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, 142 insertions, 0 deletions

diff --git a/drivers/dio/dio-driver.c b/drivers/dio/dio-driver.c
new file mode 100644
index 000000000..12fa2d209
--- /dev/null
+++ b/drivers/dio/dio-driver.c
@@ -0,0 +1,142 @@
+/*
+ *  DIO Driver Services
+ *
+ *  Copyright (C) 2004 Jochen Friedrich
+ *
+ *  Loosely based on drivers/pci/pci-driver.c and drivers/zorro/zorro-driver.c
+ *
+ *  This file is subject to the terms and conditions of the GNU General Public
+ *  License.  See the file COPYING in the main directory of this archive
+ *  for more details.
+ */
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/dio.h>
+
+
+/**
+ *  dio_match_device - Tell if a DIO device structure has a matching DIO device id structure
+ *  @ids: array of DIO device id structures to search in
+ *  @d: the DIO device structure to match against
+ *
+ *  Used by a driver to check whether a DIO device present in the
+ *  system is in its list of supported devices. Returns the matching
+ *  dio_device_id structure or %NULL if there is no match.
+ */
+
+static const struct dio_device_id *
+dio_match_device(const struct dio_device_id *ids,
+		   const struct dio_dev *d)
+{
+	while (ids->id) {
+		if (ids->id == DIO_WILDCARD)
+			return ids;
+		if (DIO_NEEDSSECID(ids->id & 0xff)) {
+			if (ids->id == d->id)
+				return ids;
+		} else {
+			if ((ids->id & 0xff) == (d->id & 0xff))
+				return ids;
+		}
+		ids++;
+	}
+	return NULL;
+}
+
+static int dio_device_probe(struct device *dev)
+{
+	int error = 0;
+	struct dio_driver *drv = to_dio_driver(dev->driver);
+	struct dio_dev *d = to_dio_dev(dev);
+
+	if (!d->driver && drv->probe) {
+		const struct dio_device_id *id;
+
+		id = dio_match_device(drv->id_table, d);
+		if (id)
+			error = drv->probe(d, id);
+		if (error >= 0) {
+			d->driver = drv;
+			error = 0;
+		}
+	}
+	return error;
+}
+
+
+/**
+ *  dio_register_driver - register a new DIO driver
+ *  @drv: the driver structure to register
+ *
+ *  Adds the driver structure to the list of registered drivers
+ *  Returns zero or a negative error value.
+ */
+
+int dio_register_driver(struct dio_driver *drv)
+{
+	/* initialize common driver fields */
+	drv->driver.name = drv->name;
+	drv->driver.bus = &dio_bus_type;
+
+	/* register with core */
+	return driver_register(&drv->driver);
+}
+
+
+/**
+ *  dio_unregister_driver - unregister a DIO driver
+ *  @drv: the driver structure to unregister
+ *
+ *  Deletes the driver structure from the list of registered DIO drivers,
+ *  gives it a chance to clean up by calling its remove() function for
+ *  each device it was responsible for, and marks those devices as
+ *  driverless.
+ */
+
+void dio_unregister_driver(struct dio_driver *drv)
+{
+	driver_unregister(&drv->driver);
+}
+
+
+/**
+ *  dio_bus_match - Tell if a DIO device structure has a matching DIO device id structure
+ *  @dev: the DIO device structure to match against
+ *  @drv: the &device_driver that points to the array of DIO device id structures to search
+ *
+ *  Used by the driver core to check whether a DIO device present in the
+ *  system is in a driver's list of supported devices. Returns 1 if supported,
+ *  and 0 if there is no match.
+ */
+
+static int dio_bus_match(struct device *dev, const struct device_driver *drv)
+{
+	struct dio_dev *d = to_dio_dev(dev);
+	const struct dio_driver *dio_drv = to_dio_driver(drv);
+	const struct dio_device_id *ids = dio_drv->id_table;
+
+	if (!ids)
+		return 0;
+
+	return dio_match_device(ids, d) ? 1 : 0;
+}
+
+
+const struct bus_type dio_bus_type = {
+	.name	= "dio",
+	.match	= dio_bus_match,
+	.probe	= dio_device_probe,
+};
+
+
+static int __init dio_driver_init(void)
+{
+	return bus_register(&dio_bus_type);
+}
+
+postcore_initcall(dio_driver_init);
+
+EXPORT_SYMBOL(dio_register_driver);
+EXPORT_SYMBOL(dio_unregister_driver);
+EXPORT_SYMBOL(dio_bus_type);