can: protect the receive lists with a per-device lock
Subsystems (6)
Oliver Hartkopp
maintainerv7 cover can: protect the receive lists with a per-device lock
This series fixes a use-after-free in CAN_ISOTP and CAN_BCM that can be triggered when a CAN device is moved to another network namespace while a socket bound to this device is released [1].
The code of patches 1-4 is unchanged. Only the commit message of patch 4 has been updated after the Sashiko review of v4 to describe the actual NULL pointer dereference in can_rx_unregister() and the CAN_ISOTP path that reaches it.
The series is based on linux-can-next/testing, which also contains "can: raw: remove redundant NULL check before netdev_hold()" from [2].
The idea behind patch 5:
The receive lists of a CAN device are located in the device's can_ml_priv and travel with the device when it is moved to another network namespace. But they are protected by the per-netns rcvlists_lock, so the lock that protects a device's lists depends on the namespace the caller passes to can_rx_register() and can_rx_unregister(). can_rx_unregister() silently returns when this namespace does not match dev_net(dev) anymore.
isotp_release() and bcm_release() leave the netdev notifier list before they can serialize against the notifier. When the device changes its namespace in this window, their filters are not removed and the remaining receivers point to the freed socket or bcm op.
Taking RTNL in these release paths is no option, as CAN_RAW already holds RTNL while adding and removing up to CAN_RAW_FILTER_MAX filters per socket, which unprivileged users can trigger repeatedly.
Instead, patch 5 protects the receive lists where they are located: A spinlock in struct can_dev_rcv_lists protects the lists of a CAN device (initialized in can_set_ml_priv()) or the 'any' CAN device lists of a namespace (initialized in can_pernet_init()). can_rx_unregister() then finds the subscription independently of the current namespace of the device, which is kept accessible by the tracked device references the protocols already hold. The per-netns rcvlists_lock only protects the receive list statistics, for which CAN_RAW, CAN_BCM and CAN_ISOTP pass the namespace that has been used for can_rx_register().
As a consequence, CAN_RAW and CAN_ISOTP do not need RTNL in bind(), setsockopt() and release() anymore. bind() is serialized against NETDEV_UNREGISTER by publishing the device inside the RCU section of its lookup, as unlist_netdevice() and synchronize_net() precede NETDEV_UNREGISTER both on unregistration and on a namespace change. The notifiers recheck the binding under lock_sock(). As RTNL does not keep NETDEV_DOWN out of bind() anymore, bind() checks IFF_UP after publishing the bound device, with an smp_mb() pair against the notifiers, so a down transition during bind() is still reported with ENETDOWN.
Changes in v7 (details in patch 5):
- report ENETDOWN in bind() and ENODEV for NETDEV_UNREGISTER under lock_sock(), so ENODEV can not be overwritten by ENETDOWN (Sashiko review)
Changes in v6:
- isotp_bind() checks the device before publishing so->dev, so a failing bind() can not get a needless ENETDOWN (Sashiko review)
Changes in v5:
- report ENETDOWN for a down transition during bind() again (Sashiko review)
- comment and kernel-doc updates (Sashiko review)
- patch 4: commit message describes the actual crash path (Sashiko review)
[1] https://lore.kernel.org/linux-can/[email protected]/
[2] https://lore.kernel.org/linux-can/[email protected]/
Previous versions of patch 5:
v6: https://lore.kernel.org/linux-can/[email protected]/
v5: https://lore.kernel.org/linux-can/[email protected]/
v4: https://lore.kernel.org/linux-can/[email protected]/
v3: https://lore.kernel.org/linux-can/[email protected]/
v2: https://lore.kernel.org/linux-can/[email protected]/
v1: https://lore.kernel.org/linux-can/[email protected]/
Oliver Hartkopp (4):
can: proc: skip stats calculation after reset in can_stat_update()
can: proc: remove pointers from CAN specific proc output
can: replace ARPHRD_CAN type checks with can_get_ml_priv()
can: protect the receive lists with a per-device lock
Runyu Xiao (1):
can: proc: reset pkg_stats atomics individually
Documentation/networking/can.rst | 16 +-- include/linux/can/can-ml.h | 3 + include/linux/can/core.h | 2 +- include/net/netns/can.h | 2 +- net/can/af_can.c | 60 ++++++----- net/can/af_can.h | 2 +- net/can/bcm.c | 64 +++++------- net/can/gw.c | 9 +- net/can/isotp.c | 71 ++++++++----- net/can/j1939/main.c | 2 +- net/can/proc.c | 95 ++++++++++------- net/can/raw.c | 169 ++++++++++++++++++++----------- 12 files changed, 289 insertions(+), 206 deletions(-)
Signature · 1 line
2.43.0
Oliver Hartkopp
maintainerPATCH v7 1/5 can: proc: reset pkg_stats atomics individually
From: Runyu Xiao <[email protected]>
Commit 80b5f90158d1 ("can: statistics: use atomic access in hot path") converted several members of struct can_pkg_stats to atomic_long_t and updated the hot TX/RX and procfs read paths to use atomic_long_*() helpers. However, can_init_stats() still clears the whole struct with memset().
can_init_stats() is reached from can_stat_update() in timer context and also from the procfs reset path. Those paths can run while the TX/RX hot paths are concurrently updating rx_frames, tx_frames, matches and their *_delta counters. Hitting the whole-struct memset() in that window performs plain writes to fields that otherwise follow an atomic_long_t access contract, which can lose or mix live statistics updates.
This issue was found by source-level API-misuse analysis looking for whole-object resets left behind after atomic_long_t conversions, then manually audited on Linux v6.18.21.
Replace the whole-struct memset() with a helper that resets the atomic_long_t counters via atomic_long_set() and clears the derived scalar statistics explicitly. This preserves the existing reset semantics for scalar fields while restoring atomic access discipline for the live counters.
Build-tested by compiling net/can/proc.o on x86_64 netdev/main. Runtime-tested with a QEMU + vcan setup on Linux v6.18.21 by driving concurrent traffic and reset_stats reads, which reproduced inconsistent exported statistics before the fix.
[Hartkopp] can_stat_update() also read the *_delta counters at the beginning and cleared them with atomic_long_set() at the end. Increments from the hot path in between were lost for both the current and the next interval. Use atomic_long_xchg() to fetch and clear the *_delta counters in one atomic step instead.
