From 69b693f0aefa0ed521e8bd02260523b5ae446ad7 Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Wed, 18 Apr 2018 15:55:57 -0700 Subject: bpf: btf: Introduce BPF Type Format (BTF) This patch introduces BPF type Format (BTF). BTF (BPF Type Format) is the meta data format which describes the data types of BPF program/map. Hence, it basically focus on the C programming language which the modern BPF is primary using. The first use case is to provide a generic pretty print capability for a BPF map. BTF has its root from CTF (Compact C-Type format). To simplify the handling of BTF data, BTF removes the differences between small and big type/struct-member. Hence, BTF consistently uses u32 instead of supporting both "one u16" and "two u32 (+padding)" in describing type and struct-member. It also raises the number of types (and functions) limit from 0x7fff to 0x7fffffff. Due to the above changes, the format is not compatible to CTF. Hence, BTF starts with a new BTF_MAGIC and version number. This patch does the first verification pass to the BTF. The first pass checks: 1. meta-data size (e.g. It does not go beyond the total btf's size) 2. name_offset is valid 3. Each BTF_KIND (e.g. int, enum, struct....) does its own check of its meta-data. Some other checks, like checking a struct's member is referring to a valid type, can only be done in the second pass. The second verification pass will be implemented in the next patch. Signed-off-by: Martin KaFai Lau Acked-by: Alexei Starovoitov Signed-off-by: Daniel Borkmann --- include/uapi/linux/btf.h | 130 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 130 insertions(+) create mode 100644 include/uapi/linux/btf.h (limited to 'include/uapi/linux') diff --git a/include/uapi/linux/btf.h b/include/uapi/linux/btf.h new file mode 100644 index 000000000000..74a30b1090df --- /dev/null +++ b/include/uapi/linux/btf.h @@ -0,0 +1,130 @@ +/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ +/* Copyright (c) 2018 Facebook */ +#ifndef _UAPI__LINUX_BTF_H__ +#define _UAPI__LINUX_BTF_H__ + +#include + +#define BTF_MAGIC 0xeB9F +#define BTF_MAGIC_SWAP 0x9FeB +#define BTF_VERSION 1 +#define BTF_FLAGS_COMPR 0x01 + +struct btf_header { + __u16 magic; + __u8 version; + __u8 flags; + + __u32 parent_label; + __u32 parent_name; + + /* All offsets are in bytes relative to the end of this header */ + __u32 label_off; /* offset of label section */ + __u32 object_off; /* offset of data object section*/ + __u32 func_off; /* offset of function section */ + __u32 type_off; /* offset of type section */ + __u32 str_off; /* offset of string section */ + __u32 str_len; /* length of string section */ +}; + +/* Max # of type identifier */ +#define BTF_MAX_TYPE 0x7fffffff +/* Max offset into the string section */ +#define BTF_MAX_NAME_OFFSET 0x7fffffff +/* Max # of struct/union/enum members or func args */ +#define BTF_MAX_VLEN 0xffff + +/* The type id is referring to a parent BTF */ +#define BTF_TYPE_PARENT(id) (((id) >> 31) & 0x1) +#define BTF_TYPE_ID(id) ((id) & BTF_MAX_TYPE) + +/* String is in the ELF string section */ +#define BTF_STR_TBL_ELF_ID(ref) (((ref) >> 31) & 0x1) +#define BTF_STR_OFFSET(ref) ((ref) & BTF_MAX_NAME_OFFSET) + +struct btf_type { + __u32 name; + /* "info" bits arrangement + * bits 0-15: vlen (e.g. # of struct's members) + * bits 16-23: unused + * bits 24-28: kind (e.g. int, ptr, array...etc) + * bits 29-30: unused + * bits 31: root + */ + __u32 info; + /* "size" is used by INT, ENUM, STRUCT and UNION. + * "size" tells the size of the type it is describing. + * + * "type" is used by PTR, TYPEDEF, VOLATILE, CONST and RESTRICT. + * "type" is a type_id referring to another type. + */ + union { + __u32 size; + __u32 type; + }; +}; + +#define BTF_INFO_KIND(info) (((info) >> 24) & 0x1f) +#define BTF_INFO_ISROOT(info) (!!