tools: bpftool: Add "inner_map" to "bpftool map create" outer maps

There is no support for creating maps of types array-of-map or
hash-of-map in bpftool. This is because the kernel needs an inner_map_fd
to collect metadata on the inner maps to be supported by the new map,
but bpftool does not provide a way to pass this file descriptor.

Add a new optional "inner_map" keyword that can be used to pass a
reference to a map, retrieve a fd to that map, and pass it as the
inner_map_fd.

Add related documentation and bash completion. Note that we can
reference the inner map by its name, meaning we can have several times
the keyword "name" with different meanings (mandatory outer map name,
and possibly a name to use to find the inner_map_fd). The bash
completion will offer it just once, and will not suggest "name" on the
following command:

    # bpftool map create /sys/fs/bpf/my_outer_map type hash_of_maps \
        inner_map name my_inner_map [TAB]

Fixing that specific case seems too convoluted. Completion will work as
expected, however, if the outer map name comes first and the "inner_map
name ..." is passed second.

Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20200910102652.10509-4-quentin@isovalent.com

diff --git a/tools/bpf/bpftool/Documentation/bpftool-map.rst b/tools/bpf/bpftool/Documentation/bpftool-map.rst
index 4b42629ade3e1ddcab5c05ddce5a83d730975417..8eac254ade48abea4c0953f2aa683107407fa926 100644 (file)
--- a/tools/bpf/bpftool/Documentation/bpftool-map.rst
+++ b/tools/bpf/bpftool/Documentation/bpftool-map.rst
@@ -23,7 +23,8 @@ MAP COMMANDS
 
 |      **bpftool** **map** { **show** | **list** }   [*MAP*]
 |      **bpftool** **map create**     *FILE* **type** *TYPE* **key** *KEY_SIZE* **value** *VALUE_SIZE* \
-|              **entries** *MAX_ENTRIES* **name** *NAME* [**flags** *FLAGS*] [**dev** *NAME*]
+|              **entries** *MAX_ENTRIES* **name** *NAME* [**flags** *FLAGS*] [**inner_map** *MAP*] \
+|              [**dev** *NAME*]
 |      **bpftool** **map dump**       *MAP*
 |      **bpftool** **map update**     *MAP* [**key** *DATA*] [**value** *VALUE*] [*UPDATE_FLAGS*]
 |      **bpftool** **map lookup**     *MAP* [**key** *DATA*]
@@ -67,7 +68,7 @@ DESCRIPTION
                  maps. On such kernels bpftool will automatically emit this
                  information as well.
 
-       **bpftool map create** *FILE* **type** *TYPE* **key** *KEY_SIZE* **value** *VALUE_SIZE*  **entries** *MAX_ENTRIES* **name** *NAME* [**flags** *FLAGS*] [**dev** *NAME*]
+       **bpftool map create** *FILE* **type** *TYPE* **key** *KEY_SIZE* **value** *VALUE_SIZE*  **entries** *MAX_ENTRIES* **name** *NAME* [**flags** *FLAGS*] [**inner_map** *MAP*] [**dev** *NAME*]
                  Create a new map with given parameters and pin it to *bpffs*
                  as *FILE*.
 
@@ -75,6 +76,11 @@ DESCRIPTION
                  desired flags, e.g. 1024 for **BPF_F_MMAPABLE** (see bpf.h
                  UAPI header for existing flags).
 
+                 To create maps of type array-of-maps or hash-of-maps, the
+                 **inner_map** keyword must be used to pass an inner map. The
+                 kernel needs it to collect metadata related to the inner maps
+                 that the new map will work with.
+
                  Keyword **dev** expects a network interface name, and is used
                  to request hardware offload for the map.
 

diff --git a/tools/bpf/bpftool/bash-completion/bpftool b/tools/bpf/bpftool/bash-completion/bpftool
index 7b68e3c0a5fbfdc12251e1cce95527f5ebf63acb..3f1da30c4da6e7b5a39001044186581d4de54f37 100644 (file)
--- a/tools/bpf/bpftool/bash-completion/bpftool
+++ b/tools/bpf/bpftool/bash-completion/bpftool
@@ -709,9 +709,26 @@ _bpftool()
                                                    "$cur" ) )
                             return 0
                             ;;
-                        key|value|flags|name|entries)
+                        key|value|flags|entries)
                             return 0
                             ;;
+                        inner_map)
+                            COMPREPLY=( $( compgen -W "$MAP_TYPE" -- "$cur" ) )
+                            return 0
+                            ;;
+                        id)
+                            _bpftool_get_map_ids
+                            ;;
+                        name)
+                            case $pprev in
+                                inner_map)
+                                    _bpftool_get_map_names
+                                    ;;
+                                *)
+                                    return 0
+                                    ;;
+                            esac
+                            ;;
                         *)
                             _bpftool_once_attr 'type'
                             _bpftool_once_attr 'key'
@@ -719,6 +736,9 @@ _bpftool()
                             _bpftool_once_attr 'entries'
                             _bpftool_once_attr 'name'
                             _bpftool_once_attr 'flags'
+                            if _bpftool_search_list 'array_of_maps' 'hash_of_maps'; then
+                                _bpftool_once_attr 'inner_map'
+                            fi
                             _bpftool_once_attr 'dev'
                             return 0
                             ;;