--- net/can/proc.c | 39 +++++++++++++++++++++++++++++---------- 1 file changed, 29 insertions(+), 10 deletions(-)
net/can/proc.c +29 -10
@@ -74,20 +74,43 @@ static const char rx_list_name[][8] = {/* * af_can statistics stuff */ +static void can_reset_pkg_stats (struct can_pkg_stats *pkg_stats) +{ +atomic_long_set (&pkg_stats->rx_frames,0 ); +atomic_long_set (&pkg_stats->tx_frames,0 ); +atomic_long_set (&pkg_stats->matches,0 ); + + pkg_stats->total_rx_rate =0 ; + pkg_stats->total_tx_rate =0 ; + pkg_stats->total_rx_match_ratio =0 ; + + pkg_stats->current_rx_rate =0 ; + pkg_stats->current_tx_rate =0 ; + pkg_stats->current_rx_match_ratio =0 ; + + pkg_stats->max_rx_rate =0 ; + pkg_stats->max_tx_rate =0 ; + pkg_stats->max_rx_match_ratio =0 ; + +atomic_long_set (&pkg_stats->rx_frames_delta,0 ); +atomic_long_set (&pkg_stats->tx_frames_delta,0 ); +atomic_long_set (&pkg_stats->matches_delta,0 ); +} +static void can_init_stats (struct net *net) {struct can_pkg_stats *pkg_stats = net->can.pkg_stats;struct can_rcv_lists_stats *rcv_lists_stats = net->can.rcv_lists_stats;/* -* This memset function is called from a timer context (when +* This stats reset is called from a timer context (when * can_stattimer is active which is the default) OR in a process * context (reading the proc_fs when can_stattimer is disabled). */ -memset (pkg_stats,0 ,sizeof (struct can_pkg_stats )); +can_reset_pkg_stats (pkg_stats); pkg_stats->jiffies_init = jiffies; rcv_lists_stats->stats_reset++;if (user_reset) {
@@ -119,13 +142,14 @@ void can_stat_update(struct timer_list *t)unsigned long j = jiffies;/* snapshot */ long rx_frames =atomic_long_read (&pkg_stats->rx_frames);long tx_frames =atomic_long_read (&pkg_stats->tx_frames);long matches =atomic_long_read (&pkg_stats->matches); -long rx_frames_delta =atomic_long_read (&pkg_stats->rx_frames_delta); -long tx_frames_delta =atomic_long_read (&pkg_stats->tx_frames_delta); -long matches_delta =atomic_long_read (&pkg_stats->matches_delta); +/* fetch and clear values for 'current rate' calculation */ +long rx_frames_delta =atomic_long_xchg (&pkg_stats->rx_frames_delta,0 ); +long tx_frames_delta =atomic_long_xchg (&pkg_stats->tx_frames_delta,0 ); +long matches_delta =atomic_long_xchg (&pkg_stats->matches_delta,0 );/* restart counting in timer context on user request */ if (user_reset)can_init_stats (net);
@@ -170,15 +194,10 @@ void can_stat_update(struct timer_list *t) pkg_stats->max_rx_rate = pkg_stats->current_rx_rate;if (pkg_stats->max_rx_match_ratio < pkg_stats->current_rx_match_ratio) pkg_stats->max_rx_match_ratio = pkg_stats->current_rx_match_ratio; -/* clear values for 'current rate' calculation */ -atomic_long_set (&pkg_stats->tx_frames_delta,0 ); -atomic_long_set (&pkg_stats->rx_frames_delta,0 ); -atomic_long_set (&pkg_stats->matches_delta,0 ); -/* restart timer (one second) */ mod_timer (&net->can.stattimer,round_jiffies (jiffies +HZ )); }/*
Signature · 1 line
2.43.0
Oliver Hartkopp
maintainerPATCH v7 2/5 can: proc: skip stats calculation after reset in can_stat_update()
In reply to Oliver HartkoppCommit 80b5f90158d1 ("can: statistics: use atomic access in hot path") introduced local snapshots of the statistics counters at the beginning of can_stat_update(). When can_init_stats() resets the statistics, these stale pre-reset snapshots are still used for the following total/current /max calculations and overwrite the freshly reset values. Additionally a jiffies tick between the snapshot and the reset could trigger a second can_init_stats() call.
Merge the reset conditions into a single check and skip the calculation for the current interval when the statistics have been reset.
--- net/can/proc.c | 30 ++++++++++++------------------ 1 file changed, 12 insertions(+), 18 deletions(-)
net/can/proc.c +12 -18
@@ -147,29 +147,22 @@ void can_stat_update(struct timer_list *t)/* fetch and clear values for 'current rate' calculation */ long rx_frames_delta =atomic_long_xchg (&pkg_stats->rx_frames_delta,0 );long tx_frames_delta =atomic_long_xchg (&pkg_stats->tx_frames_delta,0 );long matches_delta =atomic_long_xchg (&pkg_stats->matches_delta,0 ); -/* restart counting in timer context on user request */ -if (user_reset) -can_init_stats (net); - -/* restart counting on jiffies overflow */ -if (j < pkg_stats->jiffies_init) -can_init_stats (net); - -/* prevent overflow in calc_rate() */ -if (rx_frames > (LONG_MAX /HZ )) -can_init_stats (net); - -/* prevent overflow in calc_rate() */ -if (tx_frames > (LONG_MAX /HZ )) -can_init_stats (net); - -/* matches overflow - very improbable */ -if (matches > (LONG_MAX /100 )) +/* +* Restart counting in timer context on user request, on jiffies +* overflow or to prevent overflows in calc_rate() and in the +* match ratio calculation. The local snapshots are stale after +* the reset, so skip the calculation for this interval. +*/ +if (user_reset || j < pkg_stats->jiffies_init || + rx_frames > (LONG_MAX /HZ ) || tx_frames > (LONG_MAX /HZ ) || + matches > (LONG_MAX /100 )) {can_init_stats (net); + goto out; + }/* calc total values */ if (rx_frames) pkg_stats->total_rx_match_ratio = (matches *100 ) / rx_frames;
@@ -194,10 +187,11 @@ void can_stat_update(struct timer_list *t) pkg_stats->max_rx_rate = pkg_stats->current_rx_rate;if (pkg_stats->max_rx_match_ratio < pkg_stats->current_rx_match_ratio) pkg_stats->max_rx_match_ratio = pkg_stats->current_rx_match_ratio; +out:/* restart timer (one second) */ mod_timer (&net->can.stattimer,round_jiffies (jiffies +HZ )); }/*
Signature · 1 line
2.43.0
Oliver Hartkopp
maintainerPATCH v7 3/5 can: proc: remove pointers from CAN specific proc output
In reply to Oliver HartkoppThe proc content in /proc/net/can/ and /proc/net/can-bcm/ is intended to check the internal filter lists (af_can, can_raw) and the efficiency and functionality of can_bcm jobs. While exposing kernel addresses was acceptable when this code was written the times have changed and multiple attempts have been made to hash or remove such now sensitive data.
This patch removes the disclosure of pointers and instead provides the function names and sock inode numbers when available. When the function name can not be resolved (e.g. KALLSYMS=n or stripped module symbols) %ps would fall back to the raw text address. Therefore (unknown) is printed instead in this case. As there's no known tooling around the CAN specific proc output breaking the ABI with this rework creates no issue.
