提交 fac4c1ea 编写于 作者: D Dave Marchevsky 提交者: Zheng Zengkai

bpf: Migrate cgroup_bpf to internal cgroup_bpf_attach_type enum

mainline inclusion
from mainline-v5.15-rc1
commit 6fc88c35
category: bugfix
bugzilla: https://gitee.com/openeuler/kernel/issues/I4GII8?from=project-issue
CVE: NA

----------

Add an enum (cgroup_bpf_attach_type) containing only valid cgroup_bpf
attach types and a function to map bpf_attach_type values to the new
enum. Inspired by netns_bpf_attach_type.

Then, migrate cgroup_bpf to use cgroup_bpf_attach_type wherever
possible.  Functionality is unchanged as attach_type_to_prog_type
switches in bpf/syscall.c were preventing non-cgroup programs from
making use of the invalid cgroup_bpf array slots.

As a result struct cgroup_bpf uses 504 fewer bytes relative to when its
arrays were sized using MAX_BPF_ATTACH_TYPE.

bpf_cgroup_storage is notably not migrated as struct
bpf_cgroup_storage_key is part of uapi and contains a bpf_attach_type
member which is not meant to be opaque. Similarly, bpf_cgroup_link
continues to report its bpf_attach_type member to userspace via fdinfo
and bpf_link_info.

To ease disambiguation, bpf_attach_type variables are renamed from
'type' to 'atype' when changed to cgroup_bpf_attach_type.
Signed-off-by: NDave Marchevsky <davemarchevsky@fb.com>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20210819092420.1984861-2-davemarchevsky@fb.com
Conflicts:
	include/linux/bpf-cgroup.h
	kernel/bpf/cgroup.c
	net/ipv4/af_inet.c
	net/ipv6/af_inet6.c
Signed-off-by: NLu Jialin <lujialin4@huawei.com>
Reviewed-by: Wei Yongjun<weiyongjun1@huawei.com>
Signed-off-by: NZheng Zengkai <zhengzengkai@huawei.com>
上级 e960ab23
......@@ -24,9 +24,73 @@ struct ctl_table_header;
struct task_struct;
#ifdef CONFIG_CGROUP_BPF
enum cgroup_bpf_attach_type {
CGROUP_BPF_ATTACH_TYPE_INVALID = -1,
CGROUP_INET_INGRESS = 0,
CGROUP_INET_EGRESS,
CGROUP_INET_SOCK_CREATE,
CGROUP_SOCK_OPS,
CGROUP_DEVICE,
CGROUP_INET4_BIND,
CGROUP_INET6_BIND,
CGROUP_INET4_CONNECT,
CGROUP_INET6_CONNECT,
CGROUP_INET4_POST_BIND,
CGROUP_INET6_POST_BIND,
CGROUP_UDP4_SENDMSG,
CGROUP_UDP6_SENDMSG,
CGROUP_SYSCTL,
CGROUP_UDP4_RECVMSG,
CGROUP_UDP6_RECVMSG,
CGROUP_GETSOCKOPT,
CGROUP_SETSOCKOPT,
CGROUP_INET4_GETPEERNAME,
CGROUP_INET6_GETPEERNAME,
CGROUP_INET4_GETSOCKNAME,
CGROUP_INET6_GETSOCKNAME,
CGROUP_INET_SOCK_RELEASE,
MAX_CGROUP_BPF_ATTACH_TYPE
};
#define CGROUP_ATYPE(type) \
case BPF_##type: return type
static inline enum cgroup_bpf_attach_type
to_cgroup_bpf_attach_type(enum bpf_attach_type attach_type)
{
switch (attach_type) {
CGROUP_ATYPE(CGROUP_INET_INGRESS);
CGROUP_ATYPE(CGROUP_INET_EGRESS);
CGROUP_ATYPE(CGROUP_INET_SOCK_CREATE);
CGROUP_ATYPE(CGROUP_SOCK_OPS);
CGROUP_ATYPE(CGROUP_DEVICE);
CGROUP_ATYPE(CGROUP_INET4_BIND);
CGROUP_ATYPE(CGROUP_INET6_BIND);
CGROUP_ATYPE(CGROUP_INET4_CONNECT);
CGROUP_ATYPE(CGROUP_INET6_CONNECT);
CGROUP_ATYPE(CGROUP_INET4_POST_BIND);
CGROUP_ATYPE(CGROUP_INET6_POST_BIND);
CGROUP_ATYPE(CGROUP_UDP4_SENDMSG);
CGROUP_ATYPE(CGROUP_UDP6_SENDMSG);
CGROUP_ATYPE(CGROUP_SYSCTL);
CGROUP_ATYPE(CGROUP_UDP4_RECVMSG);
CGROUP_ATYPE(CGROUP_UDP6_RECVMSG);
CGROUP_ATYPE(CGROUP_GETSOCKOPT);
CGROUP_ATYPE(CGROUP_SETSOCKOPT);
CGROUP_ATYPE(CGROUP_INET4_GETPEERNAME);
CGROUP_ATYPE(CGROUP_INET6_GETPEERNAME);
CGROUP_ATYPE(CGROUP_INET4_GETSOCKNAME);
CGROUP_ATYPE(CGROUP_INET6_GETSOCKNAME);
CGROUP_ATYPE(CGROUP_INET_SOCK_RELEASE);
default:
return CGROUP_BPF_ATTACH_TYPE_INVALID;
}
}
extern struct static_key_false cgroup_bpf_enabled_key[MAX_BPF_ATTACH_TYPE];
#define cgroup_bpf_enabled(type) static_branch_unlikely(&cgroup_bpf_enabled_key[type])
#undef CGROUP_ATYPE
extern struct static_key_false cgroup_bpf_enabled_key[MAX_CGROUP_BPF_ATTACH_TYPE];
#define cgroup_bpf_enabled(atype) static_branch_unlikely(&cgroup_bpf_enabled_key[atype])
