summaryrefslogtreecommitdiffstats
path: root/kernel/bpf/arraymap.c
diff options
context:
space:
mode:
authorDaniel Borkmann <daniel@iogearbox.net>2019-11-22 21:07:58 +0100
committerAlexei Starovoitov <ast@kernel.org>2019-11-24 17:04:11 -0800
commitda765a2f599304a81a25e77908d1790414ecdbb6 (patch)
tree73e3edfabd9a026157bfaf4a73764f351be7dcf3 /kernel/bpf/arraymap.c
parenta66886fe6c24ebeeb6dc10fbd9b75158029eacf7 (diff)
downloadlinux-da765a2f599304a81a25e77908d1790414ecdbb6.tar.gz
linux-da765a2f599304a81a25e77908d1790414ecdbb6.tar.bz2
linux-da765a2f599304a81a25e77908d1790414ecdbb6.zip
bpf: Add poke dependency tracking for prog array maps
This work adds program tracking to prog array maps. This is needed such that upon prog array updates/deletions we can fix up all programs which make use of this tail call map. We add ops->map_poke_{un,}track() helpers to maps to maintain the list of programs and ops->map_poke_run() for triggering the actual update. bpf_array_aux is extended to contain the list head and poke_mutex in order to serialize program patching during updates/deletions. bpf_free_used_maps() will untrack the program shortly before dropping the reference to the map. For clearing out the prog array once all urefs are dropped we need to use schedule_work() to have a sleepable context. The prog_array_map_poke_run() is triggered during updates/deletions and walks the maintained prog list. It checks in their poke_tabs whether the map and key is matching and runs the actual bpf_arch_text_poke() for patching in the nop or new jmp location. Depending on the type of update, we use one of BPF_MOD_{NOP_TO_JUMP,JUMP_TO_NOP,JUMP_TO_JUMP}. Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Acked-by: Andrii Nakryiko <andriin@fb.com> Link: https://lore.kernel.org/bpf/1fb364bb3c565b3e415d5ea348f036ff379e779d.1574452833.git.daniel@iogearbox.net
Diffstat (limited to 'kernel/bpf/arraymap.c')
-rw-r--r--kernel/bpf/arraymap.c183
1 files changed, 179 insertions, 4 deletions
diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c
index 57da950ee55b..58bdf5fd24cc 100644
--- a/kernel/bpf/arraymap.c
+++ b/kernel/bpf/arraymap.c
@@ -586,10 +586,17 @@ int bpf_fd_array_map_update_elem(struct bpf_map *map, struct file *map_file,
if (IS_ERR(new_ptr))
return PTR_ERR(new_ptr);
- old_ptr = xchg(array->ptrs + index, new_ptr);
+ if (map->ops->map_poke_run) {
+ mutex_lock(&array->aux->poke_mutex);
+ old_ptr = xchg(array->ptrs + index, new_ptr);
+ map->ops->map_poke_run(map, index, old_ptr, new_ptr);
+ mutex_unlock(&array->aux->poke_mutex);
+ } else {
+ old_ptr = xchg(array->ptrs + index, new_ptr);
+ }
+
if (old_ptr)
map->ops->map_fd_put_ptr(old_ptr);
-
return 0;
}
@@ -602,7 +609,15 @@ static int fd_array_map_delete_elem(struct bpf_map *map, void *key)
if (index >= array->map.max_entries)
return -E2BIG;
- old_ptr = xchg(array->ptrs + index, NULL);
+ if (map->ops->map_poke_run) {
+ mutex_lock(&array->aux->poke_mutex);
+ old_ptr = xchg(array->ptrs + index, NULL);
+ map->ops->map_poke_run(map, index, old_ptr, NULL);
+ mutex_unlock(&array->aux->poke_mutex);
+ } else {
+ old_ptr = xchg(array->ptrs + index, NULL);
+ }
+
if (old_ptr) {
map->ops->map_fd_put_ptr(old_ptr);
return 0;
@@ -671,6 +686,152 @@ static void prog_array_map_seq_show_elem(struct bpf_map *map, void *key,
rcu_read_unlock();
}
+struct prog_poke_elem {
+ struct list_head list;
+ struct bpf_prog_aux *aux;
+};
+
+static int prog_array_map_poke_track(struct bpf_map *map,
+ struct bpf_prog_aux *prog_aux)
+{
+ struct prog_poke_elem *elem;
+ struct bpf_array_aux *aux;
+ int ret = 0;
+
+ aux = container_of(map, struct bpf_array, map)->aux;
+ mutex_lock(&aux->poke_mutex);
+ list_for_each_entry(elem, &aux->poke_progs, list) {
+ if (elem->aux == prog_aux)
+ goto out;
+ }
+
+ elem = kmalloc(sizeof(*elem), GFP_KERNEL);
+ if (!elem) {
+ ret = -ENOMEM;
+ goto out;
+ }
+
+ INIT_LIST_HEAD(&elem->list);
+ /* We must track the program's aux info at this point in time
+ * since the program pointer itself may not be stable yet, see
+ * also comment in prog_array_map_poke_run().
