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, 152 insertions, 0 deletions

diff --git a/drivers/mfd/max77714.c b/drivers/mfd/max77714.c
new file mode 100644
index 000000000..c618605a3
--- /dev/null
+++ b/drivers/mfd/max77714.c
@@ -0,0 +1,152 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Maxim MAX77714 Core Driver
+ *
+ * Copyright (C) 2022 Luca Ceresoli
+ * Author: Luca Ceresoli <luca.ceresoli@bootlin.com>
+ */
+
+#include <linux/i2c.h>
+#include <linux/interrupt.h>
+#include <linux/mfd/core.h>
+#include <linux/mfd/max77714.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/regmap.h>
+
+static const struct mfd_cell max77714_cells[] = {
+	{ .name = "max77714-watchdog" },
+	{ .name = "max77714-rtc" },
+};
+
+static const struct regmap_range max77714_readable_ranges[] = {
+	regmap_reg_range(MAX77714_INT_TOP,     MAX77714_INT_TOP),
+	regmap_reg_range(MAX77714_INT_TOPM,    MAX77714_INT_TOPM),
+	regmap_reg_range(MAX77714_32K_STATUS,  MAX77714_32K_CONFIG),
+	regmap_reg_range(MAX77714_CNFG_GLBL2,  MAX77714_CNFG2_ONOFF),
+};
+
+static const struct regmap_range max77714_writable_ranges[] = {
+	regmap_reg_range(MAX77714_INT_TOPM,    MAX77714_INT_TOPM),
+	regmap_reg_range(MAX77714_32K_CONFIG,  MAX77714_32K_CONFIG),
+	regmap_reg_range(MAX77714_CNFG_GLBL2,  MAX77714_CNFG2_ONOFF),
+};
+
+static const struct regmap_access_table max77714_readable_table = {
+	.yes_ranges = max77714_readable_ranges,
+	.n_yes_ranges = ARRAY_SIZE(max77714_readable_ranges),
+};
+
+static const struct regmap_access_table max77714_writable_table = {
+	.yes_ranges = max77714_writable_ranges,
+	.n_yes_ranges = ARRAY_SIZE(max77714_writable_ranges),
+};
+
+static const struct regmap_config max77714_regmap_config = {
+	.reg_bits = 8,
+	.val_bits = 8,
+	.max_register = MAX77714_CNFG2_ONOFF,
+	.rd_table = &max77714_readable_table,
+	.wr_table = &max77714_writable_table,
+};
+
+static const struct regmap_irq max77714_top_irqs[] = {
+	REGMAP_IRQ_REG(MAX77714_IRQ_TOP_ONOFF,   0, MAX77714_INT_TOP_ONOFF),
+	REGMAP_IRQ_REG(MAX77714_IRQ_TOP_RTC,     0, MAX77714_INT_TOP_RTC),
+	REGMAP_IRQ_REG(MAX77714_IRQ_TOP_GPIO,    0, MAX77714_INT_TOP_GPIO),
+	REGMAP_IRQ_REG(MAX77714_IRQ_TOP_LDO,     0, MAX77714_INT_TOP_LDO),
+	REGMAP_IRQ_REG(MAX77714_IRQ_TOP_SD,      0, MAX77714_INT_TOP_SD),
+	REGMAP_IRQ_REG(MAX77714_IRQ_TOP_GLBL,    0, MAX77714_INT_TOP_GLBL),
+};
+
+static const struct regmap_irq_chip max77714_irq_chip = {
+	.name			= "max77714-pmic",
+	.status_base		= MAX77714_INT_TOP,
+	.mask_base		= MAX77714_INT_TOPM,
+	.num_regs		= 1,
+	.irqs			= max77714_top_irqs,
+	.num_irqs		= ARRAY_SIZE(max77714_top_irqs),
+};
+
+/*
+ * MAX77714 initially uses the internal, low precision oscillator. Enable
+ * the external oscillator by setting the XOSC_RETRY bit. If the external
+ * oscillator is not OK (probably not installed) this has no effect.
+ */
+static int max77714_setup_xosc(struct device *dev, struct regmap *regmap)
+{
+	/* Internal Crystal Load Capacitance, indexed by value of 32KLOAD bits */
+	static const unsigned int load_cap[4] = {0, 10, 12, 22}; /* pF */
+	unsigned int load_cap_idx;
+	unsigned int status;
+	int err;
+
+	err = regmap_update_bits(regmap, MAX77714_32K_CONFIG,
+				 MAX77714_32K_CONFIG_XOSC_RETRY,
+				 MAX77714_32K_CONFIG_XOSC_RETRY);
+	if (err)
+		return dev_err_probe(dev, err, "Failed to configure the external oscillator\n");
+
+	err = regmap_read(regmap, MAX77714_32K_STATUS, &status);
+	if (err)
+		return dev_err_probe(dev, err, "Failed to read external oscillator status\n");
+
+	load_cap_idx = (status >> MAX77714_32K_STATUS_32KLOAD_SHF)
+		& MAX77714_32K_STATUS_32KLOAD_MSK;
+
+	dev_info(dev, "Using %s oscillator, %d pF load cap\n",
+		 status & MAX77714_32K_STATUS_32KSOURCE ? "internal" : "external",
+		 load_cap[load_cap_idx]);
+
+	return 0;
+}
+
+static int max77714_probe(struct i2c_client *client)
+{
+	struct device *dev = &client->dev;
+	struct regmap *regmap;
+	struct regmap_irq_chip_data *irq_data;
+	int err;
+
+	regmap = devm_regmap_init_i2c(client, &max77714_regmap_config);
+	if (IS_ERR(regmap))
+		return dev_err_probe(dev, PTR_ERR(regmap),
+				     "Failed to initialise regmap\n");
+
+	err = max77714_setup_xosc(dev, regmap);
+	if (err)
+		return err;
+
+	err = devm_regmap_add_irq_chip(dev, regmap, client->irq,
+				       IRQF_ONESHOT | IRQF_SHARED, 0,
+				       &max77714_irq_chip, &irq_data);
+	if (err)
+		return dev_err_probe(dev, err, "Failed to add PMIC IRQ chip\n");
+
+	err =  devm_mfd_add_devices(dev, PLATFORM_DEVID_NONE,
+				    max77714_cells, ARRAY_SIZE(max77714_cells),
+				    NULL, 0, NULL);
+	if (err)
+		return dev_err_probe(dev, err, "Failed to register child devices\n");
+
+	return 0;
+}
+
+static const struct of_device_id max77714_dt_match[] = {
+	{ .compatible = "maxim,max77714" },
+	{},
+};
+MODULE_DEVICE_TABLE(of, max77714_dt_match);
+
+static struct i2c_driver max77714_driver = {
+	.driver = {
+		.name = "max77714",
+		.of_match_table = max77714_dt_match,
+	},
+	.probe = max77714_probe,
+};
+module_i2c_driver(max77714_driver);
+
+MODULE_DESCRIPTION("Maxim MAX77714 MFD core driver");
+MODULE_AUTHOR("Luca Ceresoli <luca.ceresoli@bootlin.com>");
+MODULE_LICENSE("GPL");