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

10 files changed, 351 insertions, 0 deletions

diff --git a/arch/openrisc/boot/.gitignore b/arch/openrisc/boot/.gitignore
new file mode 100644
index 000000000..007d6fea3
--- /dev/null
+++ b/arch/openrisc/boot/.gitignore
@@ -0,0 +1,2 @@
+# SPDX-License-Identifier: GPL-2.0
+vmlinux.bin
diff --git a/arch/openrisc/boot/Makefile b/arch/openrisc/boot/Makefile
new file mode 100644
index 000000000..5b28538f4
--- /dev/null
+++ b/arch/openrisc/boot/Makefile
@@ -0,0 +1,10 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# Makefile for bootable kernel images
+#
+
+targets += vmlinux.bin
+
+OBJCOPYFLAGS_vmlinux.bin := -O binary
+$(obj)/vmlinux.bin: vmlinux FORCE
+	$(call if_changed,objcopy)
diff --git a/arch/openrisc/boot/dts/Makefile b/arch/openrisc/boot/dts/Makefile
new file mode 100644
index 000000000..3a66e0ef3
--- /dev/null
+++ b/arch/openrisc/boot/dts/Makefile
@@ -0,0 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
+dtb-y += $(addsuffix .dtb, $(CONFIG_BUILTIN_DTB_NAME))
+
+#DTC_FLAGS ?= -p 1024
diff --git a/arch/openrisc/boot/dts/de0-nano-common.dtsi b/arch/openrisc/boot/dts/de0-nano-common.dtsi
new file mode 100644
index 000000000..02e329e28
--- /dev/null
+++ b/arch/openrisc/boot/dts/de0-nano-common.dtsi
@@ -0,0 +1,42 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <dt-bindings/gpio/gpio.h>
+#include <dt-bindings/leds/common.h>
+
+/ {
+	leds0: leds {
+		compatible = "gpio-leds";
+
+		led-heartbeat {
+			gpios = <&gpio0 0 GPIO_ACTIVE_HIGH>;
+			color = <LED_COLOR_ID_GREEN>;
+			function = LED_FUNCTION_HEARTBEAT;
+			linux,default-trigger = "heartbeat";
+			label = "heartbeat";
+		};
+	};
+
+	memory@0 {
+		device_type = "memory";
+		reg = <0x00000000 0x02000000>;
+	};
+
+	/* 8 Green LEDs */
+	gpio0: gpio@91000000 {
+		compatible = "opencores,gpio";
+		reg = <0x91000000 0x1>, <0x91000001 0x1>;
+		reg-names = "dat", "dirout";
+		gpio-controller;
+		#gpio-cells = <2>;
+	};
+
+	/* 4 DIP Switches */
+	gpio1: gpio@92000000 {
+		compatible = "opencores,gpio";
+		reg = <0x92000000 0x1>, <0x92000001 0x1>;
+		reg-names = "dat", "dirout";
+		gpio-controller;
+		#gpio-cells = <2>;
+		status = "disabled";
+	};
+};
diff --git a/arch/openrisc/boot/dts/de0-nano-multicore.dts b/arch/openrisc/boot/dts/de0-nano-multicore.dts
new file mode 100644
index 000000000..b6cf286af
--- /dev/null
+++ b/arch/openrisc/boot/dts/de0-nano-multicore.dts
@@ -0,0 +1,25 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <dt-bindings/gpio/gpio.h>
+#include <dt-bindings/leds/common.h>
+
+/dts-v1/;
+
+#include "simple-smp.dtsi"
+#include "de0-nano-common.dtsi"
+
+/ {
+	model = "Terasic DE0 Nano - Multicore";
+};
+
+&cpu0 {
+	clock-frequency = <50000000>;
+};
+
+&cpu1 {
+	clock-frequency = <50000000>;
+};
+
+&serial0 {
+	clock-frequency = <50000000>;
+};
diff --git a/arch/openrisc/boot/dts/de0-nano.dts b/arch/openrisc/boot/dts/de0-nano.dts
new file mode 100644
index 000000000..b5b854e7e
--- /dev/null
+++ b/arch/openrisc/boot/dts/de0-nano.dts
@@ -0,0 +1,54 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/dts-v1/;
+
+#include "de0-nano-common.dtsi"
+
+/ {
+	model = "Terasic DE0 Nano";
+	compatible = "opencores,or1ksim";
+	#address-cells = <1>;
+	#size-cells = <1>;
+	interrupt-parent = <&pic>;
+
+	aliases {
+		uart0 = &serial0;
+	};
+
+	chosen {
+		stdout-path = "uart0:115200";
+	};
+
+	cpus {
+		#address-cells = <1>;
+		#size-cells = <0>;
+
+		cpu@0 {
+			compatible = "opencores,or1200-rtlsvn481";
+			reg = <0>;
+			clock-frequency = <50000000>;
+		};
+	};
+
+	/*
+	 * OR1K PIC is built into CPU and accessed via special purpose
+	 * registers.  It is not addressable and, hence, has no 'reg'
+	 * property.
