check_btf(const struct bpf_map *map, const struct btf *btf,
- u32 btf_key_id, u32 btf_value_id)
+static int array_map_check_btf(const struct bpf_map *map,
+ const struct btf_type *key_type,
+ const struct btf_type *value_type)
{
- const struct btf_type *key_type, *value_type;
- u32 key_size, value_size;
u32 int_data;
- key_type = btf_type_id_size(btf, &btf_key_id, &key_size);
- if (!key_type || BTF_INFO_KIND(key_type->info) != BTF_KIND_INT)
+ if (BTF_INFO_KIND(key_type->info) != BTF_KIND_INT)
return -EINVAL;
int_data = *(u32 *)(key_type + 1);
- /* bpf array can only take a u32 key. This check makes
- * sure that the btf matches the attr used during map_create.
+ /* bpf array can only take a u32 key. This check makes sure
+ * that the btf matches the attr used during map_create.
*/
- if (BTF_INT_BITS(int_data) != 32 || key_size != 4 ||
- BTF_INT_OFFSET(int_data))
- return -EINVAL;
-
- value_type = btf_type_id_size(btf, &btf_value_id, &value_size);
- if (!value_type || value_size != map->value_size)
+ if (BTF_INT_BITS(int_data) != 32 || BTF_INT_OFFSET(int_data))
return -EINVAL;
return 0;
@@ -405,6 +398,7 @@ const struct bpf_map_ops percpu_array_map_ops = {
.map_lookup_elem = percpu_array_map_lookup_elem,
.map_update_elem = array_map_update_elem,
.map_delete_elem = array_map_delete_elem,
+ .map_check_btf = array_map_check_btf,
};
static int fd_array_map_alloc_check(union bpf_attr *attr)
@@ -546,6 +540,7 @@ const struct bpf_map_ops prog_array_map_ops = {
.map_fd_put_ptr = prog_fd_array_put_ptr,
.map_fd_sys_lookup_elem = prog_fd_array_sys_lookup_elem,
.map_release_uref = bpf_fd_array_map_clear,
+ .map_check_btf = map_check_no_btf,
};
static struct bpf_event_entry *bpf_event_entry_gen(struct file *perf_file,
@@ -634,6 +629,7 @@ const struct bpf_map_ops perf_event_array_map_ops = {
.map_fd_get_ptr = perf_event_fd_array_get_ptr,
.map_fd_put_ptr = perf_event_fd_array_put_ptr,
.map_release = perf_event_fd_array_release,
+ .map_check_btf = map_check_no_btf,
};
#ifdef CONFIG_CGROUPS
@@ -665,6 +661,7 @@ const struct bpf_map_ops cgroup_array_map_ops = {
.map_delete_elem = fd_array_map_delete_elem,
.map_fd_get_ptr = cgroup_fd_array_get_ptr,
.map_fd_put_ptr = cgroup_fd_array_put_ptr,
+ .map_check_btf = map_check_no_btf,
};
#endif
@@ -749,4 +746,5 @@ const struct bpf_map_ops array_of_maps_map_ops = {
.map_fd_put_ptr = bpf_map_fd_put_ptr,
.map_fd_sys_lookup_elem = bpf_map_fd_sys_lookup_elem,
.map_gen_lookup = array_of_map_gen_lookup,
+ .map_check_btf = map_check_no_btf,
};
diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c
index 46f5f29605d4..620bc5024d7d 100644
--- a/kernel/bpf/cpumap.c
+++ b/kernel/bpf/cpumap.c
@@ -555,6 +555,7 @@ const struct bpf_map_ops cpu_map_ops = {
.map_update_elem = cpu_map_update_elem,
.map_lookup_elem = cpu_map_lookup_elem,
.map_get_next_key = cpu_map_get_next_key,
+ .map_check_btf = map_check_no_btf,
};
static int bq_flush_to_queue(struct bpf_cpu_map_entry *rcpu,
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index 750d45edae79..ac1df79f3788 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -488,6 +488,7 @@ const struct bpf_map_ops dev_map_ops = {
.map_lookup_elem = dev_map_lookup_elem,
.map_update_elem = dev_map_update_elem,
.map_delete_elem = dev_map_delete_elem,
+ .map_check_btf = map_check_no_btf,
};
static int dev_map_notification(struct notifier_block *notifier,
diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
index 513d9dfcf4ee..04b8eda94e7d 100644
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -11,9 +11,11 @@
* General Public License for more details.
