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, 172 insertions, 0 deletions

diff --git a/drivers/net/ppp/pppox.c b/drivers/net/ppp/pppox.c
new file mode 100644
index 000000000..a6f72c813
--- /dev/null
+++ b/drivers/net/ppp/pppox.c
@@ -0,0 +1,172 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/** -*- linux-c -*- ***********************************************************
+ * Linux PPP over X/Ethernet (PPPoX/PPPoE) Sockets
+ *
+ * PPPoX --- Generic PPP encapsulation socket family
+ * PPPoE --- PPP over Ethernet (RFC 2516)
+ *
+ * Version:	0.5.2
+ *
+ * Author:	Michal Ostrowski <mostrows@speakeasy.net>
+ *
+ * 051000 :	Initialization cleanup
+ *
+ * License:
+ */
+
+#include <linux/string.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/compat.h>
+#include <linux/errno.h>
+#include <linux/netdevice.h>
+#include <linux/net.h>
+#include <linux/init.h>
+#include <linux/if_pppox.h>
+#include <linux/ppp_defs.h>
+#include <linux/ppp-ioctl.h>
+#include <linux/ppp_channel.h>
+#include <linux/kmod.h>
+
+#include <net/sock.h>
+
+#include <linux/uaccess.h>
+
+static const struct pppox_proto *pppox_protos[PX_MAX_PROTO + 1];
+
+int register_pppox_proto(int proto_num, const struct pppox_proto *pp)
+{
+	if (proto_num < 0 || proto_num > PX_MAX_PROTO)
+		return -EINVAL;
+	if (pppox_protos[proto_num])
+		return -EALREADY;
+	pppox_protos[proto_num] = pp;
+	return 0;
+}
+
+void unregister_pppox_proto(int proto_num)
+{
+	if (proto_num >= 0 && proto_num <= PX_MAX_PROTO)
+		pppox_protos[proto_num] = NULL;
+}
+
+void pppox_unbind_sock(struct sock *sk)
+{
+	/* Clear connection to ppp device, if attached. */
+
+	if (sk->sk_state & (PPPOX_BOUND | PPPOX_CONNECTED)) {
+		ppp_unregister_channel(&pppox_sk(sk)->chan);
+		sk->sk_state = PPPOX_DEAD;
+	}
+}
+
+EXPORT_SYMBOL(register_pppox_proto);
+EXPORT_SYMBOL(unregister_pppox_proto);
+EXPORT_SYMBOL(pppox_unbind_sock);
+
+int pppox_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
+{
+	struct sock *sk = sock->sk;
+	struct pppox_sock *po = pppox_sk(sk);
+	int rc;
+
+	lock_sock(sk);
+
+	switch (cmd) {
+	case PPPIOCGCHAN: {
+		struct sk_buff *skb;
+		int index;
+
+		rc = -ENOTCONN;
+		if (!(sk->sk_state & PPPOX_CONNECTED))
+			break;
+
+		rc = -EINVAL;
+		index = ppp_channel_index(&po->chan);
+		if (put_user(index , (int __user *) arg))
+			break;
+
+		rc = 0;
+		/* PPPIOCGCHAN historically marks the userspace handoff to
+		 * generic PPP; pppd then attaches the returned channel to
+		 * /dev/ppp.
+		 */
+		sk->sk_state |= PPPOX_BOUND;
+		/* Let lockless receive paths finish queueing against the old
+		 * state.
+		 */
+		synchronize_net();
+		/* Drain packets queued before the handoff because a bound
+		 * socket is no longer readable.
+		 */
+		while ((skb = skb_dequeue(&sk->sk_receive_queue))) {
+			skb_orphan(skb);
+			ppp_input(&po->chan, skb);
+		}
+		break;
+	}
+	default:
+		rc = pppox_protos[sk->sk_protocol]->ioctl ?
+			pppox_protos[sk->sk_protocol]->ioctl(sock, cmd, arg) : -ENOTTY;
+	}
+
+	release_sock(sk);
+	return rc;
+}
+
+EXPORT_SYMBOL(pppox_ioctl);
+
+#ifdef CONFIG_COMPAT
+int pppox_compat_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
+{
+	return pppox_ioctl(sock, cmd, (unsigned long)compat_ptr(arg));
+}
+
+EXPORT_SYMBOL(pppox_compat_ioctl);
+#endif
+
+static int pppox_create(struct net *net, struct socket *sock, int protocol,
+			int kern)
+{
+	int rc = -EPROTOTYPE;
+
+	if (protocol < 0 || protocol > PX_MAX_PROTO)
+		goto out;
+
+	rc = -EPROTONOSUPPORT;
+	if (!pppox_protos[protocol])
+		request_module("net-pf-%d-proto-%d", PF_PPPOX, protocol);
+	if (!pppox_protos[protocol] ||
+	    !try_module_get(pppox_protos[protocol]->owner))
+		goto out;
+
+	rc = pppox_protos[protocol]->create(net, sock, kern);
+
+	module_put(pppox_protos[protocol]->owner);
+out:
+	return rc;
+}
+
+static const struct net_proto_family pppox_proto_family = {
+	.family	= PF_PPPOX,
+	.create	= pppox_create,
+	.owner	= THIS_MODULE,
+};
+
+static int __init pppox_init(void)
+{
+	return sock_register(&pppox_proto_family);
+}
+
+static void __exit pppox_exit(void)
+{
+	sock_unregister(PF_PPPOX);
+}
+
+module_init(pppox_init);
+module_exit(pppox_exit);
+
+MODULE_AUTHOR("Michal Ostrowski <mostrows@speakeasy.net>");
+MODULE_DESCRIPTION("PPP over Ethernet driver (generic socket layer)");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_NETPROTO(PF_PPPOX);