#define for_each_cgroup_storage_type(stype) \
for (stype = 0; stype < MAX_BPF_CGROUP_STORAGE_TYPE; stype++)
......@@ -68,15 +132,15 @@ struct bpf_prog_array;
struct cgroup_bpf {
/* array of effective progs in this cgroup */
struct bpf_prog_array __rcu *effective[MAX_BPF_ATTACH_TYPE];
struct bpf_prog_array __rcu *effective[MAX_CGROUP_BPF_ATTACH_TYPE];
/* attached progs to this cgroup and attach flags
* when flags == 0 or BPF_F_ALLOW_OVERRIDE the progs list will
* have either zero or one element
* when BPF_F_ALLOW_MULTI the list can have up to BPF_CGROUP_MAX_PROGS
*/
struct list_head progs[MAX_BPF_ATTACH_TYPE];
u32 flags[MAX_BPF_ATTACH_TYPE];
struct list_head progs[MAX_CGROUP_BPF_ATTACH_TYPE];
u32 flags[MAX_CGROUP_BPF_ATTACH_TYPE];
/* list of cgroup shared storages */
struct list_head storages;
......@@ -133,27 +197,27 @@ int cgroup_bpf_query(struct cgroup *cgrp, const union bpf_attr *attr,
int __cgroup_bpf_run_filter_skb(struct sock *sk,
struct sk_buff *skb,
enum bpf_attach_type type);
enum cgroup_bpf_attach_type atype);
int __cgroup_bpf_run_filter_sk(struct sock *sk,
enum bpf_attach_type type);
enum cgroup_bpf_attach_type atype);
int __cgroup_bpf_run_filter_sock_addr(struct sock *sk,
struct sockaddr *uaddr,
enum bpf_attach_type type,
enum cgroup_bpf_attach_type atype,
void *t_ctx);
int __cgroup_bpf_run_filter_sock_ops(struct sock *sk,
struct bpf_sock_ops_kern *sock_ops,
enum bpf_attach_type type);
enum cgroup_bpf_attach_type atype);
int __cgroup_bpf_check_dev_permission(short dev_type, u32 major, u32 minor,
short access, enum bpf_attach_type type);
short access, enum cgroup_bpf_attach_type atype);
int __cgroup_bpf_run_filter_sysctl(struct ctl_table_header *head,
struct ctl_table *table, int write,
char **buf, size_t *pcount, loff_t *ppos,
enum bpf_attach_type type);
enum cgroup_bpf_attach_type atype);
int __cgroup_bpf_run_filter_setsockopt(struct sock *sock, int *level,
int *optname, char __user *optval,
......@@ -192,9 +256,9 @@ int bpf_percpu_cgroup_storage_update(struct bpf_map *map, void *key,
#define BPF_CGROUP_RUN_PROG_INET_INGRESS(sk, skb) \
({ \
int __ret = 0; \
if (cgroup_bpf_enabled(BPF_CGROUP_INET_INGRESS)) \
if (cgroup_bpf_enabled(CGROUP_INET_INGRESS)) \
__ret = __cgroup_bpf_run_filter_skb(sk, skb, \
BPF_CGROUP_INET_INGRESS); \
CGROUP_INET_INGRESS); \
\
__ret; \
})
......@@ -202,51 +266,51 @@ int bpf_percpu_cgroup_storage_update(struct bpf_map *map, void *key,
#define BPF_CGROUP_RUN_PROG_INET_EGRESS(sk, skb) \
({ \
int __ret = 0; \
if (cgroup_bpf_enabled(BPF_CGROUP_INET_EGRESS) && sk && sk == skb->sk) { \
if (cgroup_bpf_enabled(CGROUP_INET_EGRESS) && sk && sk == skb->sk) { \
typeof(sk) __sk = sk_to_full_sk(sk); \
if (sk_fullsock(__sk)) \
__ret = __cgroup_bpf_run_filter_skb(__sk, skb, \
BPF_CGROUP_INET_EGRESS); \
CGROUP_INET_EGRESS); \
} \
__ret; \
})
#define BPF_CGROUP_RUN_SK_PROG(sk, type) \
#define BPF_CGROUP_RUN_SK_PROG(sk, atype) \
({ \
int __ret = 0; \
if (cgroup_bpf_enabled(type)) { \
__ret = __cgroup_bpf_run_filter_sk(sk, type); \
if (cgroup_bpf_enabled(atype)) { \
__ret = __cgroup_bpf_run_filter_sk(sk, atype); \
} \
__ret; \
})
#define BPF_CGROUP_RUN_PROG_INET_SOCK(sk) \
BPF_CGROUP_RUN_SK_PROG(sk, BPF_CGROUP_INET_SOCK_CREATE)
BPF_CGROUP_RUN_SK_PROG(sk, CGROUP_INET_SOCK_CREATE)
#define BPF_CGROUP_RUN_PROG_INET_SOCK_RELEASE(sk) \
BPF_CGROUP_RUN_SK_PROG(sk, BPF_CGROUP_INET_SOCK_RELEASE)
BPF_CGROUP_RUN_SK_PROG(sk, CGROUP_INET_SOCK_RELEASE)
#define BPF_CGROUP_RUN_PROG_INET4_POST_BIND(sk) \
BPF_CGROUP_RUN_SK_PROG(sk, BPF_CGROUP_INET4_POST_BIND)
BPF_CGROUP_RUN_SK_PROG(sk, CGROUP_INET4_POST_BIND)
#define BPF_CGROUP_RUN_PROG_INET6_POST_BIND(sk) \
BPF_CGROUP_RUN_SK_PROG(sk, BPF_CGROUP_INET6_POST_BIND)