diff --git a/tools/bpf/bpftool/map.c b/tools/bpf/bpftool/map.c
index d8581d5e98a12f2234dd1394c53ce160470f43d0..a7efbd84fbcc44310f25d2d2139a999fca7b3eea 100644 (file)
--- a/tools/bpf/bpftool/map.c
+++ b/tools/bpf/bpftool/map.c
@@ -1250,7 +1250,7 @@ static int do_create(int argc, char **argv)
 {
        struct bpf_create_map_attr attr = { NULL, };
        const char *pinfile;
-       int err, fd;
+       int err = -1, fd;
 
        if (!REQ_ARGS(7))
                return -1;
@@ -1265,13 +1265,13 @@ static int do_create(int argc, char **argv)
 
                        if (attr.map_type) {
                                p_err("map type already specified");
-                               return -1;
+                               goto exit;
                        }
 
                        attr.map_type = map_type_from_str(*argv);
                        if ((int)attr.map_type < 0) {
                                p_err("unrecognized map type: %s", *argv);
-                               return -1;
+                               goto exit;
                        }
                        NEXT_ARG();
                } else if (is_prefix(*argv, "name")) {
@@ -1280,43 +1280,56 @@ static int do_create(int argc, char **argv)
                } else if (is_prefix(*argv, "key")) {
                        if (parse_u32_arg(&argc, &argv, &attr.key_size,
                                          "key size"))
-                               return -1;
+                               goto exit;
                } else if (is_prefix(*argv, "value")) {
                        if (parse_u32_arg(&argc, &argv, &attr.value_size,
                                          "value size"))
-                               return -1;
+                               goto exit;
                } else if (is_prefix(*argv, "entries")) {
                        if (parse_u32_arg(&argc, &argv, &attr.max_entries,
                                          "max entries"))
-                               return -1;
+                               goto exit;
                } else if (is_prefix(*argv, "flags")) {
                        if (parse_u32_arg(&argc, &argv, &attr.map_flags,
                                          "flags"))
-                               return -1;
+                               goto exit;
                } else if (is_prefix(*argv, "dev")) {
                        NEXT_ARG();
 
                        if (attr.map_ifindex) {
                                p_err("offload device already specified");
-                               return -1;
+                               goto exit;
                        }
 
                        attr.map_ifindex = if_nametoindex(*argv);
                        if (!attr.map_ifindex) {
                                p_err("unrecognized netdevice '%s': %s",
                                      *argv, strerror(errno));
-                               return -1;
+                               goto exit;
                        }
                        NEXT_ARG();
+               } else if (is_prefix(*argv, "inner_map")) {
+                       struct bpf_map_info info = {};
+                       __u32 len = sizeof(info);
+                       int inner_map_fd;
+
+                       NEXT_ARG();
+                       if (!REQ_ARGS(2))
+                               usage();
+                       inner_map_fd = map_parse_fd_and_info(&argc, &argv,
+                                                            &info, &len);
+                       if (inner_map_fd < 0)
+                               return -1;
+                       attr.inner_map_fd = inner_map_fd;
                } else {
                        p_err("unknown arg %s", *argv);
-                       return -1;
+                       goto exit;
                }
        }
 
        if (!attr.name) {
                p_err("map name not specified");
-               return -1;
+               goto exit;
        }
 
        set_max_rlimit();
@@ -1324,17 +1337,22 @@ static int do_create(int argc, char **argv)
        fd = bpf_create_map_xattr(&attr);
        if (fd < 0) {
                p_err("map create failed: %s", strerror(errno));
-               return -1;
+               goto exit;
        }
 
        err = do_pin_fd(fd, pinfile);
        close(fd);
        if (err)
-               return err;
+               goto exit;
 
        if (json_output)
                jsonw_null(json_wtr);
-       return 0;
+
+exit:
+       if (attr.inner_map_fd > 0)
+               close(attr.inner_map_fd);
+
+       return err;
 }
 
 static int do_pop_dequeue(int argc, char **argv)
@@ -1420,7 +1438,7 @@ static int do_help(int argc, char **argv)
                "Usage: %1$s %2$s { show | list }   [MAP]\n"
                "       %1$s %2$s create     FILE type TYPE key KEY_SIZE value VALUE_SIZE \\\n"
                "                                  entries MAX_ENTRIES name NAME [flags FLAGS] \\\n"
-               "                                  [dev NAME]\n"
+               "                                  [inner_map MAP] [dev NAME]\n"
                "       %1$s %2$s dump       MAP\n"
                "       %1$s %2$s update     MAP [key DATA] [value VALUE] [UPDATE_FLAGS]\n"
                "       %1$s %2$s lookup     MAP [key DATA]\n"