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, 234 insertions, 0 deletions

diff --git a/drivers/cpufreq/scpi-cpufreq.c b/drivers/cpufreq/scpi-cpufreq.c
new file mode 100644
index 000000000..09e29b2a6
--- /dev/null
+++ b/drivers/cpufreq/scpi-cpufreq.c
@@ -0,0 +1,234 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * System Control and Power Interface (SCPI) based CPUFreq Interface driver
+ *
+ * Copyright (C) 2015 ARM Ltd.
+ * Sudeep Holla <sudeep.holla@arm.com>
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/clk.h>
+#include <linux/cpu.h>
+#include <linux/cpufreq.h>
+#include <linux/cpumask.h>
+#include <linux/export.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/pm_opp.h>
+#include <linux/scpi_protocol.h>
+#include <linux/slab.h>
+#include <linux/types.h>
+
+struct scpi_data {
+	struct clk *clk;
+	struct device *cpu_dev;
+};
+
+static struct scpi_ops *scpi_ops;
+
+static unsigned int scpi_cpufreq_get_rate(unsigned int cpu)
+{
+	struct cpufreq_policy *policy;
+	struct scpi_data *priv;
+	unsigned long rate;
+
+	policy = cpufreq_cpu_get_raw(cpu);
+	if (unlikely(!policy))
+		return 0;
+
+	priv = policy->driver_data;
+	rate = clk_get_rate(priv->clk);
+
+	return rate / 1000;
+}
+
+static int
+scpi_cpufreq_set_target(struct cpufreq_policy *policy, unsigned int index)
+{
+	unsigned long freq_khz = policy->freq_table[index].frequency;
+	struct scpi_data *priv = policy->driver_data;
+	unsigned long rate = freq_khz * 1000;
+	int ret;
+
+	ret = clk_set_rate(priv->clk, rate);
+
+	if (ret)
+		return ret;
+
+	if (clk_get_rate(priv->clk) / 1000 != freq_khz)
+		return -EIO;
+
+	return 0;
+}
+
+static int
+scpi_get_sharing_cpus(struct device *cpu_dev, struct cpumask *cpumask)
+{
+	int cpu, domain, tdomain;
+	struct device *tcpu_dev;
+
+	domain = scpi_ops->device_domain_id(cpu_dev);
+	if (domain < 0)
+		return domain;
+
+	for_each_present_cpu(cpu) {
+		if (cpu == cpu_dev->id)
+			continue;
+
+		tcpu_dev = get_cpu_device(cpu);
+		if (!tcpu_dev)
+			continue;
+
+		tdomain = scpi_ops->device_domain_id(tcpu_dev);
+		if (tdomain == domain)
+			cpumask_set_cpu(cpu, cpumask);
+	}
+
+	return 0;
+}
+
+static int scpi_cpufreq_init(struct cpufreq_policy *policy)
+{
+	int ret;
+	unsigned int latency;
+	struct device *cpu_dev;
+	struct scpi_data *priv;
+	struct cpufreq_frequency_table *freq_table;
+
+	cpu_dev = get_cpu_device(policy->cpu);
+	if (!cpu_dev) {
+		pr_err("failed to get cpu%d device\n", policy->cpu);
+		return -ENODEV;
+	}
+
+	ret = scpi_ops->add_opps_to_device(cpu_dev);
+	if (ret) {
+		dev_warn(cpu_dev, "failed to add opps to the device\n");
+		return ret;
+	}
+
+	ret = scpi_get_sharing_cpus(cpu_dev, policy->cpus);
+	if (ret) {
+		dev_warn(cpu_dev, "failed to get sharing cpumask\n");
+		return ret;
+	}
+
+	ret = dev_pm_opp_set_sharing_cpus(cpu_dev, policy->cpus);
+	if (ret) {
+		dev_err(cpu_dev, "%s: failed to mark OPPs as shared: %d\n",
