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, 168 insertions, 0 deletions

diff --git a/arch/m68k/mac/psc.c b/arch/m68k/mac/psc.c
new file mode 100644
index 000000000..b4183cf66
--- /dev/null
+++ b/arch/m68k/mac/psc.c
@@ -0,0 +1,168 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ *	Apple Peripheral System Controller (PSC)
+ *
+ *	The PSC is used on the AV Macs to control IO functions not handled
+ *	by the VIAs (Ethernet, DSP, SCC).
+ *
+ * TO DO:
+ *
+ * Try to figure out what's going on in pIFR5 and pIFR6. There seem to be
+ * persisant interrupt conditions in those registers and I have no idea what
+ * they are. Granted it doesn't affect since we're not enabling any interrupts
+ * on those levels at the moment, but it would be nice to know. I have a feeling
+ * they aren't actually interrupt lines but data lines (to the DSP?)
+ */
+
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/delay.h>
+#include <linux/init.h>
+#include <linux/irq.h>
+
+#include <asm/traps.h>
+#include <asm/macintosh.h>
+#include <asm/macints.h>
+#include <asm/mac_psc.h>
+
+#include "mac.h"
+
+#define DEBUG_PSC
+
+volatile __u8 *psc;
+EXPORT_SYMBOL_GPL(psc);
+
+/*
+ * Debugging dump, used in various places to see what's going on.
+ */
+
+static void psc_debug_dump(void)
+{
+	int	i;
+
+	if (!psc)
+		return;
+
+	for (i = 0x30 ; i < 0x70 ; i += 0x10) {
+		printk(KERN_DEBUG "PSC #%d:  IFR = 0x%02X IER = 0x%02X\n",
+			i >> 4,
+			(int) psc_read_byte(pIFRbase + i),
+			(int) psc_read_byte(pIERbase + i));
+	}
+}
+
+/*
+ * Try to kill all DMA channels on the PSC. Not sure how this his
+ * supposed to work; this is code lifted from macmace.c and then
+ * expanded to cover what I think are the other 7 channels.
+ */
+
+static __init void psc_dma_die_die_die(void)
+{
+	int i;
+
+	for (i = 0 ; i < 9 ; i++) {
+		psc_write_word(PSC_CTL_BASE + (i << 4), 0x8800);
+		psc_write_word(PSC_CTL_BASE + (i << 4), 0x1000);
+		psc_write_word(PSC_CMD_BASE + (i << 5), 0x1100);
+		psc_write_word(PSC_CMD_BASE + (i << 5) + 0x10, 0x1100);
+	}
+}
+
+/*
+ * Initialize the PSC. For now this just involves shutting down all
+ * interrupt sources using the IERs.
+ */
+
+void __init psc_init(void)
+{
+	int i;
+
+	if (macintosh_config->ident != MAC_MODEL_C660
+	 && macintosh_config->ident != MAC_MODEL_Q840)
+	{
+		psc = NULL;
+		return;
+	}
+
+	/*
+	 * The PSC is always at the same spot, but using psc
+	 * keeps things consistent with the psc_xxxx functions.
+	 */
+
+	psc = (void *) PSC_BASE;
+
+	pr_debug("PSC detected at %p\n", psc);
+
+	psc_dma_die_die_die();
+
+#ifdef DEBUG_PSC
+	psc_debug_dump();
+#endif
+	/*
+	 * Mask and clear all possible interrupts
+	 */
+
+	for (i = 0x30 ; i < 0x70 ; i += 0x10) {
+		psc_write_byte(pIERbase + i, 0x0F);
+		psc_write_byte(pIFRbase + i, 0x0F);
+	}
+}
+
+/*
+ * PSC interrupt handler. It's a lot like the VIA interrupt handler.
+ */
+
+static void psc_irq(struct irq_desc *desc)
+{
+	unsigned int offset = (unsigned int)irq_desc_get_handler_data(desc);
+	unsigned int irq = irq_desc_get_irq(desc);
+	int pIFR	= pIFRbase + offset;
+	int pIER	= pIERbase + offset;
+	int irq_num;
+	unsigned char irq_bit, events;
+
+	events = psc_read_byte(pIFR) & psc_read_byte(pIER) & 0xF;
+	if (!events)
+		return;
+
+	irq_num = irq << 3;
+	irq_bit = 1;
+	do {
+		if (events & irq_bit) {
+			psc_write_byte(pIFR, irq_bit);
+			generic_handle_irq(irq_num);
+		}
+		irq_num++;
+		irq_bit <<= 1;
+	} while (events >= irq_bit);
+}
+
+/*
+ * Register the PSC interrupt dispatchers for autovector interrupts 3-6.
+ */
+
+void __init psc_register_interrupts(void)
+{
+	irq_set_chained_handler_and_data(IRQ_AUTO_3, psc_irq, (void *)0x30);
+	irq_set_chained_handler_and_data(IRQ_AUTO_4, psc_irq, (void *)0x40);
+	irq_set_chained_handler_and_data(IRQ_AUTO_5, psc_irq, (void *)0x50);
+	irq_set_chained_handler_and_data(IRQ_AUTO_6, psc_irq, (void *)0x60);
+}
+
+void psc_irq_enable(int irq) {
+	int irq_src	= IRQ_SRC(irq);
+	int irq_idx	= IRQ_IDX(irq);
+	int pIER	= pIERbase + (irq_src << 4);
+
+	psc_write_byte(pIER, (1 << irq_idx) | 0x80);
+}
+
+void psc_irq_disable(int irq) {
+	int irq_src	= IRQ_SRC(irq);
+	int irq_idx	= IRQ_IDX(irq);
+	int pIER	= pIERbase + (irq_src << 4);
+
+	psc_write_byte(pIER, 1 << irq_idx);
+}