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, 465 insertions, 0 deletions

diff --git a/drivers/clk/clk-gate_test.c b/drivers/clk/clk-gate_test.c
new file mode 100644
index 000000000..e276cd974
--- /dev/null
+++ b/drivers/clk/clk-gate_test.c
@@ -0,0 +1,465 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Kunit tests for clk gate
+ */
+#include <linux/clk.h>
+#include <linux/clk-provider.h>
+#include <linux/platform_device.h>
+
+#include <kunit/test.h>
+
+static void clk_gate_register_test_dev(struct kunit *test)
+{
+	struct clk_hw *ret;
+	struct platform_device *pdev;
+
+	pdev = platform_device_register_simple("test_gate_device", -1, NULL, 0);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, pdev);
+
+	ret = clk_hw_register_gate(&pdev->dev, "test_gate", NULL, 0, NULL,
+				   0, 0, NULL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ret);
+	KUNIT_EXPECT_STREQ(test, "test_gate", clk_hw_get_name(ret));
+	KUNIT_EXPECT_EQ(test, 0UL, clk_hw_get_flags(ret));
+
+	clk_hw_unregister_gate(ret);
+	platform_device_put(pdev);
+}
+
+static void clk_gate_register_test_parent_names(struct kunit *test)
+{
+	struct clk_hw *parent;
+	struct clk_hw *ret;
+
+	parent = clk_hw_register_fixed_rate(NULL, "test_parent", NULL, 0,
+					    1000000);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, parent);
+
+	ret = clk_hw_register_gate(NULL, "test_gate", "test_parent", 0, NULL,
+				   0, 0, NULL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ret);
+	KUNIT_EXPECT_PTR_EQ(test, parent, clk_hw_get_parent(ret));
+
+	clk_hw_unregister_gate(ret);
+	clk_hw_unregister_fixed_rate(parent);
+}
+
+static void clk_gate_register_test_parent_data(struct kunit *test)
+{
+	struct clk_hw *parent;
+	struct clk_hw *ret;
+	struct clk_parent_data pdata = { };
+
+	parent = clk_hw_register_fixed_rate(NULL, "test_parent", NULL, 0,
+					    1000000);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, parent);
+	pdata.hw = parent;
+
+	ret = clk_hw_register_gate_parent_data(NULL, "test_gate", &pdata, 0,
+					       NULL, 0, 0, NULL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ret);
+	KUNIT_EXPECT_PTR_EQ(test, parent, clk_hw_get_parent(ret));
+
+	clk_hw_unregister_gate(ret);
+	clk_hw_unregister_fixed_rate(parent);
+}
+
+static void clk_gate_register_test_parent_data_legacy(struct kunit *test)
+{
+	struct clk_hw *parent;
+	struct clk_hw *ret;
+	struct clk_parent_data pdata = { };
+
+	parent = clk_hw_register_fixed_rate(NULL, "test_parent", NULL, 0,
+					    1000000);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, parent);
+	pdata.name = "test_parent";
+
+	ret = clk_hw_register_gate_parent_data(NULL, "test_gate", &pdata, 0,
+					       NULL, 0, 0, NULL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ret);
+	KUNIT_EXPECT_PTR_EQ(test, parent, clk_hw_get_parent(ret));
+
+	clk_hw_unregister_gate(ret);
+	clk_hw_unregister_fixed_rate(parent);
