![Daniel Borkmann](/assets/img/avatar_default.png)
diag socket's sock_diag_put_filterinfo() dumps classic BPF programs
upon request to user space (ss -0 -b). However, native eBPF programs
attached to sockets (SO_ATTACH_BPF) cannot be dumped with this method:
Their orig_prog is always NULL. However, sock_diag_put_filterinfo()
unconditionally tries to access its filter length resp. wants to copy
the filter insns from there. Internal cBPF to eBPF transformations
attached to sockets don't have this issue, as orig_prog state is kept.
It's currently only used by packet sockets. If we would want to add
native eBPF support in the future, this needs to be done through
a different attribute than PACKET_DIAG_FILTER to not confuse possible
user space disassemblers that work on diag data.
Fixes: 89aa075832
("net: sock: allow eBPF programs to be attached to sockets")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Acked-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
337 lines
8 KiB
C
337 lines
8 KiB
C
#include <linux/mutex.h>
|
|
#include <linux/socket.h>
|
|
#include <linux/skbuff.h>
|
|
#include <net/netlink.h>
|
|
#include <net/net_namespace.h>
|
|
#include <linux/module.h>
|
|
#include <net/sock.h>
|
|
#include <linux/kernel.h>
|
|
#include <linux/tcp.h>
|
|
#include <linux/workqueue.h>
|
|
|
|
#include <linux/inet_diag.h>
|
|
#include <linux/sock_diag.h>
|
|
|
|
static const struct sock_diag_handler *sock_diag_handlers[AF_MAX];
|
|
static int (*inet_rcv_compat)(struct sk_buff *skb, struct nlmsghdr *nlh);
|
|
static DEFINE_MUTEX(sock_diag_table_mutex);
|
|
static struct workqueue_struct *broadcast_wq;
|
|
|
|
static u64 sock_gen_cookie(struct sock *sk)
|
|
{
|
|
while (1) {
|
|
u64 res = atomic64_read(&sk->sk_cookie);
|
|
|
|
if (res)
|
|
return res;
|
|
res = atomic64_inc_return(&sock_net(sk)->cookie_gen);
|
|
atomic64_cmpxchg(&sk->sk_cookie, 0, res);
|
|
}
|
|
}
|
|
|
|
int sock_diag_check_cookie(struct sock *sk, const __u32 *cookie)
|
|
{
|
|
u64 res;
|
|
|
|
if (cookie[0] == INET_DIAG_NOCOOKIE && cookie[1] == INET_DIAG_NOCOOKIE)
|
|
return 0;
|
|
|
|
res = sock_gen_cookie(sk);
|
|
if ((u32)res != cookie[0] || (u32)(res >> 32) != cookie[1])
|
|
return -ESTALE;
|
|
|
|
return 0;
|
|
}
|
|
EXPORT_SYMBOL_GPL(sock_diag_check_cookie);
|
|
|
|
void sock_diag_save_cookie(struct sock *sk, __u32 *cookie)
|
|
{
|
|
u64 res = sock_gen_cookie(sk);
|
|
|
|
cookie[0] = (u32)res;
|
|
cookie[1] = (u32)(res >> 32);
|
|
}
|
|
EXPORT_SYMBOL_GPL(sock_diag_save_cookie);
|
|
|
|