BPF_CGROUP_RUN_SK_PROG(sk, CGROUP_INET6_POST_BIND)
#define BPF_CGROUP_RUN_SA_PROG(sk, uaddr, type) \
#define BPF_CGROUP_RUN_SA_PROG(sk, uaddr, atype) \
({ \
int __ret = 0; \
if (cgroup_bpf_enabled(type)) \
__ret = __cgroup_bpf_run_filter_sock_addr(sk, uaddr, type, \
if (cgroup_bpf_enabled(atype)) \
__ret = __cgroup_bpf_run_filter_sock_addr(sk, uaddr, atype, \
NULL); \
__ret; \
})
#define BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, type, t_ctx) \
#define BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, atype, t_ctx) \
({ \
int __ret = 0; \
if (cgroup_bpf_enabled(type)) { \
if (cgroup_bpf_enabled(atype)) { \
lock_sock(sk); \
__ret = __cgroup_bpf_run_filter_sock_addr(sk, uaddr, type, \
__ret = __cgroup_bpf_run_filter_sock_addr(sk, uaddr, atype, \
t_ctx); \
release_sock(sk); \
} \
......@@ -254,39 +318,39 @@ int bpf_percpu_cgroup_storage_update(struct bpf_map *map, void *key,
})
#define BPF_CGROUP_RUN_PROG_INET4_BIND_LOCK(sk, uaddr) \
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, BPF_CGROUP_INET4_BIND, NULL)
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, CGROUP_INET4_BIND, NULL)
#define BPF_CGROUP_RUN_PROG_INET6_BIND_LOCK(sk, uaddr) \
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, BPF_CGROUP_INET6_BIND, NULL)
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, CGROUP_INET6_BIND, NULL)
#define BPF_CGROUP_PRE_CONNECT_ENABLED(sk) \
((cgroup_bpf_enabled(BPF_CGROUP_INET4_CONNECT) || \
cgroup_bpf_enabled(BPF_CGROUP_INET6_CONNECT)) && \
((cgroup_bpf_enabled(CGROUP_INET4_CONNECT) || \
cgroup_bpf_enabled(CGROUP_INET6_CONNECT)) && \
(sk)->sk_prot->pre_connect)
#define BPF_CGROUP_RUN_PROG_INET4_CONNECT(sk, uaddr) \
BPF_CGROUP_RUN_SA_PROG(sk, uaddr, BPF_CGROUP_INET4_CONNECT)
BPF_CGROUP_RUN_SA_PROG(sk, uaddr, CGROUP_INET4_CONNECT)
#define BPF_CGROUP_RUN_PROG_INET6_CONNECT(sk, uaddr) \
BPF_CGROUP_RUN_SA_PROG(sk, uaddr, BPF_CGROUP_INET6_CONNECT)
BPF_CGROUP_RUN_SA_PROG(sk, uaddr, CGROUP_INET6_CONNECT)
#define BPF_CGROUP_RUN_PROG_INET4_CONNECT_LOCK(sk, uaddr) \
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, BPF_CGROUP_INET4_CONNECT, NULL)
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, CGROUP_INET4_CONNECT, NULL)
#define BPF_CGROUP_RUN_PROG_INET6_CONNECT_LOCK(sk, uaddr) \
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, BPF_CGROUP_INET6_CONNECT, NULL)
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, CGROUP_INET6_CONNECT, NULL)
#define BPF_CGROUP_RUN_PROG_UDP4_SENDMSG_LOCK(sk, uaddr, t_ctx) \
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, BPF_CGROUP_UDP4_SENDMSG, t_ctx)
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, CGROUP_UDP4_SENDMSG, t_ctx)
#define BPF_CGROUP_RUN_PROG_UDP6_SENDMSG_LOCK(sk, uaddr, t_ctx) \
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, BPF_CGROUP_UDP6_SENDMSG, t_ctx)
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, CGROUP_UDP6_SENDMSG, t_ctx)
#define BPF_CGROUP_RUN_PROG_UDP4_RECVMSG_LOCK(sk, uaddr) \
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, BPF_CGROUP_UDP4_RECVMSG, NULL)
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, CGROUP_UDP4_RECVMSG, NULL)
#define BPF_CGROUP_RUN_PROG_UDP6_RECVMSG_LOCK(sk, uaddr) \
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, BPF_CGROUP_UDP6_RECVMSG, NULL)
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, CGROUP_UDP6_RECVMSG, NULL)
/* The SOCK_OPS"_SK" macro should be used when sock_ops->sk is not a
* fullsock and its parent fullsock cannot be traced by
......@@ -306,33 +370,33 @@ int bpf_percpu_cgroup_storage_update(struct bpf_map *map, void *key,
#define BPF_CGROUP_RUN_PROG_SOCK_OPS_SK(sock_ops, sk) \
({ \
int __ret = 0; \
if (cgroup_bpf_enabled(BPF_CGROUP_SOCK_OPS)) \
if (cgroup_bpf_enabled(CGROUP_SOCK_OPS)) \
