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, 184 insertions, 0 deletions

diff --git a/drivers/acpi/event.c b/drivers/acpi/event.c
new file mode 100644
index 000000000..4d840d2e7
--- /dev/null
+++ b/drivers/acpi/event.c
@@ -0,0 +1,184 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * event.c - exporting ACPI events via procfs
+ *
+ *  Copyright (C) 2001, 2002 Andy Grover <andrew.grover@intel.com>
+ *  Copyright (C) 2001, 2002 Paul Diefenbaugh <paul.s.diefenbaugh@intel.com>
+ *
+ */
+
+#define pr_fmt(fmt) "ACPI: " fmt
+
+#include <linux/spinlock.h>
+#include <linux/export.h>
+#include <linux/proc_fs.h>
+#include <linux/init.h>
+#include <linux/poll.h>
+#include <linux/gfp.h>
+#include <linux/acpi.h>
+#include <net/netlink.h>
+#include <net/genetlink.h>
+
+#include "internal.h"
+
+/* ACPI notifier chain */
+static BLOCKING_NOTIFIER_HEAD(acpi_chain_head);
+
+int acpi_notifier_call_chain(const char *device_class,
+			     const char *bus_id, u32 type, u32 data)
+{
+	struct acpi_bus_event event;
+
+	strscpy(event.device_class, device_class);
+	strscpy(event.bus_id, bus_id);
+	event.type = type;
+	event.data = data;
+	return (blocking_notifier_call_chain(&acpi_chain_head, 0, (void *)&event)
+			== NOTIFY_BAD) ? -EINVAL : 0;
+}
+EXPORT_SYMBOL(acpi_notifier_call_chain);
+
+int register_acpi_notifier(struct notifier_block *nb)
+{
+	return blocking_notifier_chain_register(&acpi_chain_head, nb);
+}
+EXPORT_SYMBOL(register_acpi_notifier);
+
+int unregister_acpi_notifier(struct notifier_block *nb)
+{
+	return blocking_notifier_chain_unregister(&acpi_chain_head, nb);
+}
+EXPORT_SYMBOL(unregister_acpi_notifier);
+
+#ifdef CONFIG_NET
+static unsigned int acpi_event_seqnum;
+struct acpi_genl_event {
+	acpi_device_class device_class;
+	char bus_id[15];
+	u32 type;
+	u32 data;
+};
+
+/* attributes of acpi_genl_family */
+enum {
+	ACPI_GENL_ATTR_UNSPEC,
+	ACPI_GENL_ATTR_EVENT,	/* ACPI event info needed by user space */
+	__ACPI_GENL_ATTR_MAX,
+};
+#define ACPI_GENL_ATTR_MAX (__ACPI_GENL_ATTR_MAX - 1)
+
+/* commands supported by the acpi_genl_family */
+enum {
+	ACPI_GENL_CMD_UNSPEC,
+	ACPI_GENL_CMD_EVENT,	/* kernel->user notifications for ACPI events */
+	__ACPI_GENL_CMD_MAX,
+};
+#define ACPI_GENL_CMD_MAX (__ACPI_GENL_CMD_MAX - 1)
+
+#define ACPI_GENL_FAMILY_NAME		"acpi_event"
+#define ACPI_GENL_VERSION		0x01
+#define ACPI_GENL_MCAST_GROUP_NAME 	"acpi_mc_group"
+
+static const struct genl_multicast_group acpi_event_mcgrps[] = {
+	{ .name = ACPI_GENL_MCAST_GROUP_NAME, },
+};
+
+static struct genl_family acpi_event_genl_family __ro_after_init = {
+	.module = THIS_MODULE,
+	.name = ACPI_GENL_FAMILY_NAME,
+	.version = ACPI_GENL_VERSION,
+	.maxattr = ACPI_GENL_ATTR_MAX,
+	.mcgrps = acpi_event_mcgrps,
+	.n_mcgrps = ARRAY_SIZE(acpi_event_mcgrps),
+};
+
+int acpi_bus_generate_netlink_event(const char *device_class,
+				      const char *bus_id,
+				      u8 type, int data)
+{
+	struct sk_buff *skb;
+	struct nlattr *attr;
+	struct acpi_genl_event *event;
+	void *msg_header;
+	int size;
+
+	/* allocate memory */
+	size = nla_total_size(sizeof(struct acpi_genl_event)) +
+	    nla_total_size(0);
+
+	skb = genlmsg_new(size, GFP_ATOMIC);
+	if (!skb)
+		return -ENOMEM;
+
+	/* add the genetlink message header */
+	msg_header = genlmsg_put(skb, 0, acpi_event_seqnum++,
+				 &acpi_event_genl_family, 0,
+				 ACPI_GENL_CMD_EVENT);
+	if (!msg_header) {
+		nlmsg_free(skb);
+		return -ENOMEM;
+	}
+
+	/* fill the data */
+	attr =
+	    nla_reserve(skb, ACPI_GENL_ATTR_EVENT,
+			sizeof(struct acpi_genl_event));
+	if (!attr) {
+		nlmsg_free(skb);
+		return -EINVAL;
+	}
+
+	event = nla_data(attr);
+	memset(event, 0, sizeof(struct acpi_genl_event));
+
+	strscpy(event->device_class, device_class, sizeof(event->device_class));
+	strscpy(event->bus_id, bus_id, sizeof(event->bus_id));
+	event->type = type;
+	event->data = data;
+
+	/* send multicast genetlink message */
+	genlmsg_end(skb, msg_header);
+
+	genlmsg_multicast(&acpi_event_genl_family, skb, 0, 0, GFP_ATOMIC);
+	return 0;
+}
+
+EXPORT_SYMBOL(acpi_bus_generate_netlink_event);
+
+static int __init acpi_event_genetlink_init(void)
+{
+	return genl_register_family(&acpi_event_genl_family);
+}
+
+#else
+int acpi_bus_generate_netlink_event(const char *device_class,
+				      const char *bus_id,
+				      u8 type, int data)
+{
+	return 0;
+}
+
+EXPORT_SYMBOL(acpi_bus_generate_netlink_event);
+
+static int acpi_event_genetlink_init(void)
+{
+	return -ENODEV;
+}
+#endif
+
+static int __init acpi_event_init(void)
+{
+	int error;
+
+	if (acpi_disabled)
+		return 0;
+
+	/* create genetlink for acpi event */
+	error = acpi_event_genetlink_init();
+	if (error)
+		pr_warn("Failed to create genetlink family for ACPI event\n");
+
+	return 0;
+}
+
+fs_initcall(acpi_event_init);