--- Documentation/networking/can.rst | 16 ++++++++++------ include/linux/can/core.h | 2 +- net/can/af_can.c | 6 +++--- net/can/af_can.h | 2 +- net/can/bcm.c | 14 +++++++------- net/can/gw.c | 2 +- net/can/isotp.c | 4 ++-- net/can/j1939/main.c | 2 +- net/can/proc.c | 26 ++++++++++++++++---------- net/can/raw.c | 4 ++-- 10 files changed, 44 insertions(+), 34 deletions(-)
Documentation/networking/can.rst +10 -6
@@ -1038,21 +1038,25 @@ procfs content As described in :ref:`socketcan-receive-lists` the SocketCAN core uses several filter lists to deliver received CAN frames to CAN protocol modules. These receive lists, their filters and the count of filter matches can be checked in the appropriate receive list. All entries contain the -device and a protocol module identifier:: +CAN network device, the filter (can_id/can_mask), the number of filter +matches, the inode number of the subscribing socket and the name of the +receive function. A sock_inode of 0 denotes a kernel internal subscriber +without a socket (e.g. CAN gateway or J1939). When the function name can +not be resolved (e.g. with CONFIG_KALLSYMS=n) "(unknown)" is printed:: foo@bar:~$ cat /proc/net/can/rcvlist_all receive list 'rx_all': - (vcan3: no entry) - (vcan2: no entry) - (vcan1: no entry) - device can_id can_mask function userdata matches ident - vcan0 000 00000000 f88e6370 f6c6f400 0 raw (any: no entry) + device can_id can_mask matches sock_inode function + vcan0 000 00000000 0 63586 raw_rcv [can_raw] + (vcan1: no entry) + (vcan2: no entry) + (vcan3: no entry) In this example an application requests any CAN traffic from vcan0:: rcvlist_all - list for unfiltered entries (no filter operations) rcvlist_eff - list for single extended frame (EFF) entries
include/linux/can/core.h +1 -1
@@ -49,11 +49,11 @@ extern int can_proto_register(const struct can_proto *cp);extern void can_proto_unregister (const struct can_proto *cp);int can_rx_register (struct net *net,struct net_device *dev,canid_t can_id,canid_t mask,void (*func)(struct sk_buff *,void *), -void *data,char *ident,struct sock *sk); +void *data,u64 ino,struct sock *sk);extern void can_rx_unregister (struct net *net,struct net_device *dev,canid_t can_id,canid_t mask,void (*func)(struct sk_buff *,void *),void *data);
net/can/af_can.c +3 -3
@@ -416,11 +416,11 @@ static struct hlist_head *can_rcv_list_find(canid_t *can_id, canid_t *mask,* @dev: pointer to netdevice (NULL => subscribe from 'all' CAN devices list) * @can_id: CAN identifier (see description) * @mask: CAN mask (see description) * @func: callback function on filter match * @data: returned parameter for callback function -* @ident: string for calling module identification +* @ino: inode number of sock (0 = unknown) * @sk: socket pointer (might be NULL) * * Description: * Invokes the callback function with the received sk_buff and the given * parameter 'data' on a matching receive filter. A filter matches, when
@@ -441,11 +441,11 @@ static struct hlist_head *can_rcv_list_find(canid_t *can_id, canid_t *mask,* -ENOMEM on missing cache mem to create subscription entry * -ENODEV unknown device */ int can_rx_register (struct net *net,struct net_device *dev,canid_t can_id,canid_t mask,void (*func)(struct sk_buff *,void *), -void *data,char *ident,struct sock *sk) +void *data,u64 ino,struct sock *sk) {struct receiver *rcv;struct hlist_head *rcv_list;struct can_dev_rcv_lists *dev_rcv_lists;struct can_rcv_lists_stats *rcv_lists_stats = net->can.rcv_lists_stats;
@@ -470,11 +470,11 @@ int can_rx_register(struct net *net, struct net_device *dev, canid_t can_id, rcv->can_id = can_id; rcv->mask = mask;atomic_long_set (&rcv->matches,0 ); rcv->func = func; rcv->data = data; - rcv->ident = ident; + rcv->ino = ino; rcv->sk = sk;hlist_add_head_rcu (&rcv->list, rcv_list); dev_rcv_lists->entries++;
net/can/af_can.h +1 -1
@@ -53,11 +53,11 @@ struct receiver {canid_t can_id;canid_t mask;atomic_long_t matches;void (*func)(struct sk_buff *skb,void *data);void *data; -char *ident; +u64 ino;struct sock *sk;struct rcu_head rcu; };/* statistic structures */
net/can/bcm.c +7 -7
@@ -218,13 +218,11 @@ static int bcm_proc_show(struct seq_file *m, void *v)struct net *net = m->private;struct sock *sk = (struct sock *)pde_data (m->file->f_inode);struct bcm_sock *bo =bcm_sk (sk);struct bcm_op *op; -seq_printf (m,">>> socket %pK" , sk->sk_socket); -seq_printf (m," / sk %pK" , sk); -seq_printf (m," / bo %pK" , bo); +seq_printf (m,">>> sock inode %s" , bo->procname);seq_printf (m," / dropped %lu" , bo->dropped_usr_msgs);seq_printf (m," / bound %s" ,bcm_proc_getifname (net, ifname, bo->ifindex));seq_printf (m," <<<\n" );rcu_read_lock ();
@@ -1534,11 +1532,11 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg,if (dev) { err =can_rx_register (sock_net (sk), dev, op->can_id,REGMASK (op->can_id), bcm_rx_handler, op, -"bcm" , sk); +sock_i_ino (sk), sk);/* keep a tracked reference so that a later * unregister can safely reach the device even * if a concurrent NETDEV_UNREGISTER has * already unlisted it by ifindex
@@ -1558,11 +1556,12 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, } }else { err =can_rx_register (sock_net (sk),NULL , op->can_id,REGMASK (op->can_id), - bcm_rx_handler, op,"bcm" , sk); + bcm_rx_handler, op, +sock_i_ino (sk), sk); }if (err) {/* newly created bcm rx op is broken -> remove it */ if (new_op) {
@@ -2038,12 +2037,13 @@ static int bcm_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int bo->ifindex =0 ; }#if IS_ENABLED (CONFIG_PROC_FS )if (net->can.bcmproc_dir) { -/* unique socket address as filename */ -sprintf (bo->procname,"%llu" ,sock_i_ino (sk)); +/* use unique socket inode number as filename */ +snprintf (bo->procname,sizeof (bo->procname), +"%llu" ,sock_i_ino (sk)); bo->bcm_proc_read =proc_create_net_single (bo->procname,0644 , net->can.bcmproc_dir, bcm_proc_show, sk);if (!bo->bcm_proc_read) { ret = -ENOMEM ;
net/can/gw.c +1 -1
@@ -577,11 +577,11 @@ static void can_can_gw_rcv(struct sk_buff *skb, void *data)static inline int cgw_register_filter (struct net *net,struct cgw_job *gwj) {return can_rx_register (net, gwj->src.dev, gwj->ccgw.filter.can_id, gwj->ccgw.filter.can_mask, can_can_gw_rcv, - gwj,"gw" ,NULL ); + gwj,0 ,NULL ); }static inline void cgw_unregister_filter (struct net *net,struct cgw_job *gwj) { can_rx_unregister(net, gwj->src.dev, gwj->ccgw.filter.can_id,
net/can/isotp.c +2 -2
@@ -1623,18 +1623,18 @@ static int isotp_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int l ifindex = dev->ifindex;if (isotp_register_rxid (so))can_rx_register (net, dev, rx_id,SINGLE_MASK (rx_id), - isotp_rcv, sk,"isotp" , sk); + isotp_rcv, sk,sock_i_ino (sk), sk);/* no consecutive frame echo skb in flight */ WRITE_ONCE (so->cfecho,0 );/* register for echo skb's */ can_rx_register (net, dev, tx_id,SINGLE_MASK (tx_id), - isotp_rcv_echo, sk,"isotpe" , sk); + isotp_rcv_echo, sk,sock_i_ino (sk), sk);/* switch to new settings */ so->ifindex = ifindex; so->rxid = rx_id; so->txid = tx_id;