__ret = __cgroup_bpf_run_filter_sock_ops(sk, \
sock_ops, \
BPF_CGROUP_SOCK_OPS); \
CGROUP_SOCK_OPS); \
__ret; \
})
#define BPF_CGROUP_RUN_PROG_SOCK_OPS(sock_ops) \
({ \
int __ret = 0; \
if (cgroup_bpf_enabled(BPF_CGROUP_SOCK_OPS) && (sock_ops)->sk) { \
if (cgroup_bpf_enabled(CGROUP_SOCK_OPS) && (sock_ops)->sk) { \
typeof(sk) __sk = sk_to_full_sk((sock_ops)->sk); \
if (__sk && sk_fullsock(__sk)) \
__ret = __cgroup_bpf_run_filter_sock_ops(__sk, \
sock_ops, \
BPF_CGROUP_SOCK_OPS); \
CGROUP_SOCK_OPS); \
} \
__ret; \
})
#define BPF_CGROUP_RUN_PROG_DEVICE_CGROUP(type, major, minor, access) \
#define BPF_CGROUP_RUN_PROG_DEVICE_CGROUP(atype, major, minor, access) \
({ \
int __ret = 0; \
if (cgroup_bpf_enabled(BPF_CGROUP_DEVICE)) \
__ret = __cgroup_bpf_check_dev_permission(type, major, minor, \
if (cgroup_bpf_enabled(CGROUP_DEVICE)) \
__ret = __cgroup_bpf_check_dev_permission(atype, major, minor, \
access, \
BPF_CGROUP_DEVICE); \
CGROUP_DEVICE); \
\
__ret; \
})
......@@ -341,10 +405,10 @@ int bpf_percpu_cgroup_storage_update(struct bpf_map *map, void *key,
#define BPF_CGROUP_RUN_PROG_SYSCTL(head, table, write, buf, count, pos) \
({ \
int __ret = 0; \
if (cgroup_bpf_enabled(BPF_CGROUP_SYSCTL)) \
if (cgroup_bpf_enabled(CGROUP_SYSCTL)) \
__ret = __cgroup_bpf_run_filter_sysctl(head, table, write, \
buf, count, pos, \
BPF_CGROUP_SYSCTL); \
CGROUP_SYSCTL); \
__ret; \
})
......@@ -352,7 +416,7 @@ int bpf_percpu_cgroup_storage_update(struct bpf_map *map, void *key,
kernel_optval) \
({ \
int __ret = 0; \
if (cgroup_bpf_enabled(BPF_CGROUP_SETSOCKOPT)) \
if (cgroup_bpf_enabled(CGROUP_SETSOCKOPT)) \
__ret = __cgroup_bpf_run_filter_setsockopt(sock, level, \
optname, optval, \
optlen, \
......@@ -363,7 +427,7 @@ int bpf_percpu_cgroup_storage_update(struct bpf_map *map, void *key,
#define BPF_CGROUP_GETSOCKOPT_MAX_OPTLEN(optlen) \
({ \
int __ret = 0; \
if (cgroup_bpf_enabled(BPF_CGROUP_GETSOCKOPT)) \
if (cgroup_bpf_enabled(CGROUP_GETSOCKOPT)) \
get_user(__ret, optlen); \
__ret; \
})
......@@ -372,7 +436,7 @@ int bpf_percpu_cgroup_storage_update(struct bpf_map *map, void *key,
max_optlen, retval) \
({ \
int __ret = retval; \
if (cgroup_bpf_enabled(BPF_CGROUP_GETSOCKOPT)) \
if (cgroup_bpf_enabled(CGROUP_GETSOCKOPT)) \
__ret = __cgroup_bpf_run_filter_getsockopt(sock, level, \
optname, optval, \
optlen, max_optlen, \
......@@ -434,8 +498,8 @@ static inline int bpf_percpu_cgroup_storage_update(struct bpf_map *map,
return 0;
}
#define cgroup_bpf_enabled(type) (0)
#define BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, type, t_ctx) ({ 0; })
#define cgroup_bpf_enabled(atype) (0)
#define BPF_CGROUP_RUN_SA_PROG_LOCK(sk, uaddr, atype, t_ctx) ({ 0; })
#define BPF_CGROUP_PRE_CONNECT_ENABLED(sk) (0)
#define BPF_CGROUP_RUN_PROG_INET_INGRESS(sk,skb) ({ 0; })
#define BPF_CGROUP_RUN_PROG_INET_EGRESS(sk,skb) ({ 0; })
......@@ -454,7 +518,7 @@ static inline int bpf_percpu_cgroup_storage_update(struct bpf_map *map,
#define BPF_CGROUP_RUN_PROG_UDP4_RECVMSG_LOCK(sk, uaddr) ({ 0; })
#define BPF_CGROUP_RUN_PROG_UDP6_RECVMSG_LOCK(sk, uaddr) ({ 0; })
#define BPF_CGROUP_RUN_PROG_SOCK_OPS(sock_ops) ({ 0; })
#define BPF_CGROUP_RUN_PROG_DEVICE_CGROUP(type,major,minor,access) ({ 0; })
#define BPF_CGROUP_RUN_PROG_DEVICE_CGROUP(atype, major, minor, access) ({ 0; })
#define BPF_CGROUP_RUN_PROG_SYSCTL(head,table,write,buf,count,pos) ({ 0; })
#define BPF_CGROUP_GETSOCKOPT_MAX_OPTLEN(optlen) ({ 0; })
#define BPF_CGROUP_RUN_PROG_GETSOCKOPT(sock, level, optname, optval, \
......
......@@ -78,7 +78,7 @@ struct bpf_lpm_trie_key {
struct bpf_cgroup_storage_key {
__u64 cgroup_inode_id; /* cgroup inode id */
__u32 attach_type; /* program attach type */
__u32 attach_type; /* program attach type (enum bpf_attach_type) */
};
union bpf_iter_link_info {
......