(((info) >> 24) & 0x80)) +#define BTF_INFO_VLEN(info) ((info) & 0xffff) + +#define BTF_KIND_UNKN 0 /* Unknown */ +#define BTF_KIND_INT 1 /* Integer */ +#define BTF_KIND_PTR 2 /* Pointer */ +#define BTF_KIND_ARRAY 3 /* Array */ +#define BTF_KIND_STRUCT 4 /* Struct */ +#define BTF_KIND_UNION 5 /* Union */ +#define BTF_KIND_ENUM 6 /* Enumeration */ +#define BTF_KIND_FWD 7 /* Forward */ +#define BTF_KIND_TYPEDEF 8 /* Typedef */ +#define BTF_KIND_VOLATILE 9 /* Volatile */ +#define BTF_KIND_CONST 10 /* Const */ +#define BTF_KIND_RESTRICT 11 /* Restrict */ +#define BTF_KIND_MAX 11 +#define NR_BTF_KINDS 12 + +/* For some specific BTF_KIND, "struct btf_type" is immediately + * followed by extra data. + */ + +/* BTF_KIND_INT is followed by a u32 and the following + * is the 32 bits arrangement: + */ +#define BTF_INT_ENCODING(VAL) (((VAL) & 0xff000000) >> 24) +#define BTF_INT_OFFSET(VAL) (((VAL & 0x00ff0000)) >> 16) +#define BTF_INT_BITS(VAL) ((VAL) & 0x0000ffff) + +/* Attributes stored in the BTF_INT_ENCODING */ +#define BTF_INT_SIGNED 0x1 +#define BTF_INT_CHAR 0x2 +#define BTF_INT_BOOL 0x4 +#define BTF_INT_VARARGS 0x8 + +/* BTF_KIND_ENUM is followed by multiple "struct btf_enum". + * The exact number of btf_enum is stored in the vlen (of the + * info in "struct btf_type"). + */ +struct btf_enum { + __u32 name; + __s32 val; +}; + +/* BTF_KIND_ARRAY is followed by one "struct btf_array" */ +struct btf_array { + __u32 type; + __u32 index_type; + __u32 nelems; +}; + +/* BTF_KIND_STRUCT and BTF_KIND_UNION are followed + * by multiple "struct btf_member". The exact number + * of btf_member is stored in the vlen (of the info in + * "struct btf_type"). + */ +struct btf_member { + __u32 name; + __u32 type; + __u32 offset; /* offset in bits */ +}; + +#endif /* _UAPI__LINUX_BTF_H__ */ -- cgit v1.3-14-g43fede From f56a653c1fd13a197076dec4461c656fd2adec73 Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Wed, 18 Apr 2018 15:56:01 -0700 Subject: bpf: btf: Add BPF_BTF_LOAD command This patch adds a BPF_BTF_LOAD command which 1) loads and verifies the BTF (implemented in earlier patches) 2) returns a BTF fd to userspace. In the next patch, the BTF fd can be specified during BPF_MAP_CREATE. It currently limits to CAP_SYS_ADMIN. Signed-off-by: Martin KaFai Lau Acked-by: Alexei Starovoitov Signed-off-by: Daniel Borkmann --- include/linux/btf.h | 4 +++ include/uapi/linux/bpf.h | 9 +++++++ kernel/bpf/btf.c | 67 ++++++++++++++++++++++++++++++++++++++++++++++++ kernel/bpf/syscall.c | 17 ++++++++++++ 4 files changed, 97 insertions(+) (limited to 'include/uapi/linux') diff --git a/include/linux/btf.h b/include/linux/btf.h index d8bdab0280ba..a7c7072535ea 100644 --- a/include/linux/btf.h +++ b/include/linux/btf.h @@ -8,7 +8,11 @@ struct btf; struct btf_type; +union bpf_attr; +void btf_put(struct btf *btf); +int btf_new_fd(const union bpf_attr *attr); +struct btf *btf_get_by_fd(int fd); /* Figure out the size of a type_id. If type_id is a modifier * (e.g. const), it will be resolved to find out the type with size. * diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 9a2d1a04eb24..795bcd577750 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -95,6 +95,7 @@ enum bpf_cmd { BPF_OBJ_GET_INFO_BY_FD, BPF_PROG_QUERY, BPF_RAW_TRACEPOINT_OPEN, + BPF_BTF_LOAD, }; enum bpf_map_type { @@ -363,6 +364,14 @@ union bpf_attr { __u64 name; __u32 prog_fd; } raw_tracepoint; + + struct { /* anonymous