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, 343 insertions, 0 deletions

diff --git a/drivers/pinctrl/berlin/berlin.c b/drivers/pinctrl/berlin/berlin.c
new file mode 100644
index 000000000..7971339e2
--- /dev/null
+++ b/drivers/pinctrl/berlin/berlin.c
@@ -0,0 +1,343 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Marvell Berlin SoC pinctrl core driver
+ *
+ * Copyright (C) 2014 Marvell Technology Group Ltd.
+ *
+ * Antoine Ténart <antoine.tenart@free-electrons.com>
+ */
+
+#include <linux/io.h>
+#include <linux/mfd/syscon.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/of_device.h>
+#include <linux/pinctrl/pinctrl.h>
+#include <linux/pinctrl/pinmux.h>
+#include <linux/platform_device.h>
+#include <linux/regmap.h>
+#include <linux/slab.h>
+
+#include "../core.h"
+#include "../pinctrl-utils.h"
+#include "berlin.h"
+
+struct berlin_pinctrl {
+	struct regmap *regmap;
+	struct device *dev;
+	const struct berlin_pinctrl_desc *desc;
+	struct pinfunction *functions;
+	unsigned nfunctions;
+	struct pinctrl_dev *pctrl_dev;
+};
+
+static int berlin_pinctrl_get_group_count(struct pinctrl_dev *pctrl_dev)
+{
+	struct berlin_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctrl_dev);
+
+	return pctrl->desc->ngroups;
+}
+
+static const char *berlin_pinctrl_get_group_name(struct pinctrl_dev *pctrl_dev,
+						 unsigned group)
+{
+	struct berlin_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctrl_dev);
+
+	return pctrl->desc->groups[group].name;
+}
+
+static int berlin_pinctrl_dt_node_to_map(struct pinctrl_dev *pctrl_dev,
+					 struct device_node *node,
+					 struct pinctrl_map **map,
+					 unsigned *num_maps)
+{
+	struct berlin_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctrl_dev);
+	struct property *prop;
+	const char *function_name, *group_name;
+	unsigned reserved_maps = 0;
+	int ret, ngroups;
+
+	*map = NULL;
+	*num_maps = 0;
+
+	ret = of_property_read_string(node, "function", &function_name);
+	if (ret) {
+		dev_err(pctrl->dev,
+			"missing function property in node %pOFn\n", node);
+		return -EINVAL;
+	}
+
+	ngroups = of_property_count_strings(node, "groups");
+	if (ngroups < 0) {
+		dev_err(pctrl->dev,
+			"missing groups property in node %pOFn\n", node);
+		return -EINVAL;
+	}
+
+	ret = pinctrl_utils_reserve_map(pctrl_dev, map, &reserved_maps,
+					num_maps, ngroups);
+	if (ret) {
+		dev_err(pctrl->dev, "can't reserve map: %d\n", ret);
+		return ret;
+	}
+
+	of_property_for_each_string(node, "groups", prop, group_name) {
+		ret = pinctrl_utils_add_map_mux(pctrl_dev, map, &reserved_maps,
+						num_maps, group_name,
+						function_name);
+		if (ret) {
+			dev_err(pctrl->dev, "can't add map: %d\n", ret);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+static const struct pinctrl_ops berlin_pinctrl_ops = {
+	.get_groups_count	= berlin_pinctrl_get_group_count,
+	.get_group_name		= berlin_pinctrl_get_group_name,
+	.dt_node_to_map		= berlin_pinctrl_dt_node_to_map,
+	.dt_free_map		= pinctrl_utils_free_map,
+};
+
+static int berlin_pinmux_get_functions_count(struct pinctrl_dev *pctrl_dev)
+{
+	struct berlin_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctrl_dev);
+
+	return pctrl->nfunctions;
+}
+
+static const char *berlin_pinmux_get_function_name(struct pinctrl_dev *pctrl_dev,
+						   unsigned function)
+{
+	struct berlin_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctrl_dev);
+
+	return pctrl->functions[function].name;
+}
+
+static int berlin_pinmux_get_function_groups(struct pinctrl_dev *pctrl_dev,
+					     unsigned function,
+					     const char * const **groups,
+					     unsigned * const ngroups)
+{
+	struct berlin_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctrl_dev);
+
+	*groups = pctrl->functions[function].groups;
+	*ngroups = pctrl->functions[function].ngroups;
+
+	return 0;
+}
+
+static struct berlin_desc_function *
+berlin_pinctrl_find_function_by_name(struct berlin_pinctrl *pctrl,
+				     const struct berlin_desc_group *group,
+				     const char *fname)
+{
+	struct berlin_desc_function *function = group->functions;
+
+	while (function->name) {
+		if (!strcmp(function->name, fname))
+			return function;
+
+		function++;
+	}
+
+	return NULL;
+}
+
+static int berlin_pinmux_set(struct pinctrl_dev *pctrl_dev,
+			     unsigned function,
+			     unsigned group)
+{
+	struct berlin_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctrl_dev);
+	const struct berlin_desc_group *group_desc = pctrl->desc->groups + group;
+	struct pinfunction *func = pctrl->functions + function;
+	struct berlin_desc_function *function_desc =