+}
+
+static void clk_gate_register_test_parent_hw(struct kunit *test)
+{
+	struct clk_hw *parent;
+	struct clk_hw *ret;
+
+	parent = clk_hw_register_fixed_rate(NULL, "test_parent", NULL, 0,
+					    1000000);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, parent);
+
+	ret = clk_hw_register_gate_parent_hw(NULL, "test_gate", parent, 0, NULL,
+					     0, 0, NULL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ret);
+	KUNIT_EXPECT_PTR_EQ(test, parent, clk_hw_get_parent(ret));
+
+	clk_hw_unregister_gate(ret);
+	clk_hw_unregister_fixed_rate(parent);
+}
+
+static void clk_gate_register_test_hiword_invalid(struct kunit *test)
+{
+	struct clk_hw *ret;
+
+	ret = clk_hw_register_gate(NULL, "test_gate", NULL, 0, NULL,
+				   20, CLK_GATE_HIWORD_MASK, NULL);
+
+	KUNIT_EXPECT_TRUE(test, IS_ERR(ret));
+}
+
+static struct kunit_case clk_gate_register_test_cases[] = {
+	KUNIT_CASE(clk_gate_register_test_dev),
+	KUNIT_CASE(clk_gate_register_test_parent_names),
+	KUNIT_CASE(clk_gate_register_test_parent_data),
+	KUNIT_CASE(clk_gate_register_test_parent_data_legacy),
+	KUNIT_CASE(clk_gate_register_test_parent_hw),
+	KUNIT_CASE(clk_gate_register_test_hiword_invalid),
+	{}
+};
+
+static struct kunit_suite clk_gate_register_test_suite = {
+	.name = "clk-gate-register-test",
+	.test_cases = clk_gate_register_test_cases,
+};
+
+struct clk_gate_test_context {
+	void __iomem *fake_mem;
+	struct clk_hw *hw;
+	struct clk_hw *parent;
+	__le32 fake_reg; /* Keep at end, KASAN can detect out of bounds */
+};
+
+static struct clk_gate_test_context *clk_gate_test_alloc_ctx(struct kunit *test)
+{
+	struct clk_gate_test_context *ctx;
+
+	test->priv = ctx = kunit_kzalloc(test, sizeof(*ctx), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx);
+	ctx->fake_mem = (void __force __iomem *)&ctx->fake_reg;
+
+	return ctx;
+}
+
+static void clk_gate_test_parent_rate(struct kunit *test)
+{
+	struct clk_gate_test_context *ctx = test->priv;
+	struct clk_hw *parent = ctx->parent;
+	struct clk_hw *hw = ctx->hw;
+	unsigned long prate = clk_hw_get_rate(parent);
+	unsigned long rate = clk_hw_get_rate(hw);
+
+	KUNIT_EXPECT_EQ(test, prate, rate);
+}
+
+static void clk_gate_test_enable(struct kunit *test)
+{
+	struct clk_gate_test_context *ctx = test->priv;
+	struct clk_hw *parent = ctx->parent;
+	struct clk_hw *hw = ctx->hw;
+	struct clk *clk = hw->clk;
+	u32 enable_val = BIT(5);
+
+	KUNIT_ASSERT_EQ(test, clk_prepare_enable(clk), 0);
+
+	KUNIT_EXPECT_EQ(test, enable_val, le32_to_cpu(ctx->fake_reg));
+	KUNIT_EXPECT_TRUE(test, clk_hw_is_enabled(hw));
+	KUNIT_EXPECT_TRUE(test, clk_hw_is_prepared(hw));
+	KUNIT_EXPECT_TRUE(test, clk_hw_is_enabled(parent));
+	KUNIT_EXPECT_TRUE(test, clk_hw_is_prepared(parent));
+}
+
+static void clk_gate_test_disable(struct kunit *test)
+{
+	struct clk_gate_test_context *ctx = test->priv;
+	struct clk_hw *parent = ctx->parent;