net/can/j1939/main.c +1 -1
@@ -182,11 +182,11 @@ static int j1939_can_rx_register(struct j1939_priv *priv)struct net_device *ndev = priv->ndev;int ret;j1939_priv_get (priv); ret =can_rx_register (dev_net (ndev), ndev,J1939_CAN_ID ,J1939_CAN_MASK , - j1939_can_recv, priv,"j1939" ,NULL ); + j1939_can_recv, priv,0 ,NULL );if (ret <0 ) {j1939_priv_put (priv);return ret; }
net/can/proc.c +16 -10
@@ -202,30 +202,36 @@ static void can_print_rcvlist(struct seq_file *m, struct hlist_head *rx_list, struct net_device *dev) {struct receiver *r;hlist_for_each_entry_rcu (r, rx_list, list) { +char symcheck[4 ];char *fmt = (r->can_id &CAN_EFF_FLAG )? -" %-5s %08x %08x %pK %pK %8ld %s\n" : -" %-5s %03x %08x %pK %pK %8ld %s\n" ; +" %6s %08x %08x %8ld %10llu" : +" %6s %03x %08x %8ld %10llu" ;seq_printf (m, fmt,DNAME (dev), r->can_id, r->mask, - r->func, r->data,atomic_long_read (&r->matches), - r->ident); +atomic_long_read (&r->matches), r->ino); + +/* %ps falls back to the raw text address when the symbol +* can not be resolved (KALLSYMS=n, stripped module symbols) +*/ +snprintf (symcheck,sizeof (symcheck),"%ps" , r->func); +if (strncmp (symcheck,"0x" ,2 )) +seq_printf (m," %ps\n" , r->func); +else +seq_puts (m," (unknown)\n" ); } }static void can_print_recv_banner (struct seq_file *m) {/* -* can1. 00000000 00000000 00000000 -* ....... 0 tp20 +* device can_id can_mask matches sock_inode function +* vcan0 000 00000000 0 43798 raw_rcv [can_raw] */ -if (IS_ENABLED (CONFIG_64BIT )) -seq_puts (m," device can_id can_mask function userdata matches ident\n" ); -else -seq_puts (m," device can_id can_mask function userdata matches ident\n" ); +seq_puts (m," device can_id can_mask matches sock_inode function\n" ); }static int can_stats_proc_show (struct seq_file *m,void *v) {struct net *net = m->private;
net/can/raw.c +2 -2
@@ -224,11 +224,11 @@ static int raw_enable_filters(struct net *net, struct net_device *dev,int i;for (i =0 ; i < count; i++) { err =can_rx_register (net, dev, filter[i].can_id, filter[i].can_mask, - raw_rcv, sk,"raw" , sk); + raw_rcv, sk,sock_i_ino (sk), sk);if (err) {/* clean up successfully registered filters */ while (--i >=0 ) can_rx_unregister(net, dev, filter[i].can_id, filter[i].can_mask,
@@ -245,11 +245,11 @@ static int raw_enable_errfilter(struct net *net, struct net_device *dev, {int err =0 ;if (err_mask) err =can_rx_register (net, dev,0 , err_mask |CAN_ERR_FLAG , - raw_rcv, sk,"raw" , sk); + raw_rcv, sk,sock_i_ino (sk), sk);return err; } static void raw_disable_filters(struct net *net, struct net_device *dev,
Signature · 1 line
2.43.0
Oliver Hartkopp
maintainerPATCH v7 4/5 can: replace ARPHRD_CAN type checks with can_get_ml_priv()
In reply to Oliver HartkoppCommit 4e096a18867a ("net: introduce CAN specific pointer in the struct net_device") introduced an explicit way to assign the midlayer private pointer (dev->ml_priv) to named users like ML_PRIV_CAN. A valid pointer returned by can_get_ml_priv() therefore identifies a real CAN device.
The referenced commit only converted the j1939 and proc code to this check, while the other parts of the CAN subsystem still rely on dev->type == ARPHRD_CAN.
The TUNSETLINK ioctl allows a down TUN/TAP device to set its hardware type to ARPHRD_CAN while dev->ml_priv remains NULL. Such a repurposed netdevice passes the ARPHRD_CAN checks.
Commit 8aa59e355949 ("can: af_can: fix NULL pointer dereference in can_rx_register()") already made can_rx_register() reject such devices. But can_rx_unregister() still only checks dev->type and dereferences the NULL ml_priv when it walks the receive lists of the device.
CAN_ISOTP reaches this path: isotp_bind() ignores the return values of can_rx_register() and binds the socket anyway. When the socket is released or the device is unregistered, isotp_release() and isotp_notify() remove the filters that have never been registered and trigger the NULL pointer dereference in can_rx_unregister(). This only requires owning a TUN/TAP device or CAP_NET_ADMIN in an unprivileged user namespace.
Convert the remaining ARPHRD_CAN checks in af_can, bcm, gw, isotp and raw to can_get_ml_priv().
--- net/can/af_can.c | 15 ++++++--------- net/can/bcm.c | 7 ++++--- net/can/gw.c | 7 ++++--- net/can/isotp.c | 5 +++-- net/can/raw.c | 4 ++-- 5 files changed, 19 insertions(+), 19 deletions(-)
net/can/af_can.c +6 -9
@@ -224,11 +224,11 @@ int can_send(struct sk_buff *skb, int loop)if (unlikely (skb->len >READ_ONCE (skb->dev->mtu))) { err = -EMSGSIZE ; goto inval_skb; } -if (unlikely (skb->dev->type !=ARPHRD_CAN )) { +if (unlikely (!can_get_ml_priv (skb->dev))) { err = -EPERM ; goto inval_skb; }if (unlikely (!(skb->dev->flags &IFF_UP ))) {
@@ -450,11 +450,11 @@ int can_rx_register(struct net *net, struct net_device *dev, canid_t can_id,struct can_dev_rcv_lists *dev_rcv_lists;struct can_rcv_lists_stats *rcv_lists_stats = net->can.rcv_lists_stats;/* insert new receiver (dev,canid,mask) -> (func,data) */ -if (dev && (dev->type !=ARPHRD_CAN || !can_get_ml_priv (dev))) +if (dev && !can_get_ml_priv (dev))return -ENODEV ;if (dev && !net_eq (net,dev_net (dev)))return -ENODEV ;
@@ -517,11 +517,11 @@ void can_rx_unregister(struct net *net, struct net_device *dev, canid_t can_id,struct receiver *rcv =NULL ;struct hlist_head *rcv_list;struct can_rcv_lists_stats *rcv_lists_stats = net->can.rcv_lists_stats;struct can_dev_rcv_lists *dev_rcv_lists; -if (dev && dev->type !=ARPHRD_CAN ) +if (dev && !can_get_ml_priv (dev))return ;if (dev && !net_eq (net,dev_net (dev)))return ;
@@ -682,12 +682,11 @@ static void can_receive(struct sk_buff *skb, struct net_device *dev)static int can_rcv (struct sk_buff *skb,struct net_device *dev,struct packet_type *pt,struct net_device *orig_dev) {struct can_skb_ext *csx =can_skb_ext_find (skb); -if (unlikely (dev->type !=ARPHRD_CAN || !can_get_ml_priv (dev) || - !csx || !can_is_can_skb (skb))) { +if (unlikely (!can_get_ml_priv (dev) || !csx || !can_is_can_skb (skb))) {pr_warn_once ("PF_CAN: dropped non conform CAN skbuff: dev type %d, len %d\n" , dev->type, skb->len);kfree_skb_reason (skb,SKB_DROP_REASON_CAN_RX_INVALID_FRAME );return NET_RX_DROP ;
@@ -708,12 +707,11 @@ static int can_rcv(struct sk_buff *skb, struct net_device *dev,static int canfd_rcv (struct sk_buff *skb,struct net_device *dev,struct packet_type *pt,struct net_device *orig_dev) {struct can_skb_ext *csx =can_skb_ext_find (skb); -if (unlikely (dev->type !=ARPHRD_CAN || !can_get_ml_priv (dev) || - !csx || !can_is_canfd_skb (skb))) { +if (unlikely (!can_get_ml_priv (dev) || !csx || !can_is_canfd_skb (skb))) {pr_warn_once ("PF_CAN: dropped non conform CAN FD skbuff: dev type %d, len %d\n" , dev->type, skb->len);kfree_skb_reason (skb,SKB_DROP_REASON_CANFD_RX_INVALID_FRAME );return NET_RX_DROP ;
