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, 299 insertions, 0 deletions

diff --git a/arch/microblaze/mm/fault.c b/arch/microblaze/mm/fault.c
new file mode 100644
index 000000000..d3c3c33b7
--- /dev/null
+++ b/arch/microblaze/mm/fault.c
@@ -0,0 +1,299 @@
+/*
+ *  arch/microblaze/mm/fault.c
+ *
+ *    Copyright (C) 2007 Xilinx, Inc.  All rights reserved.
+ *
+ *  Derived from "arch/ppc/mm/fault.c"
+ *    Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org)
+ *
+ *  Derived from "arch/i386/mm/fault.c"
+ *    Copyright (C) 1991, 1992, 1993, 1994  Linus Torvalds
+ *
+ *  Modified by Cort Dougan and Paul Mackerras.
+ *
+ * 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.
+ *
+ */
+
+#include <linux/extable.h>
+#include <linux/signal.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/ptrace.h>
+#include <linux/mman.h>
+#include <linux/mm.h>
+#include <linux/interrupt.h>
+#include <linux/perf_event.h>
+
+#include <asm/page.h>
+#include <asm/mmu.h>
+#include <linux/mmu_context.h>
+#include <linux/uaccess.h>
+#include <asm/exceptions.h>
+
+static unsigned long pte_misses;	/* updated by do_page_fault() */
+static unsigned long pte_errors;	/* updated by do_page_fault() */
+
+/*
+ * Check whether the instruction at regs->pc is a store using
+ * an update addressing form which will update r1.
+ */
+static int store_updates_sp(struct pt_regs *regs)
+{
+	unsigned int inst;
+
+	if (get_user(inst, (unsigned int __user *)regs->pc))
+		return 0;
+	/* check for 1 in the rD field */
+	if (((inst >> 21) & 0x1f) != 1)
+		return 0;
+	/* check for store opcodes */
+	if ((inst & 0xd0000000) == 0xd0000000)
+		return 1;
+	return 0;
+}
+
+
+/*
+ * bad_page_fault is called when we have a bad access from the kernel.
+ * It is called from do_page_fault above and from some of the procedures
+ * in traps.c.
+ */
+void bad_page_fault(struct pt_regs *regs, unsigned long address, int sig)
+{
+	const struct exception_table_entry *fixup;
+/* MS: no context */
+	/* Are we prepared to handle this fault?  */
+	fixup = search_exception_tables(regs->pc);
+	if (fixup) {
+		regs->pc = fixup->fixup;
+		return;
+	}
+
+	/* kernel has accessed a bad area */
+	die("kernel access of bad area", regs, sig);
+}
+
+/*
+ * The error_code parameter is ESR for a data fault,
+ * 0 for an instruction fault.
+ */
+void do_page_fault(struct pt_regs *regs, unsigned long address,
+		   unsigned long error_code)
+{
+	struct vm_area_struct *vma;
+	struct mm_struct *mm = current->mm;
+	int code = SEGV_MAPERR;
+	int is_write = error_code & ESR_S;
+	vm_fault_t fault;
+	unsigned int flags = FAULT_FLAG_DEFAULT;
+
+	regs->ear = address;
+	regs->esr = error_code;
+
+	/* On a kernel SLB miss we can only check for a valid exception entry */
+	if (unlikely(kernel_mode(regs) && (address >= TASK_SIZE))) {
+		pr_warn("kernel task_size exceed");
+		_exception(SIGSEGV, regs, code, address);
+	}
+
+	/* for instr TLB miss and instr storage exception ESR_S is undefined */
+	if ((error_code & 0x13) == 0x13 || (error_code & 0x11) == 0x11)
+		is_write = 0;
+
+	if (unlikely(faulthandler_disabled() || !mm)) {
+		if (kernel_mode(regs))
+			goto bad_area_nosemaphore;
+
+		/* faulthandler_disabled() in user mode is really bad,
+		   as is current->mm == NULL. */
+		pr_emerg("Page fault in user mode with faulthandler_disabled(), mm = %p\n",
+			 mm);
+		pr_emerg("r15 = %lx  MSR = %lx\n",
+		       regs->r15, regs->msr);
+		die("Weird page fault", regs, SIGSEGV);
+	}
+
+	if (user_mode(regs))
+		flags |= FAULT_FLAG_USER;
+
+	perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, address);
+
+	/* When running in the kernel we expect faults to occur only to
+	 * addresses in user space.  All other faults represent errors in the
+	 * kernel and should generate an OOPS.  Unfortunately, in the case of an
+	 * erroneous fault occurring in a code path which already holds mmap_lock
+	 * we will deadlock attempting to validate the fault against the
+	 * address space.  Luckily the kernel only validly references user
+	 * space from well defined areas of code, which are listed in the
+	 * exceptions table.
+	 *
+	 * As the vast majority of faults will be valid we will only perform
+	 * the source reference check when there is a possibility of a deadlock.