*/
#include <linux/bpf.h>
+#include <linux/btf.h>
#include <linux/jhash.h>
#include <linux/filter.h>
#include <linux/rculist_nulls.h>
+#include <uapi/linux/btf.h>
#include "percpu_freelist.h"
#include "bpf_lru_list.h"
#include "map_in_map.h"
@@ -1162,6 +1164,27 @@ static void htab_map_free(struct bpf_map *map)
kfree(htab);
}
+static void htab_map_seq_show_elem(struct bpf_map *map, void *key,
+ struct seq_file *m)
+{
+ void *value;
+
+ rcu_read_lock();
+
+ value = htab_map_lookup_elem(map, key);
+ if (!value) {
+ rcu_read_unlock();
+ return;
+ }
+
+ btf_type_seq_show(map->btf, map->btf_key_type_id, key, m);
+ seq_puts(m, ": ");
+ btf_type_seq_show(map->btf, map->btf_value_type_id, value, m);
+ seq_puts(m, "\n");
+
+ rcu_read_unlock();
+}
+
const struct bpf_map_ops htab_map_ops = {
.map_alloc_check = htab_map_alloc_check,
.map_alloc = htab_map_alloc,
@@ -1171,6 +1194,7 @@ const struct bpf_map_ops htab_map_ops = {
.map_update_elem = htab_map_update_elem,
.map_delete_elem = htab_map_delete_elem,
.map_gen_lookup = htab_map_gen_lookup,
+ .map_seq_show_elem = htab_map_seq_show_elem,
};
const struct bpf_map_ops htab_lru_map_ops = {
@@ -1182,6 +1206,7 @@ const struct bpf_map_ops htab_lru_map_ops = {
.map_update_elem = htab_lru_map_update_elem,
.map_delete_elem = htab_lru_map_delete_elem,
.map_gen_lookup = htab_lru_map_gen_lookup,
+ .map_seq_show_elem = htab_map_seq_show_elem,
};
/* Called from eBPF program */
@@ -1408,4 +1433,5 @@ const struct bpf_map_ops htab_of_maps_map_ops = {
.map_fd_put_ptr = bpf_map_fd_put_ptr,
.map_fd_sys_lookup_elem = bpf_map_fd_sys_lookup_elem,
.map_gen_lookup = htab_of_map_gen_lookup,
+ .map_check_btf = map_check_no_btf,
};
diff --git a/kernel/bpf/inode.c b/kernel/bpf/inode.c
index 76efe9a183f5..2ada5e21dfa6 100644
--- a/kernel/bpf/inode.c
+++ b/kernel/bpf/inode.c
@@ -196,19 +196,21 @@ static void *map_seq_next(struct seq_file *m, void *v, loff_t *pos)
{
struct bpf_map *map = seq_file_to_map(m);
void *key = map_iter(m)->key;
+ void *prev_key;
if (map_iter(m)->done)
return NULL;
if (unlikely(v == SEQ_START_TOKEN))
- goto done;
+ prev_key = NULL;
+ else
+ prev_key = key;
- if (map->ops->map_get_next_key(map, key, key)) {
+ if (map->ops->map_get_next_key(map, prev_key, key)) {
map_iter(m)->done = true;
return NULL;
}
-done:
++(*pos);
return key;
}
@@ -332,7 +334,8 @@ static int bpf_mkmap(struct dentry *dentry, umode_t mode, void *arg)
struct bpf_map *map = arg;
return bpf_mkobj_ops(dentry, mode, arg, &bpf_map_iops,
- map->btf ? &bpffs_map_fops : &bpffs_obj_fops);
+ bpf_map_support_seq_show(map) ?