struct for BPF_BTF_LOAD */ + __aligned_u64 btf; + __aligned_u64 btf_log_buf; + __u32 btf_size; + __u32 btf_log_size; + __u32 btf_log_level; + }; } __attribute__((aligned(8))); /* BPF helper function descriptions: diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c index 10ee41589da2..2322340694cf 100644 --- a/kernel/bpf/btf.c +++ b/kernel/bpf/btf.c @@ -7,6 +7,8 @@ #include #include #include +#include +#include #include #include #include @@ -190,6 +192,7 @@ struct btf { u32 nr_types; u32 types_size; u32 data_size; + refcount_t refcnt; }; enum verifier_phase { @@ -604,6 +607,17 @@ static void btf_free(struct btf *btf) kfree(btf); } +static void btf_get(struct btf *btf) +{ + refcount_inc(&btf->refcnt); +} + +void btf_put(struct btf *btf) +{ + if (btf && refcount_dec_and_test(&btf->refcnt)) + btf_free(btf); +} + static int env_resolve_init(struct btf_verifier_env *env) { struct btf *btf = env->btf; @@ -1963,6 +1977,7 @@ static struct btf *btf_parse(void __user *btf_data, u32 btf_data_size, if (!err) { btf_verifier_env_free(env); + btf_get(btf); return btf; } @@ -1980,3 +1995,55 @@ void btf_type_seq_show(const struct btf *btf, u32 type_id, void *obj, btf_type_ops(t)->seq_show(btf, t, type_id, obj, 0, m); } + +static int btf_release(struct inode *inode, struct file *filp) +{ + btf_put(filp->private_data); + return 0; +} + +static const struct file_operations btf_fops = { + .release = btf_release, +}; + +int btf_new_fd(const union bpf_attr *attr) +{ + struct btf *btf; + int fd; + + btf = btf_parse(u64_to_user_ptr(attr->btf), + attr->btf_size, attr->btf_log_level, + u64_to_user_ptr(attr->btf_log_buf), + attr->btf_log_size); + if (IS_ERR(btf)) + return PTR_ERR(btf); + + fd = anon_inode_getfd("btf", &btf_fops, btf, + O_RDONLY | O_CLOEXEC); + if (fd < 0) + btf_put(btf); + + return fd; +} + +struct btf *btf_get_by_fd(int fd) +{ + struct btf *btf; + struct fd f; + + f = fdget(fd); + + if (!f.file) + return ERR_PTR(-EBADF); + + if (f.file->f_op != &btf_fops) { + fdput(f); + return ERR_PTR(-EINVAL); + } + + btf = f.file->private_data; + btf_get(btf); + fdput(f); + + return btf; +} diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index 4ca46df19c9a..cd8ebadc66eb 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -11,6 +11,7 @@ */ #include #include +#include #include #include #include @@ -2023,6 +2024,19 @@ static int bpf_obj_get_info_by_fd(const union bpf_attr *attr, return err; } +#define BPF_BTF_LOAD_LAST_FIELD btf_log_level + +static int bpf_btf_load(const union bpf_attr *attr) +{ + if (CHECK_ATTR(BPF_BTF_LOAD)) + return -EINVAL; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + return btf_new_fd(attr); +} + SYSCALL_DEFINE3(bpf, int, cmd, union bpf_attr __user *, uattr, unsigned int, size) { union bpf_attr attr = {}; @@ -2103,6 +2117,9 @@ SYSCALL_DEFINE3(bpf, int, cmd, union bpf_attr __user *, uattr, unsigned int, siz case BPF_RAW_TRACEPOINT_OPEN: err = bpf_raw_tracepoint_open(&attr); break; + case BPF_BTF_LOAD: + err = bpf_btf_load(&attr); + break; default: err = -EINVAL; break; -- cgit v1.3-14-g43fede From a26ca7c982cb576749cbdd01e8ecde4bf010d60a Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Wed, 18 Apr 2018 15:56:03 -0700 Subject: bpf: btf: Add pretty print support to the basic arraymap This patch adds pretty print support to the basic arraymap. Support for other bpf maps can be added later. This patch adds new attrs to the BPF_MAP_CREATE command to allow specifying the btf_fd, btf_key_id and btf_value_id. The BPF_MAP_CREATE can then associate the btf to the map if the creating map supports BTF. A BTF supported map needs to implement two