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, 162 insertions, 0 deletions

diff --git a/net/atm/pvc.c b/net/atm/pvc.c
new file mode 100644
index 000000000..8b2c3e515
--- /dev/null
+++ b/net/atm/pvc.c
@@ -0,0 +1,162 @@
+// SPDX-License-Identifier: GPL-2.0
+/* net/atm/pvc.c - ATM PVC sockets */
+
+/* Written 1995-2000 by Werner Almesberger, EPFL LRC/ICA */
+
+
+#include <linux/net.h>		/* struct socket, struct proto_ops */
+#include <linux/atm.h>		/* ATM stuff */
+#include <linux/atmdev.h>	/* ATM devices */
+#include <linux/errno.h>	/* error codes */
+#include <linux/kernel.h>	/* printk */
+#include <linux/init.h>
+#include <linux/skbuff.h>
+#include <linux/bitops.h>
+#include <linux/export.h>
+#include <net/sock.h>		/* for sock_no_* */
+
+#include "resources.h"		/* devs and vccs */
+#include "common.h"		/* common for PVCs and SVCs */
+
+
+static int pvc_shutdown(struct socket *sock, int how)
+{
+	return 0;
+}
+
+static int pvc_bind(struct socket *sock, struct sockaddr_unsized *sockaddr,
+		    int sockaddr_len)
+{
+	struct sock *sk = sock->sk;
+	struct sockaddr_atmpvc *addr;
+	struct atm_vcc *vcc;
+	int error;
+
+	if (sockaddr_len != sizeof(struct sockaddr_atmpvc))
+		return -EINVAL;
+	addr = (struct sockaddr_atmpvc *)sockaddr;
+	if (addr->sap_family != AF_ATMPVC)
+		return -EAFNOSUPPORT;
+	lock_sock(sk);
+	vcc = ATM_SD(sock);
+	if (!test_bit(ATM_VF_HASQOS, &vcc->flags)) {
+		error = -EBADFD;
+		goto out;
+	}
+	if (test_bit(ATM_VF_PARTIAL, &vcc->flags)) {
+		if (vcc->vpi != ATM_VPI_UNSPEC)
+			addr->sap_addr.vpi = vcc->vpi;
+		if (vcc->vci != ATM_VCI_UNSPEC)
+			addr->sap_addr.vci = vcc->vci;
+	}
+	error = vcc_connect(sock, addr->sap_addr.itf, addr->sap_addr.vpi,
+			    addr->sap_addr.vci);
+out:
+	release_sock(sk);
+	return error;
+}
+
+static int pvc_connect(struct socket *sock, struct sockaddr_unsized *sockaddr,
+		       int sockaddr_len, int flags)
+{
+	return pvc_bind(sock, sockaddr, sockaddr_len);
+}
+
+static int pvc_setsockopt(struct socket *sock, int level, int optname,
+			  sockptr_t optval, unsigned int optlen)
+{
+	struct sock *sk = sock->sk;
+	int error;
+
+	lock_sock(sk);
+	error = vcc_setsockopt(sock, level, optname, optval, optlen);
+	release_sock(sk);
+	return error;
+}
+
+static int pvc_getsockopt(struct socket *sock, int level, int optname,
+			  sockopt_t *opt)
+{
+	struct sock *sk = sock->sk;
+	int error;
+
+	lock_sock(sk);
+	error = vcc_getsockopt(sock, level, optname, opt);
+	release_sock(sk);
+	return error;
+}
+
+static int pvc_getname(struct socket *sock, struct sockaddr *sockaddr,
+		       int peer)
+{
+	struct sockaddr_atmpvc *addr;
+	struct atm_vcc *vcc = ATM_SD(sock);
+
+	if (!vcc->dev || !test_bit(ATM_VF_ADDR, &vcc->flags))
+		return -ENOTCONN;
+	addr = (struct sockaddr_atmpvc *)sockaddr;
+	memset(addr, 0, sizeof(*addr));
+	addr->sap_family = AF_ATMPVC;
+	addr->sap_addr.itf = vcc->dev->number;
+	addr->sap_addr.vpi = vcc->vpi;
+	addr->sap_addr.vci = vcc->vci;
+	return sizeof(struct sockaddr_atmpvc);
+}
+
+static const struct proto_ops pvc_proto_ops = {
+	.family =	PF_ATMPVC,
+	.owner =	THIS_MODULE,
+
+	.release =	vcc_release,
+	.bind =		pvc_bind,
+	.connect =	pvc_connect,
+	.socketpair =	sock_no_socketpair,
+	.accept =	sock_no_accept,
+	.getname =	pvc_getname,
+	.poll =		vcc_poll,
+	.ioctl =	vcc_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl = vcc_compat_ioctl,
+#endif
+	.gettstamp =	sock_gettstamp,
+	.listen =	sock_no_listen,
+	.shutdown =	pvc_shutdown,
+	.setsockopt =	pvc_setsockopt,
+	.getsockopt_iter = pvc_getsockopt,
+	.sendmsg =	vcc_sendmsg,
+	.recvmsg =	vcc_recvmsg,
+	.mmap =		sock_no_mmap,
+};
+
+
+static int pvc_create(struct net *net, struct socket *sock, int protocol,
+		      int kern)
+{
+	if (net != &init_net)
+		return -EAFNOSUPPORT;
+
+	sock->ops = &pvc_proto_ops;
+	return vcc_create(net, sock, protocol, PF_ATMPVC, kern);
+}
+
+static const struct net_proto_family pvc_family_ops = {
+	.family = PF_ATMPVC,
+	.create = pvc_create,
+	.owner = THIS_MODULE,
+};
+
+
+/*
+ *	Initialize the ATM PVC protocol family
+ */
+
+
+int __init atmpvc_init(void)
+{
+	return sock_register(&pvc_family_ops);
+}
+
+void atmpvc_exit(void)
+{
+	sock_unregister(PF_ATMPVC);
+}