+	 */
+	pic: pic {
+		compatible = "opencores,or1k-pic";
+		#interrupt-cells = <1>;
+		interrupt-controller;
+	};
+
+	serial0: serial@90000000 {
+		compatible = "opencores,uart16550-rtlsvn105", "ns16550a";
+		reg = <0x90000000 0x100>;
+		interrupts = <2>;
+		clock-frequency = <50000000>;
+	};
+};
+
+&gpio1 {
+	status = "okay";
+};
diff --git a/arch/openrisc/boot/dts/or1klitex.dts b/arch/openrisc/boot/dts/or1klitex.dts
new file mode 100644
index 000000000..91c7173c5
--- /dev/null
+++ b/arch/openrisc/boot/dts/or1klitex.dts
@@ -0,0 +1,64 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * LiteX-based System on Chip
+ *
+ * Copyright (C) 2019 Antmicro <www.antmicro.com>
+ */
+
+/dts-v1/;
+/ {
+	compatible = "opencores,or1ksim";
+	#address-cells = <1>;
+	#size-cells = <1>;
+	interrupt-parent = <&pic>;
+
+	aliases {
+		serial0 = &serial0;
+	};
+
+	chosen {
+		bootargs = "console=liteuart";
+	};
+
+	memory@0 {
+		device_type = "memory";
+		reg = <0x00000000 0x10000000>;
+	};
+
+	cpus {
+		#address-cells = <1>;
+		#size-cells = <0>;
+		cpu@0 {
+			compatible = "opencores,or1200-rtlsvn481";
+			reg = <0>;
+			clock-frequency = <100000000>;
+		};
+	};
+
+	pic: pic {
+		compatible = "opencores,or1k-pic";
+		#interrupt-cells = <1>;
+		interrupt-controller;
+	};
+
+	serial0: serial@e0006800 {
+		device_type = "serial";
+		compatible = "litex,liteuart";
+		reg = <0xe0006800 0x100>;
+	};
+
+	soc_ctrl0: soc_controller@e0000000 {
+			compatible = "litex,soc-controller";
+			reg = <0xe0000000 0xc>;
+			status = "okay";
+	};
+
+	ethernet@e0001000 {
+		compatible = "litex,liteeth";
+		reg = <0xe0001000 0x7c>,
+		      <0xe0001800 0x0a>,
+		      <0x80000000 0x2000>;
+		reg-names = "mac", "mdio", "buffer";
+		interrupts = <2>;
+	};
+};
diff --git a/arch/openrisc/boot/dts/or1ksim.dts b/arch/openrisc/boot/dts/or1ksim.dts
new file mode 100644
index 000000000..c0cb74e52
--- /dev/null
+++ b/arch/openrisc/boot/dts/or1ksim.dts
@@ -0,0 +1,57 @@
+// SPDX-License-Identifier: GPL-2.0
+/dts-v1/;
+/ {
+	compatible = "opencores,or1ksim";
+	#address-cells = <1>;
+	#size-cells = <1>;
+	interrupt-parent = <&pic>;
+
+	aliases {
+		uart0 = &serial0;
+	};
+
+	chosen {
+		bootargs = "earlycon";
+		stdout-path = "uart0:115200";
+	};
+
+	memory@0 {
+		device_type = "memory";
+		reg = <0x00000000 0x02000000>;
+	};
+
+	cpus {
+		#address-cells = <1>;
+		#size-cells = <0>;
+		cpu@0 {
+			compatible = "opencores,or1200-rtlsvn481";
+			reg = <0>;
+			clock-frequency = <20000000>;
+		};
+	};
+
+	/*
+	 * OR1K PIC is built into CPU and accessed via special purpose
+	 * registers.  It is not addressable and, hence, has no 'reg'
+	 * property.