int sock_diag_put_meminfo(struct sock *sk, struct sk_buff *skb, int attrtype)
|
|
{
|
|
u32 mem[SK_MEMINFO_VARS];
|
|
|
|
mem[SK_MEMINFO_RMEM_ALLOC] = sk_rmem_alloc_get(sk);
|
|
mem[SK_MEMINFO_RCVBUF] = sk->sk_rcvbuf;
|
|
mem[SK_MEMINFO_WMEM_ALLOC] = sk_wmem_alloc_get(sk);
|
|
mem[SK_MEMINFO_SNDBUF] = sk->sk_sndbuf;
|
|
mem[SK_MEMINFO_FWD_ALLOC] = sk->sk_forward_alloc;
|
|
mem[SK_MEMINFO_WMEM_QUEUED] = sk->sk_wmem_queued;
|
|
mem[SK_MEMINFO_OPTMEM] = atomic_read(&sk->sk_omem_alloc);
|
|
mem[SK_MEMINFO_BACKLOG] = sk->sk_backlog.len;
|
|
|
|
return nla_put(skb, attrtype, sizeof(mem), &mem);
|
|
}
|
|
EXPORT_SYMBOL_GPL(sock_diag_put_meminfo);
|
|
|
|
int sock_diag_put_filterinfo(bool may_report_filterinfo, struct sock *sk,
|
|
struct sk_buff *skb, int attrtype)
|
|
{
|
|
struct sock_fprog_kern *fprog;
|
|
struct sk_filter *filter;
|
|
struct nlattr *attr;
|
|
unsigned int flen;
|
|
int err = 0;
|
|
|
|
if (!may_report_filterinfo) {
|
|
nla_reserve(skb, attrtype, 0);
|
|
return 0;
|
|
}
|
|
|
|
rcu_read_lock();
|
|
filter = rcu_dereference(sk->sk_filter);
|
|
if (!filter)
|
|
goto out;
|
|
|
|
fprog = filter->prog->orig_prog;
|
|
if (!fprog)
|
|
goto out;
|
|
|
|
flen = bpf_classic_proglen(fprog);
|
|
|
|
attr = nla_reserve(skb, attrtype, flen);
|
|
if (attr == NULL) {
|
|
err = -EMSGSIZE;
|
|
goto out;
|
|
}
|
|
|
|
memcpy(nla_data(attr), fprog->filter, flen);
|
|
out:
|
|
rcu_read_unlock();
|
|
return err;
|
|
}
|
|
EXPORT_SYMBOL(sock_diag_put_filterinfo);
|
|
|
|
struct broadcast_sk {
|
|
struct sock *sk;
|
|
struct work_struct work;
|
|
};
|
|
|
|
static size_t sock_diag_nlmsg_size(void)
|
|
{
|
|
return NLMSG_ALIGN(sizeof(struct inet_diag_msg)
|
|
+ nla_total_size(sizeof(u8)) /* INET_DIAG_PROTOCOL */
|
|
+ nla_total_size(sizeof(struct tcp_info))); /* INET_DIAG_INFO */
|
|
}
|
|
|
|
static void sock_diag_broadcast_destroy_work(struct work_struct *work)
|
|
{
|
|
struct broadcast_sk *bsk =
|
|
container_of(work, struct broadcast_sk, work);
|
|
struct sock *sk = bsk->sk;
|
|
const struct sock_diag_handler *hndl;
|
|
struct sk_buff *skb;
|
|
const enum sknetlink_groups group = sock_diag_destroy_group(sk);
|
|
int err = -1;
|
|
|
|
WARN_ON(group == SKNLGRP_NONE);
|
|
|
|
skb = nlmsg_new(sock_diag_nlmsg_size(), GFP_KERNEL);
|
|
if (!skb)
|
|
goto out;
|
|
|
|
mutex_lock(&sock_diag_table_mutex);
|
|
hndl = sock_diag_handlers[sk->sk_family];
|
|
if (hndl && hndl->get_info)
|
|
err = hndl->get_info(skb, sk);
|
|
mutex_unlock(&sock_diag_table_mutex);
|
|
|
|
if (!err)
|
|
nlmsg_multicast(sock_net(sk)->diag_nlsk, skb, 0, group,
|
|
GFP_KERNEL);