@@ -734,12 +732,11 @@ static int canfd_rcv(struct sk_buff *skb, struct net_device *dev,static int canxl_rcv (struct sk_buff *skb,struct net_device *dev,struct packet_type *pt,struct net_device *orig_dev) {struct can_skb_ext *csx =can_skb_ext_find (skb); -if (unlikely (dev->type !=ARPHRD_CAN || !can_get_ml_priv (dev) || - !csx || !can_is_canxl_skb (skb))) { +if (unlikely (!can_get_ml_priv (dev) || !csx || !can_is_canxl_skb (skb))) {pr_warn_once ("PF_CAN: dropped non conform CAN XL skbuff: dev type %d, len %d\n" , dev->type, skb->len);kfree_skb_reason (skb,SKB_DROP_REASON_CANXL_RX_INVALID_FRAME );return NET_RX_DROP ;
net/can/bcm.c +4 -3
@@ -52,10 +52,11 @@#include <linux/netdevice.h> #include <linux/socket.h> #include <linux/if_arp.h> #include <linux/skbuff.h> #include <linux/can.h> +#include <linux/can/can-ml.h> #include <linux/can/core.h> #include <linux/can/skb.h> #include <linux/can/bcm.h> #include <linux/slab.h> #include <linux/workqueue.h>
@@ -1716,11 +1717,11 @@ static int bcm_sendmsg(struct socket *sock, struct msghdr *msg, size_t size)if (!dev) { ret = -ENODEV ; goto out_release; } -if (dev->type !=ARPHRD_CAN ) { +if (!can_get_ml_priv (dev)) {dev_put (dev); ret = -ENODEV ; goto out_release; }
@@ -1864,11 +1865,11 @@ static void bcm_notify(struct bcm_sock *bo, unsigned long msg,static int bcm_notifier (struct notifier_block *nb,unsigned long msg,void *ptr) {struct net_device *dev =netdev_notifier_info_to_dev (ptr); -if (dev->type !=ARPHRD_CAN ) +if (!can_get_ml_priv (dev))return NOTIFY_DONE ;if (msg !=NETDEV_UNREGISTER && msg !=NETDEV_DOWN )return NOTIFY_DONE ;if (unlikely (bcm_busy_notifier))/* Check for reentrant bug. */ return NOTIFY_DONE ;
@@ -2021,11 +2022,11 @@ static int bcm_connect(struct socket *sock, struct sockaddr_unsized *uaddr, int dev =dev_get_by_index (net, addr->can_ifindex);if (!dev) { ret = -ENODEV ; goto fail; } -if (dev->type !=ARPHRD_CAN ) { +if (!can_get_ml_priv (dev)) {dev_put (dev); ret = -ENODEV ; goto fail; }
net/can/gw.c +4 -3
@@ -50,10 +50,11 @@#include <linux/net.h> #include <linux/netdevice.h> #include <linux/if_arp.h> #include <linux/skbuff.h> #include <linux/can.h> +#include <linux/can/can-ml.h> #include <linux/can/core.h> #include <linux/can/skb.h> #include <linux/can/gw.h> #include <net/can.h> #include <net/rtnetlink.h>
@@ -610,11 +611,11 @@ static int cgw_notifier(struct notifier_block *nb, unsigned long msg, void *ptr) { struct net_device *dev = netdev_notifier_info_to_dev(ptr);struct net *net =dev_net (dev); -if (dev->type !=ARPHRD_CAN ) +if (!can_get_ml_priv (dev))return NOTIFY_DONE ;if (msg ==NETDEV_UNREGISTER ) {struct cgw_job *gwj =NULL ;struct hlist_node *nx;
@@ -1161,19 +1162,19 @@ static int cgw_create_job(struct sk_buff *skb, struct nlmsghdr *nlh, gwj->src.dev =__dev_get_by_index (net, gwj->ccgw.src_idx);if (!gwj->src.dev) goto out; -if (gwj->src.dev->type !=ARPHRD_CAN ) +if (!can_get_ml_priv (gwj->src.dev)) goto out; gwj->dst.dev =__dev_get_by_index (net, gwj->ccgw.dst_idx);if (!gwj->dst.dev) goto out; -if (gwj->dst.dev->type !=ARPHRD_CAN ) +if (!can_get_ml_priv (gwj->dst.dev)) goto out;/* is sending the skb back to the incoming interface intended? */ if (gwj->src.dev == gwj->dst.dev && !(gwj->flags &CGW_FLAGS_CAN_IIF_TX_OK )) {
net/can/isotp.c +3 -2
@@ -63,10 +63,11 @@#include <linux/netdevice.h> #include <linux/socket.h> #include <linux/if_arp.h> #include <linux/skbuff.h> #include <linux/can.h> +#include <linux/can/can-ml.h> #include <linux/can/core.h> #include <linux/can/skb.h> #include <linux/can/isotp.h> #include <linux/slab.h> #include <net/can.h>
@@ -1608,11 +1609,11 @@ static int isotp_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int l dev =dev_get_by_index (net, addr->can_ifindex);if (!dev) { err = -ENODEV ; goto out; } -if (dev->type !=ARPHRD_CAN ) { +if (!can_get_ml_priv (dev)) { err = -ENODEV ; goto out_put_dev; }if (READ_ONCE (dev->mtu) < so->ll.mtu) { err = -EINVAL ;
@@ -1895,11 +1896,11 @@ static void isotp_notify(struct isotp_sock *so, unsigned long msg,static int isotp_notifier (struct notifier_block *nb,unsigned long msg,void *ptr) {struct net_device *dev =netdev_notifier_info_to_dev (ptr); -if (dev->type !=ARPHRD_CAN ) +if (!can_get_ml_priv (dev))return NOTIFY_DONE ;if (msg !=NETDEV_UNREGISTER && msg !=NETDEV_DOWN )return NOTIFY_DONE ;if (unlikely (isotp_busy_notifier))/* Check for reentrant bug. */ return NOTIFY_DONE ;
net/can/raw.c +2 -2
@@ -346,11 +346,11 @@ static void raw_notify(struct raw_sock *ro, unsigned long msg,static int raw_notifier (struct notifier_block *nb,unsigned long msg,void *ptr) {struct net_device *dev =netdev_notifier_info_to_dev (ptr); -if (dev->type !=ARPHRD_CAN ) +if (!can_get_ml_priv (dev))return NOTIFY_DONE ;if (msg !=NETDEV_UNREGISTER && msg !=NETDEV_DOWN )return NOTIFY_DONE ;if (unlikely (raw_busy_notifier))/* Check for reentrant bug. */ return NOTIFY_DONE ;
@@ -489,11 +489,11 @@ static int raw_bind(struct socket *sock, struct sockaddr_unsized *uaddr, int len dev =dev_get_by_index (sock_net (sk), addr->can_ifindex);if (!dev) { err = -ENODEV ; goto out; } -if (dev->type !=ARPHRD_CAN ) { +if (!can_get_ml_priv (dev)) { err = -ENODEV ; goto out_put_dev; }if (!(dev->flags &IFF_UP ))
Signature · 1 line
2.43.0
Oliver Hartkopp
maintainerPATCH v7 5/5 can: protect the receive lists with a per-device lock
In reply to Oliver HartkoppThe receive lists of a CAN device are located in the device's can_ml_priv and travel with the device when it is moved to another network namespace. But they are protected by the per-netns rcvlists_lock. So can_rx_unregister() needs dev_net(dev) to pick the right lock and silently returns when the passed namespace does not match dev_net(dev) anymore.
isotp_release() and bcm_release() leave their notifier list before they take any lock that serializes against the netdev notifier, which runs under RTNL. When the bound device is moved to another namespace in this window, the notifier skips the socket and the release path has to remove the filters itself. isotp_release() passes the socket's namespace, which does not match dev_net(dev) anymore. bcm_rx_unreg() passes dev_net(dev), which can_rx_unregister() reads again for its check, so a namespace change between both reads makes the check fail as well. In both cases can_rx_unregister() silently returns. As can_rx_register() stores rcv->sk without a reference, the remaining receivers point to the freed socket (isotp) or the freed op (bcm):
BUG: KASAN: use-after-free in isotp_rcv+0x1570/0x24d0 Read of size 1 at addr ffff888118130552 by task isotp_ns_uaf/578 can_rcv_filter+0x4af/0x8c0 can_receive+0x28d/0x3c0 can_rcv+0x2a9/0x310 __netif_receive_skb_one_core+0x21a/0x260 process_backlog+0x210/0x760
Besides this, the lists of a device are protected by the rcvlists_lock of the namespace that is passed in. Callers which pass dev_net(dev) read it at different times, so the same lists may be modified under the locks of different namespaces while the device is moved.