+ &bpffs_map_fops : &bpffs_obj_fops);
}
static struct dentry *
diff --git a/kernel/bpf/local_storage.c b/kernel/bpf/local_storage.c
index fc4e37f68f2a..22ad967d1e5f 100644
--- a/kernel/bpf/local_storage.c
+++ b/kernel/bpf/local_storage.c
@@ -246,6 +246,7 @@ const struct bpf_map_ops cgroup_storage_map_ops = {
.map_lookup_elem = cgroup_storage_lookup_elem,
.map_update_elem = cgroup_storage_update_elem,
.map_delete_elem = cgroup_storage_delete_elem,
+ .map_check_btf = map_check_no_btf,
};
int bpf_cgroup_storage_assign(struct bpf_prog *prog, struct bpf_map *_map)
diff --git a/kernel/bpf/lpm_trie.c b/kernel/bpf/lpm_trie.c
index 1603492c9cc7..9058317ba9de 100644
--- a/kernel/bpf/lpm_trie.c
+++ b/kernel/bpf/lpm_trie.c
@@ -10,11 +10,13 @@
*/
#include <linux/bpf.h>
+#include <linux/btf.h>
#include <linux/err.h>
#include <linux/slab.h>
#include <linux/spinlock.h>
#include <linux/vmalloc.h>
#include <net/ipv6.h>
+#include <uapi/linux/btf.h>
/* Intermediate node */
#define LPM_TREE_NODE_FLAG_IM BIT(0)
@@ -686,6 +688,15 @@ free_stack:
return err;
}
+static int trie_check_btf(const struct bpf_map *map,
+ const struct btf_type *key_type,
+ const struct btf_type *value_type)
+{
+ /* Keys must have struct bpf_lpm_trie_key embedded. */
+ return BTF_INFO_KIND(key_type->info) != BTF_KIND_STRUCT ?
+ -EINVAL : 0;
+}
+
const struct bpf_map_ops trie_map_ops = {
.map_alloc = trie_alloc,
.map_free = trie_free,
@@ -693,4 +704,5 @@ const struct bpf_map_ops trie_map_ops = {
.map_lookup_elem = trie_lookup_elem,
.map_update_elem = trie_update_elem,
.map_delete_elem = trie_delete_elem,
+ .map_check_btf = trie_check_btf,
};
diff --git a/kernel/bpf/reuseport_array.c b/kernel/bpf/reuseport_array.c
new file mode 100644
index 000000000000..18e225de80ff
--- /dev/null
+++ b/kernel/bpf/reuseport_array.c
@@ -0,0 +1,363 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2018 Facebook
+ */
+#include <linux/bpf.h>
+#include <linux/err.h>
+#include <linux/sock_diag.h>
+#include <net/sock_reuseport.h>
+
+struct reuseport_array {
+ struct bpf_map map;
+ struct sock __rcu *ptrs[];
+};
+
+static struct reuseport_array *reuseport_array(struct bpf_map *map)
+{
+ return (struct reuseport_array *)map;
+}
+
+/* The caller must hold the reuseport_lock */
+void bpf_sk_reuseport_detach(struct sock *sk)
+{
+ struct sock __rcu **socks;
+
+ write_lock_bh(&sk->sk_callback_lock);
+ socks = sk->sk_user_data;
+ if (socks) {
+ WRITE_ONCE(sk->sk_user_data, NULL);
+ /*
+ * Do not move this NULL assignment outside of
+ * sk->sk_callback_lock because there is
+ * a race with reuseport_array_free()
+ * which does not hold the reuseport_lock.
+ */
+ RCU_INIT_POINTER(*socks, NULL);
+ }
+ write_unlock_bh(&sk->sk_callback_lock);
+}
+
+static int reuseport_array_alloc_check(union bpf_attr *attr)
+{
+ if (attr->value_size != sizeof(u32) &&
+ attr->value_size != sizeof(u64))
+ return -EINVAL;
+
+ return array_map_alloc_check(attr);
+}
+
+static void *reuseport_array_lookup_elem(struct bpf_map *map, void *key)
+{
+ struct reuseport_array *array = reuseport_array(map);
+ u32 index = *(u32 *)key;
+
+ if (unlikely(index >= array->map.max_entries))
+ return NULL;
+
+ return rcu_dereference(array->ptrs[index]);
+}
+
+/* Called from syscall only */
+static int reuseport_array_delete_elem(struct bpf_map *map, void *key)
+{
+ struct reuseport_array *array = reuseport_array(map);
+ u32 index = *(u32 *)key;
+ struct sock *sk;
+ int err;
+
+ if (index >= map->max_entries)
+ return -E2BIG;
+
+ if (!rcu_access_pointer(array->ptrs[index]))
+ return -ENOENT;
+
+ spin_lock_bh(&reuseport_lock);
+
+ sk = rcu_dereference_protected(array->ptrs[index],
+ lockdep_is_held(&reuseport_lock));
+ if (sk) {
+ write_lock_bh(&sk->sk_callback_lock);
+ WRITE_ONCE(sk->sk_user_data, NULL);
+ RCU_INIT_POINTER(array->ptrs[index], NULL);
+ write_unlock_bh(&sk->sk_callback_lock);
+ err = 0;
+ } else {
+ err = -ENOENT;
+ }
+
+ spin_unlock_bh(&reuseport_lock);
+
+ return err;
+}
+
+static void reuseport_array_free(struct bpf_map *map)
+{
+ struct reuseport_array *array = reuseport_array(map);
+ struct sock *sk;
+ u32 i;
+
+ synchronize_rcu();
+
+ /*
+ * ops->map_*_elem() will not be able to access this
+ * array now. Hence, this function only races with
+ * bpf_sk_reuseport_detach() which was triggerred by
+ * close() or disconnect().
