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, 201 insertions, 0 deletions

diff --git a/arch/mips/jazz/setup.c b/arch/mips/jazz/setup.c
new file mode 100644
index 000000000..d21e5d441
--- /dev/null
+++ b/arch/mips/jazz/setup.c
@@ -0,0 +1,201 @@
+/*
+ * Setup pointers to hardware-dependent routines.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1996, 1997, 1998, 2001, 07, 08 by Ralf Baechle
+ * Copyright (C) 2001 MIPS Technologies, Inc.
+ * Copyright (C) 2007 by Thomas Bogendoerfer
+ */
+#include <linux/eisa.h>
+#include <linux/init.h>
+#include <linux/ioport.h>
+#include <linux/console.h>
+#include <linux/platform_device.h>
+#include <linux/serial_8250.h>
+#include <linux/dma-mapping.h>
+#include <linux/pgtable.h>
+
+#include <asm/jazz.h>
+#include <asm/jazzdma.h>
+#include <asm/reboot.h>
+#include <asm/tlbmisc.h>
+
+extern void jazz_machine_restart(char *command);
+
+static struct resource jazz_io_resources[] = {
+	{
+		.start	= 0x00,
+		.end	= 0x1f,
+		.name	= "dma1",
+		.flags	= IORESOURCE_IO | IORESOURCE_BUSY
+	}, {
+		.start	= 0x40,
+		.end	= 0x5f,
+		.name	= "timer",
+		.flags	= IORESOURCE_IO | IORESOURCE_BUSY
+	}, {
+		.start	= 0x80,
+		.end	= 0x8f,
+		.name	= "dma page reg",
+		.flags	= IORESOURCE_IO | IORESOURCE_BUSY
+	}, {
+		.start	= 0xc0,
+		.end	= 0xdf,
+		.name	= "dma2",
+		.flags	= IORESOURCE_IO | IORESOURCE_BUSY
+	}
+};
+
+void __init plat_mem_setup(void)
+{
+	int i;
+
+	/* Map 0xe0000000 -> 0x0:800005C0, 0xe0010000 -> 0x1:30000580 */
+	add_wired_entry(0x02000017, 0x03c00017, 0xe0000000, PM_64K);
+	/* Map 0xe2000000 -> 0x0:900005C0, 0xe3010000 -> 0x0:910005C0 */
+	add_wired_entry(0x02400017, 0x02440017, 0xe2000000, PM_16M);
+	/* Map 0xe4000000 -> 0x0:600005C0, 0xe4100000 -> 400005C0 */
+	add_wired_entry(0x01800017, 0x01000017, 0xe4000000, PM_4M);
+
+	set_io_port_base(JAZZ_PORT_BASE);
+#ifdef CONFIG_EISA
+	EISA_bus = 1;
+#endif
+
+	/* request I/O space for devices used on all i[345]86 PCs */
+	for (i = 0; i < ARRAY_SIZE(jazz_io_resources); i++)
+		request_resource(&ioport_resource, jazz_io_resources + i);
+
+	/* The RTC is outside the port address space */
+
+	_machine_restart = jazz_machine_restart;
+
+	add_preferred_console("ttyS", 0, "9600");
+}
+
+#ifdef CONFIG_OLIVETTI_M700
+#define UART_CLK  1843200
+#else
+/* Some Jazz machines seem to have an 8MHz crystal clock but I don't know
+   exactly which ones ... XXX */
+#define UART_CLK (8000000 / 16) /* ( 3072000 / 16) */
+#endif
+
+#define MEMPORT(_base, _irq)				\
+	{						\
+		.mapbase	= (_base),		\
+		.membase	= (void *)(_base),	\
+		.irq		= (_irq),		\
+		.uartclk	= UART_CLK,		\
+		.iotype		= UPIO_MEM,		\
+		.flags		= UPF_BOOT_AUTOCONF,	\
+	}
+
+static struct plat_serial8250_port jazz_serial_data[] = {
+	MEMPORT(JAZZ_SERIAL1_BASE, JAZZ_SERIAL1_IRQ),
+	MEMPORT(JAZZ_SERIAL2_BASE, JAZZ_SERIAL2_IRQ),
+	{ },
+};
+
+static struct platform_device jazz_serial8250_device = {
+	.name			= "serial8250",
+	.id			= PLAT8250_DEV_PLATFORM,
+	.dev			= {
+		.platform_data	= jazz_serial_data,
+	},
+};
+
+static struct resource jazz_esp_rsrc[] = {
+	{
+		.start = JAZZ_SCSI_BASE,
+		.end   = JAZZ_SCSI_BASE + 31,
+		.flags = IORESOURCE_MEM
+	},
+	{
+		.start = JAZZ_SCSI_DMA,
+		.end   = JAZZ_SCSI_DMA,
+		.flags = IORESOURCE_MEM
+	},
+	{
+		.start = JAZZ_SCSI_IRQ,
+		.end   = JAZZ_SCSI_IRQ,
+		.flags = IORESOURCE_IRQ
+	}
+};
+
+static u64 jazz_esp_dma_mask = DMA_BIT_MASK(32);
+
+static struct platform_device jazz_esp_pdev = {
+	.name		= "jazz_esp",
+	.num_resources	= ARRAY_SIZE(jazz_esp_rsrc),
+	.resource	= jazz_esp_rsrc,
+	.dev = {
+		.dma_mask	   = &jazz_esp_dma_mask,
+		.coherent_dma_mask = DMA_BIT_MASK(32),
+	}
+};
+
+static struct resource jazz_sonic_rsrc[] = {
+	{
+		.start = JAZZ_ETHERNET_BASE,
+		.end   = JAZZ_ETHERNET_BASE + 0xff,
+		.flags = IORESOURCE_MEM
+	},
+	{
+		.start = JAZZ_ETHERNET_IRQ,
+		.end   = JAZZ_ETHERNET_IRQ,
+		.flags = IORESOURCE_IRQ
+	}
+};
+
+static u64 jazz_sonic_dma_mask = DMA_BIT_MASK(32);
+
+static struct platform_device jazz_sonic_pdev = {
+	.name		= "jazzsonic",
+	.num_resources	= ARRAY_SIZE(jazz_sonic_rsrc),
+	.resource	= jazz_sonic_rsrc,
+	.dev = {
+		.dma_mask	   = &jazz_sonic_dma_mask,
+		.coherent_dma_mask = DMA_BIT_MASK(32),
+	}
+};
+
+static struct resource jazz_cmos_rsrc[] = {
+	{
+		.start = 0x70,
+		.end   = 0x71,
+		.flags = IORESOURCE_IO
+	},
+	{
+		.start = 8,
+		.end   = 8,
+		.flags = IORESOURCE_IRQ
+	}
+};
+
+static struct platform_device jazz_cmos_pdev = {
+	.name		= "rtc_cmos",
+	.num_resources	= ARRAY_SIZE(jazz_cmos_rsrc),
+	.resource	= jazz_cmos_rsrc
+};
+
+static struct platform_device pcspeaker_pdev = {
+	.name		= "pcspkr",
+	.id		= -1,
+};
+
+static int __init jazz_setup_devinit(void)
+{
+	platform_device_register(&jazz_serial8250_device);
+	platform_device_register(&jazz_esp_pdev);
+	platform_device_register(&jazz_sonic_pdev);
+	platform_device_register(&jazz_cmos_pdev);
+	platform_device_register(&pcspeaker_pdev);
+
+	return 0;
+}
+
+device_initcall(jazz_setup_devinit);