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, 338 insertions, 0 deletions

diff --git a/drivers/interconnect/imx/imx.c b/drivers/interconnect/imx/imx.c
new file mode 100644
index 000000000..9511f80cf
--- /dev/null
+++ b/drivers/interconnect/imx/imx.c
@@ -0,0 +1,338 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Interconnect framework driver for i.MX SoC
+ *
+ * Copyright (c) 2019, BayLibre
+ * Copyright (c) 2019-2020, NXP
+ * Author: Alexandre Bailon <abailon@baylibre.com>
+ * Author: Leonard Crestez <leonard.crestez@nxp.com>
+ */
+
+#include <linux/device.h>
+#include <linux/interconnect-provider.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_platform.h>
+#include <linux/platform_device.h>
+#include <linux/pm_qos.h>
+
+#include "imx.h"
+
+/* private icc_node data */
+struct imx_icc_node {
+	const struct imx_icc_node_desc *desc;
+	const struct imx_icc_noc_setting *setting;
+	struct device *qos_dev;
+	struct dev_pm_qos_request qos_req;
+	struct imx_icc_provider *imx_provider;
+};
+
+static int imx_icc_get_bw(struct icc_node *node, u32 *avg, u32 *peak)
+{
+	*avg = 0;
+	*peak = 0;
+
+	return 0;
+}
+
+static int imx_icc_node_set(struct icc_node *node)
+{
+	struct device *dev = node->provider->dev;
+	struct imx_icc_node *node_data = node->data;
+	void __iomem *base;
+	u32 prio;
+	u64 freq;
+
+	if (node_data->setting && node->peak_bw) {
+		base = node_data->setting->reg + node_data->imx_provider->noc_base;
+		if (node_data->setting->mode == IMX_NOC_MODE_FIXED) {
+			prio = node_data->setting->prio_level;
+			prio = PRIORITY_COMP_MARK | (prio << 8) | prio;
+			writel(prio, base + IMX_NOC_PRIO_REG);
+			writel(node_data->setting->mode, base + IMX_NOC_MODE_REG);
+			writel(node_data->setting->ext_control, base + IMX_NOC_EXT_CTL_REG);
+			dev_dbg(dev, "%s: mode: 0x%x, prio: 0x%x, ext_control: 0x%x\n",
+				node_data->desc->name, node_data->setting->mode, prio,
+				node_data->setting->ext_control);
+		} else if (node_data->setting->mode == IMX_NOC_MODE_UNCONFIGURED) {
+			dev_dbg(dev, "%s: mode not unconfigured\n", node_data->desc->name);
+		} else {
+			dev_info(dev, "%s: mode: %d not supported\n",
+				 node_data->desc->name, node_data->setting->mode);
+			return -EOPNOTSUPP;
+		}
+	}
+
+	if (!node_data->qos_dev)
+		return 0;
+
+	freq = (node->avg_bw + node->peak_bw) * node_data->desc->adj->bw_mul;
+	do_div(freq, node_data->desc->adj->bw_div);
+	dev_dbg(dev, "node %s device %s avg_bw %ukBps peak_bw %ukBps min_freq %llukHz\n",
+		node->name, dev_name(node_data->qos_dev),
+		node->avg_bw, node->peak_bw, freq);
+
+	if (freq > S32_MAX) {
+		dev_err(dev, "%s can't request more than S32_MAX freq\n",
+				node->name);
+		return -ERANGE;
+	}
+
+	dev_pm_qos_update_request(&node_data->qos_req, freq);
+
+	return 0;
+}
+
+static int imx_icc_set(struct icc_node *src, struct icc_node *dst)
+{
+	int ret;
+
+	ret = imx_icc_node_set(src);
+	if (ret)
+		return ret;
+
+	return imx_icc_node_set(dst);
+}
+
+/* imx_icc_node_destroy() - Destroy an imx icc_node, including private data */
+static void imx_icc_node_destroy(struct icc_node *node)
+{
+	struct imx_icc_node *node_data = node->data;
+	int ret;
+