+ *
+ * This function and bpf_sk_reuseport_detach() are
+ * both removing sk from "array". Who removes it
+ * first does not matter.
+ *
+ * The only concern here is bpf_sk_reuseport_detach()
+ * may access "array" which is being freed here.
+ * bpf_sk_reuseport_detach() access this "array"
+ * through sk->sk_user_data _and_ with sk->sk_callback_lock
+ * held which is enough because this "array" is not freed
+ * until all sk->sk_user_data has stopped referencing this "array".
+ *
+ * Hence, due to the above, taking "reuseport_lock" is not
+ * needed here.
+ */
+
+ /*
+ * Since reuseport_lock is not taken, sk is accessed under
+ * rcu_read_lock()
+ */
+ rcu_read_lock();
+ for (i = 0; i < map->max_entries; i++) {
+ sk = rcu_dereference(array->ptrs[i]);
+ if (sk) {
+ write_lock_bh(&sk->sk_callback_lock);
+ /*
+ * No need for WRITE_ONCE(). At this point,
+ * no one is reading it without taking the
+ * sk->sk_callback_lock.
+ */
+ sk->sk_user_data = NULL;
+ write_unlock_bh(&sk->sk_callback_lock);
+ RCU_INIT_POINTER(array->ptrs[i], NULL);
+ }
+ }
+ rcu_read_unlock();
+
+ /*
+ * Once reaching here, all sk->sk_user_data is not
+ * referenceing this "array". "array" can be freed now.
+ */
+ bpf_map_area_free(array);
+}
+
+static struct bpf_map *reuseport_array_alloc(union bpf_attr *attr)
+{
+ int err, numa_node = bpf_map_attr_numa_node(attr);
+ struct reuseport_array *array;
+ u64 cost, array_size;
+
+ if (!capable(CAP_SYS_ADMIN))
+ return ERR_PTR(-EPERM);
+
+ array_size = sizeof(*array);
+ array_size += (u64)attr->max_entries * sizeof(struct sock *);
+
+ /* make sure there is no u32 overflow later in round_up() */
+ cost = array_size;
+ if (cost >= U32_MAX - PAGE_SIZE)
+ return ERR_PTR(-ENOMEM);
+ cost = round_up(cost, PAGE_SIZE) >> PAGE_SHIFT;
+
+ err = bpf_map_precharge_memlock(cost);
+ if (err)
+ return ERR_PTR(err);
+
+ /* allocate all map elements and zero-initialize them */
+ array = bpf_map_area_alloc(array_size, numa_node);
+ if (!array)
+ return ERR_PTR(-ENOMEM);
+
+ /* copy mandatory map attributes */
+ bpf_map_init_from_attr(&array->map, attr);
+ array->map.pages = cost;
+
+ return &array->map;
+}
+
+int bpf_fd_reuseport_array_lookup_elem(struct bpf_map *map, void *key,
+ void *value)
+{
+ struct sock *sk;
+ int err;
+
+ if (map->value_size != sizeof(u64))
+ return -ENOSPC;
+
+ rcu_read_lock();
+ sk = reuseport_array_lookup_elem(map, key);
+ if (sk) {
+ *(u64 *)value = sock_gen_cookie(sk);
+ err = 0;
+ } else {
+ err = -ENOENT;
+ }
+ rcu_read_unlock();
+
+ return err;
+}
+
+static int
+reuseport_array_update_check(const struct reuseport_array *array,
+ const struct sock *nsk,
+ const struct sock *osk,
+ const struct sock_reuseport *nsk_reuse,
+ u32 map_flags)
+{
+ if (osk && map_flags == BPF_NOEXIST)
+ return -EEXIST;
+
+ if (!osk && map_flags == BPF_EXIST)
+ return -ENOENT;
+
+ if (nsk->sk_protocol != IPPROTO_UDP && nsk->sk_protocol != IPPROTO_TCP)
+ return -ENOTSUPP;
+
+ if (nsk->sk_family != AF_INET && nsk->sk_family != AF_INET6)
+ return -ENOTSUPP;
+
+ if (nsk->sk_type != SOCK_STREAM && nsk->sk_type != SOCK_DGRAM)
+ return -ENOTSUPP;
+
+ /*
+ * sk must be hashed (i.e. listening in the TCP case or binded
+ * in the UDP case) and
+ * it must also be a SO_REUSEPORT sk (i.e. reuse cannot be NULL).
