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, 127 insertions, 0 deletions

diff --git a/drivers/ras/ras.c b/drivers/ras/ras.c
new file mode 100644
index 000000000..03df3db62
--- /dev/null
+++ b/drivers/ras/ras.c
@@ -0,0 +1,127 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2014 Intel Corporation
+ *
+ * Authors:
+ *	Chen, Gong <gong.chen@linux.intel.com>
+ */
+
+#include <linux/init.h>
+#include <linux/ras.h>
+#include <linux/uuid.h>
+
+#if IS_ENABLED(CONFIG_AMD_ATL)
+/*
+ * Once set, this function pointer should never be unset.
+ *
+ * The library module will set this pointer if it successfully loads. The module
+ * should not be unloaded except for testing and debug purposes.
+ */
+static unsigned long (*amd_atl_umc_na_to_spa)(struct atl_err *err);
+
+void amd_atl_register_decoder(unsigned long (*f)(struct atl_err *))
+{
+	amd_atl_umc_na_to_spa = f;
+}
+EXPORT_SYMBOL_GPL(amd_atl_register_decoder);
+
+void amd_atl_unregister_decoder(void)
+{
+	amd_atl_umc_na_to_spa = NULL;
+}
+EXPORT_SYMBOL_GPL(amd_atl_unregister_decoder);
+
+unsigned long amd_convert_umc_mca_addr_to_sys_addr(struct atl_err *err)
+{
+	if (!amd_atl_umc_na_to_spa)
+		return -EINVAL;
+
+	return amd_atl_umc_na_to_spa(err);
+}
+EXPORT_SYMBOL_GPL(amd_convert_umc_mca_addr_to_sys_addr);
+#endif /* CONFIG_AMD_ATL */
+
+#define CREATE_TRACE_POINTS
+#define TRACE_INCLUDE_PATH ../../include/ras
+#include <ras/ras_event.h>
+
+void log_non_standard_event(const guid_t *sec_type, const guid_t *fru_id,
+			    const char *fru_text, const u8 sev, const u8 *err,
+			    const u32 len)
+{
+	trace_non_standard_event(sec_type, fru_id, fru_text, sev, err, len);
+}
+EXPORT_SYMBOL_GPL(log_non_standard_event);
+
+void log_arm_hw_error(struct cper_sec_proc_arm *err, const u8 sev)
+{
+	struct cper_arm_err_info *err_info;
+	struct cper_arm_ctx_info *ctx_info;
+	u8 *ven_err_data;
+	u32 ctx_len = 0;
+	int n, sz, cpu;
+	s32 vsei_len;
+	u32 pei_len;
+	u8 *pei_err, *ctx_err;
+
+	pei_len = sizeof(struct cper_arm_err_info) * err->err_info_num;
+	pei_err = (u8 *)(err + 1);
+
+	err_info = (struct cper_arm_err_info *)(err + 1);
+	ctx_info = (struct cper_arm_ctx_info *)(err_info + err->err_info_num);
+	ctx_err = (u8 *)ctx_info;
+
+	for (n = 0; n < err->context_info_num; n++) {
+		sz = sizeof(struct cper_arm_ctx_info);
+
+		if (sz + (long)ctx_info - (long)err >= err->section_length)
+			sz += ctx_info->size;
+
+		ctx_info = (struct cper_arm_ctx_info *)((long)ctx_info + sz);
+		ctx_len += sz;
+	}
+
+	vsei_len = err->section_length - (sizeof(struct cper_sec_proc_arm) + pei_len + ctx_len);
+	if (vsei_len < 0) {
+		pr_warn(FW_BUG "section length: %d\n", err->section_length);
+		pr_warn(FW_BUG "section length is too small\n");
+		pr_warn(FW_BUG "firmware-generated error record is incorrect\n");
+		vsei_len = 0;
+	}
+	ven_err_data = (u8 *)ctx_info;
+
+	cpu = GET_LOGICAL_INDEX(err->mpidr);
+	if (cpu < 0)
+		cpu = -1;
+
+	trace_arm_event(err, pei_err, pei_len, ctx_err, ctx_len,
+			ven_err_data, (u32)vsei_len, sev, cpu);
+}
+
+static int __init ras_init(void)
+{
+	int rc = 0;
+
+	ras_debugfs_init();
+	rc = ras_add_daemon_trace();
+
+	return rc;
+}
+subsys_initcall(ras_init);
+
+#if defined(CONFIG_ACPI_EXTLOG) || defined(CONFIG_ACPI_EXTLOG_MODULE)
+EXPORT_TRACEPOINT_SYMBOL_GPL(extlog_mem_event);
+#endif
+EXPORT_TRACEPOINT_SYMBOL_GPL(mc_event);
+EXPORT_TRACEPOINT_SYMBOL_GPL(non_standard_event);
+EXPORT_TRACEPOINT_SYMBOL_GPL(arm_event);
+
+static int __init parse_ras_param(char *str)
+{
+#ifdef CONFIG_RAS_CEC
+	parse_cec_param(str);
+#endif
+
+	return 1;
+}
+__setup("ras", parse_ras_param);