......@@ -19,7 +19,7 @@
#include "../cgroup/cgroup-internal.h"
DEFINE_STATIC_KEY_ARRAY_FALSE(cgroup_bpf_enabled_key, MAX_BPF_ATTACH_TYPE);
DEFINE_STATIC_KEY_ARRAY_FALSE(cgroup_bpf_enabled_key, MAX_CGROUP_BPF_ATTACH_TYPE);
EXPORT_SYMBOL(cgroup_bpf_enabled_key);
void cgroup_bpf_offline(struct cgroup *cgrp)
......@@ -113,12 +113,12 @@ static void cgroup_bpf_release(struct work_struct *work)
struct list_head *storages = &cgrp->bpf.storages;
struct bpf_cgroup_storage *storage, *stmp;
unsigned int type;
unsigned int atype;
mutex_lock(&cgroup_mutex);
for (type = 0; type < ARRAY_SIZE(cgrp->bpf.progs); type++) {
struct list_head *progs = &cgrp->bpf.progs[type];
for (atype = 0; atype < ARRAY_SIZE(cgrp->bpf.progs); atype++) {
struct list_head *progs = &cgrp->bpf.progs[atype];
struct bpf_prog_list *pl, *pltmp;
list_for_each_entry_safe(pl, pltmp, progs, node) {
......@@ -128,10 +128,10 @@ static void cgroup_bpf_release(struct work_struct *work)
if (pl->link)
bpf_cgroup_link_auto_detach(pl->link);
kfree(pl);
static_branch_dec(&cgroup_bpf_enabled_key[type]);
static_branch_dec(&cgroup_bpf_enabled_key[atype]);
}
old_array = rcu_dereference_protected(
cgrp->bpf.effective[type],
cgrp->bpf.effective[atype],
lockdep_is_held(&cgroup_mutex));
bpf_prog_array_free(old_array);
}
......@@ -196,7 +196,7 @@ static u32 prog_list_length(struct list_head *head)
* if parent has overridable or multi-prog, allow attaching
*/
static bool hierarchy_allows_attach(struct cgroup *cgrp,
enum bpf_attach_type type)
enum cgroup_bpf_attach_type atype)
{
struct cgroup *p;
......@@ -204,12 +204,12 @@ static bool hierarchy_allows_attach(struct cgroup *cgrp,
if (!p)
return true;
do {
u32 flags = p->bpf.flags[type];
u32 flags = p->bpf.flags[atype];
u32 cnt;
if (flags & BPF_F_ALLOW_MULTI)
return true;
cnt = prog_list_length(&p->bpf.progs[type]);
cnt = prog_list_length(&p->bpf.progs[atype]);
WARN_ON_ONCE(cnt > 1);
if (cnt == 1)
return !!(flags & BPF_F_ALLOW_OVERRIDE);
......@@ -225,7 +225,7 @@ static bool hierarchy_allows_attach(struct cgroup *cgrp,
* to programs in this cgroup
*/
static int compute_effective_progs(struct cgroup *cgrp,
enum bpf_attach_type type,
enum cgroup_bpf_attach_type atype,
struct bpf_prog_array **array)
{
struct bpf_prog_array_item *item;
......@@ -236,8 +236,8 @@ static int compute_effective_progs(struct cgroup *cgrp,
/* count number of effective programs by walking parents */
do {
if (cnt == 0 || (p->bpf.flags[type] & BPF_F_ALLOW_MULTI))
cnt += prog_list_length(&p->bpf.progs[type]);
if (cnt == 0 || (p->bpf.flags[atype] & BPF_F_ALLOW_MULTI))
cnt += prog_list_length(&p->bpf.progs[atype]);
p = cgroup_parent(p);
} while (p);
......@@ -249,10 +249,10 @@ static int compute_effective_progs(struct cgroup *cgrp,
cnt = 0;
p = cgrp;
do {
if (cnt > 0 && !(p->bpf.flags[type] & BPF_F_ALLOW_MULTI))
if (cnt > 0 && !(p->bpf.flags[atype] & BPF_F_ALLOW_MULTI))
continue;
list_for_each_entry(pl, &p->bpf.progs[type], node) {
list_for_each_entry(pl, &p->bpf.progs[atype], node) {
if (!prog_list_prog(pl))
continue;
......@@ -269,10 +269,10 @@ static int compute_effective_progs(struct cgroup *cgrp,
}
static void activate_effective_progs(struct cgroup *cgrp,
enum bpf_attach_type type,
enum cgroup_bpf_attach_type atype,
struct bpf_prog_array *old_array)
{
old_array = rcu_replace_pointer(cgrp->bpf.effective[type], old_array,
old_array = rcu_replace_pointer(cgrp->bpf.effective[atype], old_array,
lockdep_is_held(&cgroup_mutex));
/* free prog array after grace period, since __cgroup_bpf_run_*()
* might be still walking the array
......@@ -328,7 +328,7 @@ int cgroup_bpf_inherit(struct cgroup *cgrp)
}
static int update_effective_progs(struct cgroup *cgrp,
enum bpf_attach_type type)
enum cgroup_bpf_attach_type atype)
{
struct cgroup_subsys_state *css;
int err;
......@@ -340,7 +340,7 @@ static int update_effective_progs(struct cgroup *cgrp,
if (percpu_ref_is_zero(&desc->bpf.refcnt))
continue;
err = compute_effective_progs(desc, type, &desc->bpf.inactive);
err = compute_effective_progs(desc, atype, &desc->bpf.inactive);
if (err)
goto cleanup;
}
......@@ -357,7 +357,7 @@ static int update_effective_progs(struct cgroup *cgrp,
continue;
}
activate_effective_progs(desc, type, desc->bpf.inactive);
activate_effective_progs(desc, atype, desc->bpf.inactive);
desc->bpf.inactive = NULL;
}
......@@ -436,11 +436,12 @@ int __cgroup_bpf_attach(struct cgroup *cgrp,
enum bpf_attach_type type, u32 flags)
{
u32 saved_flags = (flags & (BPF_F_ALLOW_OVERRIDE | BPF_F_ALLOW_MULTI));
struct list_head *progs = &cgrp->bpf.progs[type];
struct bpf_prog *old_prog = NULL;
struct bpf_cgroup_storage *storage[MAX_BPF_CGROUP_STORAGE_TYPE] = {};
struct bpf_cgroup_storage *new_storage[MAX_BPF_CGROUP_STORAGE_TYPE] = {};
enum cgroup_bpf_attach_type atype;
struct bpf_prog_list *pl;
struct list_head *progs;
int err;
if (((flags & BPF_F_ALLOW_OVERRIDE) && (flags & BPF_F_ALLOW_MULTI)) ||
......@@ -454,10 +455,16 @@ int __cgroup_bpf_attach(struct cgroup *cgrp,
/* replace_prog implies BPF_F_REPLACE, and vice versa */
return -EINVAL;
if (!hierarchy_allows_attach(cgrp, type))
atype = to_cgroup_bpf_attach_type(type);
if (atype < 0)
return -EINVAL;
progs = &cgrp->bpf.progs[atype];
if (!hierarchy_allows_attach(cgrp, atype))
return -EPERM;
if (!list_empty(progs) && cgrp->bpf.flags[type] != saved_flags)
if (!list_empty(progs) && cgrp->bpf.flags[atype] != saved_flags)
/* Disallow attaching non-overridable on top
* of existing overridable in this cgroup.