+ */
+ elem->aux = prog_aux;
+
+ list_add_tail(&elem->list, &aux->poke_progs);
+out:
+ mutex_unlock(&aux->poke_mutex);
+ return ret;
+}
+
+static void prog_array_map_poke_untrack(struct bpf_map *map,
+ struct bpf_prog_aux *prog_aux)
+{
+ struct prog_poke_elem *elem, *tmp;
+ struct bpf_array_aux *aux;
+
+ aux = container_of(map, struct bpf_array, map)->aux;
+ mutex_lock(&aux->poke_mutex);
+ list_for_each_entry_safe(elem, tmp, &aux->poke_progs, list) {
+ if (elem->aux == prog_aux) {
+ list_del_init(&elem->list);
+ kfree(elem);
+ break;
+ }
+ }
+ mutex_unlock(&aux->poke_mutex);
+}
+
+static void prog_array_map_poke_run(struct bpf_map *map, u32 key,
+ struct bpf_prog *old,
+ struct bpf_prog *new)
+{
+ enum bpf_text_poke_type type;
+ struct prog_poke_elem *elem;
+ struct bpf_array_aux *aux;
+
+ if (!old && new)
+ type = BPF_MOD_NOP_TO_JUMP;
+ else if (old && !new)
+ type = BPF_MOD_JUMP_TO_NOP;
+ else if (old && new)
+ type = BPF_MOD_JUMP_TO_JUMP;
+ else
+ return;
+
+ aux = container_of(map, struct bpf_array, map)->aux;
+ WARN_ON_ONCE(!mutex_is_locked(&aux->poke_mutex));
+
+ list_for_each_entry(elem, &aux->poke_progs, list) {
+ struct bpf_jit_poke_descriptor *poke;
+ int i, ret;
+
+ for (i = 0; i < elem->aux->size_poke_tab; i++) {
+ poke = &elem->aux->poke_tab[i];
+
+ /* Few things to be aware of:
+ *
+ * 1) We can only ever access aux in this context, but
+ * not aux->prog since it might not be stable yet and
+ * there could be danger of use after free otherwise.
+ * 2) Initially when we start tracking aux, the program
+ * is not JITed yet and also does not have a kallsyms
+ * entry. We skip these as poke->ip_stable is not
+ * active yet. The JIT will do the final fixup before
+ * setting it stable. The various poke->ip_stable are
+ * successively activated, so tail call updates can
+ * arrive from here while JIT is still finishing its
+ * final fixup for non-activated poke entries.
+ * 3) On program teardown, the program's kallsym entry gets
+ * removed out of RCU callback, but we can only untrack
+ * from sleepable context, therefore bpf_arch_text_poke()
+ * might not see that this is in BPF text section and
+ * bails out with -EINVAL. As these are unreachable since
+ * RCU grace period already passed, we simply skip them.
+ * 4) Also programs reaching refcount of zero while patching
+ * is in progress is okay since we're protected under
+ * poke_mutex and untrack the programs before the JIT
+ * buffer is freed. When we're still in the middle of
+ * patching and suddenly kallsyms entry of the program
+ * gets evicted, we just skip the rest which is fine due
+ * to point 3).
+ * 5) Any other error happening below from bpf_arch_text_poke()
+ * is a unexpected bug.
+ */
+ if (!READ_ONCE(poke->ip_stable))
+ continue;
+ if (poke->reason != BPF_POKE_REASON_TAIL_CALL)
+ continue;
+ if (poke->tail_call.map != map ||
+ poke->tail_call.key != key)
+ continue;
+
+ ret = bpf_arch_text_poke(poke->ip, type,
+ old ? (u8 *)old->bpf_func +
+ poke->adj_off : NULL,
+ new ? (u8 *)new->bpf_func +
+ poke->adj_off : NULL);
+ BUG_ON(ret < 0 && ret != -EINVAL);
+ }
+ }
+}
+
+static void prog_array_map_clear_deferred(struct work_struct *work)
+{
+ struct bpf_map *map = container_of(work, struct bpf_array_aux,
+ work)->map;
+ bpf_fd_array_map_clear(map);
+ bpf_map_put(map);
+}
+
+static void prog_array_map_clear(struct bpf_map *map)
+{
+ struct bpf_array_aux *aux = container_of(map, struct bpf_array,
+ map)->aux;
+ bpf_map_inc(map);
+ schedule_work(&aux->work);
+}
+
static struct bpf_map *prog_array_map_alloc(union bpf_attr *attr)
{
struct bpf_array_aux *aux;
@@ -680,6 +841,10 @@ static struct bpf_map *prog_array_map_alloc(union bpf_attr *attr)
if (!aux)
return ERR_PTR(-ENOMEM);
+ INIT_WORK(&aux->work, prog_array_map_clear_deferred);
+ INIT_LIST_HEAD(&aux->poke_progs);
+ mutex_init(&aux->poke_mutex);
+
map = array_map_alloc(attr);
if (IS_ERR(map)) {
kfree(aux);
@@ -687,14 +852,21 @@ static struct bpf_map *prog_array_map_alloc(union bpf_attr *attr)
}
container_of(map, struct bpf_array, map)->aux = aux;
+ aux->map = map;
+
return map;
}
static void prog_array_map_free(struct bpf_map *map)
{
+ struct prog_poke_elem *elem, *tmp;
struct bpf_array_aux *aux;
aux = container_of(map, struct bpf_array, map)->aux;
+ list_for_each_entry_safe(elem, tmp, &aux->poke_progs, list) {
+ list_del_init(&elem->list);
+ kfree(elem);
+ }
kfree(aux);
fd_array_map_free(map);
}
@@ -703,13 +875,16 @@ const struct bpf_map_ops prog_array_map_ops = {
.map_alloc_check = fd_array_map_alloc_check,
.map_alloc = prog_array_map_alloc,
.map_free = prog_array_map_free,
+ .map_poke_track = prog_array_map_poke_track,
+ .map_poke_untrack = prog_array_map_poke_untrack,
+ .map_poke_run = prog_array_map_poke_run,
.map_get_next_key = array_map_get_next_key,
.map_lookup_elem = fd_array_map_lookup_elem,
.map_delete_elem = fd_array_map_delete_elem,
.map_fd_get_ptr = prog_fd_array_get_ptr,
.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_release_uref = prog_array_map_clear,
.map_seq_show_elem = prog_array_map_seq_show_elem,
};