+	struct clk_hw *hw = ctx->hw;
+	struct clk *clk = hw->clk;
+	u32 enable_val = BIT(5);
+	u32 disable_val = 0;
+
+	KUNIT_ASSERT_EQ(test, clk_prepare_enable(clk), 0);
+	KUNIT_ASSERT_EQ(test, enable_val, le32_to_cpu(ctx->fake_reg));
+
+	clk_disable_unprepare(clk);
+	KUNIT_EXPECT_EQ(test, disable_val, le32_to_cpu(ctx->fake_reg));
+	KUNIT_EXPECT_FALSE(test, clk_hw_is_enabled(hw));
+	KUNIT_EXPECT_FALSE(test, clk_hw_is_prepared(hw));
+	KUNIT_EXPECT_FALSE(test, clk_hw_is_enabled(parent));
+	KUNIT_EXPECT_FALSE(test, clk_hw_is_prepared(parent));
+}
+
+static struct kunit_case clk_gate_test_cases[] = {
+	KUNIT_CASE(clk_gate_test_parent_rate),
+	KUNIT_CASE(clk_gate_test_enable),
+	KUNIT_CASE(clk_gate_test_disable),
+	{}
+};
+
+static int clk_gate_test_init(struct kunit *test)
+{
+	struct clk_hw *parent;
+	struct clk_hw *hw;
+	struct clk_gate_test_context *ctx;
+
+	ctx = clk_gate_test_alloc_ctx(test);
+	parent = clk_hw_register_fixed_rate(NULL, "test_parent", NULL, 0,
+					    2000000);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, parent);
+
+	hw = clk_hw_register_gate_parent_hw(NULL, "test_gate", parent, 0,
+					    ctx->fake_mem, 5, 0, NULL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, hw);
+
+	ctx->hw = hw;
+	ctx->parent = parent;
+
+	return 0;
+}
+
+static void clk_gate_test_exit(struct kunit *test)
+{
+	struct clk_gate_test_context *ctx = test->priv;
+
+	clk_hw_unregister_gate(ctx->hw);
+	clk_hw_unregister_fixed_rate(ctx->parent);
+}
+
+static struct kunit_suite clk_gate_test_suite = {
+	.name = "clk-gate-test",
+	.init = clk_gate_test_init,
+	.exit = clk_gate_test_exit,
+	.test_cases = clk_gate_test_cases,
+};
+
+static void clk_gate_test_invert_enable(struct kunit *test)
+{
+	struct clk_gate_test_context *ctx = test->priv;
+	struct clk_hw *parent = ctx->parent;
+	struct clk_hw *hw = ctx->hw;
+	struct clk *clk = hw->clk;
+	u32 enable_val = 0;
+
+	KUNIT_ASSERT_EQ(test, clk_prepare_enable(clk), 0);
+
+	KUNIT_EXPECT_EQ(test, enable_val, le32_to_cpu(ctx->fake_reg));
+	KUNIT_EXPECT_TRUE(test, clk_hw_is_enabled(hw));
+	KUNIT_EXPECT_TRUE(test, clk_hw_is_prepared(hw));
+	KUNIT_EXPECT_TRUE(test, clk_hw_is_enabled(parent));
+	KUNIT_EXPECT_TRUE(test, clk_hw_is_prepared(parent));
+}
+
+static void clk_gate_test_invert_disable(struct kunit *test)
+{
+	struct clk_gate_test_context *ctx = test->priv;
+	struct clk_hw *parent = ctx->parent;
+	struct clk_hw *hw = ctx->hw;
+	struct clk *clk = hw->clk;
+	u32 enable_val = 0;
+	u32 disable_val = BIT(15);
+
+	KUNIT_ASSERT_EQ(test, clk_prepare_enable(clk), 0);
+	KUNIT_ASSERT_EQ(test, enable_val, le32_to_cpu(ctx->fake_reg));
+
+	clk_disable_unprepare(clk);
+	KUNIT_EXPECT_EQ(test, disable_val, le32_to_cpu(ctx->fake_reg));
+	KUNIT_EXPECT_FALSE(test, clk_hw_is_enabled(hw));
+	KUNIT_EXPECT_FALSE(test, clk_hw_is_prepared(hw));
+	KUNIT_EXPECT_FALSE(test, clk_hw_is_enabled(parent));
+	KUNIT_EXPECT_FALSE(test, clk_hw_is_prepared(parent));