+	if (dev_pm_qos_request_active(&node_data->qos_req)) {
+		ret = dev_pm_qos_remove_request(&node_data->qos_req);
+		if (ret)
+			dev_warn(node->provider->dev,
+				 "failed to remove qos request for %s\n",
+				 dev_name(node_data->qos_dev));
+	}
+
+	put_device(node_data->qos_dev);
+	icc_node_del(node);
+	icc_node_destroy(node->id);
+}
+
+static int imx_icc_node_init_qos(struct icc_provider *provider,
+				 struct icc_node *node)
+{
+	struct imx_icc_node *node_data = node->data;
+	const struct imx_icc_node_adj_desc *adj = node_data->desc->adj;
+	struct device *dev = provider->dev;
+	struct device_node *dn = NULL;
+	struct platform_device *pdev;
+
+	if (adj->main_noc) {
+		node_data->qos_dev = dev;
+		dev_dbg(dev, "icc node %s[%d] is main noc itself\n",
+			node->name, node->id);
+	} else {
+		dn = of_parse_phandle(dev->of_node, adj->phandle_name, 0);
+		if (!dn) {
+			dev_warn(dev, "Failed to parse %s\n",
+				 adj->phandle_name);
+			return -ENODEV;
+		}
+		/* Allow scaling to be disabled on a per-node basis */
+		if (!of_device_is_available(dn)) {
+			dev_warn(dev, "Missing property %s, skip scaling %s\n",
+				 adj->phandle_name, node->name);
+			of_node_put(dn);
+			return 0;
+		}
+
+		pdev = of_find_device_by_node(dn);
+		of_node_put(dn);
+		if (!pdev) {
+			dev_warn(dev, "node %s[%d] missing device for %pOF\n",
+				 node->name, node->id, dn);
+			return -EPROBE_DEFER;
+		}
+		node_data->qos_dev = &pdev->dev;
+		dev_dbg(dev, "node %s[%d] has device node %pOF\n",
+			node->name, node->id, dn);
+	}
+
+	return dev_pm_qos_add_request(node_data->qos_dev,
+				      &node_data->qos_req,
+				      DEV_PM_QOS_MIN_FREQUENCY, 0);
+}
+
+static struct icc_node *imx_icc_node_add(struct imx_icc_provider *imx_provider,
+					 const struct imx_icc_node_desc *node_desc,
+					 const struct imx_icc_noc_setting *setting)
+{
+	struct icc_provider *provider = &imx_provider->provider;
+	struct device *dev = provider->dev;
+	struct imx_icc_node *node_data;
+	struct icc_node *node;
+	int ret;
+
+	node = icc_node_create(node_desc->id);
+	if (IS_ERR(node)) {
+		dev_err(dev, "failed to create node %d\n", node_desc->id);
+		return node;
+	}
+
+	if (node->data) {
+		dev_err(dev, "already created node %s id=%d\n",
+			node_desc->name, node_desc->id);
+		return ERR_PTR(-EEXIST);
+	}
+
+	node_data = devm_kzalloc(dev, sizeof(*node_data), GFP_KERNEL);
+	if (!node_data) {
+		icc_node_destroy(node->id);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	node->name = node_desc->name;
+	node->data = node_data;
+	node_data->desc = node_desc;
+	node_data->setting = setting;
+	node_data->imx_provider = imx_provider;
+	icc_node_add(node, provider);
+
+	if (node_desc->adj) {
+		ret = imx_icc_node_init_qos(provider, node);
+		if (ret < 0) {
+			imx_icc_node_destroy(node);
+			return ERR_PTR(ret);
+		}
+	}
+
+	return node;
+}
+
+static void imx_icc_unregister_nodes(struct icc_provider *provider)
+{
+	struct icc_node *node, *tmp;
+
+	list_for_each_entry_safe(node, tmp, &provider->nodes, node_list)
+		imx_icc_node_destroy(node);
+}
+
+static int imx_icc_register_nodes(struct imx_icc_provider *imx_provider,
+				  const struct imx_icc_node_desc *descs,
+				  int count,
