[PATCH v3 bpf-next 2/3] bpf: Relax max_entries check for most of the inner map types
From: Martin KaFai Lau <hidden>
Date: 2020-08-28 01:18:17
Also in:
bpf
Subsystem:
bpf [general] (safe dynamic programs and tools), networking [general], the rest, xdp sockets (af_xdp) · Maintainers:
Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko, Eduard Zingerman, Kumar Kartikeya Dwivedi, "David S. Miller", Eric Dumazet, Jakub Kicinski, Paolo Abeni, Linus Torvalds, Magnus Karlsson, Maciej Fijalkowski
Most of the maps do not use max_entries during verification time. Thus, those map_meta_equal() do not need to enforce max_entries when it is inserted as an inner map during runtime. The max_entries check is removed from the default implementation bpf_map_meta_equal(). The prog_array_map and xsk_map are exception. Its map_gen_lookup uses max_entries to generate inline lookup code. Thus, they will implement its own map_meta_equal() to enforce max_entries. Since there are only two cases now, the max_entries check is not refactored and stays in its own .c file. Signed-off-by: Martin KaFai Lau <redacted> --- kernel/bpf/arraymap.c | 9 ++++++++- kernel/bpf/map_in_map.c | 3 +-- net/xdp/xskmap.c | 9 ++++++++- 3 files changed, 17 insertions(+), 4 deletions(-)
diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c
index 40d1f7f94307..d851ebbcf302 100644
--- a/kernel/bpf/arraymap.c
+++ b/kernel/bpf/arraymap.c@@ -487,6 +487,13 @@ static int array_map_mmap(struct bpf_map *map, struct vm_area_struct *vma) vma->vm_pgoff + pgoff); } +static bool array_map_meta_equal(const struct bpf_map *meta0, + const struct bpf_map *meta1) +{ + return meta0->max_entries == meta1->max_entries && + bpf_map_meta_equal(meta0, meta1); +} + struct bpf_iter_seq_array_map_info { struct bpf_map *map; void *percpu_value_buf;
@@ -625,7 +632,7 @@ static const struct bpf_iter_seq_info iter_seq_info = { static int array_map_btf_id; const struct bpf_map_ops array_map_ops = { - .map_meta_equal = bpf_map_meta_equal, + .map_meta_equal = array_map_meta_equal, .map_alloc_check = array_map_alloc_check, .map_alloc = array_map_alloc, .map_free = array_map_free,
diff --git a/kernel/bpf/map_in_map.c b/kernel/bpf/map_in_map.c
index e97a22dd3232..39ab0b68cade 100644
--- a/kernel/bpf/map_in_map.c
+++ b/kernel/bpf/map_in_map.c@@ -75,8 +75,7 @@ bool bpf_map_meta_equal(const struct bpf_map *meta0, return meta0->map_type == meta1->map_type && meta0->key_size == meta1->key_size && meta0->value_size == meta1->value_size && - meta0->map_flags == meta1->map_flags && - meta0->max_entries == meta1->max_entries; + meta0->map_flags == meta1->map_flags; } void *bpf_map_fd_get_ptr(struct bpf_map *map,
diff --git a/net/xdp/xskmap.c b/net/xdp/xskmap.c
index f45f29f04151..2a4fd6677155 100644
--- a/net/xdp/xskmap.c
+++ b/net/xdp/xskmap.c@@ -254,9 +254,16 @@ void xsk_map_try_sock_delete(struct xsk_map *map, struct xdp_sock *xs, spin_unlock_bh(&map->lock); } +static bool xsk_map_meta_equal(const struct bpf_map *meta0, + const struct bpf_map *meta1) +{ + return meta0->max_entries == meta1->max_entries && + bpf_map_meta_equal(meta0, meta1); +} + static int xsk_map_btf_id; const struct bpf_map_ops xsk_map_ops = { - .map_meta_equal = bpf_map_meta_equal, + .map_meta_equal = xsk_map_meta_equal, .map_alloc = xsk_map_alloc, .map_free = xsk_map_free, .map_get_next_key = xsk_map_get_next_key,
--
2.24.1