+}
+
+static struct kunit_case clk_gate_test_invert_cases[] = {
+	KUNIT_CASE(clk_gate_test_invert_enable),
+	KUNIT_CASE(clk_gate_test_invert_disable),
+	{}
+};
+
+static int clk_gate_test_invert_init(struct kunit *test)
+{
+	struct clk_hw *parent;
+	struct clk_hw *hw;
+	struct clk_gate_test_context *ctx;
+
+	ctx = clk_gate_test_alloc_ctx(test);
+	parent = clk_hw_register_fixed_rate(NULL, "test_parent", NULL, 0,
+					    2000000);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, parent);
+
+	ctx->fake_reg = cpu_to_le32(BIT(15)); /* Default to off */
+	hw = clk_hw_register_gate_parent_hw(NULL, "test_gate", parent, 0,
+					    ctx->fake_mem, 15,
+					    CLK_GATE_SET_TO_DISABLE, NULL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, hw);
+
+	ctx->hw = hw;
+	ctx->parent = parent;
+
+	return 0;
+}
+
+static struct kunit_suite clk_gate_test_invert_suite = {
+	.name = "clk-gate-invert-test",
+	.init = clk_gate_test_invert_init,
+	.exit = clk_gate_test_exit,
+	.test_cases = clk_gate_test_invert_cases,
+};
+
+static void clk_gate_test_hiword_enable(struct kunit *test)
+{
+	struct clk_gate_test_context *ctx = test->priv;
+	struct clk_hw *parent = ctx->parent;
+	struct clk_hw *hw = ctx->hw;
+	struct clk *clk = hw->clk;
+	u32 enable_val = BIT(9) | BIT(9 + 16);
+
+	KUNIT_ASSERT_EQ(test, clk_prepare_enable(clk), 0);
+
+	KUNIT_EXPECT_EQ(test, enable_val, le32_to_cpu(ctx->fake_reg));
+	KUNIT_EXPECT_TRUE(test, clk_hw_is_enabled(hw));
+	KUNIT_EXPECT_TRUE(test, clk_hw_is_prepared(hw));
+	KUNIT_EXPECT_TRUE(test, clk_hw_is_enabled(parent));
+	KUNIT_EXPECT_TRUE(test, clk_hw_is_prepared(parent));
+}
+
+static void clk_gate_test_hiword_disable(struct kunit *test)
+{
+	struct clk_gate_test_context *ctx = test->priv;
+	struct clk_hw *parent = ctx->parent;
+	struct clk_hw *hw = ctx->hw;
+	struct clk *clk = hw->clk;
+	u32 enable_val = BIT(9) | BIT(9 + 16);
+	u32 disable_val = BIT(9 + 16);
+
+	KUNIT_ASSERT_EQ(test, clk_prepare_enable(clk), 0);
+	KUNIT_ASSERT_EQ(test, enable_val, le32_to_cpu(ctx->fake_reg));
+
+	clk_disable_unprepare(clk);
+	KUNIT_EXPECT_EQ(test, disable_val, le32_to_cpu(ctx->fake_reg));
+	KUNIT_EXPECT_FALSE(test, clk_hw_is_enabled(hw));
+	KUNIT_EXPECT_FALSE(test, clk_hw_is_prepared(hw));
+	KUNIT_EXPECT_FALSE(test, clk_hw_is_enabled(parent));
+	KUNIT_EXPECT_FALSE(test, clk_hw_is_prepared(parent));
+}
+
+static struct kunit_case clk_gate_test_hiword_cases[] = {
+	KUNIT_CASE(clk_gate_test_hiword_enable),
+	KUNIT_CASE(clk_gate_test_hiword_disable),
+	{}
+};
+
+static int clk_gate_test_hiword_init(struct kunit *test)
+{
+	struct clk_hw *parent;
+	struct clk_hw *hw;
+	struct clk_gate_test_context *ctx;
+
+	ctx = clk_gate_test_alloc_ctx(test);
+	parent = clk_hw_register_fixed_rate(NULL, "test_parent", NULL, 0,
+					    2000000);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, parent);
+