* Disallow attaching multi-prog if overridable or none
......@@ -490,16 +497,16 @@ int __cgroup_bpf_attach(struct cgroup *cgrp,
pl->prog = prog;
pl->link = link;
bpf_cgroup_storages_assign(pl->storage, storage);
cgrp->bpf.flags[type] = saved_flags;
cgrp->bpf.flags[atype] = saved_flags;
err = update_effective_progs(cgrp, type);
err = update_effective_progs(cgrp, atype);
if (err)
goto cleanup;
if (old_prog)
bpf_prog_put(old_prog);
else
static_branch_inc(&cgroup_bpf_enabled_key[type]);
static_branch_inc(&cgroup_bpf_enabled_key[atype]);
bpf_cgroup_storages_link(new_storage, cgrp, type);
return 0;
......@@ -520,7 +527,7 @@ int __cgroup_bpf_attach(struct cgroup *cgrp,
* all descendant cgroups. This function is guaranteed to succeed.
*/
static void replace_effective_prog(struct cgroup *cgrp,
enum bpf_attach_type type,
enum cgroup_bpf_attach_type atype,
struct bpf_cgroup_link *link)
{
struct bpf_prog_array_item *item;
......@@ -539,10 +546,10 @@ static void replace_effective_prog(struct cgroup *cgrp,
/* find position of link in effective progs array */
for (pos = 0, cg = desc; cg; cg = cgroup_parent(cg)) {
if (pos && !(cg->bpf.flags[type] & BPF_F_ALLOW_MULTI))
if (pos && !(cg->bpf.flags[atype] & BPF_F_ALLOW_MULTI))
continue;
head = &cg->bpf.progs[type];
head = &cg->bpf.progs[atype];
list_for_each_entry(pl, head, node) {
if (!prog_list_prog(pl))
continue;
......@@ -554,7 +561,7 @@ static void replace_effective_prog(struct cgroup *cgrp,
found:
BUG_ON(!cg);
progs = rcu_dereference_protected(
desc->bpf.effective[type],
desc->bpf.effective[atype],
lockdep_is_held(&cgroup_mutex));
item = &progs->items[pos];
WRITE_ONCE(item->prog, link->link.prog);
......@@ -574,11 +581,18 @@ static int __cgroup_bpf_replace(struct cgroup *cgrp,
struct bpf_cgroup_link *link,
struct bpf_prog *new_prog)
{
struct list_head *progs = &cgrp->bpf.progs[link->type];
enum cgroup_bpf_attach_type atype;
struct bpf_prog *old_prog;
struct bpf_prog_list *pl;
struct list_head *progs;
bool found = false;
atype = to_cgroup_bpf_attach_type(link->type);
if (atype < 0)
return -EINVAL;
progs = &cgrp->bpf.progs[atype];
if (link->link.prog->type != new_prog->type)
return -EINVAL;
......@@ -592,7 +606,7 @@ static int __cgroup_bpf_replace(struct cgroup *cgrp,
return -ENOENT;
old_prog = xchg(&link->link.prog, new_prog);
replace_effective_prog(cgrp, link->type, link);
replace_effective_prog(cgrp, atype, link);
bpf_prog_put(old_prog);
return 0;
}
......@@ -667,12 +681,20 @@ static struct bpf_prog_list *find_detach_entry(struct list_head *progs,
int __cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog,
struct bpf_cgroup_link *link, enum bpf_attach_type type)
{
struct list_head *progs = &cgrp->bpf.progs[type];
u32 flags = cgrp->bpf.flags[type];
struct bpf_prog_list *pl;
enum cgroup_bpf_attach_type atype;
struct bpf_prog *old_prog;
struct bpf_prog_list *pl;
struct list_head *progs;
u32 flags;
int err;
atype = to_cgroup_bpf_attach_type(type);
if (atype < 0)
return -EINVAL;
progs = &cgrp->bpf.progs[atype];
flags = cgrp->bpf.flags[atype];
if (prog && link)
/* only one of prog or link can be specified */
return -EINVAL;
......@@ -686,7 +708,7 @@ int __cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog,
pl->prog = NULL;
pl->link = NULL;
err = update_effective_progs(cgrp, type);
err = update_effective_progs(cgrp, atype);
if (err)
goto cleanup;
......@@ -695,10 +717,10 @@ int __cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog,
kfree(pl);
if (list_empty(progs))
/* last program was detached, reset flags to zero */
cgrp->bpf.flags[type] = 0;
cgrp->bpf.flags[atype] = 0;
if (old_prog)
bpf_prog_put(old_prog);
static_branch_dec(&cgroup_bpf_enabled_key[type]);
static_branch_dec(&cgroup_bpf_enabled_key[atype]);
return 0;
cleanup:
......@@ -714,13 +736,21 @@ int __cgroup_bpf_query(struct cgroup *cgrp, const union bpf_attr *attr,
{
__u32 __user *prog_ids = u64_to_user_ptr(attr->query.prog_ids);
enum bpf_attach_type type = attr->query.attach_type;
struct list_head *progs = &cgrp->bpf.progs[type];
u32 flags = cgrp->bpf.flags[type];
enum cgroup_bpf_attach_type atype;
struct bpf_prog_array *effective;
struct list_head *progs;
struct bpf_prog *prog;
int cnt, ret = 0, i;
u32 flags;
effective = rcu_dereference_protected(cgrp->bpf.effective[type],
atype = to_cgroup_bpf_attach_type(type);
if (atype < 0)
return -EINVAL;
progs = &cgrp->bpf.progs[atype];
flags = cgrp->bpf.flags[atype];
effective = rcu_dereference_protected(cgrp->bpf.effective[atype],
lockdep_is_held(&cgroup_mutex));
if (attr->query.query_flags & BPF_F_QUERY_EFFECTIVE)
......@@ -931,8 +961,8 @@ int cgroup_bpf_link_attach(const union bpf_attr *attr, struct bpf_prog *prog)