+				  const struct imx_icc_noc_setting *settings)
+{
+	struct icc_provider *provider = &imx_provider->provider;
+	struct icc_onecell_data *provider_data = provider->data;
+	int ret;
+	int i;
+
+	for (i = 0; i < count; i++) {
+		struct icc_node *node;
+		const struct imx_icc_node_desc *node_desc = &descs[i];
+		size_t j;
+
+		node = imx_icc_node_add(imx_provider, node_desc,
+					settings ? &settings[node_desc->id] : NULL);
+		if (IS_ERR(node)) {
+			ret = dev_err_probe(provider->dev, PTR_ERR(node),
+					    "failed to add %s\n", node_desc->name);
+			goto err;
+		}
+		provider_data->nodes[node->id] = node;
+
+		for (j = 0; j < node_desc->num_links; j++) {
+			ret = icc_link_create(node, node_desc->links[j]);
+			if (ret) {
+				dev_err(provider->dev, "failed to link node %d to %d: %d\n",
+					node->id, node_desc->links[j], ret);
+				goto err;
+			}
+		}
+	}
+
+	return 0;
+
+err:
+	imx_icc_unregister_nodes(provider);
+
+	return ret;
+}
+
+static int get_max_node_id(struct imx_icc_node_desc *nodes, int nodes_count)
+{
+	int i, ret = 0;
+
+	for (i = 0; i < nodes_count; ++i)
+		if (nodes[i].id > ret)
+			ret = nodes[i].id;
+
+	return ret;
+}
+
+int imx_icc_register(struct platform_device *pdev,
+		     struct imx_icc_node_desc *nodes, int nodes_count,
+		     struct imx_icc_noc_setting *settings)
+{
+	struct device *dev = &pdev->dev;
+	struct icc_onecell_data *data;
+	struct imx_icc_provider *imx_provider;
+	struct icc_provider *provider;
+	int num_nodes;
+	int ret;
+
+	/* icc_onecell_data is indexed by node_id, unlike nodes param */
+	num_nodes = get_max_node_id(nodes, nodes_count) + 1;
+	data = devm_kzalloc(dev, struct_size(data, nodes, num_nodes),
+			    GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+	data->num_nodes = num_nodes;
+
+	imx_provider = devm_kzalloc(dev, sizeof(*imx_provider), GFP_KERNEL);
+	if (!imx_provider)
+		return -ENOMEM;
+	provider = &imx_provider->provider;
+	provider->set = imx_icc_set;
+	provider->get_bw = imx_icc_get_bw;
+	provider->aggregate = icc_std_aggregate;
+	provider->xlate = of_icc_xlate_onecell;
+	provider->data = data;
+	provider->dev = dev->parent;
+
+	icc_provider_init(provider);
+
+	platform_set_drvdata(pdev, imx_provider);
+
+	if (settings) {
+		imx_provider->noc_base = devm_of_iomap(dev, provider->dev->of_node, 0, NULL);
+		if (IS_ERR(imx_provider->noc_base)) {
+			ret = PTR_ERR(imx_provider->noc_base);
+			dev_err(dev, "Error mapping NoC: %d\n", ret);
+			return ret;
+		}
+	}
+
+	ret = imx_icc_register_nodes(imx_provider, nodes, nodes_count, settings);
+	if (ret)
+		return ret;
+
+	ret = icc_provider_register(provider);
+	if (ret)
+		goto err_unregister_nodes;
+
+	return 0;
+
+err_unregister_nodes:
+	imx_icc_unregister_nodes(&imx_provider->provider);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(imx_icc_register);
+
+void imx_icc_unregister(struct platform_device *pdev)
+{
+	struct imx_icc_provider *imx_provider = platform_get_drvdata(pdev);
+
+	icc_provider_deregister(&imx_provider->provider);
+	imx_icc_unregister_nodes(&imx_provider->provider);
+}
+EXPORT_SYMBOL_GPL(imx_icc_unregister);
+
+MODULE_DESCRIPTION("Interconnect framework driver for i.MX SoC");
+MODULE_LICENSE("GPL v2");