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, 120 insertions, 0 deletions

diff --git a/drivers/of/overlay_test.c b/drivers/of/overlay_test.c
new file mode 100644
index 000000000..c787524c5
--- /dev/null
+++ b/drivers/of/overlay_test.c
@@ -0,0 +1,120 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * KUnit tests for device tree overlays
+ */
+#include <linux/device/bus.h>
+#include <linux/kconfig.h>
+#include <linux/of.h>
+#include <linux/of_platform.h>
+#include <linux/platform_device.h>
+
+#include <kunit/of.h>
+#include <kunit/test.h>
+
+#include "of_private.h"
+
+static const char * const kunit_node_name = "kunit-test";
+static const char * const kunit_compatible = "test,empty";
+
+/* Test that of_overlay_apply_kunit() adds a node to the live tree */
+static void of_overlay_apply_kunit_apply(struct kunit *test)
+{
+	struct device_node *np;
+
+	KUNIT_ASSERT_EQ(test, 0,
+			of_overlay_apply_kunit(test, kunit_overlay_test));
+
+	np = of_find_node_by_name(NULL, kunit_node_name);
+	KUNIT_EXPECT_NOT_ERR_OR_NULL(test, np);
+	of_node_put(np);
+}
+
+/*
+ * Test that of_overlay_apply_kunit() creates platform devices with the
+ * expected device_node
+ */
+static void of_overlay_apply_kunit_platform_device(struct kunit *test)
+{
+	struct platform_device *pdev;
+	struct device_node *np;
+
+	KUNIT_ASSERT_EQ(test, 0,
+			of_overlay_apply_kunit(test, kunit_overlay_test));
+
+	np = of_find_node_by_name(NULL, kunit_node_name);
+	of_node_put_kunit(test, np);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, np);
+
+	pdev = of_find_device_by_node(np);
+	KUNIT_EXPECT_NOT_ERR_OR_NULL(test, pdev);
+	if (pdev)
+		put_device(&pdev->dev);
+}
+
+static int of_overlay_bus_match_compatible(struct device *dev, const void *data)
+{
+	return of_device_is_compatible(dev->of_node, data);
+}
+
+/* Test that of_overlay_apply_kunit() cleans up after the test is finished */
+static void of_overlay_apply_kunit_cleanup(struct kunit *test)
+{
+	struct kunit fake;
+	struct platform_device *pdev;
+	struct device *dev;
+	struct device_node *np;
+
+	of_root_kunit_skip(test);
+	if (!IS_ENABLED(CONFIG_OF_OVERLAY))
+		kunit_skip(test, "requires CONFIG_OF_OVERLAY to apply overlay");
+	if (!IS_ENABLED(CONFIG_OF_EARLY_FLATTREE))
+		kunit_skip(test, "requires CONFIG_OF_EARLY_FLATTREE for root node");
+
+	kunit_init_test(&fake, "fake test", NULL);
+	KUNIT_ASSERT_EQ(test, fake.status, KUNIT_SUCCESS);
+
+	KUNIT_ASSERT_EQ(test, 0,
+			of_overlay_apply_kunit(&fake, kunit_overlay_test));
+
+	np = of_find_node_by_name(NULL, kunit_node_name);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, np);
+	of_node_put_kunit(&fake, np);
+
+	pdev = of_find_device_by_node(np);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, pdev);
+	put_device(&pdev->dev); /* Not derefing 'pdev' after this */
+
+	/* Remove overlay */
+	kunit_cleanup(&fake);
+
+	/* The node and device should be removed */
+	np = of_find_node_by_name(NULL, kunit_node_name);
+	KUNIT_EXPECT_PTR_EQ(test, NULL, np);
+	of_node_put(np);
+
+	dev = bus_find_device(&platform_bus_type, NULL, kunit_compatible,
+			      of_overlay_bus_match_compatible);
+	KUNIT_EXPECT_PTR_EQ(test, NULL, dev);
+	put_device(dev);
+}
+
+static struct kunit_case of_overlay_apply_kunit_test_cases[] = {
+	KUNIT_CASE(of_overlay_apply_kunit_apply),
+	KUNIT_CASE(of_overlay_apply_kunit_platform_device),
+	KUNIT_CASE(of_overlay_apply_kunit_cleanup),
+	{}
+};
+
+/*
+ * Test suite for test managed device tree overlays.
+ */
+static struct kunit_suite of_overlay_apply_kunit_suite = {
+	.name = "of_overlay_apply_kunit",
+	.test_cases = of_overlay_apply_kunit_test_cases,
+};
+
+kunit_test_suites(
+	&of_overlay_apply_kunit_suite,
+);
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("KUnit tests for device tree overlays");