x86/kprobes: Fix crash when probing CS CALL instructionsgrafted

When using eBPF to probe CS CALL instructions within a function,
a crash can be triggered.

The eBPF tool probes offset 257 of the __hrtimer_run_queues()
function:

<__hrtimer_run_queues+249>:  nopl   0x0(%rax,%rax,1)
<__hrtimer_run_queues+254>:  mov    %r14,%rdi
<__hrtimer_run_queues+257>:  cs call <__x86_indirect_thunk_r12>
<__hrtimer_run_queues+263>:  mov    %eax,%r12d
<__hrtimer_run_queues+266>:  xchg   %ax,%ax
<__hrtimer_run_queues+268>:  mov    %r13,%rdi

Which triggers this crash:

  BUG: unable to handle page fault for address: 00000000000f41c9
  #PF: supervisor write access in kernel mode
  #PF: error_code(0x0002) - not-present page
  PGD 0 P4D 0
  Oops: 0002 [#1] SMP NOPTI
  CPU: 1 PID: 0 Comm: swapper/1 Kdump: loaded Tainted: P
  RIP: 0010:__hrtimer_run_queues+0x106/0x230

Note that __hrtimer_run_queues+0x106 is __hrtimer_run_queues+262, which is
at the 6th byte of the above CS CALL instruction. Since the CS CALL
instruction occupies 6 bytes, the exception occurred in the middle of that
call instruction.

The root cause is that when using eBPF tools to probe in the middle of a
function, a kprobe with INT3 is used as the underlying implementation.

During single-step emulation of the original CALL instruction,
int3_emulate_call() assumes that the probed CALL instruction is 5 bytes
long. However, the actual CS-prefixed CALL instruction occupies 6 bytes,
so it constructs an incorrect exception return address. When the CPU
returns from the kprobe handler, the next instruction to be executed is at
the address of the last byte of that CS CALL instruction. Coincidentally,
starting from that address, the CPU fetches and decodes a completely
different instruction, which ultimately triggers a kernel crash.

Fix the issue by using the actual instruction length obtained from
the instruction decoder when constructing the exception return
address, rather than relying on the hardcoded CALL_INSN_SIZE macro.

[ mingo: Refined the changelog ]

Fixes: 6256e668b7af ("x86/kprobes: Use int3 instead of debug trap for single-step")
Suggested-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Jinke Han <jinkehan@didiglobal.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Acked-by: Yafang Shao <laoar.shao@gmail.com>
Acked-by: Borislav Petkov <bp@alien8.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://patch.msgid.link/20260908073742.GA10517@didi-ThinkCentre-M920t-N000

