On Mon, Oct 10, 2022 at 4:32 AM Donald Hunter <donald.hunter@xxxxxxxxx> wrote: > > Add documentation for the ARRAY_OF_MAPS and HASH_OF_MAPS map types, > including usage and examples. > > Signed-off-by: Donald Hunter <donald.hunter@xxxxxxxxx> > --- subject suggestion (as it's pretty long): bpf, docs: document BPF_MAP_TYPE_{ARRAY,HASH}_OF_MAPS > Documentation/bpf/map_of_maps.rst | 145 ++++++++++++++++++++++++++++++ > 1 file changed, 145 insertions(+) > create mode 100644 Documentation/bpf/map_of_maps.rst > > diff --git a/Documentation/bpf/map_of_maps.rst b/Documentation/bpf/map_of_maps.rst > new file mode 100644 > index 000000000000..16fcda8720de > --- /dev/null > +++ b/Documentation/bpf/map_of_maps.rst > @@ -0,0 +1,145 @@ > +.. SPDX-License-Identifier: GPL-2.0-only > +.. Copyright (C) 2022 Red Hat, Inc. > + > +======================================================== > +BPF_MAP_TYPE_ARRAY_OF_MAPS and BPF_MAP_TYPE_HASH_OF_MAPS > +======================================================== > + > +.. note:: > + - ``BPF_MAP_TYPE_ARRAY_OF_MAPS`` and ``BPF_MAP_TYPE_HASH_OF_MAPS`` were > + introduced in kernel version 4.12. > + > +``BPF_MAP_TYPE_ARRAY_OF_MAPS`` and ``BPF_MAP_TYPE_HASH_OF_MAPS`` provide general > +purpose support for map in map storage. One level of nesting is supported, where > +an outer map contains instances of a single type of inner map, for example > +``array_of_maps->sock_map``. > + > +When creating an outer map, an inner map instance is used to initialize the > +metadata that the outer map holds about its inner maps. This inner map has a > +separate lifetime from the outer map and can be deleted after the outer map has > +been created. > + > +The outer map supports element update and delete from user space using the > +syscall API. A BPF program is only allowed to do element lookup in the outer > +map. > + > +.. note:: > + - Multi-level nesting is not supported. > + - Any BPF map type can be used as an inner map, except for > + ``BPF_MAP_TYPE_PROG_ARRAY``. > + - A BPF program cannot update or delete outer map entries. > + > +Array of Maps > +------------- > + > +For ``BPF_MAP_TYPE_ARRAY_OF_MAPS`` the key is an unsigned 32-bit integer index > +into the array. The array is a fixed size with `max_entries` elements that are > +zero initialized when created. > + > +Hash of Maps > +------------ > + > +For ``BPF_MAP_TYPE_HASH_OF_MAPS`` the key type can be chosen when defining the > +map. > + > +The kernel is responsible for allocating and freeing key/value pairs, up > +to the max_entries limit that you specify. Hash maps use pre-allocation > +of hash table elements by default. The ``BPF_F_NO_PREALLOC`` flag can be > +used to disable pre-allocation when it is too memory expensive. > + > +Usage > +===== > + > +.. c:function:: > + void *bpf_map_lookup_elem(struct bpf_map *map, const void *key) > + > +Inner maps can be retrieved using the ``bpf_map_lookup_elem()`` helper. This > +helper returns a pointer to the inner map, or ``NULL`` if no entry was found. > + > +Examples > +======== > + > +Kernel BPF > +---------- > + > +This snippet shows how to create an array of devmaps in a BPF program. Note that > +the outer array can only be modified from user space using the syscall API. > + > +.. code-block:: c > + > + struct redirect_map { > + __uint(type, BPF_MAP_TYPE_DEVMAP); > + __uint(max_entries, 32); > + __type(key, enum skb_drop_reason); > + __type(value, __u64); > + } redirect_map SEC(".maps"); > + > + struct { > + __uint(type, BPF_MAP_TYPE_ARRAY_OF_MAPS); > + __uint(max_entries, 2); > + __uint(key_size, sizeof(int)); > + __uint(value_size, sizeof(int)); > + __array(values, struct redirect_map); > + } outer_map SEC(".maps"); > + Let's also demonstrate libbpf's declarative way to initialize entries in outer map? See progs/test_btf_map_in_map.c under selftests/bpf for various examples. > +This snippet shows how to lookup an outer map to retrieve an inner map. > + > +.. code-block:: c > + > + SEC("xdp") > + int redirect_by_priority(struct xdp_md *ctx) { > + struct bpf_map *devmap; > + int action = XDP_PASS; > + int index = 0; > + > + devmap = bpf_map_lookup_elem(&outer_arr, &index); > + if (!devmap) > + return XDP_PASS; > + > + /* use inner devmap here */ > + > + return action; > + } > + > +User Space > +---------- > + > +This snippet shows how to create an array based outer map: > + > +.. code-block:: c > + > + int create_outer_array(int inner_fd) { > + int fd; > + LIBBPF_OPTS(bpf_map_create_opts, opts); > + opts.inner_map_fd = inner_fd; LIBBPF_OPTS(bpf_map_create_opts, opts, .inner_map_fd = inner_fd); > + fd = bpf_map_create(BPF_MAP_TYPE_ARRAY_OF_MAPS, > + "example_array", /* name */ > + sizeof(__u32), /* key size */ > + sizeof(__u32), /* value size */ > + 256, /* max entries */ > + &opts); /* create opts */ > + return fd; > + } > + > + > +This snippet shows how to add an inner map to an outer map: > + > +.. code-block:: c > + > + int add_devmap(int outer_fd, int index, const char *name) { > + int fd, ret; > + > + fd = bpf_map_create(BPF_MAP_TYPE_DEVMAP, name, > + sizeof(__u32), sizeof(__u32), 256, NULL); > + if (fd < 0) > + return fd; > + > + ret = bpf_map_update_elem(outer_fd, &index, &fd, BPF_NOEXIST); > + return ret; > + } > + > +References > +========== > + > +- https://lore.kernel.org/netdev/20170322170035.923581-3-kafai@xxxxxx/ > +- https://lore.kernel.org/netdev/20170322170035.923581-4-kafai@xxxxxx/ > -- > 2.35.1 >