Taking RTNL in these release paths is no option: raw_bind(), raw_setsockopt() and raw_release() already hold RTNL while adding and removing up to CAN_RAW_FILTER_MAX filters per socket, which unprivileged users can trigger repeatedly and which makes RTNL a global bottleneck.
Instead fix the root cause and add a spinlock to struct can_dev_rcv_lists, which protects the receive lists of a CAN device or the 'any' CAN device lists of a namespace. It is initialized in can_set_ml_priv(), which is used by all CAN network devices. With this, can_rx_unregister() finds the subscription independently of the current namespace of the device. The protocols keep the device valid with their tracked references. The per-netns rcvlists_lock only protects the receive list statistics anymore, and CAN_RAW, CAN_BCM and CAN_ISOTP pass the namespace that has been used for can_rx_register().
So isotp_release() and bcm_release() now remove their filters correctly without RTNL, and raw_release(), raw_setsockopt(), raw_bind() and isotp_bind() do not need RTNL anymore and add and remove their filters under lock_sock(). To serialize bind() with NETDEV_UNREGISTER, bind() publishes the device inside the RCU section of its lookup, and raw_notify() and isotp_notify() recheck the bound device under lock_sock(). As RTNL does not keep NETDEV_DOWN out of bind() anymore, bind() checks IFF_UP after publishing the bound device, with a barrier pair against raw_notify() and isotp_notify(), so a down transition during bind() is still reported with ENETDOWN.
---
Changes in v7:
- raw_bind() / isotp_bind(): report ENETDOWN before release_sock(), so that a concurrent NETDEV_UNREGISTER waiting for lock_sock() in raw_notify() / isotp_notify() can not have its ENODEV overwritten by ENETDOWN (Sashiko review)
- raw_notify() / isotp_notify(): report ENODEV for NETDEV_UNREGISTER before release_sock() for consistency (Sashiko review)
- v6: https://lore.kernel.org/linux-can/[email protected]/
Changes in v6:
- isotp_bind(): check the device (CAN device, MTU) before publishing so->dev, so bind() can not fail after so->dev has become visible to isotp_notify(). A failing bind() can not get a needless ENETDOWN from
a concurrent NETDEV_DOWN anymore (Sashiko review)
- v5: https://lore.kernel.org/linux-can/[email protected]/
Changes in v5:
- raw_bind() / isotp_bind(): check IFF_UP after publishing the bound device, with an smp_mb() pair against raw_notify() / isotp_notify() for NETDEV_DOWN, so a down transition during bind() is reported with ENETDOWN again, as with RTNL before (Sashiko review). isotp_bind() already published so->dev before the IFF_UP check, but without a barrier the load could still pass the store.
- raw_bind() no longer reports ENETDOWN when bind() fails
- rcvlists_lock comment names the rcv_entries(_max) stats (Sashiko review)
- can_rx_unregister() kernel-doc and commit message: @net should be the namespace of can_rx_register(), otherwise only the statistics become imprecise (j1939, Sashiko review)
- v4: https://lore.kernel.org/linux-can/[email protected]/
Changes in v4:
- rebased on linux-can-next/testing: the context in raw_bind() changed with "can: raw: remove redundant NULL check before netdev_hold()", which made v3 fail to apply
- added "can: proc: skip stats calculation after reset in can_stat_update()" from linux-can-next/testing to the base sequence
- no functional change
- v3: https://lore.kernel.org/linux-can/[email protected]/
Changes in v3:
- not RFC anymore, sent as new series with the patches 1-3 from the linux-can-next pull request to build on top of mainline (see cover letter)
- commit message: describe the isotp and bcm cases separately (bcm passes dev_net(dev), which can_rx_unregister() reads again for its check) and mention that the lists of a device could be modified under the locks of different namespaces
- can_rx_unregister() kernel-doc: @net also selects the 'any' CAN device lists for @dev == NULL
- v2: https://lore.kernel.org/linux-can/[email protected]/
Changes in v2:
- v1 relied on the NETDEV_UNREGISTER rebroadcast in netdev_wait_allrefs_any() to catch a bind() racing with the notifier, which is not sent on a namespace change. raw_bind() and isotp_bind() now publish the device inside the RCU section of dev_get_by_index_rcu() (CAN_RAW: new ro->bind_dev hint with smp_store_release() / smp_load_acquire(), CAN_ISOTP: so->dev is published early and reset when bind() fails)
- moved the implementation details from the commit message to the notes
- shortened the code comments
- replaced the rebind rate measurement with an RTNL latency measurement
- v1: https://lore.kernel.org/linux-can/[email protected]/
Notes:
- bind() vs. NETDEV_UNREGISTER: Both on unregistration and on a namespace change, unlist_netdevice() and synchronize_net() precede NETDEV_UNREGISTER. bind() publishes the device (ro->bind_dev / so->dev) inside the RCU section of dev_get_by_index_rcu(). So either the notifier sees it after synchronize_net() and handles the socket under lock_sock(), or the lookup does not find the unlisted device and bind() fails with -ENODEV. The NETDEV_UNREGISTER rebroadcast in netdev_wait_allrefs_any() is not relied on, as it does not happen on a namespace change.
- raw_notify() / isotp_notify() keep a lockless READ_ONCE() fast path, so RTNL only waits for lock_sock() of sockets that are bound or being bound to the vanishing device. CAN_RAW needs the separate ro->bind_dev hint, as ro->dev still points to the previously bound device until a rebind completes. ro->bind_dev is never dereferenced; a stale value only causes a needless lock_sock(). raw_notify() reads ro->bind_dev with smp_load_acquire() before ro->dev, paired with smp_store_release() in raw_bind(): A later bind() can only overwrite ro->bind_dev after the previous bind() has set ro->dev and released lock_sock(). Otherwise the notifier could see a stale ro->dev and the ro->bind_dev of a subsequent failing rebind to another device, and miss the binding. CAN_ISOTP does not support rebinding and publishes so->dev directly after all checks that can fail. bcm_notify() always takes lock_sock() and bcm_rx_setup() looks up the device under lock_sock(), so bcm needs no extra handling.
- raw_setsockopt() copies the filters from user space before lock_sock(), so a stalled copy can not block a NETDEV_UNREGISTER that waits for lock_sock() under RTNL. The dev->reg_state check is only a shortcut now (READ_ONCE()) and does not catch a namespace change.