+ *
+ * Also, sk will be used in bpf helper that is protected by
+ * rcu_read_lock().
+ */
+ if (!sock_flag(nsk, SOCK_RCU_FREE) || !sk_hashed(nsk) || !nsk_reuse)
+ return -EINVAL;
+
+ /* READ_ONCE because the sk->sk_callback_lock may not be held here */
+ if (READ_ONCE(nsk->sk_user_data))
+ return -EBUSY;
+
+ return 0;
+}
+
+/*
+ * Called from syscall only.
+ * The "nsk" in the fd refcnt.
+ * The "osk" and "reuse" are protected by reuseport_lock.
+ */
+int bpf_fd_reuseport_array_update_elem(struct bpf_map *map, void *key,
+ void *value, u64 map_flags)
+{
+ struct reuseport_array *array = reuseport_array(map);
+ struct sock *free_osk = NULL, *osk, *nsk;
+ struct sock_reuseport *reuse;
+ u32 index = *(u32 *)key;
+ struct socket *socket;
+ int err, fd;
+
+ if (map_flags > BPF_EXIST)
+ return -EINVAL;
+
+ if (index >= map->max_entries)
+ return -E2BIG;
+
+ if (map->value_size == sizeof(u64)) {
+ u64 fd64 = *(u64 *)value;
+
+ if (fd64 > S32_MAX)
+ return -EINVAL;
+ fd = fd64;
+ } else {
+ fd = *(int *)value;
+ }
+
+ socket = sockfd_lookup(fd, &err);
+ if (!socket)
+ return err;
+
+ nsk = socket->sk;
+ if (!nsk) {
+ err = -EINVAL;
+ goto put_file;
+ }
+
+ /* Quick checks before taking reuseport_lock */
+ err = reuseport_array_update_check(array, nsk,
+ rcu_access_pointer(array->ptrs[index]),
+ rcu_access_pointer(nsk->sk_reuseport_cb),
+ map_flags);
+ if (err)
+ goto put_file;
+
+ spin_lock_bh(&reuseport_lock);
+ /*
+ * Some of the checks only need reuseport_lock
+ * but it is done under sk_callback_lock also
+ * for simplicity reason.