goto out_put_cgroup;
}
err = cgroup_bpf_attach(cgrp, NULL, NULL, link, link->type,
BPF_F_ALLOW_MULTI);
err = cgroup_bpf_attach(cgrp, NULL, NULL, link,
link->type, BPF_F_ALLOW_MULTI);
if (err) {
bpf_link_cleanup(&link_primer);
goto out_put_cgroup;
......@@ -986,7 +1016,7 @@ int cgroup_bpf_prog_query(const union bpf_attr *attr,
*/
int __cgroup_bpf_run_filter_skb(struct sock *sk,
struct sk_buff *skb,
enum bpf_attach_type type)
enum cgroup_bpf_attach_type atype)
{
unsigned int offset = skb->data - skb_network_header(skb);
struct sock *save_sk;
......@@ -1008,11 +1038,11 @@ int __cgroup_bpf_run_filter_skb(struct sock *sk,
/* compute pointers for the bpf prog */
bpf_compute_and_save_data_end(skb, &saved_data_end);
if (type == BPF_CGROUP_INET_EGRESS) {
if (atype == CGROUP_INET_EGRESS) {
ret = BPF_PROG_CGROUP_INET_EGRESS_RUN_ARRAY(
cgrp->bpf.effective[type], skb, __bpf_prog_run_save_cb);
cgrp->bpf.effective[atype], skb, __bpf_prog_run_save_cb);
} else {
ret = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[type], skb,
ret = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[atype], skb,
__bpf_prog_run_save_cb);
ret = (ret == 1 ? 0 : -EPERM);
}
......@@ -1038,12 +1068,12 @@ EXPORT_SYMBOL(__cgroup_bpf_run_filter_skb);
* and if it returned != 1 during execution. In all other cases, 0 is returned.
*/
int __cgroup_bpf_run_filter_sk(struct sock *sk,
enum bpf_attach_type type)
enum cgroup_bpf_attach_type atype)
{
struct cgroup *cgrp = sock_cgroup_ptr(&sk->sk_cgrp_data);
int ret;
ret = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[type], sk, BPF_PROG_RUN);
ret = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[atype], sk, BPF_PROG_RUN);
return ret == 1 ? 0 : -EPERM;
}
EXPORT_SYMBOL(__cgroup_bpf_run_filter_sk);
......@@ -1063,7 +1093,7 @@ EXPORT_SYMBOL(__cgroup_bpf_run_filter_sk);
*/
int __cgroup_bpf_run_filter_sock_addr(struct sock *sk,
struct sockaddr *uaddr,
enum bpf_attach_type type,
enum cgroup_bpf_attach_type atype,
void *t_ctx)
{
struct bpf_sock_addr_kern ctx = {
......@@ -1087,7 +1117,7 @@ int __cgroup_bpf_run_filter_sock_addr(struct sock *sk,
}
cgrp = sock_cgroup_ptr(&sk->sk_cgrp_data);
ret = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[type], &ctx, BPF_PROG_RUN);
ret = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[atype], &ctx, BPF_PROG_RUN);
return ret == 1 ? 0 : -EPERM;
}
......@@ -1111,19 +1141,19 @@ EXPORT_SYMBOL(__cgroup_bpf_run_filter_sock_addr);
*/
int __cgroup_bpf_run_filter_sock_ops(struct sock *sk,
struct bpf_sock_ops_kern *sock_ops,
enum bpf_attach_type type)
enum cgroup_bpf_attach_type atype)
{
struct cgroup *cgrp = sock_cgroup_ptr(&sk->sk_cgrp_data);
int ret;
ret = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[type], sock_ops,
ret = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[atype], sock_ops,
BPF_PROG_RUN);
return ret == 1 ? 0 : -EPERM;
}
EXPORT_SYMBOL(__cgroup_bpf_run_filter_sock_ops);
int __cgroup_bpf_check_dev_permission(short dev_type, u32 major, u32 minor,
short access, enum bpf_attach_type type)
short access, enum cgroup_bpf_attach_type atype)
{
struct cgroup *cgrp;
struct bpf_cgroup_dev_ctx ctx = {
......@@ -1135,7 +1165,7 @@ int __cgroup_bpf_check_dev_permission(short dev_type, u32 major, u32 minor,
rcu_read_lock();
cgrp = task_dfl_cgroup(current);
allow = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[type], &ctx,
allow = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[atype], &ctx,
BPF_PROG_RUN);
rcu_read_unlock();
......@@ -1227,7 +1257,7 @@ const struct bpf_verifier_ops cg_dev_verifier_ops = {
int __cgroup_bpf_run_filter_sysctl(struct ctl_table_header *head,
struct ctl_table *table, int write,
char **buf, size_t *pcount, loff_t *ppos,
enum bpf_attach_type type)
enum cgroup_bpf_attach_type atype)
{
struct bpf_sysctl_kern ctx = {
.head = head,
......@@ -1267,7 +1297,7 @@ int __cgroup_bpf_run_filter_sysctl(struct ctl_table_header *head,
rcu_read_lock();
cgrp = task_dfl_cgroup(current);
ret = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[type], &ctx, BPF_PROG_RUN);
ret = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[atype], &ctx, BPF_PROG_RUN);
rcu_read_unlock();
kfree(ctx.cur_val);
......@@ -1285,7 +1315,7 @@ int __cgroup_bpf_run_filter_sysctl(struct ctl_table_header *head,
#ifdef CONFIG_NET
static bool __cgroup_bpf_prog_array_is_empty(struct cgroup *cgrp,
enum bpf_attach_type attach_type)
enum cgroup_bpf_attach_type attach_type)
{
struct bpf_prog_array *prog_array;
bool empty;
......@@ -1360,7 +1390,7 @@ int __cgroup_bpf_run_filter_setsockopt(struct sock *sk, int *level,
* attached to the hook so we don't waste time allocating
* memory and locking the socket.
