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, 193 insertions, 0 deletions

diff --git a/arch/arc/mm/fault.c b/arch/arc/mm/fault.c
new file mode 100644
index 000000000..95119a5e7
--- /dev/null
+++ b/arch/arc/mm/fault.c
@@ -0,0 +1,193 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Page Fault Handling for ARC (TLB Miss / ProtV)
+ *
+ * Copyright (C) 2004, 2007-2010, 2011-2012 Synopsys, Inc. (www.synopsys.com)
+ */
+
+#include <linux/signal.h>
+#include <linux/interrupt.h>
+#include <linux/sched/signal.h>
+#include <linux/errno.h>
+#include <linux/ptrace.h>
+#include <linux/uaccess.h>
+#include <linux/kdebug.h>
+#include <linux/perf_event.h>
+#include <linux/mm_types.h>
+#include <asm/entry.h>
+#include <asm/mmu.h>
+
+/*
+ * kernel virtual address is required to implement vmalloc/pkmap/fixmap
+ * Refer to asm/processor.h for System Memory Map
+ *
+ * It simply copies the PMD entry (pointer to 2nd level page table or hugepage)
+ * from swapper pgdir to task pgdir. The 2nd level table/page is thus shared
+ */
+noinline static int handle_kernel_vaddr_fault(unsigned long address)
+{
+	/*
+	 * Synchronize this task's top level page-table
+	 * with the 'reference' page table.
+	 */
+	pgd_t *pgd, *pgd_k;
+	p4d_t *p4d, *p4d_k;
+	pud_t *pud, *pud_k;
+	pmd_t *pmd, *pmd_k;
+
+	pgd = pgd_offset(current->active_mm, address);
+	pgd_k = pgd_offset_k(address);
+
+	if (pgd_none (*pgd_k))
+		goto bad_area;
+	if (!pgd_present(*pgd))
+		set_pgd(pgd, *pgd_k);
+
+	p4d = p4d_offset(pgd, address);
+	p4d_k = p4d_offset(pgd_k, address);
+	if (p4d_none(*p4d_k))
+		goto bad_area;
+	if (!p4d_present(*p4d))
+		set_p4d(p4d, *p4d_k);
+
+	pud = pud_offset(p4d, address);
+	pud_k = pud_offset(p4d_k, address);
+	if (pud_none(*pud_k))
+		goto bad_area;
+	if (!pud_present(*pud))
+		set_pud(pud, *pud_k);
+
+	pmd = pmd_offset(pud, address);
+	pmd_k = pmd_offset(pud_k, address);
+	if (pmd_none(*pmd_k))
+		goto bad_area;
+	if (!pmd_present(*pmd))
+		set_pmd(pmd, *pmd_k);
+
+	/* XXX: create the TLB entry here */
+	return 0;
+
+bad_area:
+	return 1;
+}
+
+void do_page_fault(unsigned long address, struct pt_regs *regs)
+{
+	struct vm_area_struct *vma = NULL;
+	struct task_struct *tsk = current;
+	struct mm_struct *mm = tsk->mm;
+	int sig, si_code = SEGV_MAPERR;
+	unsigned int write = 0, exec = 0, mask;
+	vm_fault_t fault = VM_FAULT_SIGSEGV;	/* handle_mm_fault() output */
+	unsigned int flags;			/* handle_mm_fault() input */
+
+	/*
+	 * NOTE! We MUST NOT take any locks for this case. We may
+	 * be in an interrupt or a critical region, and should
+	 * only copy the information from the master page table,
+	 * nothing more.
+	 */
+	if (address >= VMALLOC_START && !user_mode(regs)) {
+		if (unlikely(handle_kernel_vaddr_fault(address)))
+			goto no_context;
+		else
+			return;
+	}
+
+	/*
+	 * If we're in an interrupt or have no user
+	 * context, we must not take the fault..
+	 */
+	if (faulthandler_disabled() || !mm)
+		goto no_context;
+
+	if (regs->ecr.cause & ECR_C_PROTV_STORE)	/* ST/EX */
+		write = 1;
+	else if ((regs->ecr.vec == ECR_V_PROTV) &&
+	         (regs->ecr.cause == ECR_C_PROTV_INST_FETCH))
+		exec = 1;
+
+	flags = FAULT_FLAG_DEFAULT;
+	if (user_mode(regs))
+		flags |= FAULT_FLAG_USER;
+	if (write)
+		flags |= FAULT_FLAG_WRITE;
+
+	perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, address);
+retry:
+	vma = lock_mm_and_find_vma(mm, address, regs);
+	if (!vma)
+		goto bad_area_nosemaphore;
+
+	/*
+	 * vm_area is good, now check permissions for this memory access
+	 */
+	mask = VM_READ;
+	if (write)
+		mask = VM_WRITE;
+	if (exec)
+		mask = VM_EXEC;
+
+	if (!(vma->vm_flags & mask)) {
+		si_code = SEGV_ACCERR;
+		goto bad_area;
+	}
+
+	fault = handle_mm_fault(vma, address, flags, regs);
+
+	/* Quick path to respond to signals */
+	if (fault_signal_pending(fault, regs)) {
+		if (!user_mode(regs))
+			goto no_context;
+		return;
+	}
+
+	/* The fault is fully completed (including releasing mmap lock) */
+	if (fault & VM_FAULT_COMPLETED)
+		return;
+
+	/*
+	 * Fault retry nuances, mmap_lock already relinquished by core mm
+	 */
+	if (unlikely(fault & VM_FAULT_RETRY)) {
+		flags |= FAULT_FLAG_TRIED;
+		goto retry;
+	}
+
+bad_area:
+	mmap_read_unlock(mm);
+
+bad_area_nosemaphore:
+	/*
+	 * Major/minor page fault accounting
+	 * (in case of retry we only land here once)
+	 */
+	if (likely(!(fault & VM_FAULT_ERROR)))
+		/* Normal return path: fault Handled Gracefully */
+		return;
+
+	if (!user_mode(regs))
+		goto no_context;
+
+	if (fault & VM_FAULT_OOM) {
+		pagefault_out_of_memory();
+		return;
+	}
+
+	if (fault & VM_FAULT_SIGBUS) {
+		sig = SIGBUS;
+		si_code = BUS_ADRERR;
+	}
+	else {
+		sig = SIGSEGV;
+	}
+
+	tsk->thread.fault_address = address;
+	force_sig_fault(sig, si_code, (void __user *)address);
+	return;
+
+no_context:
+	if (fixup_exception(regs))
+		return;
+
+	die("Oops", regs, address);
+}