+ */
+ write_lock_bh(&nsk->sk_callback_lock);
+
+ osk = rcu_dereference_protected(array->ptrs[index],
+ lockdep_is_held(&reuseport_lock));
+ reuse = rcu_dereference_protected(nsk->sk_reuseport_cb,
+ lockdep_is_held(&reuseport_lock));
+ err = reuseport_array_update_check(array, nsk, osk, reuse, map_flags);
+ if (err)
+ goto put_file_unlock;
+
+ /* Ensure reuse->reuseport_id is set */
+ err = reuseport_get_id(reuse);
+ if (err < 0)
+ goto put_file_unlock;
+
+ WRITE_ONCE(nsk->sk_user_data, &array->ptrs[index]);
+ rcu_assign_pointer(array->ptrs[index], nsk);
+ free_osk = osk;
+ err = 0;
+
+put_file_unlock:
+ write_unlock_bh(&nsk->sk_callback_lock);
+
+ if (free_osk) {
+ write_lock_bh(&free_osk->sk_callback_lock);
+ WRITE_ONCE(free_osk->sk_user_data, NULL);
+ write_unlock_bh(&free_osk->sk_callback_lock);
+ }
+
+ spin_unlock_bh(&reuseport_lock);
+put_file:
+ fput(socket->file);
+ return err;
+}
+
+/* Called from syscall */
+static int reuseport_array_get_next_key(struct bpf_map *map, void *key,
+ void *next_key)
+{
+ struct reuseport_array *array = reuseport_array(map);
+ u32 index = key ? *(u32 *)key : U32_MAX;
+ u32 *next = (u32 *)next_key;
+
+ if (index >= array->map.max_entries) {
+ *next = 0;
+ return 0;
+ }
+
+ if (index == array->map.max_entries - 1)
+ return -ENOENT;
+
+ *next = index + 1;
+ return 0;
+}
+
+const struct bpf_map_ops reuseport_array_ops = {
+ .map_alloc_check = reuseport_array_alloc_check,
+ .map_alloc = reuseport_array_alloc,
+ .map_free = reuseport_array_free,
+ .map_lookup_elem = reuseport_array_lookup_elem,
+ .map_get_next_key = reuseport_array_get_next_key,
+ .map_delete_elem = reuseport_array_delete_elem,
+};
diff --git a/kernel/bpf/sockmap.c b/kernel/bpf/sockmap.c
index 70c0755e8fc4..0c1a696b041b 100644
--- a/kernel/bpf/sockmap.c
+++ b/kernel/bpf/sockmap.c
@@ -2498,6 +2498,7 @@ const struct bpf_map_ops sock_map_ops = {
.map_update_elem = sock_map_update_elem,
.map_delete_elem = sock_map_delete_elem,
.map_release_uref = sock_map_release,
+ .map_check_btf = map_check_no_btf,
};
const struct bpf_map_ops sock_hash_ops = {
@@ -2508,6 +2509,7 @@ const struct bpf_map_ops sock_hash_ops = {
.map_update_elem = sock_hash_update_elem,
.map_delete_elem = sock_hash_delete_elem,
.map_release_uref = sock_map_release,
+ .map_check_btf = map_check_no_btf,
};
BPF_CALL_4(bpf_sock_map_update, struct bpf_sock_ops_kern *, bpf_sock,
diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c
index b675a3f3d141..8061a439ef18 100644
--- a/kernel/bpf/stackmap.c
+++ b/kernel/bpf/stackmap.c
@@ -607,6 +607,7 @@ const struct bpf_map_ops stack_map_ops = {
.map_lookup_elem = stack_map_lookup_elem,
.map_update_elem = stack_map_update_elem,
.map_delete_elem = stack_map_delete_elem,
+ .map_check_btf = map_check_no_btf,
};
static int __init stack_map_init(void)
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 5af4e9e2722d..43727ed0d94a 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -103,6 +103,7 @@ int bpf_check_uarg_tail_zero(void __user *uaddr,
const struct bpf_map_ops bpf_map_offload_ops = {
.map_alloc = bpf_map_offload_map_alloc,
.map_free = bpf_map_offload_map_free,
+ .map_check_btf = map_check_no_btf,
};
static struct bpf_map *find_and_alloc_map(union bpf_attr *attr)
@@ -455,6 +456,34 @@ static int bpf_obj_name_cpy(char *dst, const char *src)
return 0;
}
+int map_check_no_btf(const struct bpf_map *map,
+ const struct btf_type *key_type,
+ const struct btf_type *value_type)
+{
+ return -ENOTSUPP;
+}
+
+static int map_check_btf(const struct bpf_map *map, const struct btf *btf,