+	 */
+	pic: pic {
+		compatible = "opencores,or1k-pic";
+		#interrupt-cells = <1>;
+		interrupt-controller;
+	};
+
+	serial0: serial@90000000 {
+		compatible = "opencores,uart16550-rtlsvn105", "ns16550a";
+		reg = <0x90000000 0x100>;
+		interrupts = <2>;
+		clock-frequency = <20000000>;
+	};
+
+	enet0: ethoc@92000000 {
+		compatible = "opencores,ethoc";
+		reg = <0x92000000 0x800>;
+		interrupts = <4>;
+		big-endian;
+	};
+};
diff --git a/arch/openrisc/boot/dts/simple-smp.dts b/arch/openrisc/boot/dts/simple-smp.dts
new file mode 100644
index 000000000..01cf219e6
--- /dev/null
+++ b/arch/openrisc/boot/dts/simple-smp.dts
@@ -0,0 +1,25 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/dts-v1/;
+
+#include "simple-smp.dtsi"
+
+/ {
+	model = "Simple SMP Board";
+};
+
+&cpu0 {
+	clock-frequency = <20000000>;
+};
+
+&cpu1 {
+	clock-frequency = <20000000>;
+};
+
+&enet0 {
+	status = "okay";
+};
+
+&serial0 {
+	clock-frequency = <20000000>;
+};
diff --git a/arch/openrisc/boot/dts/simple-smp.dtsi b/arch/openrisc/boot/dts/simple-smp.dtsi
new file mode 100644
index 000000000..42d6eda33
--- /dev/null
+++ b/arch/openrisc/boot/dts/simple-smp.dtsi
@@ -0,0 +1,68 @@
+/ {
+	compatible = "opencores,or1ksim";
+	#address-cells = <1>;
+	#size-cells = <1>;
+	interrupt-parent = <&pic>;
+
+	aliases {
+		uart0 = &serial0;
+	};
+
+	chosen {
+		bootargs = "earlycon";
+		stdout-path = "uart0:115200";
+	};
+
+	memory@0 {
+		device_type = "memory";
+		reg = <0x00000000 0x02000000>;
+	};
+
+	cpus {
+		#address-cells = <1>;
+		#size-cells = <0>;
+
+		cpu0: cpu@0 {
+			compatible = "opencores,or1200-rtlsvn481";
+			reg = <0>;
+		};
+
+		cpu1: cpu@1 {
+			compatible = "opencores,or1200-rtlsvn481";
+			reg = <1>;
+		};
+	};
+
+	ompic: ompic@98000000 {
+		compatible = "openrisc,ompic";
+		reg = <0x98000000 16>;
+		interrupt-controller;
+		#interrupt-cells = <0>;
+		interrupts = <1>;
+	};
+
+	/*
+	 * OR1K PIC is built into CPU and accessed via special purpose
+	 * registers.  It is not addressable and, hence, has no 'reg'
+	 * property.
+	 */
+	pic: pic {
+		compatible = "opencores,or1k-pic-level";
+		#interrupt-cells = <1>;
+		interrupt-controller;
+	};
+
+	serial0: serial@90000000 {
+		compatible = "opencores,uart16550-rtlsvn105", "ns16550a";
+		reg = <0x90000000 0x100>;
+		interrupts = <2>;
+	};
+
+	enet0: ethoc@92000000 {
+		compatible = "opencores,ethoc";
+		reg = <0x92000000 0x800>;
+		interrupts = <4>;
+		big-endian;
+		status = "disabled";
+	};
+};