linux/net/sched/cls_cgroup.c
Matteo Croce 594725db0c cls_cgroup: avoid panic when receiving a packet before filter set
When a cgroup classifier is added, there is a small time interval in
which tp->root is NULL. If we receive a packet in this small time slice
a NULL pointer dereference will happen, leading to a kernel panic:

    # mkdir /sys/fs/cgroup/net_cls/0
    # echo 0x100001 >  /sys/fs/cgroup/net_cls/0/net_cls.classid
    # echo $$ >/sys/fs/cgroup/net_cls/0/tasks
    # ping -qfb 255.255.255.255 -I eth0 &>/dev/null &
    # tc qdisc add dev eth0 root handle 10: htb
    # while : ; do
    > tc filter add dev eth0 parent 10: protocol ip prio 10 handle 1: cgroup
    > tc filter delete dev eth0
    > done
    Unable to handle kernel NULL pointer dereference at virtual address 0000000000000028
    Mem abort info:
      ESR = 0x96000005
      Exception class = DABT (current EL), IL = 32 bits
      SET = 0, FnV = 0
      EA = 0, S1PTW = 0
    Data abort info:
      ISV = 0, ISS = 0x00000005
      CM = 0, WnR = 0
    user pgtable: 4k pages, 39-bit VAs, pgdp = 0000000098a7ff91
    [0000000000000028] pgd=0000000000000000, pud=0000000000000000
    Internal error: Oops: 96000005 [#1] SMP
    Modules linked in: sch_htb cls_cgroup algif_hash af_alg nls_iso8859_1 nls_cp437 vfat fat xhci_plat_hcd m25p80 spi_nor xhci_hcd mtd usbcore usb_common spi_orion sfp i2c_mv64xxx phy_generic mdio_i2c marvell10g i2c_core mvpp2 mvmdio phylink sbsa_gwdt ip_tables x_tables autofs4
    Process ping (pid: 5421, stack limit = 0x00000000b20b1505)
    CPU: 3 PID: 5421 Comm: ping Not tainted 5.1.0-rc6 #31
    Hardware name: Marvell 8040 MACCHIATOBin Double-shot (DT)
    pstate: 60000005 (nZCv daif -PAN -UAO)
    pc : cls_cgroup_classify+0x80/0xec [cls_cgroup]
    lr : cls_cgroup_classify+0x34/0xec [cls_cgroup]
    sp : ffffff8012e6b850
    x29: ffffff8012e6b850 x28: ffffffc423dd3c00
    x27: ffffff801093ebc0 x26: ffffffc425a85b00
    x25: 0000000020000000 x24: 0000000000000000
    x23: ffffff8012e6b910 x22: ffffffc428db4900
    x21: ffffff8012e6b910 x20: 0000000000100001
    x19: 0000000000000000 x18: 0000000000000000
    x17: 0000000000000000 x16: 0000000000000000
    x15: 0000000000000000 x14: 0000000000000000
    x13: 0000000000000000 x12: 000000000000001c
    x11: 0000000000000018 x10: ffffff8012e6b840
    x9 : 0000000000003580 x8 : 000000000000009d
    x7 : 0000000000000002 x6 : ffffff8012e6b860
    x5 : 000000007cd66ffe x4 : 000000009742a193
    x3 : ffffff800865b4d8 x2 : ffffff8012e6b910
    x1 : 0000000000000400 x0 : ffffffc42c38f300
    Call trace:
     cls_cgroup_classify+0x80/0xec [cls_cgroup]
     tcf_classify+0x78/0x138
     htb_enqueue+0x74/0x320 [sch_htb]
     __dev_queue_xmit+0x3e4/0x9d0
     dev_queue_xmit+0x24/0x30
     ip_finish_output2+0x2e4/0x4d0
     ip_finish_output+0x1d8/0x270
     ip_mc_output+0xa8/0x240
     ip_local_out+0x58/0x68
     ip_send_skb+0x2c/0x88
     ip_push_pending_frames+0x44/0x50
     raw_sendmsg+0x458/0x830
     inet_sendmsg+0x54/0xe8
     sock_sendmsg+0x34/0x50
     __sys_sendto+0xd0/0x120
     __arm64_sys_sendto+0x30/0x40
     el0_svc_common.constprop.0+0x88/0xf8
     el0_svc_handler+0x2c/0x38
     el0_svc+0x8/0xc
    Code: 39496001 360002a1 b9425c14 34000274 (79405260)

Fixes: ed76f5edcc ("net: sched: protect filter_chain list with filter_chain_lock mutex")
Suggested-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: Matteo Croce <mcroce@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-05 10:00:17 -07:00