+ u32 btf_key_id, u32 btf_value_id)
+{
+ const struct btf_type *key_type, *value_type;
+ u32 key_size, value_size;
+ int ret = 0;
+
+ key_type = btf_type_id_size(btf, &btf_key_id, &key_size);
+ if (!key_type || key_size != map->key_size)
+ return -EINVAL;
+
+ value_type = btf_type_id_size(btf, &btf_value_id, &value_size);
+ if (!value_type || value_size != map->value_size)
+ return -EINVAL;
+
+ if (map->ops->map_check_btf)
+ ret = map->ops->map_check_btf(map, key_type, value_type);
+
+ return ret;
+}
+
#define BPF_MAP_CREATE_LAST_FIELD btf_value_type_id
/* called via syscall */
static int map_create(union bpf_attr *attr)
@@ -489,8 +518,7 @@ static int map_create(union bpf_attr *attr)
atomic_set(&map->refcnt, 1);
atomic_set(&map->usercnt, 1);
- if (bpf_map_support_seq_show(map) &&
- (attr->btf_key_type_id || attr->btf_value_type_id)) {
+ if (attr->btf_key_type_id || attr->btf_value_type_id) {
struct btf *btf;
if (!attr->btf_key_type_id || !attr->btf_value_type_id) {
@@ -504,8 +532,8 @@ static int map_create(union bpf_attr *attr)
goto free_map_nouncharge;
}
- err = map->ops->map_check_btf(map, btf, attr->btf_key_type_id,
- attr->btf_value_type_id);
+ err = map_check_btf(map, btf, attr->btf_key_type_id,
+ attr->btf_value_type_id);
if (err) {
btf_put(btf);
goto free_map_nouncharge;
@@ -684,6 +712,8 @@ static int map_lookup_elem(union bpf_attr *attr)
err = bpf_fd_array_map_lookup_elem(map, key, value);
} else if (IS_FD_HASH(map)) {
err = bpf_fd_htab_map_lookup_elem(map, key, value);
+ } else if (map->map_type == BPF_MAP_TYPE_REUSEPORT_SOCKARRAY) {
+ err = bpf_fd_reuseport_array_lookup_elem(map, key, value);
} else {
rcu_read_lock();
ptr = map->ops->map_lookup_elem(map, key);
@@ -790,6 +820,10 @@ static int map_update_elem(union bpf_attr *attr)
err = bpf_fd_htab_map_update_elem(map, f.file, key, value,
attr->flags);
rcu_read_unlock();
+ } else if (map->map_type == BPF_MAP_TYPE_REUSEPORT_SOCKARRAY) {
+ /* rcu_read_lock() is not needed */
+ err = bpf_fd_reuseport_array_update_elem(map, key, value,
+ attr->flags);
} else {
rcu_read_lock();
err = map->ops->map_update_elem(map, key, value, attr->flags);
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 587468a9c37d..ca90679a7fe5 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -1310,6 +1310,7 @@ static bool may_access_direct_pkt_data(struct bpf_verifier_env *env,
case BPF_PROG_TYPE_LWT_IN:
case BPF_PROG_TYPE_LWT_OUT:
case BPF_PROG_TYPE_LWT_SEG6LOCAL:
+ case BPF_PROG_TYPE_SK_REUSEPORT:
/* dst_input() and dst_output() can't write for now */
if (t == BPF_WRITE)
return false;
@@ -2166,6 +2167,10 @@ static int check_map_func_compatibility(struct bpf_verifier_env *env,
func_id != BPF_FUNC_msg_redirect_hash)
goto error;
break;
+ case BPF_MAP_TYPE_REUSEPORT_SOCKARRAY:
+ if (func_id != BPF_FUNC_sk_select_reuseport)
+ goto error;
+ break;
default:
break;
}
@@ -2217,6 +2222,10 @@ static int check_map_func_compatibility(struct bpf_verifier_env *env,
if (map->map_type != BPF_MAP_TYPE_CGROUP_STORAGE)
goto error;
break;
+ case BPF_FUNC_sk_select_reuseport:
+ if (map->map_type != BPF_MAP_TYPE_REUSEPORT_SOCKARRAY)
+ goto error;
+ break;
default:
break;
}
diff --git a/kernel/bpf/xskmap.c b/kernel/bpf/xskmap.c
index b3c557476a8d..4ddf61e158f6 100644
--- a/kernel/bpf/xskmap.c
+++ b/kernel/bpf/xskmap.c
@@ -227,6 +227,5 @@ const struct bpf_map_ops xsk_map_ops = {
.map_lookup_elem = xsk_map_lookup_elem,
.map_update_elem = xsk_map_update_elem,
.map_delete_elem = xsk_map_delete_elem,
+ .map_check_btf = map_check_no_btf,
};
-
-