+	hw = clk_hw_register_gate_parent_hw(NULL, "test_gate", parent, 0,
+					    ctx->fake_mem, 9,
+					    CLK_GATE_HIWORD_MASK, NULL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, hw);
+
+	ctx->hw = hw;
+	ctx->parent = parent;
+
+	return 0;
+}
+
+static struct kunit_suite clk_gate_test_hiword_suite = {
+	.name = "clk-gate-hiword-test",
+	.init = clk_gate_test_hiword_init,
+	.exit = clk_gate_test_exit,
+	.test_cases = clk_gate_test_hiword_cases,
+};
+
+static void clk_gate_test_is_enabled(struct kunit *test)
+{
+	struct clk_hw *hw;
+	struct clk_gate_test_context *ctx;
+
+	ctx = clk_gate_test_alloc_ctx(test);
+	ctx->fake_reg = cpu_to_le32(BIT(7));
+	hw = clk_hw_register_gate(NULL, "test_gate", NULL, 0, ctx->fake_mem, 7,
+				  0, NULL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, hw);
+	KUNIT_ASSERT_TRUE(test, clk_hw_is_enabled(hw));
+
+	clk_hw_unregister_gate(hw);
+}
+
+static void clk_gate_test_is_disabled(struct kunit *test)
+{
+	struct clk_hw *hw;
+	struct clk_gate_test_context *ctx;
+
+	ctx = clk_gate_test_alloc_ctx(test);
+	ctx->fake_reg = cpu_to_le32(BIT(4));
+	hw = clk_hw_register_gate(NULL, "test_gate", NULL, 0, ctx->fake_mem, 7,
+				  0, NULL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, hw);
+	KUNIT_ASSERT_FALSE(test, clk_hw_is_enabled(hw));
+
+	clk_hw_unregister_gate(hw);
+}
+
+static void clk_gate_test_is_enabled_inverted(struct kunit *test)
+{
+	struct clk_hw *hw;
+	struct clk_gate_test_context *ctx;
+
+	ctx = clk_gate_test_alloc_ctx(test);
+	ctx->fake_reg = cpu_to_le32(BIT(31));
+	hw = clk_hw_register_gate(NULL, "test_gate", NULL, 0, ctx->fake_mem, 2,
+				  CLK_GATE_SET_TO_DISABLE, NULL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, hw);
+	KUNIT_ASSERT_TRUE(test, clk_hw_is_enabled(hw));
+
+	clk_hw_unregister_gate(hw);
+}
+
+static void clk_gate_test_is_disabled_inverted(struct kunit *test)
+{
+	struct clk_hw *hw;
+	struct clk_gate_test_context *ctx;
+
+	ctx = clk_gate_test_alloc_ctx(test);
+	ctx->fake_reg = cpu_to_le32(BIT(29));
+	hw = clk_hw_register_gate(NULL, "test_gate", NULL, 0, ctx->fake_mem, 29,
+				  CLK_GATE_SET_TO_DISABLE, NULL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, hw);
+	KUNIT_ASSERT_FALSE(test, clk_hw_is_enabled(hw));
+
+	clk_hw_unregister_gate(hw);
+}
+
+static struct kunit_case clk_gate_test_enabled_cases[] = {
+	KUNIT_CASE(clk_gate_test_is_enabled),
+	KUNIT_CASE(clk_gate_test_is_disabled),
+	KUNIT_CASE(clk_gate_test_is_enabled_inverted),
+	KUNIT_CASE(clk_gate_test_is_disabled_inverted),
+	{}
+};
+
+static struct kunit_suite clk_gate_test_enabled_suite = {
+	.name = "clk-gate-is_enabled-test",
+	.test_cases = clk_gate_test_enabled_cases,
+};
+
+kunit_test_suites(
+	&clk_gate_register_test_suite,
+	&clk_gate_test_suite,
+	&clk_gate_test_invert_suite,
+	&clk_gate_test_hiword_suite,
+	&clk_gate_test_enabled_suite
+);
+MODULE_DESCRIPTION("Kunit tests for clk gate");
+MODULE_LICENSE("GPL v2");