new map ops, map_seq_show_elem() and map_check_btf(). This patch has implemented these new map ops for the basic arraymap. It also adds file_operations, bpffs_map_fops, to the pinned map such that the pinned map can be opened and read. After that, the user has an intuitive way to do "cat bpffs/pathto/a-pinned-map" instead of getting an error. bpffs_map_fops should not be extended further to support other operations. Other operations (e.g. write/key-lookup...) should be realized by the userspace tools (e.g. bpftool) through the BPF_OBJ_GET_INFO_BY_FD, map's lookup/update interface...etc. Follow up patches will allow the userspace to obtain the BTF from a map-fd. Here is a sample output when reading a pinned arraymap with the following map's value: struct map_value { int count_a; int count_b; }; cat /sys/fs/bpf/pinned_array_map: 0: {1,2} 1: {3,4} 2: {5,6} ... Signed-off-by: Martin KaFai Lau Acked-by: Alexei Starovoitov Signed-off-by: Daniel Borkmann --- include/linux/bpf.h | 20 +++++- include/uapi/linux/bpf.h | 3 + kernel/bpf/arraymap.c | 50 +++++++++++++++ kernel/bpf/inode.c | 156 ++++++++++++++++++++++++++++++++++++++++++++++- kernel/bpf/syscall.c | 32 +++++++++- 5 files changed, 254 insertions(+), 7 deletions(-) (limited to 'include/uapi/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 95a7abd0ee92..ee5275e7d4df 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -22,6 +22,8 @@ struct perf_event; struct bpf_prog; struct bpf_map; struct sock; +struct seq_file; +struct btf; /* map is generic key/value storage optionally accesible by eBPF programs */ struct bpf_map_ops { @@ -43,10 +45,14 @@ struct bpf_map_ops { void (*map_fd_put_ptr)(void *ptr); u32 (*map_gen_lookup)(struct bpf_map *map, struct bpf_insn *insn_buf); u32 (*map_fd_sys_lookup_elem)(void *ptr); + void (*map_seq_show_elem)(struct bpf_map *map, void *key, + struct seq_file *m); + int (*map_check_btf)(const struct bpf_map *map, const struct btf *btf, + u32 key_type_id, u32 value_type_id); }; struct bpf_map { - /* 1st cacheline with read-mostly members of which some + /* The first two cachelines with read-mostly members of which some * are also accessed in fast-path (e.g. ops, max_entries). */ const struct bpf_map_ops *ops ____cacheline_aligned; @@ -62,10 +68,13 @@ struct bpf_map { u32 pages; u32 id; int numa_node; + u32 btf_key_id; + u32 btf_value_id; + struct btf *btf; bool unpriv_array; - /* 7 bytes hole */ + /* 55 bytes hole */ - /* 2nd cacheline with misc members to avoid false sharing + /* The 3rd and 4th cacheline with misc members to avoid false sharing * particularly with refcounting. */ struct user_struct *user ____cacheline_aligned; @@ -100,6 +109,11 @@ static inline struct bpf_offloaded_map *map_to_offmap(struct bpf_map *map) return container_of(map, struct bpf_offloaded_map, map); } +static inline bool bpf_map_support_seq_show(const struct bpf_map *map) +{ + return map->ops->map_seq_show_elem && map->ops->map_check_btf; +} + extern const struct bpf_map_ops bpf_map_offload_ops; /* function argument constraints */ diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 795bcd577750..c8383a289f7b 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -280,6 +280,9 @@ union bpf_attr { */ char map_name[BPF_OBJ_NAME_LEN]; __u32 map_ifindex; /* ifindex of netdev to create on */ + __u32 btf_fd; /* fd pointing to a BTF type data */ + __u32 btf_key_id; /* BTF type_id of the key */ + __u32 btf_value_id; /* BTF type_id of the value */ }; struct { /* anonymous struct used by BPF_MAP_*_ELEM commands */ diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index 14750e7c5ee4..02a189339381 