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, 205 insertions, 0 deletions

diff --git a/drivers/acpi/acpica/dsdebug.c b/drivers/acpi/acpica/dsdebug.c
new file mode 100644
index 000000000..691c00159
--- /dev/null
+++ b/drivers/acpi/acpica/dsdebug.c
@@ -0,0 +1,205 @@
+// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0
+/******************************************************************************
+ *
+ * Module Name: dsdebug - Parser/Interpreter interface - debugging
+ *
+ * Copyright (C) 2000 - 2026, Intel Corp.
+ *
+ *****************************************************************************/
+
+#include <acpi/acpi.h>
+#include "accommon.h"
+#include "acdispat.h"
+#include "acnamesp.h"
+#ifdef ACPI_DISASSEMBLER
+#include "acdisasm.h"
+#endif
+#include "acinterp.h"
+
+#define _COMPONENT          ACPI_DISPATCHER
+ACPI_MODULE_NAME("dsdebug")
+
+#if defined(ACPI_DEBUG_OUTPUT) || defined(ACPI_DEBUGGER)
+/* Local prototypes */
+static void
+acpi_ds_print_node_pathname(struct acpi_namespace_node *node,
+			    const char *message);
+
+/*******************************************************************************
+ *
+ * FUNCTION:    acpi_ds_print_node_pathname
+ *
+ * PARAMETERS:  node            - Object
+ *              message         - Prefix message
+ *
+ * DESCRIPTION: Print an object's full namespace pathname
+ *              Manages allocation/freeing of a pathname buffer
+ *
+ ******************************************************************************/
+
+static void
+acpi_ds_print_node_pathname(struct acpi_namespace_node *node,
+			    const char *message)
+{
+	struct acpi_buffer buffer;
+	acpi_status status;
+
+	ACPI_FUNCTION_TRACE(ds_print_node_pathname);
+
+	if (!node) {
+		ACPI_DEBUG_PRINT_RAW((ACPI_DB_DISPATCH, "[NULL NAME]"));
+		return_VOID;
+	}
+
+	/* Convert handle to full pathname and print it (with supplied message) */
+
+	buffer.length = ACPI_ALLOCATE_LOCAL_BUFFER;
+
+	status = acpi_ns_handle_to_pathname(node, &buffer, TRUE);
+	if (ACPI_SUCCESS(status)) {
+		if (message) {
+			ACPI_DEBUG_PRINT_RAW((ACPI_DB_DISPATCH, "%s ",
+					      message));
+		}
+
+		ACPI_DEBUG_PRINT_RAW((ACPI_DB_DISPATCH, "[%s] (Node %p)",
+				      (char *)buffer.pointer, node));
+		ACPI_FREE(buffer.pointer);
+	}
+
+	return_VOID;
+}
+
+/*******************************************************************************
+ *
+ * FUNCTION:    acpi_ds_dump_method_stack
+ *
+ * PARAMETERS:  status          - Method execution status
+ *              walk_state      - Current state of the parse tree walk
+ *              op              - Executing parse op
+ *
+ * RETURN:      None
+ *
+ * DESCRIPTION: Called when a method has been aborted because of an error.
+ *              Dumps the method execution stack.
+ *
+ ******************************************************************************/
+
+void
+acpi_ds_dump_method_stack(acpi_status status,
+			  struct acpi_walk_state *walk_state,
+			  union acpi_parse_object *op)
+{
+	union acpi_parse_object *next;
+	struct acpi_thread_state *thread;
+	struct acpi_walk_state *next_walk_state;
+	struct acpi_namespace_node *previous_method = NULL;
+	union acpi_operand_object *method_desc;
+
+	ACPI_FUNCTION_TRACE(ds_dump_method_stack);
+
+	/* Ignore control codes, they are not errors */
+
+	if (ACPI_CNTL_EXCEPTION(status)) {
+		return_VOID;
+	}
+
+	/* We may be executing a deferred opcode */
+
+	if (walk_state->deferred_node) {
+		ACPI_DEBUG_PRINT((ACPI_DB_DISPATCH,
+				  "Executing subtree for Buffer/Package/Region\n"));
+		return_VOID;
+	}
+
+	/*
+	 * If there is no Thread, we are not actually executing a method.
+	 * This can happen when the iASL compiler calls the interpreter
+	 * to perform constant folding.
+	 */
+	thread = walk_state->thread;
+	if (!thread) {
+		return_VOID;
+	}
+
+	/* Display exception and method name */
+
+	ACPI_DEBUG_PRINT((ACPI_DB_DISPATCH,
+			  "\n**** Exception %s during execution of method ",
+			  acpi_format_exception(status)));
+
+	acpi_ds_print_node_pathname(walk_state->method_node, NULL);
+
+	/* Display stack of executing methods */
+
+	ACPI_DEBUG_PRINT_RAW((ACPI_DB_DISPATCH,
+			      "\n\nMethod Execution Stack:\n"));
+	next_walk_state = thread->walk_state_list;
+
+	/* Walk list of linked walk states */
+
+	while (next_walk_state) {
+		method_desc = next_walk_state->method_desc;
+		if (method_desc) {
+			acpi_ex_stop_trace_method((struct acpi_namespace_node *)
+						  method_desc->method.node,
+						  method_desc, walk_state);
+		}
+
+		ACPI_DEBUG_PRINT((ACPI_DB_DISPATCH,
+				  "    Method [%4.4s] executing: ",
+				  acpi_ut_get_node_name(next_walk_state->
+							method_node)));
+
+		/* First method is the currently executing method */
+
+		if (next_walk_state == walk_state) {
+			if (op) {
+
+				/* Display currently executing ASL statement */
+
+				next = op->common.next;
+				op->common.next = NULL;
+
+#ifdef ACPI_DISASSEMBLER
+				if (walk_state->method_node !=
+				    acpi_gbl_root_node) {
+
+					/* More verbose if not module-level code */
+
+					acpi_os_printf("Failed at ");
+					acpi_dm_disassemble(next_walk_state, op,
+							    ACPI_UINT32_MAX);
+				}
+#endif
+				op->common.next = next;
+			}
+		} else {
+			/*
+			 * This method has called another method
+			 * NOTE: the method call parse subtree is already deleted at
+			 * this point, so we cannot disassemble the method invocation.
+			 */
+			ACPI_DEBUG_PRINT_RAW((ACPI_DB_DISPATCH,
+					      "Call to method "));
+			acpi_ds_print_node_pathname(previous_method, NULL);
+		}
+
+		previous_method = next_walk_state->method_node;
+		next_walk_state = next_walk_state->next;
+		ACPI_DEBUG_PRINT_RAW((ACPI_DB_DISPATCH, "\n"));
+	}
+
+	return_VOID;
+}
+
+#else
+void
+acpi_ds_dump_method_stack(acpi_status status,
+			  struct acpi_walk_state *walk_state,
+			  union acpi_parse_object *op)
+{
+	return;
+}
+
+#endif