227 lines
5.2 KiB
C

/*
* net/sched/cls_cgroup.c Control Group Classifier
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version
* 2 of the License, or (at your option) any later version.
*
* Authors: Thomas Graf <tgraf@suug.ch>
*/
#include <linux/module.h>
#include <linux/slab.h>
#include <linux/skbuff.h>
#include <linux/rcupdate.h>
#include <net/rtnetlink.h>
#include <net/pkt_cls.h>
#include <net/sock.h>
#include <net/cls_cgroup.h>
struct cls_cgroup_head {
u32 handle;
struct tcf_exts exts;
struct tcf_ematch_tree ematches;
struct tcf_proto *tp;
struct rcu_work rwork;
};
static int cls_cgroup_classify(struct sk_buff *skb, const struct tcf_proto *tp,
struct tcf_result *res)
{
struct cls_cgroup_head *head = rcu_dereference_bh(tp->root);
u32 classid = task_get_classid(skb);
if (unlikely(!head))
return -1;
if (!classid)
return -1;
if (!tcf_em_tree_match(skb, &head->ematches, NULL))
return -1;
res->classid = classid;
res->class = 0;
return tcf_exts_exec(skb, &head->exts, res);
}
static void *cls_cgroup_get(struct tcf_proto *tp, u32 handle)
{
return NULL;
}
static int cls_cgroup_init(struct tcf_proto *tp)
{
return 0;
}
static const struct nla_policy cgroup_policy[TCA_CGROUP_MAX + 1] = {
[TCA_CGROUP_EMATCHES] = { .type = NLA_NESTED },
};
static void __cls_cgroup_destroy(struct cls_cgroup_head *head)
{
tcf_exts_destroy(&head->exts);
tcf_em_tree_destroy(&head->ematches);
tcf_exts_put_net(&head->exts);
kfree(head);
}
static void cls_cgroup_destroy_work(struct work_struct *work)
{
struct cls_cgroup_head *head = container_of(to_rcu_work(work),
struct cls_cgroup_head,
rwork);
rtnl_lock();
__cls_cgroup_destroy(head);
rtnl_unlock();
}
static int cls_cgroup_change(struct net *net, struct sk_buff *in_skb,
struct tcf_proto *tp, unsigned long base,
u32 handle, struct nlattr **tca,
void **arg, bool ovr, bool rtnl_held,
struct netlink_ext_ack *extack)
{
struct nlattr *tb[TCA_CGROUP_MAX + 1];
struct cls_cgroup_head *head = rtnl_dereference(tp->root);
struct cls_cgroup_head *new;
int err;
if (!tca[TCA_OPTIONS])
return -EINVAL;
if (!head && !handle)
return -EINVAL;
if (head && handle != head->handle)
return -ENOENT;
new = kzalloc(sizeof(*head), GFP_KERNEL);
if (!new)
return -ENOBUFS;
err = tcf_exts_init(&new->exts, net, TCA_CGROUP_ACT, TCA_CGROUP_POLICE);
if (err < 0)
goto errout;
new->handle = handle;
new->tp = tp;
err = nla_parse_nested(tb, TCA_CGROUP_MAX, tca[TCA_OPTIONS],
cgroup_policy, NULL);
if (err < 0)
goto errout;
err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &new->exts, ovr,
true, extack);
if (err < 0)
goto errout;
err = tcf_em_tree_validate(tp, tb[TCA_CGROUP_EMATCHES], &new->ematches);
if (err < 0)
goto errout;
rcu_assign_pointer(tp->root, new);
if (head) {
tcf_exts_get_net(&head->exts);
tcf_queue_work(&head->rwork, cls_cgroup_destroy_work);
}
return 0;
errout:
tcf_exts_destroy(&new->exts);
kfree(new);
return err;
}
static void cls_cgroup_destroy(struct tcf_proto *tp, bool rtnl_held,
struct netlink_ext_ack *extack)
{
struct cls_cgroup_head *head = rtnl_dereference(tp->root);
/* Head can still be NULL due to cls_cgroup_init(). */
if (head) {
if (tcf_exts_get_net(&head->exts))
tcf_queue_work(&head->rwork, cls_cgroup_destroy_work);
else
__cls_cgroup_destroy(head);
}
}
static int cls_cgroup_delete(struct tcf_proto *tp, void *arg, bool *last,
bool rtnl_held, struct netlink_ext_ack *extack)
{
return -EOPNOTSUPP;
}
static void cls_cgroup_walk(struct tcf_proto *tp, struct tcf_walker *arg,
bool rtnl_held)
{
struct cls_cgroup_head *head = rtnl_dereference(tp->root);
if (arg->count < arg->skip)
goto skip;
if (!head)
return;
if (arg->fn(tp, head, arg) < 0) {
arg->stop = 1;
return;
}
skip:
arg->count++;
}
static int cls_cgroup_dump(struct net *net, struct tcf_proto *tp, void *fh,
struct sk_buff *skb, struct tcmsg *t, bool rtnl_held)
{
struct cls_cgroup_head *head = rtnl_dereference(tp->root);
struct nlattr *nest;
t->tcm_handle = head->handle;
nest = nla_nest_start(skb, TCA_OPTIONS);
if (nest == NULL)
goto nla_put_failure;
if (tcf_exts_dump(skb, &head->exts) < 0 ||
tcf_em_tree_dump(skb, &head->ematches, TCA_CGROUP_EMATCHES) < 0)
goto nla_put_failure;
nla_nest_end(skb, nest);
if (tcf_exts_dump_stats(skb, &head->exts) < 0)
goto nla_put_failure;
return skb->len;
nla_put_failure:
nla_nest_cancel(skb, nest);
return -1;
}
static struct tcf_proto_ops cls_cgroup_ops __read_mostly = {
.kind = "cgroup",
.init = cls_cgroup_init,
.change = cls_cgroup_change,
.classify = cls_cgroup_classify,
.destroy = cls_cgroup_destroy,
.get = cls_cgroup_get,
.delete = cls_cgroup_delete,
.walk = cls_cgroup_walk,
.dump = cls_cgroup_dump,
.owner = THIS_MODULE,
};
static int __init init_cgroup_cls(void)
{
return register_tcf_proto_ops(&cls_cgroup_ops);
}
static void __exit exit_cgroup_cls(void)
{
unregister_tcf_proto_ops(&cls_cgroup_ops);
}
module_init(init_cgroup_cls);
module_exit(exit_cgroup_cls);
MODULE_LICENSE("GPL");