|
|
else
|
|
kfree_skb(skb);
|
|
out:
|
|
sk_destruct(sk);
|
|
kfree(bsk);
|
|
}
|
|
|
|
void sock_diag_broadcast_destroy(struct sock *sk)
|
|
{
|
|
/* Note, this function is often called from an interrupt context. */
|
|
struct broadcast_sk *bsk =
|
|
kmalloc(sizeof(struct broadcast_sk), GFP_ATOMIC);
|
|
if (!bsk)
|
|
return sk_destruct(sk);
|
|
bsk->sk = sk;
|
|
INIT_WORK(&bsk->work, sock_diag_broadcast_destroy_work);
|
|
queue_work(broadcast_wq, &bsk->work);
|
|
}
|
|
|
|
void sock_diag_register_inet_compat(int (*fn)(struct sk_buff *skb, struct nlmsghdr *nlh))
|
|
{
|
|
mutex_lock(&sock_diag_table_mutex);
|
|
inet_rcv_compat = fn;
|
|
mutex_unlock(&sock_diag_table_mutex);
|
|
}
|
|
EXPORT_SYMBOL_GPL(sock_diag_register_inet_compat);
|
|
|
|
void sock_diag_unregister_inet_compat(int (*fn)(struct sk_buff *skb, struct nlmsghdr *nlh))
|
|
{
|
|
mutex_lock(&sock_diag_table_mutex);
|
|
inet_rcv_compat = NULL;
|
|
mutex_unlock(&sock_diag_table_mutex);
|
|
}
|
|
EXPORT_SYMBOL_GPL(sock_diag_unregister_inet_compat);
|
|
|
|
int sock_diag_register(const struct sock_diag_handler *hndl)
|
|
{
|
|
int err = 0;
|
|
|
|
if (hndl->family >= AF_MAX)
|
|
return -EINVAL;
|
|
|
|
mutex_lock(&sock_diag_table_mutex);
|
|
if (sock_diag_handlers[hndl->family])
|
|
err = -EBUSY;
|
|
else
|
|
sock_diag_handlers[hndl->family] = hndl;
|
|
mutex_unlock(&sock_diag_table_mutex);
|
|
|
|
return err;
|
|
}
|
|
EXPORT_SYMBOL_GPL(sock_diag_register);
|
|
|
|
void sock_diag_unregister(const struct sock_diag_handler *hnld)
|
|
{
|
|
int family = hnld->family;
|
|
|
|
if (family >= AF_MAX)
|
|
return;
|
|
|
|
mutex_lock(&sock_diag_table_mutex);
|
|
BUG_ON(sock_diag_handlers[family] != hnld);
|
|
sock_diag_handlers[family] = NULL;
|
|
mutex_unlock(&sock_diag_table_mutex);
|
|
}
|
|
EXPORT_SYMBOL_GPL(sock_diag_unregister);
|
|
|
|
static int __sock_diag_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
|
|
{
|
|
int err;
|
|
struct sock_diag_req *req = nlmsg_data(nlh);
|
|
const struct sock_diag_handler *hndl;
|
|
|
|
if (nlmsg_len(nlh) < sizeof(*req))
|
|
return -EINVAL;
|
|
|
|
if (req->sdiag_family >= AF_MAX)
|
|
return -EINVAL;
|
|
|
|
if (sock_diag_handlers[req->sdiag_family] == NULL)
|
|
request_module("net-pf-%d-proto-%d-type-%d", PF_NETLINK,
|
|
NETLINK_SOCK_DIAG, req->sdiag_family);
|
|
|
|
mutex_lock(&sock_diag_table_mutex);
|
|
hndl = sock_diag_handlers[req->sdiag_family];
|
|
if (hndl == NULL)
|
|
err = -ENOENT;
|
|
else
|
|
err = hndl->dump(skb, nlh);
|
|
mutex_unlock(&sock_diag_table_mutex);
|
|
|
|
return err;
|
|
}