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -11,11 +11,13 @@ * General Public License for more details. */ #include +#include #include #include #include #include #include +#include #include "map_in_map.h" @@ -336,6 +338,52 @@ static void array_map_free(struct bpf_map *map) bpf_map_area_free(array); } +static void array_map_seq_show_elem(struct bpf_map *map, void *key, + struct seq_file *m) +{ + void *value; + + rcu_read_lock(); + + value = array_map_lookup_elem(map, key); + if (!value) { + rcu_read_unlock(); + return; + } + + seq_printf(m, "%u: ", *(u32 *)key); + btf_type_seq_show(map->btf, map->btf_value_id, value, m); + seq_puts(m, "\n"); + + 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) +{ + 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) + 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. + */ + 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) + return -EINVAL; + + return 0; +} + const struct bpf_map_ops array_map_ops = { .map_alloc_check = array_map_alloc_check, .map_alloc = array_map_alloc, @@ -345,6 +393,8 @@ const struct bpf_map_ops array_map_ops = { .map_update_elem = array_map_update_elem, .map_delete_elem = array_map_delete_elem, .map_gen_lookup = array_map_gen_lookup, + .map_seq_show_elem = array_map_seq_show_elem, + .map_check_btf = array_map_check_btf, }; const struct bpf_map_ops percpu_array_map_ops = { diff --git a/kernel/bpf/inode.c b/kernel/bpf/inode.c index bf6da59ae0d0..a41343009ccc 100644 --- a/kernel/bpf/inode.c +++ b/kernel/bpf/inode.c @@ -150,8 +150,154 @@ static int bpf_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) return 0; } +struct map_iter { + void *key; + bool done; +}; + +static struct map_iter *map_iter(struct seq_file *m) +{ + return m->private; +} + +static struct bpf_map *seq_file_to_map(struct seq_file *m) +{ + return file_inode(m->file)->i_private; +} + +static void map_iter_free(struct map_iter *iter) +{ + if (iter) { + kfree(iter->key); + kfree(iter); + } +} + +static struct map_iter *map_iter_alloc(struct bpf_map *map) +{ + struct map_iter *iter; + + iter = kzalloc(sizeof(*iter), GFP_KERNEL | __GFP_NOWARN); + if (!iter) + goto error; + + iter->key = kzalloc(map->key_size, GFP_KERNEL | __GFP_NOWARN); + if (!iter->key) + goto error; + + return iter; + +error: + map_iter_free(iter); + return NULL; +} + +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; + + if (map_iter(m)->done) + return NULL; + + if (unlikely(v == SEQ_START_TOKEN)) + goto done; + + if (map->ops->map_get_next_key(map, key, key)) { + map_iter(m)->done = true; + return NULL; + } + +done: + ++(*pos); + return key; +} + +static void *map_seq_start(struct seq_file *m, loff_t *pos) +{ + if (map_iter(m)->done) + return NULL; + + return *pos ? map_iter(m)->key : SEQ_START_TOKEN; +} + +static void map_seq_stop(struct seq_file *m, void *v) +{ +} + +static int map_seq_show(struct seq_file *m, void *v) +{ + struct bpf_map *map = seq_file_to_map(m); + void *key = map_iter(m)->key; + + if (unlikely(v == SEQ_START_TOKEN)) { + seq_puts(m, "# WARNING!! The output is for debug purpose only\n"); + seq_puts(m, "# WARNING!! The output format will change\n"); + } else { + map->ops->map_seq_show_elem(map, key, m); + } + + return 0; +} + +static const struct seq_operations bpffs_map_seq_ops = { + .start = map_seq_start, + .next = map_seq_next, + .show = map_seq_show, + .stop = map_seq_stop, +}; + +static int bpffs_map_open(struct inode *inode, struct file *file) +{ + struct bpf_map *map = inode->i_private; + struct map_iter *iter; + struct seq_file *m; + int err; + + iter = map_iter_alloc(map); + if (!iter) + return -ENOMEM; + + err = seq_open(file, &bpffs_map_seq_ops); + if (err) { + map_iter_free(iter); + return err; + } + + m = file->private_data; + m->private = iter; + + return 0; +} + +static int bpffs_map_release(struct inode *inode, struct file *file) +{ + struct seq_file *m = file->private_data; + + map_iter_free(map_iter(m)); + + return seq_release(inode, file); +} + +/* bpffs_map_fops should only implement the basic + * read operation for a BPF map. The purpose is to + * provide a simple user intuitive way to do + * "cat bpffs/pathto/a-pinned-map". + * + * Other operations (e.g. write, lookup...) should be realized by + * the userspace tools (e.g. bpftool) through the + * BPF_OBJ_GET_INFO_BY_FD and the map's lookup/update + * interface. + */ +static const struct file_operations bpffs_map_fops = { + .open = bpffs_map_open, + .read = seq_read, + .release = bpffs_map_release, +}; + static int bpf_mkobj_ops(struct dentry *dentry, umode_t mode, void *raw, - const struct inode_operations *iops) + const struct inode_operations *iops, + const struct file_operations *fops) { struct inode *dir = dentry->d_parent->d_inode; struct inode *inode = bpf_get_inode(dir->i_sb, dir, mode); @@ -159,6 +305,7 @@ static int bpf_mkobj_ops(struct dentry *dentry, umode_t mode, void *raw, return PTR_ERR(inode); inode->i_op = iops; + inode->i_fop = fops; inode->i_private = raw; bpf_dentry_finalize(dentry, inode, dir); @@ -167,12 +314,15 @@ static int bpf_mkobj_ops(struct dentry *dentry, umode_t mode, void *raw, static int bpf_mkprog(struct dentry *dentry, umode_t mode, void *arg) { - return bpf_mkobj_ops(dentry, mode, arg, &bpf_prog_iops); + return bpf_mkobj_ops(dentry, mode, arg, &bpf_prog_iops, NULL); } static int bpf_mkmap(struct dentry *dentry, umode_t mode, void *arg) { - return bpf_mkobj_ops(dentry, mode, arg, &bpf_map_iops); + struct bpf_map *map = arg; + + return bpf_mkobj_ops(dentry, mode, arg, &bpf_map_iops, + map->btf ? &bpffs_map_fops : NULL); } static struct dentry * diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index 0a4924a0a8da..fe23dc5a3ec4 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -27,6 +27,7 @@ #include #include #include +#include #define IS_FD_ARRAY(map) ((map)->map_type == BPF_MAP_TYPE_PROG_ARRAY || \ (map)->map_type == BPF_MAP_TYPE_PERF_EVENT_ARRAY || \ @@ -251,6 +252,7 @@ static void bpf_map_free_deferred(struct work_struct *work) bpf_map_uncharge_memlock(map); security_bpf_map_free(map); + btf_put(map->btf); /* implementation dependent freeing */ map->ops->map_free(map); } @@ -416,7 +418,7 @@ static int bpf_obj_name_cpy(char *dst, const char *src) return 0; } -#define BPF_MAP_CREATE_LAST_FIELD map_ifindex +#define BPF_MAP_CREATE_LAST_FIELD btf_value_id /* called via syscall */ static int map_create(union bpf_attr *attr) { @@ -450,6 +452,33 @@ 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_id || attr->btf_value_id)) { + struct btf *btf; + + if (!attr->btf_key_id || !attr->btf_value_id) { + err = -EINVAL; + goto free_map_nouncharge; + } + + btf = btf_get_by_fd(attr->btf_fd); + if (IS_ERR(btf)) { + err = PTR_ERR(btf); + goto free_map_nouncharge; + } + + err = map->ops->map_check_btf(map, btf, attr->btf_key_id, + attr->btf_value_id); + if (err) { + btf_put(btf); + goto free_map_nouncharge; + } + + map->btf = btf; + map->btf_key_id = attr->btf_key_id; + map->btf_value_id = attr->btf_value_id; + } + err = security_bpf_map_alloc(map); if (err) goto free_map_nouncharge; @@ -482,6 +511,7 @@ free_map: free_map_sec: security_bpf_map_free(map); free_map_nouncharge: + btf_put(map->btf); map->ops->map_free(map); return err; } -- cgit v1.3-14-g43fede