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

3 files changed, 170 insertions, 0 deletions

diff --git a/drivers/soc/fujitsu/Kconfig b/drivers/soc/fujitsu/Kconfig
new file mode 100644
index 000000000..987731e80
--- /dev/null
+++ b/drivers/soc/fujitsu/Kconfig
@@ -0,0 +1,16 @@
+# SPDX-License-Identifier: GPL-2.0-only
+menu "fujitsu SoC drivers"
+
+config A64FX_DIAG
+	bool "A64FX diag driver"
+	depends on ARM64
+	depends on ACPI
+	help
+	  Say Y here if you want to enable diag interrupt on Fujitsu A64FX.
+	  This driver enables BMC's diagnostic requests and enables
+	  A64FX-specific interrupts. This allows administrators to obtain
+	  kernel dumps via diagnostic requests using ipmitool, etc.
+
+	  If unsure, say N.
+
+endmenu
diff --git a/drivers/soc/fujitsu/Makefile b/drivers/soc/fujitsu/Makefile
new file mode 100644
index 000000000..945bc1c14
--- /dev/null
+++ b/drivers/soc/fujitsu/Makefile
@@ -0,0 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0
+
+obj-$(CONFIG_A64FX_DIAG)	+= a64fx-diag.o
diff --git a/drivers/soc/fujitsu/a64fx-diag.c b/drivers/soc/fujitsu/a64fx-diag.c
new file mode 100644
index 000000000..76cb0b6a2
--- /dev/null
+++ b/drivers/soc/fujitsu/a64fx-diag.c
@@ -0,0 +1,151 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * A64FX diag driver.
+ * Copyright (c) 2022 Fujitsu Ltd.
+ */
+
+#include <linux/acpi.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+
+#define A64FX_DIAG_IRQ 1
+#define BMC_DIAG_INTERRUPT_ENABLE 0x40
+#define BMC_DIAG_INTERRUPT_STATUS 0x44
+#define BMC_DIAG_INTERRUPT_MASK BIT(31)
+
+struct a64fx_diag_priv {
+	void __iomem *mmsc_reg_base;
+	int irq;
+	bool has_nmi;
+};
+
+static irqreturn_t a64fx_diag_handler_nmi(int irq, void *dev_id)
+{
+	nmi_panic(NULL, "a64fx_diag: interrupt received\n");
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t a64fx_diag_handler_irq(int irq, void *dev_id)
+{
+	panic("a64fx_diag: interrupt received\n");
+
+	return IRQ_HANDLED;
+}
+
+static void a64fx_diag_interrupt_clear(struct a64fx_diag_priv *priv)
+{
+	void __iomem *diag_status_reg_addr;
+	u32 mmsc;
+
+	diag_status_reg_addr = priv->mmsc_reg_base + BMC_DIAG_INTERRUPT_STATUS;
+	mmsc = readl(diag_status_reg_addr);
+	if (mmsc & BMC_DIAG_INTERRUPT_MASK)
+		writel(BMC_DIAG_INTERRUPT_MASK, diag_status_reg_addr);
+}
+
+static void a64fx_diag_interrupt_enable(struct a64fx_diag_priv *priv)
+{
+	void __iomem *diag_enable_reg_addr;
+	u32 mmsc;
+
+	diag_enable_reg_addr = priv->mmsc_reg_base + BMC_DIAG_INTERRUPT_ENABLE;
+	mmsc = readl(diag_enable_reg_addr);
+	if (!(mmsc & BMC_DIAG_INTERRUPT_MASK)) {
+		mmsc |= BMC_DIAG_INTERRUPT_MASK;
+		writel(mmsc, diag_enable_reg_addr);
+	}
+}
+
+static void a64fx_diag_interrupt_disable(struct a64fx_diag_priv *priv)
+{
+	void __iomem *diag_enable_reg_addr;
+	u32 mmsc;
+
+	diag_enable_reg_addr = priv->mmsc_reg_base + BMC_DIAG_INTERRUPT_ENABLE;
+	mmsc = readl(diag_enable_reg_addr);
+	if (mmsc & BMC_DIAG_INTERRUPT_MASK) {
+		mmsc &= ~BMC_DIAG_INTERRUPT_MASK;
+		writel(mmsc, diag_enable_reg_addr);
+	}
+}
+
+static int a64fx_diag_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct a64fx_diag_priv *priv;
+	unsigned long irq_flags;
+	int ret;
+
+	priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
+	if (priv == NULL)
+		return -ENOMEM;
+
+	priv->mmsc_reg_base = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(priv->mmsc_reg_base))
+		return PTR_ERR(priv->mmsc_reg_base);
+
+	priv->irq = platform_get_irq(pdev, A64FX_DIAG_IRQ);
+	if (priv->irq < 0)
+		return priv->irq;
+
+	platform_set_drvdata(pdev, priv);
+
+	irq_flags = IRQF_PERCPU | IRQF_NOBALANCING | IRQF_NO_AUTOEN |
+		   IRQF_NO_THREAD;
+	ret = request_nmi(priv->irq, &a64fx_diag_handler_nmi, irq_flags,
+			"a64fx_diag_nmi", NULL);
+	if (ret) {
+		ret = request_irq(priv->irq, &a64fx_diag_handler_irq,
+				irq_flags, "a64fx_diag_irq", NULL);
+		if (ret) {
+			dev_err(dev, "cannot register IRQ %d\n", ret);
+			return ret;
+		}
+		enable_irq(priv->irq);
+	} else {
+		enable_nmi(priv->irq);
+		priv->has_nmi = true;
+	}
+
+	a64fx_diag_interrupt_clear(priv);
+	a64fx_diag_interrupt_enable(priv);
+
+	return 0;
+}
+
+static void a64fx_diag_remove(struct platform_device *pdev)
+{
+	struct a64fx_diag_priv *priv = platform_get_drvdata(pdev);
+
+	a64fx_diag_interrupt_disable(priv);
+	a64fx_diag_interrupt_clear(priv);
+
+	if (priv->has_nmi)
+		free_nmi(priv->irq, NULL);
+	else
+		free_irq(priv->irq, NULL);
+}
+
+static const struct acpi_device_id a64fx_diag_acpi_match[] = {
+	{ "FUJI2007", 0 },
+	{ },
+};
+MODULE_DEVICE_TABLE(acpi, a64fx_diag_acpi_match);
+
+
+static struct platform_driver a64fx_diag_driver = {
+	.driver = {
+		.name = "a64fx_diag_driver",
+		.acpi_match_table = ACPI_PTR(a64fx_diag_acpi_match),
+	},
+	.probe = a64fx_diag_probe,
+	.remove = a64fx_diag_remove,
+};
+
+module_platform_driver(a64fx_diag_driver);
+
+MODULE_AUTHOR("Hitomi Hasegawa <hasegawa-hitomi@fujitsu.com>");
+MODULE_DESCRIPTION("A64FX diag driver");