1 files changed, 514 insertions, 0 deletions

diff --git a/net/bridge/br.c b/net/bridge/br.c
new file mode 100644
index 000000000..a5e5b2db1
--- /dev/null
+++ b/net/bridge/br.c
@@ -0,0 +1,514 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ *	Generic parts
+ *	Linux ethernet bridge
+ *
+ *	Authors:
+ *	Lennert Buytenhek		<buytenh@gnu.org>
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/init.h>
+#include <linux/llc.h>
+#include <net/llc.h>
+#include <net/stp.h>
+#include <net/switchdev.h>
+
+#include "br_private.h"
+
+/*
+ * Handle changes in state of network devices enslaved to a bridge.
+ *
+ * Note: don't care about up/down if bridge itself is down, because
+ *     port state is checked when bridge is brought up.
+ */
+static int br_device_event(struct notifier_block *unused, unsigned long event, void *ptr)
+{
+	struct netlink_ext_ack *extack = netdev_notifier_info_to_extack(ptr);
+	struct netdev_notifier_pre_changeaddr_info *prechaddr_info;
+	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
+	struct net_bridge_port *p;
+	struct net_bridge *br;
+	bool notified = false;
+	bool changed_addr;
+	int err;
+
+	if (netif_is_bridge_master(dev)) {
+		struct net_bridge *br = netdev_priv(dev);
+
+		if (event == NETDEV_REGISTER)
+			br_fdb_change_mac_address(br, dev->dev_addr);
+
+		err = br_vlan_bridge_event(dev, event, ptr);
+		if (err)
+			return notifier_from_errno(err);
+
+		if (event == NETDEV_REGISTER) {
+			/* register of bridge completed, add sysfs entries */
+			err = br_sysfs_addbr(dev);
+			if (err)
+				return notifier_from_errno(err);
+
+			return NOTIFY_DONE;
+		}
+	}
+
+	if (is_vlan_dev(dev)) {
+		struct net_device *real_dev = vlan_dev_real_dev(dev);
+
+		if (netif_is_bridge_master(real_dev))
+			br_vlan_vlan_upper_event(real_dev, dev, event);
+	}
+
+	/* not a port of a bridge */
+	p = br_port_get_rtnl(dev);
+	if (!p)
+		return NOTIFY_DONE;
+
+	br = p->br;
+
+	switch (event) {
+	case NETDEV_CHANGEMTU:
+		br_mtu_auto_adjust(br);
+		break;
+
+	case NETDEV_PRE_CHANGEADDR:
+		if (br->dev->addr_assign_type == NET_ADDR_SET)
+			break;
+		prechaddr_info = ptr;
+		err = netif_pre_changeaddr_notify(br->dev,
+						  prechaddr_info->dev_addr,
+						  extack);
+		if (err)
+			return notifier_from_errno(err);
+		break;
+
+	case NETDEV_CHANGEADDR:
+		spin_lock_bh(&br->lock);
+		br_fdb_changeaddr(p, dev->dev_addr);
+		changed_addr = br_stp_recalculate_bridge_id(br);
+		spin_unlock_bh(&br->lock);
+
+		if (changed_addr)
+			call_netdevice_notifiers(NETDEV_CHANGEADDR, br->dev);
+
+		break;
+
+	case NETDEV_CHANGE:
+		br_port_carrier_check(p, &notified);
+		break;
+
+	case NETDEV_FEAT_CHANGE:
+		netdev_update_features(br->dev);
+		break;
+
+	case NETDEV_DOWN:
+		spin_lock_bh(&br->lock);
+		if (br->dev->flags & IFF_UP) {
+			br_stp_disable_port(p);
+			notified = true;
+		}
+		spin_unlock_bh(&br->lock);
+		break;
+
+	case NETDEV_UP:
+		if (netif_running(br->dev) && netif_oper_up(dev)) {
+			spin_lock_bh(&br->lock);
+			br_stp_enable_port(p);
+			notified = true;
+			spin_unlock_bh(&br->lock);
+		}
+		break;
+
+	case NETDEV_UNREGISTER:
+		br_del_if(br, dev);
+		break;
+
+	case NETDEV_CHANGENAME:
+		err = br_sysfs_renameif(p);