*/
if (__cgroup_bpf_prog_array_is_empty(cgrp, BPF_CGROUP_SETSOCKOPT))
if (__cgroup_bpf_prog_array_is_empty(cgrp, CGROUP_SETSOCKOPT))
return 0;
/* Allocate a bit more than the initial user buffer for
......@@ -1381,7 +1411,7 @@ int __cgroup_bpf_run_filter_setsockopt(struct sock *sk, int *level,
}
lock_sock(sk);
ret = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[BPF_CGROUP_SETSOCKOPT],
ret = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[CGROUP_SETSOCKOPT],
&ctx, BPF_PROG_RUN);
release_sock(sk);
......@@ -1456,7 +1486,7 @@ int __cgroup_bpf_run_filter_getsockopt(struct sock *sk, int level,
* attached to the hook so we don't waste time allocating
* memory and locking the socket.
*/
if (__cgroup_bpf_prog_array_is_empty(cgrp, BPF_CGROUP_GETSOCKOPT))
if (__cgroup_bpf_prog_array_is_empty(cgrp, CGROUP_GETSOCKOPT))
return retval;
ctx.optlen = max_optlen;
......@@ -1491,7 +1521,7 @@ int __cgroup_bpf_run_filter_getsockopt(struct sock *sk, int level,
}
lock_sock(sk);
ret = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[BPF_CGROUP_GETSOCKOPT],
ret = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[CGROUP_GETSOCKOPT],
&ctx, BPF_PROG_RUN);
release_sock(sk);
......
......@@ -778,7 +778,7 @@ int inet_getname(struct socket *sock, struct sockaddr *uaddr,
sin->sin_port = inet->inet_dport;
sin->sin_addr.s_addr = inet->inet_daddr;
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, (struct sockaddr *)sin,
BPF_CGROUP_INET4_GETPEERNAME,
CGROUP_INET4_GETPEERNAME,
NULL);
} else {
__be32 addr = inet->inet_rcv_saddr;
......@@ -787,7 +787,7 @@ int inet_getname(struct socket *sock, struct sockaddr *uaddr,
sin->sin_port = inet->inet_sport;
sin->sin_addr.s_addr = addr;
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, (struct sockaddr *)sin,
BPF_CGROUP_INET4_GETSOCKNAME,
CGROUP_INET4_GETSOCKNAME,
NULL);
}
memset(sin->sin_zero, 0, sizeof(sin->sin_zero));
......
......@@ -1126,7 +1126,7 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
rcu_read_unlock();
}
if (cgroup_bpf_enabled(BPF_CGROUP_UDP4_SENDMSG) && !connected) {
if (cgroup_bpf_enabled(CGROUP_UDP4_SENDMSG) && !connected) {
err = BPF_CGROUP_RUN_PROG_UDP4_SENDMSG_LOCK(sk,
(struct sockaddr *)usin, &ipc.addr);
if (err)
......
......@@ -528,7 +528,7 @@ int inet6_getname(struct socket *sock, struct sockaddr *uaddr,
if (np->sndflow)
sin->sin6_flowinfo = np->flow_label;
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, (struct sockaddr *)sin,
BPF_CGROUP_INET6_GETPEERNAME,
CGROUP_INET6_GETPEERNAME,
NULL);
} else {
if (ipv6_addr_any(&sk->sk_v6_rcv_saddr))
......@@ -537,7 +537,7 @@ int inet6_getname(struct socket *sock, struct sockaddr *uaddr,
sin->sin6_addr = sk->sk_v6_rcv_saddr;
sin->sin6_port = inet->inet_sport;
BPF_CGROUP_RUN_SA_PROG_LOCK(sk, (struct sockaddr *)sin,
BPF_CGROUP_INET6_GETSOCKNAME,
CGROUP_INET6_GETSOCKNAME,
NULL);
}
sin->sin6_scope_id = ipv6_iface_scope_id(&sin->sin6_addr,
......
......@@ -1460,7 +1460,7 @@ int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
fl6.saddr = np->saddr;
fl6.fl6_sport = inet->inet_sport;
if (cgroup_bpf_enabled(BPF_CGROUP_UDP6_SENDMSG) && !connected) {
if (cgroup_bpf_enabled(CGROUP_UDP6_SENDMSG) && !connected) {
err = BPF_CGROUP_RUN_PROG_UDP6_SENDMSG_LOCK(sk,
(struct sockaddr *)sin6, &fl6.saddr);
if (err)
......
......@@ -78,7 +78,7 @@ struct bpf_lpm_trie_key {
struct bpf_cgroup_storage_key {
__u64 cgroup_inode_id; /* cgroup inode id */
__u32 attach_type; /* program attach type */
__u32 attach_type; /* program attach type (enum bpf_attach_type) */
};
union bpf_iter_link_info {
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册