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, 186 insertions, 0 deletions

diff --git a/drivers/cpufreq/bmips-cpufreq.c b/drivers/cpufreq/bmips-cpufreq.c
new file mode 100644
index 000000000..a8e35bc75
--- /dev/null
+++ b/drivers/cpufreq/bmips-cpufreq.c
@@ -0,0 +1,186 @@
+/*
+ * CPU frequency scaling for Broadcom BMIPS SoCs
+ *
+ * Copyright (c) 2017 Broadcom
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/cpufreq.h>
+#include <linux/module.h>
+#include <linux/of_address.h>
+#include <linux/slab.h>
+
+/* for mips_hpt_frequency */
+#include <asm/time.h>
+
+#define BMIPS_CPUFREQ_PREFIX	"bmips"
+#define BMIPS_CPUFREQ_NAME	BMIPS_CPUFREQ_PREFIX "-cpufreq"
+
+#define TRANSITION_LATENCY	(25 * 1000)	/* 25 us */
+
+#define BMIPS5_CLK_DIV_SET_SHIFT	0x7
+#define BMIPS5_CLK_DIV_SHIFT		0x4
+#define BMIPS5_CLK_DIV_MASK		0xf
+
+enum bmips_type {
+	BMIPS5000,
+	BMIPS5200,
+};
+
+struct cpufreq_compat {
+	const char *compatible;
+	unsigned int bmips_type;
+	unsigned int clk_mult;
+	unsigned int max_freqs;
+};
+
+#define BMIPS(c, t, m, f) { \
+	.compatible = c, \
+	.bmips_type = (t), \
+	.clk_mult = (m), \
+	.max_freqs = (f), \
+}
+
+static struct cpufreq_compat bmips_cpufreq_compat[] = {
+	BMIPS("brcm,bmips5000", BMIPS5000, 8, 4),
+	BMIPS("brcm,bmips5200", BMIPS5200, 8, 4),
+	{ }
+};
+
+static struct cpufreq_compat *priv;
+
+static int htp_freq_to_cpu_freq(unsigned int clk_mult)
+{
+	return mips_hpt_frequency * clk_mult / 1000;
+}
+
+static struct cpufreq_frequency_table *
+bmips_cpufreq_get_freq_table(const struct cpufreq_policy *policy)
+{
+	struct cpufreq_frequency_table *table;
+	unsigned long cpu_freq;
+	int i;
+
+	cpu_freq = htp_freq_to_cpu_freq(priv->clk_mult);
+
+	table = kmalloc_objs(*table, priv->max_freqs + 1);
+	if (!table)
+		return ERR_PTR(-ENOMEM);
+
+	for (i = 0; i < priv->max_freqs; i++) {
+		table[i].frequency = cpu_freq / (1 << i);
+		table[i].driver_data = i;
+	}
+	table[i].frequency = CPUFREQ_TABLE_END;
+
+	return table;
+}
+
+static unsigned int bmips_cpufreq_get(unsigned int cpu)
+{
+	unsigned int div;
+	uint32_t mode;
+
+	switch (priv->bmips_type) {
+	case BMIPS5200:
+	case BMIPS5000:
+		mode = read_c0_brcm_mode();
+		div = ((mode >> BMIPS5_CLK_DIV_SHIFT) & BMIPS5_CLK_DIV_MASK);
+		break;
+	default:
+		div = 0;
+	}
+
+	return htp_freq_to_cpu_freq(priv->clk_mult) / (1 << div);
+}
+
+static int bmips_cpufreq_target_index(struct cpufreq_policy *policy,
+				      unsigned int index)
+{
+	unsigned int div = policy->freq_table[index].driver_data;
+
+	switch (priv->bmips_type) {
+	case BMIPS5200:
+	case BMIPS5000:
+		change_c0_brcm_mode(BMIPS5_CLK_DIV_MASK << BMIPS5_CLK_DIV_SHIFT,
+				    (1 << BMIPS5_CLK_DIV_SET_SHIFT) |
+				    (div << BMIPS5_CLK_DIV_SHIFT));
+		break;
+	default:
+		return -ENOTSUPP;
+	}
+
+	return 0;
+}
+
+static void bmips_cpufreq_exit(struct cpufreq_policy *policy)
+{
+	kfree(policy->freq_table);
+}
+
+static int bmips_cpufreq_init(struct cpufreq_policy *policy)
+{
+	struct cpufreq_frequency_table *freq_table;
+
+	freq_table = bmips_cpufreq_get_freq_table(policy);
+	if (IS_ERR(freq_table)) {
+		pr_err("%s: couldn't determine frequency table (%ld).\n",
+			BMIPS_CPUFREQ_NAME, PTR_ERR(freq_table));
+		return PTR_ERR(freq_table);
+	}
+
+	cpufreq_generic_init(policy, freq_table, TRANSITION_LATENCY);
+	pr_info("%s: registered\n", BMIPS_CPUFREQ_NAME);
+
+	return 0;
+}
+
+static struct cpufreq_driver bmips_cpufreq_driver = {
+	.flags		= CPUFREQ_NEED_INITIAL_FREQ_CHECK,
+	.verify		= cpufreq_generic_frequency_table_verify,
+	.target_index	= bmips_cpufreq_target_index,
+	.get		= bmips_cpufreq_get,
+	.init		= bmips_cpufreq_init,
+	.exit		= bmips_cpufreq_exit,
+	.name		= BMIPS_CPUFREQ_PREFIX,
+};
+
+static int __init bmips_cpufreq_driver_init(void)
+{
+	struct cpufreq_compat *cc;
+	struct device_node *np;
+
+	for (cc = bmips_cpufreq_compat; cc->compatible; cc++) {
+		np = of_find_compatible_node(NULL, "cpu", cc->compatible);
+		if (np) {
+			of_node_put(np);
+			priv = cc;
+			break;
+		}
+	}
+
+	/* We hit the guard element of the array. No compatible CPU found. */
+	if (!cc->compatible)
+		return -ENODEV;
+
+	return cpufreq_register_driver(&bmips_cpufreq_driver);
+}
+module_init(bmips_cpufreq_driver_init);
+
+static void __exit bmips_cpufreq_driver_exit(void)
+{
+	cpufreq_unregister_driver(&bmips_cpufreq_driver);
+}
+module_exit(bmips_cpufreq_driver_exit);
+
+MODULE_AUTHOR("Markus Mayer <mmayer@broadcom.com>");
+MODULE_DESCRIPTION("CPUfreq driver for Broadcom BMIPS SoCs");
+MODULE_LICENSE("GPL");