+		if (err)
+			return notifier_from_errno(err);
+		break;
+
+	case NETDEV_PRE_TYPE_CHANGE:
+		/* Forbid underlying device to change its type. */
+		return NOTIFY_BAD;
+
+	case NETDEV_RESEND_IGMP:
+		/* Propagate to master device */
+		call_netdevice_notifiers(event, br->dev);
+		break;
+	}
+
+	if (event != NETDEV_UNREGISTER)
+		br_vlan_port_event(p, event);
+
+	/* Events that may cause spanning tree to refresh */
+	if (!notified && (event == NETDEV_CHANGEADDR || event == NETDEV_UP ||
+			  event == NETDEV_CHANGE || event == NETDEV_DOWN))
+		br_ifinfo_notify(RTM_NEWLINK, NULL, p);
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block br_device_notifier = {
+	.notifier_call = br_device_event
+};
+
+/* called with RTNL or RCU */
+static int br_switchdev_event(struct notifier_block *unused,
+			      unsigned long event, void *ptr)
+{
+	struct net_device *dev = switchdev_notifier_info_to_dev(ptr);
+	struct net_bridge_port *p;
+	struct net_bridge *br;
+	struct switchdev_notifier_fdb_info *fdb_info;
+	int err = NOTIFY_DONE;
+
+	p = br_port_get_rtnl_rcu(dev);
+	if (!p)
+		goto out;
+
+	br = p->br;
+
+	switch (event) {
+	case SWITCHDEV_FDB_ADD_TO_BRIDGE:
+		fdb_info = ptr;
+		err = br_fdb_external_learn_add(br, p, fdb_info->addr,
+						fdb_info->vid,
+						fdb_info->locked, false);
+		if (err) {
+			err = notifier_from_errno(err);
+			break;
+		}
+		br_fdb_offloaded_set(br, p, fdb_info->addr,
+				     fdb_info->vid, fdb_info->offloaded);
+		break;
+	case SWITCHDEV_FDB_DEL_TO_BRIDGE:
+		fdb_info = ptr;
+		err = br_fdb_external_learn_del(br, p, fdb_info->addr,
+						fdb_info->vid, false);
+		if (err)
+			err = notifier_from_errno(err);
+		break;
+	case SWITCHDEV_FDB_OFFLOADED:
+		fdb_info = ptr;
+		br_fdb_offloaded_set(br, p, fdb_info->addr,
+				     fdb_info->vid, fdb_info->offloaded);
+		break;
+	case SWITCHDEV_FDB_FLUSH_TO_BRIDGE:
+		fdb_info = ptr;
+		/* Don't delete static entries */
+		br_fdb_delete_by_port(br, p, fdb_info->vid, 0);
+		break;
+	}
+
+out:
+	return err;
+}
+
+static struct notifier_block br_switchdev_notifier = {
+	.notifier_call = br_switchdev_event,
+};
+
+/* called under rtnl_mutex */
+static int br_switchdev_blocking_event(struct notifier_block *nb,
+				       unsigned long event, void *ptr)
+{
+	struct netlink_ext_ack *extack = netdev_notifier_info_to_extack(ptr);
+	struct net_device *dev = switchdev_notifier_info_to_dev(ptr);
+	struct switchdev_notifier_brport_info *brport_info;
+	const struct switchdev_brport *b;
+	struct net_bridge_port *p;
+	int err = NOTIFY_DONE;
+
+	p = br_port_get_rtnl(dev);
+	if (!p)
+		goto out;
+
+	switch (event) {
+	case SWITCHDEV_BRPORT_OFFLOADED:
+		brport_info = ptr;
+		b = &brport_info->brport;
+
+		err = br_switchdev_port_offload(p, b->dev, b->ctx,
+						b->atomic_nb, b->blocking_nb,
+						b->tx_fwd_offload, extack);
+		err = notifier_from_errno(err);
+		break;
+	case SWITCHDEV_BRPORT_UNOFFLOADED:
+		brport_info = ptr;
+		b = &brport_info->brport;
+
+		br_switchdev_port_unoffload(p, b->ctx, b->atomic_nb,
+					    b->blocking_nb);
+		break;
+	case SWITCHDEV_BRPORT_REPLAY:
+		brport_info = ptr;
+		b = &brport_info->brport;
+