- NETDEV_DOWN: raw_notify() / isotp_notify() check ro->dev / so->dev locklessly. They are written with WRITE_ONCE() in bind() and in the notifier. IFF_UP is cleared in __dev_close_many() before netif_close_many() or __dev_notify_flags() send NETDEV_DOWN in the same context, and bind() checks IFF_UP after publishing the bound device. With an smp_mb() on both sides, either the notifier sees the binding or bind() sees the cleared IFF_UP.
- j1939 benefits as well: After a namespace change its can_rx_unregister() with dev_net(ndev) picked the rcvlists_lock of the wrong namespace. The receive list statistics may still become imprecise in this case, as before.
- Tested on a lockdep + KASAN kernel:
- 10 min CAN_RAW / CAN_BCM / CAN_ISOTP traffic and procfs reads in two namespaces while moving the CAN device between them (2272 moves)
- CAN_RAW / CAN_ISOTP bind and CAN_RAW rebind churn (incl. failing rebinds to a non-CAN device) against 1007 namespace moves, checking that no filters are left on the device in the other namespace, and against 611 device deletions
- bcm_release() with a stalled procfs reader does not block RTNL
- RTNL latency (no-op SIOCSIFMTU on lo) under CAN_RAW 512 filter
churn: avg ~1 us, max ~1.2 ms (before: avg ~250 us, max ~24 ms)
include/linux/can/can-ml.h | 3 + include/net/netns/can.h | 2 +- net/can/af_can.c | 39 +++++---- net/can/bcm.c | 43 +++------- net/can/isotp.c | 64 +++++++++------ net/can/raw.c | 161 ++++++++++++++++++++++++------------- 6 files changed, 186 insertions(+), 126 deletions(-)
include/linux/can/can-ml.h +3 -0
@@ -43,10 +43,11 @@#define CAN_ML_H #include <linux/can.h> #include <linux/list.h> #include <linux/netdevice.h> +#include <linux/spinlock.h> /* exposed CAN device capabilities for network layer */ #define CAN_CAP_CC BIT(0)/* CAN CC aka Classical CAN */ #define CAN_CAP_FD BIT(1)/* CAN FD */ #define CAN_CAP_XL BIT(2)/* CAN XL */
@@ -57,10 +58,11 @@#define CAN_EFF_RCV_ARRAY_SZ (1 << CAN_EFF_RCV_HASH_BITS)enum {RX_ERR ,RX_ALL ,RX_FIL ,RX_INV ,RX_MAX };struct can_dev_rcv_lists { +spinlock_t lock;/* protects the receive lists and entries */ struct hlist_head rx[RX_MAX ];struct hlist_head rx_sff[CAN_SFF_RCV_ARRAY_SZ ];struct hlist_head rx_eff[CAN_EFF_RCV_ARRAY_SZ ];int entries; };
@@ -79,10 +81,11 @@ static inline struct can_ml_priv *can_get_ml_priv(struct net_device *dev) }static inline void can_set_ml_priv (struct net_device *dev,struct can_ml_priv *ml_priv) { +spin_lock_init (&ml_priv->dev_rcv_lists.lock);netdev_set_ml_priv (dev, ml_priv,ML_PRIV_CAN ); }static inline bool can_cap_enabled (struct net_device *dev,u32 cap) {
include/net/netns/can.h +1 -1
@@ -27,11 +27,11 @@ struct netns_can {struct proc_dir_entry *bcmproc_dir;#endif /* receive filters subscribed for 'all' CAN devices */ struct can_dev_rcv_lists *rx_alldev_list; -spinlock_t rcvlists_lock; +spinlock_t rcvlists_lock;/* protects the rcv_entries(_max) stats */ struct timer_list stattimer;/* timer for statistics update */ struct can_pkg_stats *pkg_stats;struct can_rcv_lists_stats *rcv_lists_stats;/* CAN GW per-net gateway jobs */
net/can/af_can.c +25 -14
@@ -460,13 +460,14 @@ int can_rx_register(struct net *net, struct net_device *dev, canid_t can_id, rcv =kmem_cache_alloc (rcv_cache,GFP_KERNEL );if (!rcv)return -ENOMEM ; -spin_lock_bh (&net->can.rcvlists_lock); - dev_rcv_lists =can_dev_rcv_lists_find (net, dev); + +spin_lock_bh (&dev_rcv_lists->lock); + rcv_list =can_rcv_list_find (&can_id, &mask, dev_rcv_lists); rcv->can_id = can_id; rcv->mask = mask;atomic_long_set (&rcv->matches,0 );
@@ -476,10 +477,13 @@ int can_rx_register(struct net *net, struct net_device *dev, canid_t can_id, rcv->sk = sk;hlist_add_head_rcu (&rcv->list, rcv_list); dev_rcv_lists->entries++; +spin_unlock_bh (&dev_rcv_lists->lock); + +spin_lock_bh (&net->can.rcvlists_lock); rcv_lists_stats->rcv_entries++; rcv_lists_stats->rcv_entries_max =max (rcv_lists_stats->rcv_entries_max, rcv_lists_stats->rcv_entries);spin_unlock_bh (&net->can.rcvlists_lock);
@@ -507,10 +511,18 @@ static void can_rx_delete_receiver(struct rcu_head *rp)* @func: callback function on filter match * @data: returned parameter for callback function * * Description: * Removes subscription entry depending on given (subscription) values. +* The receive lists of a CAN device are protected by a per-device lock, +* so the subscription is found even when the device has been moved to +* another network namespace after can_rx_register(). The caller has to +* ensure that @dev stays valid, e.g. by holding a reference to @dev or +* RTNL. @net should be the namespace that has been used for +* can_rx_register(). It selects the 'any' CAN device lists for +* @dev == NULL and is otherwise only used for the receive list +* statistics, which become imprecise when another namespace is passed. */ void can_rx_unregister (struct net *net,struct net_device *dev,canid_t can_id,canid_t mask,void (*func)(struct sk_buff *,void *),void *data) {
@@ -520,16 +532,14 @@ void can_rx_unregister(struct net *net, struct net_device *dev, canid_t can_id,struct can_dev_rcv_lists *dev_rcv_lists;if (dev && !can_get_ml_priv (dev))return ; -if (dev && !net_eq (net,dev_net (dev))) -return ; + dev_rcv_lists =can_dev_rcv_lists_find (net, dev); -spin_lock_bh (&net->can.rcvlists_lock); +spin_lock_bh (&dev_rcv_lists->lock); - dev_rcv_lists =can_dev_rcv_lists_find (net, dev); rcv_list =can_rcv_list_find (&can_id, &mask, dev_rcv_lists);/* Search the receiver list for the item to delete. This should * exist, since no receiver may be unregistered that hasn't * been registered before.