+		berlin_pinctrl_find_function_by_name(pctrl, group_desc,
+						     func->name);
+	u32 mask, val;
+
+	if (!function_desc)
+		return -EINVAL;
+
+	mask = GENMASK(group_desc->lsb + group_desc->bit_width - 1,
+		       group_desc->lsb);
+	val = function_desc->muxval << group_desc->lsb;
+	regmap_update_bits(pctrl->regmap, group_desc->offset, mask, val);
+
+	return 0;
+}
+
+static const struct pinmux_ops berlin_pinmux_ops = {
+	.get_functions_count	= &berlin_pinmux_get_functions_count,
+	.get_function_name	= &berlin_pinmux_get_function_name,
+	.get_function_groups	= &berlin_pinmux_get_function_groups,
+	.set_mux		= &berlin_pinmux_set,
+};
+
+static int berlin_pinctrl_add_function(struct berlin_pinctrl *pctrl,
+				       const char *name)
+{
+	struct pinfunction *function = pctrl->functions;
+
+	while (function->name) {
+		if (!strcmp(function->name, name)) {
+			function->ngroups++;
+			return -EEXIST;
+		}
+		function++;
+	}
+
+	function->name = name;
+	function->ngroups = 1;
+
+	pctrl->nfunctions++;
+
+	return 0;
+}
+
+static int berlin_pinctrl_build_state(struct platform_device *pdev)
+{
+	struct berlin_pinctrl *pctrl = platform_get_drvdata(pdev);
+	const struct berlin_desc_group *desc_group;
+	const struct berlin_desc_function *desc_function;
+	int i, max_functions = 0;
+	struct pinfunction *new_functions;
+
+	pctrl->nfunctions = 0;
+
+	for (i = 0; i < pctrl->desc->ngroups; i++) {
+		desc_group = pctrl->desc->groups + i;
+		/* compute the maximum number of functions a group can have */
+		max_functions += 1 << (desc_group->bit_width + 1);
+	}
+
+	/* we will reallocate later */
+	pctrl->functions = kzalloc_objs(*pctrl->functions, max_functions);
+	if (!pctrl->functions)
+		return -ENOMEM;
+
+	/* register all functions */
+	for (i = 0; i < pctrl->desc->ngroups; i++) {
+		desc_group = pctrl->desc->groups + i;
+		desc_function = desc_group->functions;
+
+		while (desc_function->name) {
+			berlin_pinctrl_add_function(pctrl, desc_function->name);
+			desc_function++;
+		}
+	}
+
+	new_functions = krealloc(pctrl->functions,
+				    pctrl->nfunctions * sizeof(*pctrl->functions),
+				    GFP_KERNEL);
+	if (!new_functions) {
+		kfree(pctrl->functions);
+		return -ENOMEM;
+	}
+
+	pctrl->functions = new_functions;
+	/* map functions to theirs groups */
+	for (i = 0; i < pctrl->desc->ngroups; i++) {
+		desc_group = pctrl->desc->groups + i;
+		desc_function = desc_group->functions;
+
+		while (desc_function->name) {
+			struct pinfunction *function = pctrl->functions;
+			const char **groups;
+			bool found = false;
+
+			while (function->name) {
+				if (!strcmp(desc_function->name, function->name)) {
+					found = true;
+					break;
+				}
+				function++;
+			}
+
+			if (!found) {
+				kfree(pctrl->functions);
+				return -EINVAL;
+			}
+
+			if (!function->groups) {
+				function->groups =
+					devm_kcalloc(&pdev->dev,
+						     function->ngroups,
+						     sizeof(*function->groups),
+						     GFP_KERNEL);
+				if (!function->groups) {
+					kfree(pctrl->functions);
+					return -ENOMEM;
+				}
+			}
+
+			groups = (const char **)function->groups;
+			while (*groups)
+				groups++;
+
+			*groups = desc_group->name;
+
+			desc_function++;
+		}
+	}
+
+	return 0;
+}
+
+static const struct pinctrl_desc berlin_pctrl_desc = {
+	.name		= "berlin-pinctrl",
+	.pctlops	= &berlin_pinctrl_ops,
+	.pmxops		= &berlin_pinmux_ops,
+	.owner		= THIS_MODULE,
+};
+
+int berlin_pinctrl_probe_regmap(struct platform_device *pdev,
+				const struct berlin_pinctrl_desc *desc,
+				struct regmap *regmap)
+{
+	struct device *dev = &pdev->dev;
+	struct berlin_pinctrl *pctrl;
+	int ret;
+
+	pctrl = devm_kzalloc(dev, sizeof(*pctrl), GFP_KERNEL);
+	if (!pctrl)
+		return -ENOMEM;
+
+	platform_set_drvdata(pdev, pctrl);
+
+	pctrl->regmap = regmap;
+	pctrl->dev = &pdev->dev;
+	pctrl->desc = desc;
+
+	ret = berlin_pinctrl_build_state(pdev);
+	if (ret) {
+		dev_err(dev, "cannot build driver state: %d\n", ret);
+		return ret;
+	}
+
+	pctrl->pctrl_dev = devm_pinctrl_register(dev, &berlin_pctrl_desc,
+						 pctrl);
+	if (IS_ERR(pctrl->pctrl_dev)) {
+		dev_err(dev, "failed to register pinctrl driver\n");
+		return PTR_ERR(pctrl->pctrl_dev);
+	}
+
+	return 0;
+}
+
+int berlin_pinctrl_probe(struct platform_device *pdev,
+			 const struct berlin_pinctrl_desc *desc)
+{
+	struct device *dev = &pdev->dev;
+	struct device_node *parent_np = of_get_parent(dev->of_node);
+	struct regmap *regmap = syscon_node_to_regmap(parent_np);
+
+	of_node_put(parent_np);
+	if (IS_ERR(regmap))
+		return PTR_ERR(regmap);
+
+	return berlin_pinctrl_probe_regmap(pdev, desc, regmap);
+}