+		err = br_switchdev_port_replay(p, b->dev, b->ctx, b->atomic_nb,
+					       b->blocking_nb, extack);
+		err = notifier_from_errno(err);
+		break;
+	}
+
+out:
+	return err;
+}
+
+static struct notifier_block br_switchdev_blocking_notifier = {
+	.notifier_call = br_switchdev_blocking_event,
+};
+
+static int
+br_toggle_fdb_local_vlan_0(struct net_bridge *br, bool on,
+			   struct netlink_ext_ack *extack)
+{
+	int err;
+
+	if (br_opt_get(br, BROPT_FDB_LOCAL_VLAN_0) == on)
+		return 0;
+
+	err = br_fdb_toggle_local_vlan_0(br, on, extack);
+	if (err)
+		return err;
+
+	br_opt_toggle(br, BROPT_FDB_LOCAL_VLAN_0, on);
+	return 0;
+}
+
+/* br_boolopt_toggle - change user-controlled boolean option
+ *
+ * @br: bridge device
+ * @opt: id of the option to change
+ * @on: new option value
+ * @extack: extack for error messages
+ *
+ * Changes the value of the respective boolean option to @on taking care of
+ * any internal option value mapping and configuration.
+ */
+int br_boolopt_toggle(struct net_bridge *br, enum br_boolopt_id opt, bool on,
+		      struct netlink_ext_ack *extack)
+{
+	int err = 0;
+
+	switch (opt) {
+	case BR_BOOLOPT_NO_LL_LEARN:
+		br_opt_toggle(br, BROPT_NO_LL_LEARN, on);
+		break;
+	case BR_BOOLOPT_MCAST_VLAN_SNOOPING:
+		err = br_multicast_toggle_vlan_snooping(br, on, extack);
+		break;
+	case BR_BOOLOPT_MST_ENABLE:
+		err = br_mst_set_enabled(br, on, extack);
+		break;
+	case BR_BOOLOPT_MDB_OFFLOAD_FAIL_NOTIFICATION:
+		br_opt_toggle(br, BROPT_MDB_OFFLOAD_FAIL_NOTIFICATION, on);
+		break;
+	case BR_BOOLOPT_FDB_LOCAL_VLAN_0:
+		err = br_toggle_fdb_local_vlan_0(br, on, extack);
+		break;
+	default:
+		/* shouldn't be called with unsupported options */
+		WARN_ON(1);
+		break;
+	}
+
+	return err;
+}
+
+int br_boolopt_get(const struct net_bridge *br, enum br_boolopt_id opt)
+{
+	switch (opt) {
+	case BR_BOOLOPT_NO_LL_LEARN:
+		return br_opt_get(br, BROPT_NO_LL_LEARN);
+	case BR_BOOLOPT_MCAST_VLAN_SNOOPING:
+		return br_opt_get(br, BROPT_MCAST_VLAN_SNOOPING_ENABLED);
+	case BR_BOOLOPT_MST_ENABLE:
+		return br_opt_get(br, BROPT_MST_ENABLED);
+	case BR_BOOLOPT_MDB_OFFLOAD_FAIL_NOTIFICATION:
+		return br_opt_get(br, BROPT_MDB_OFFLOAD_FAIL_NOTIFICATION);
+	case BR_BOOLOPT_FDB_LOCAL_VLAN_0:
+		return br_opt_get(br, BROPT_FDB_LOCAL_VLAN_0);
+	default:
+		/* shouldn't be called with unsupported options */
+		WARN_ON(1);
+		break;
+	}
+
+	return 0;
+}
+
+int br_boolopt_multi_toggle(struct net_bridge *br,
+			    struct br_boolopt_multi *bm,
+			    struct netlink_ext_ack *extack)
+{
+	unsigned long bitmap = bm->optmask;
+	int err = 0;
+	int opt_id;
+
+	opt_id = find_next_bit(&bitmap, BITS_PER_LONG, BR_BOOLOPT_MAX);
+	if (opt_id != BITS_PER_LONG) {
+		NL_SET_ERR_MSG_FMT_MOD(extack, "Unknown boolean option %d",
+				       opt_id);
+		return -EINVAL;
+	}
+
+	for_each_set_bit(opt_id, &bitmap, BR_BOOLOPT_MAX) {
+		bool on = !!(bm->optval & BIT(opt_id));
+
+		err = br_boolopt_toggle(br, opt_id, on, extack);
+		if (err) {
+			br_debug(br, "boolopt multi-toggle error: option: %d current: %d new: %d error: %d\n",
+				 opt_id, br_boolopt_get(br, opt_id), on, err);
+			break;
+		}
+	}
+
+	return err;
+}
+
+void br_boolopt_multi_get(const struct net_bridge *br,