@@ -545,30 +555,30 @@ void can_rx_unregister(struct net *net, struct net_device *dev, canid_t can_id,* As this case may potentially happen when closing a socket while * the notifier for removing the CAN netdev is running we just print * a warning here. */ if (!rcv) { +spin_unlock_bh (&dev_rcv_lists->lock);pr_warn ("can: receive list entry not found for dev %s, id %03X, mask %03X\n" ,DNAME (dev), can_id, mask); - goto out; +return ; }hlist_del_rcu (&rcv->list); dev_rcv_lists->entries--; +spin_unlock_bh (&dev_rcv_lists->lock); + +spin_lock_bh (&net->can.rcvlists_lock);if (rcv_lists_stats->rcv_entries >0 ) rcv_lists_stats->rcv_entries--; - - out:spin_unlock_bh (&net->can.rcvlists_lock);/* schedule the receiver item for deletion */ -if (rcv) { -if (rcv->sk) -sock_hold (rcv->sk); -call_rcu (&rcv->rcu, can_rx_delete_receiver); - } +if (rcv->sk) +sock_hold (rcv->sk); +call_rcu (&rcv->rcu, can_rx_delete_receiver); }EXPORT_SYMBOL (can_rx_unregister);static inline void deliver (struct sk_buff *skb,struct receiver *rcv) {
@@ -821,10 +831,11 @@ static int can_pernet_init(struct net *net) {spin_lock_init (&net->can.rcvlists_lock); net->can.rx_alldev_list =kzalloc_obj (*net->can.rx_alldev_list);if (!net->can.rx_alldev_list) goto out; +spin_lock_init (&net->can.rx_alldev_list->lock); net->can.pkg_stats =kzalloc_obj (*net->can.pkg_stats);if (!net->can.pkg_stats) goto out_free_rx_alldev_list; net->can.rcv_lists_stats =kzalloc_obj (*net->can.rcv_lists_stats);if (!net->can.rcv_lists_stats)
net/can/bcm.c +12 -31
91 lines not on this page · raw message
net/can/isotp.c +39 -25
159 lines not on this page · raw message
net/can/raw.c +106 -55
389 lines not on this page · raw message
Signature · 1 line
2.43.0
Oliver Hartkopp
maintainerHi Marc,
as the pull-request: can-next 2026-10-22 will be processed after the merge into Linus' tree we get some more time. Sorry for the nightly stress.
In this v7 update some release_sock() calls have been moved down behind sk_error_report(sk) as suggested by sachiko-bot.
So this patch set can hang around to be cross reviewed once more with less pressure.
There are two things that always have been remarked by sashiko:
1. the missing -ENOMEM handling for can_rx_register() in isptp_bind()
2. copying user-space data in raw.c and isotp.c under lock_sock() which can block rtnl in the notifiers
Would it make sense to you that I provide some patches/fixes to address these two points so that they can additionally go upstream when Paolo opens the fixes-phase?
Best regards,
Oliver
Quoted 102 lines · Oliver Hartkopp
This series fixes a use-after-free in
CAN_ISOTPandCAN_BCMthat can be triggered when a CAN device is moved to another network namespace while a socket bound to this device is released [1].The code of patches 1-4 is unchanged. Only the commit message of patch 4 has been updated after the Sashiko review of v4 to describe the actual NULL pointer dereference in
can_rx_unregister()and theCAN_ISOTPpath that reaches it.The series is based on linux-can-next/testing, which also contains "can: raw: remove redundant NULL check before
netdev_hold()" from [2].The idea behind patch 5:
The receive lists of a CAN device are located in the device's
can_ml_privand travel with the device when it is moved to another network namespace. But they are protected by the per-netnsrcvlists_lock, so the lock that protects a device's lists depends on the namespace the caller passes tocan_rx_register()andcan_rx_unregister().can_rx_unregister()silently returns when this namespace does not matchdev_net(dev)anymore.
isotp_release()andbcm_release()leave the netdev notifier list before they can serialize against the notifier. When the device changes its namespace in this window, their filters are not removed and the remaining receivers point to the freed socket or bcm op.Taking RTNL in these release paths is no option, as
CAN_RAWalready holds RTNL while adding and removing up toCAN_RAW_FILTER_MAXfilters per socket, which unprivileged users can trigger repeatedly.Instead, patch 5 protects the receive lists where they are located: A spinlock in struct
can_dev_rcv_listsprotects the lists of a CAN device (initialized incan_set_ml_priv()) or the 'any' CAN device lists of a namespace (initialized incan_pernet_init()).can_rx_unregister()then finds the subscription independently of the current namespace of the device, which is kept accessible by the tracked device references the protocols already hold. The per-netnsrcvlists_lockonly protects the receive list statistics, for whichCAN_RAW,CAN_BCMandCAN_ISOTPpass the namespace that has been used forcan_rx_register().As a consequence,
CAN_RAWandCAN_ISOTPdo not need RTNL inbind(),setsockopt()andrelease()anymore.bind()is serialized againstNETDEV_UNREGISTERby publishing the device inside the RCU section of its lookup, asunlist_netdevice()andsynchronize_net()precedeNETDEV_UNREGISTERboth on unregistration and on a namespace change. The notifiers recheck the binding underlock_sock(). As RTNL does not keepNETDEV_DOWNout ofbind()anymore,bind()checksIFF_UPafter publishing the bound device, with ansmp_mb()pair against the notifiers, so a down transition duringbind()is still reported with ENETDOWN.Changes in v7 (details in patch 5):
- report ENETDOWN inbind()and ENODEV forNETDEV_UNREGISTERunderlock_sock(), so ENODEV can not be overwritten by ENETDOWN (Sashiko review)Changes in v6:
-isotp_bind()checks the device before publishingso->dev, so a failingbind()can not get a needless ENETDOWN (Sashiko review)Changes in v5:
- report ENETDOWN for a down transition duringbind()again (Sashiko review)
- comment and kernel-doc updates (Sashiko review)
- patch 4: commit message describes the actual crash path (Sashiko review)[1] https://lore.kernel.org/linux-can/[email protected]/
[2] https://lore.kernel.org/linux-can/[email protected]/Previous versions of patch 5:
v6: https://lore.kernel.org/linux-can/[email protected]/
v5: https://lore.kernel.org/linux-can/[email protected]/
v4: https://lore.kernel.org/linux-can/[email protected]/
v3: https://lore.kernel.org/linux-can/[email protected]/
v2: https://lore.kernel.org/linux-can/[email protected]/
v1: https://lore.kernel.org/linux-can/[email protected]/Oliver Hartkopp (4):
can: proc: skip stats calculation after reset incan_stat_update()
can: proc: remove pointers from CAN specific proc output
can: replaceARPHRD_CANtype checks withcan_get_ml_priv()
can: protect the receive lists with a per-device lockRunyu Xiao (1):
can: proc: resetpkg_statsatomics individuallyDocumentation/networking/can.rst | 16 +-- include/linux/can/can-ml.h | 3 + include/linux/can/core.h | 2 +- include/net/netns/can.h | 2 +- net/can/af_can.c | 60 ++++++----- net/can/af_can.h | 2 +- net/can/bcm.c | 64 +++++------- net/can/gw.c | 9 +- net/can/isotp.c | 71 ++++++++----- net/can/j1939/main.c | 2 +- net/can/proc.c | 95 ++++++++++------- net/can/raw.c | 169 ++++++++++++++++++++----------- 12 files changed, 289 insertions(+), 206 deletions(-)
Marc Kleine-Budde
maintainerQuoted 2 lines · Oliver Hartkopp
as the pull-request: can-next 2026-10-22 will be processed after the merge into Linus' tree we get some more time. Sorry for the nightly stress.
:)
Quoted 8 lines
In this v7 update some
release_sock()calls have been moved down behindsk_error_report(sk)as suggested by sachiko-bot.So this patch set can hang around to be cross reviewed once more with less pressure.
There are two things that always have been remarked by sashiko:
Quoted 7 lines
1. the missing
-ENOMEMhandling forcan_rx_register()inisptp_bind()
2. copying user-space data inraw.candisotp.cunderlock_sock()which can block rtnl in the notifiersWould it make sense to you that I provide some patches/fixes to address these two points so that they can additionally go upstream when Paolo opens the fixes-phase?
Yes, please. Please include the new patches in the vN+1 and adjust your cover letter.
regards,
Marc
Signature · 4 lines
Pengutronix e.K. | Marc Kleine-Budde | Embedded Linux | https://www.pengutronix.de | Vertretung Nürnberg | Phone: +49-5121-206917-129 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-9 |