|
|
|
|
static int sock_diag_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
|
|
{
|
|
int ret;
|
|
|
|
switch (nlh->nlmsg_type) {
|
|
case TCPDIAG_GETSOCK:
|
|
case DCCPDIAG_GETSOCK:
|
|
if (inet_rcv_compat == NULL)
|
|
request_module("net-pf-%d-proto-%d-type-%d", PF_NETLINK,
|
|
NETLINK_SOCK_DIAG, AF_INET);
|
|
|
|
mutex_lock(&sock_diag_table_mutex);
|
|
if (inet_rcv_compat != NULL)
|
|
ret = inet_rcv_compat(skb, nlh);
|
|
else
|
|
ret = -EOPNOTSUPP;
|
|
mutex_unlock(&sock_diag_table_mutex);
|
|
|
|
return ret;
|
|
case SOCK_DIAG_BY_FAMILY:
|
|
return __sock_diag_rcv_msg(skb, nlh);
|
|
default:
|
|
return -EINVAL;
|
|
}
|
|
}
|
|
|
|
static DEFINE_MUTEX(sock_diag_mutex);
|
|
|
|
static void sock_diag_rcv(struct sk_buff *skb)
|
|
{
|
|
mutex_lock(&sock_diag_mutex);
|
|
netlink_rcv_skb(skb, &sock_diag_rcv_msg);
|
|
mutex_unlock(&sock_diag_mutex);
|
|
}
|
|
|
|
static int sock_diag_bind(struct net *net, int group)
|
|
{
|
|
switch (group) {
|
|
case SKNLGRP_INET_TCP_DESTROY:
|
|
case SKNLGRP_INET_UDP_DESTROY:
|
|
if (!sock_diag_handlers[AF_INET])
|
|
request_module("net-pf-%d-proto-%d-type-%d", PF_NETLINK,
|
|
NETLINK_SOCK_DIAG, AF_INET);
|
|
break;
|
|
case SKNLGRP_INET6_TCP_DESTROY:
|
|
case SKNLGRP_INET6_UDP_DESTROY:
|
|
if (!sock_diag_handlers[AF_INET6])
|
|
request_module("net-pf-%d-proto-%d-type-%d", PF_NETLINK,
|
|
NETLINK_SOCK_DIAG, AF_INET);
|
|
break;
|
|
}
|
|
return 0;
|
|
}
|
|
|
|
static int __net_init diag_net_init(struct net *net)
|
|
{
|
|
struct netlink_kernel_cfg cfg = {
|
|
.groups = SKNLGRP_MAX,
|
|
.input = sock_diag_rcv,
|
|
.bind = sock_diag_bind,
|
|
.flags = NL_CFG_F_NONROOT_RECV,
|
|
};
|
|
|
|
net->diag_nlsk = netlink_kernel_create(net, NETLINK_SOCK_DIAG, &cfg);
|
|
return net->diag_nlsk == NULL ? -ENOMEM : 0;
|
|
}
|
|
|
|
static void __net_exit diag_net_exit(struct net *net)
|
|
{
|
|
netlink_kernel_release(net->diag_nlsk);
|
|
net->diag_nlsk = NULL;
|
|
}
|
|
|
|
static struct pernet_operations diag_net_ops = {
|
|
.init = diag_net_init,
|
|
.exit = diag_net_exit,
|
|
};
|
|
|
|
static int __init sock_diag_init(void)
|
|
{
|
|
broadcast_wq = alloc_workqueue("sock_diag_events", 0, 0);
|
|
BUG_ON(!broadcast_wq);
|
|
return register_pernet_subsys(&diag_net_ops);
|
|
}
|
|
|
|
static void __exit sock_diag_exit(void)
|
|
{
|
|
unregister_pernet_subsys(&diag_net_ops);
|
|
destroy_workqueue(broadcast_wq);
|
|
}
|
|
|
|
module_init(sock_diag_init);
|
|
module_exit(sock_diag_exit);
|
|
MODULE_LICENSE("GPL");
|
|
MODULE_ALIAS_NET_PF_PROTO(PF_NETLINK, NETLINK_SOCK_DIAG);
|