+			  struct br_boolopt_multi *bm)
+{
+	u32 optval = 0;
+	int opt_id;
+
+	for (opt_id = 0; opt_id < BR_BOOLOPT_MAX; opt_id++)
+		optval |= (br_boolopt_get(br, opt_id) << opt_id);
+
+	bm->optval = optval;
+	bm->optmask = GENMASK((BR_BOOLOPT_MAX - 1), 0);
+}
+
+/* private bridge options, controlled by the kernel */
+void br_opt_toggle(struct net_bridge *br, enum net_bridge_opts opt, bool on)
+{
+	bool cur = !!br_opt_get(br, opt);
+
+	br_debug(br, "toggle option: %d state: %d -> %d\n",
+		 opt, cur, on);
+
+	if (cur == on)
+		return;
+
+	if (on)
+		set_bit(opt, &br->options);
+	else
+		clear_bit(opt, &br->options);
+}
+
+static void __net_exit br_net_exit_rtnl(struct net *net,
+					struct list_head *dev_to_kill)
+{
+	struct net_device *dev;
+
+	ASSERT_RTNL_NET(net);
+
+	for_each_netdev(net, dev)
+		if (netif_is_bridge_master(dev))
+			br_dev_delete(dev, dev_to_kill);
+}
+
+static struct pernet_operations br_net_ops = {
+	.exit_rtnl = br_net_exit_rtnl,
+};
+
+static const struct stp_proto br_stp_proto = {
+	.rcv	= br_stp_rcv,
+};
+
+static int __init br_init(void)
+{
+	int err;
+
+	BUILD_BUG_ON(sizeof(struct br_input_skb_cb) > sizeof_field(struct sk_buff, cb));
+
+	err = stp_proto_register(&br_stp_proto);
+	if (err < 0) {
+		pr_err("bridge: can't register sap for STP\n");
+		return err;
+	}
+
+	err = br_fdb_init();
+	if (err)
+		goto err_out;
+
+	err = register_pernet_subsys(&br_net_ops);
+	if (err)
+		goto err_out1;
+
+	err = br_nf_core_init();
+	if (err)
+		goto err_out2;
+
+	err = register_netdevice_notifier(&br_device_notifier);
+	if (err)
+		goto err_out3;
+
+	err = register_switchdev_notifier(&br_switchdev_notifier);
+	if (err)
+		goto err_out4;
+
+	err = register_switchdev_blocking_notifier(&br_switchdev_blocking_notifier);
+	if (err)
+		goto err_out5;
+
+	err = br_netlink_init();
+	if (err)
+		goto err_out6;
+
+	brioctl_set(br_ioctl_stub);
+
+#if IS_MODULE(CONFIG_BRIDGE_NETFILTER)
+	pr_info("bridge: filtering via arp/ip/ip6tables is no longer available "
+		"by default. Update your scripts to load br_netfilter if you "
+		"need this.\n");
+#endif
+
+	return 0;
+
+err_out6:
+	unregister_switchdev_blocking_notifier(&br_switchdev_blocking_notifier);
+err_out5:
+	unregister_switchdev_notifier(&br_switchdev_notifier);
+err_out4:
+	unregister_netdevice_notifier(&br_device_notifier);
+err_out3:
+	br_nf_core_fini();
+err_out2:
+	unregister_pernet_subsys(&br_net_ops);
+err_out1:
+	br_fdb_fini();
+err_out:
+	stp_proto_unregister(&br_stp_proto);
+	return err;
+}
+
+static void __exit br_deinit(void)
+{
+	stp_proto_unregister(&br_stp_proto);
+	br_netlink_fini();
+	unregister_switchdev_blocking_notifier(&br_switchdev_blocking_notifier);
+	unregister_switchdev_notifier(&br_switchdev_notifier);
+	unregister_netdevice_notifier(&br_device_notifier);
+	brioctl_set(NULL);
+	unregister_pernet_subsys(&br_net_ops);
+
+	rcu_barrier(); /* Wait for completion of call_rcu()'s */
+
+	br_nf_core_fini();
+	br_fdb_fini();
+}
+
+module_init(br_init)
+module_exit(br_deinit)
+MODULE_LICENSE("GPL");
+MODULE_VERSION(BR_VERSION);
+MODULE_ALIAS_RTNL_LINK("bridge");
+MODULE_DESCRIPTION("Ethernet bridge driver");
+MODULE_IMPORT_NS("NETDEV_INTERNAL");