+			__func__, ret);
+		return ret;
+	}
+
+	ret = dev_pm_opp_get_opp_count(cpu_dev);
+	if (ret <= 0) {
+		dev_dbg(cpu_dev, "OPP table is not ready, deferring probe\n");
+		ret = -EPROBE_DEFER;
+		goto out_free_opp;
+	}
+
+	priv = kzalloc_obj(*priv);
+	if (!priv) {
+		ret = -ENOMEM;
+		goto out_free_opp;
+	}
+
+	ret = dev_pm_opp_init_cpufreq_table(cpu_dev, &freq_table);
+	if (ret) {
+		dev_err(cpu_dev, "failed to init cpufreq table: %d\n", ret);
+		goto out_free_priv;
+	}
+
+	priv->cpu_dev = cpu_dev;
+	priv->clk = clk_get(cpu_dev, NULL);
+	if (IS_ERR(priv->clk)) {
+		dev_err(cpu_dev, "%s: Failed to get clk for cpu: %d\n",
+			__func__, cpu_dev->id);
+		ret = PTR_ERR(priv->clk);
+		goto out_free_cpufreq_table;
+	}
+
+	policy->driver_data = priv;
+	policy->freq_table = freq_table;
+
+	/* scpi allows DVFS request for any domain from any CPU */
+	policy->dvfs_possible_from_any_cpu = true;
+
+	latency = scpi_ops->get_transition_latency(cpu_dev);
+	if (!latency)
+		latency = CPUFREQ_DEFAULT_TRANSITION_LATENCY_NS;
+
+	policy->cpuinfo.transition_latency = latency;
+
+	policy->fast_switch_possible = false;
+
+	return 0;
+
+out_free_cpufreq_table:
+	dev_pm_opp_free_cpufreq_table(cpu_dev, &freq_table);
+out_free_priv:
+	kfree(priv);
+out_free_opp:
+	dev_pm_opp_remove_all_dynamic(cpu_dev);
+
+	return ret;
+}
+
+static void scpi_cpufreq_exit(struct cpufreq_policy *policy)
+{
+	struct scpi_data *priv = policy->driver_data;
+
+	clk_put(priv->clk);
+	dev_pm_opp_free_cpufreq_table(priv->cpu_dev, &policy->freq_table);
+	dev_pm_opp_remove_all_dynamic(priv->cpu_dev);
+	kfree(priv);
+}
+
+static struct cpufreq_driver scpi_cpufreq_driver = {
+	.name	= "scpi-cpufreq",
+	.flags	= CPUFREQ_HAVE_GOVERNOR_PER_POLICY |
+		  CPUFREQ_NEED_INITIAL_FREQ_CHECK |
+		  CPUFREQ_IS_COOLING_DEV,
+	.verify	= cpufreq_generic_frequency_table_verify,
+	.get	= scpi_cpufreq_get_rate,
+	.init	= scpi_cpufreq_init,
+	.exit	= scpi_cpufreq_exit,
+	.target_index	= scpi_cpufreq_set_target,
+	.register_em	= cpufreq_register_em_with_opp,
+};
+
+static int scpi_cpufreq_probe(struct platform_device *pdev)
+{
+	int ret;
+
+	scpi_ops = get_scpi_ops();
+	if (!scpi_ops)
+		return -EIO;
+
+	ret = cpufreq_register_driver(&scpi_cpufreq_driver);
+	if (ret)
+		dev_err(&pdev->dev, "%s: registering cpufreq failed, err: %d\n",
+			__func__, ret);
+	return ret;
+}
+
+static void scpi_cpufreq_remove(struct platform_device *pdev)
+{
+	cpufreq_unregister_driver(&scpi_cpufreq_driver);
+	scpi_ops = NULL;
+}
+
+static struct platform_driver scpi_cpufreq_platdrv = {
+	.driver = {
+		.name	= "scpi-cpufreq",
+	},
+	.probe		= scpi_cpufreq_probe,
+	.remove		= scpi_cpufreq_remove,
+};
+module_platform_driver(scpi_cpufreq_platdrv);
+
+MODULE_ALIAS("platform:scpi-cpufreq");
+MODULE_AUTHOR("Sudeep Holla <sudeep.holla@arm.com>");
+MODULE_DESCRIPTION("ARM SCPI CPUFreq interface driver");
+MODULE_LICENSE("GPL v2");