+	 * Attempt to lock the address space, if we cannot we then validate the
+	 * source.  If this is invalid we can skip the address space check,
+	 * thus avoiding the deadlock.
+	 */
+	if (unlikely(!mmap_read_trylock(mm))) {
+		if (kernel_mode(regs) && !search_exception_tables(regs->pc))
+			goto bad_area_nosemaphore;
+
+retry:
+		mmap_read_lock(mm);
+	}
+
+	vma = find_vma(mm, address);
+	if (unlikely(!vma))
+		goto bad_area;
+
+	if (vma->vm_start <= address)
+		goto good_area;
+
+	if (unlikely(!(vma->vm_flags & VM_GROWSDOWN)))
+		goto bad_area;
+
+	if (unlikely(!is_write))
+		goto bad_area;
+
+	/*
+	 * N.B. The ABI allows programs to access up to
+	 * a few hundred bytes below the stack pointer (TBD).
+	 * The kernel signal delivery code writes up to about 1.5kB
+	 * below the stack pointer (r1) before decrementing it.
+	 * The exec code can write slightly over 640kB to the stack
+	 * before setting the user r1.  Thus we allow the stack to
+	 * expand to 1MB without further checks.
+	 */
+	if (unlikely(address + 0x100000 < vma->vm_end)) {
+
+		/* get user regs even if this fault is in kernel mode */
+		struct pt_regs *uregs = current->thread.regs;
+		if (uregs == NULL)
+			goto bad_area;
+
+		/*
+		 * A user-mode access to an address a long way below
+		 * the stack pointer is only valid if the instruction
+		 * is one which would update the stack pointer to the
+		 * address accessed if the instruction completed,
+		 * i.e. either stwu rs,n(r1) or stwux rs,r1,rb
+		 * (or the byte, halfword, float or double forms).
+		 *
+		 * If we don't check this then any write to the area
+		 * between the last mapped region and the stack will
+		 * expand the stack rather than segfaulting.
+		 */
+		if (address + 2048 < uregs->r1
+			&& (kernel_mode(regs) || !store_updates_sp(regs)))
+				goto bad_area;
+	}
+	vma = expand_stack(mm, address);
+	if (!vma)
+		goto bad_area_nosemaphore;
+
+good_area:
+	code = SEGV_ACCERR;
+
+	/* a write */
+	if (unlikely(is_write)) {
+		if (unlikely(!(vma->vm_flags & VM_WRITE)))
+			goto bad_area;
+		flags |= FAULT_FLAG_WRITE;
+	/* a read */
+	} else {
+		/* protection fault */
+		if (unlikely(error_code & 0x08000000))
+			goto bad_area;
+		if (unlikely(!(vma->vm_flags & (VM_READ | VM_EXEC))))
+			goto bad_area;
+	}
+
+	/*
+	 * If for any reason at all we couldn't handle the fault,
+	 * make sure we exit gracefully rather than endlessly redo
+	 * the fault.
+	 */
+	fault = handle_mm_fault(vma, address, flags, regs);
+
+	if (fault_signal_pending(fault, regs)) {
+		if (!user_mode(regs))
+			bad_page_fault(regs, address, SIGBUS);
+		return;
+	}
+
+	/* The fault is fully completed (including releasing mmap lock) */
+	if (fault & VM_FAULT_COMPLETED)
+		return;
+
+	if (unlikely(fault & VM_FAULT_ERROR)) {
+		if (fault & VM_FAULT_OOM)
+			goto out_of_memory;
+		else if (fault & VM_FAULT_SIGSEGV)
+			goto bad_area;
+		else if (fault & VM_FAULT_SIGBUS)
+			goto do_sigbus;
+		BUG();
+	}
+
+	if (fault & VM_FAULT_RETRY) {
+		flags |= FAULT_FLAG_TRIED;
+
+		/*
+		 * No need to mmap_read_unlock(mm) as we would
+		 * have already released it in __lock_page_or_retry
+		 * in mm/filemap.c.
+		 */
+
+		goto retry;
+	}
+
+	mmap_read_unlock(mm);
+
+	/*
+	 * keep track of tlb+htab misses that are good addrs but
+	 * just need pte's created via handle_mm_fault()
+	 * -- Cort
+	 */
+	pte_misses++;
+	return;
+
+bad_area:
+	mmap_read_unlock(mm);
+
+bad_area_nosemaphore:
+	pte_errors++;
+
+	/* User mode accesses cause a SIGSEGV */
+	if (user_mode(regs)) {
+		_exception(SIGSEGV, regs, code, address);
+		return;
+	}
+
+	bad_page_fault(regs, address, SIGSEGV);
+	return;
+
+/*
+ * We ran out of memory, or some other thing happened to us that made
+ * us unable to handle the page fault gracefully.
+ */
+out_of_memory:
+	mmap_read_unlock(mm);
+	if (!user_mode(regs))
+		bad_page_fault(regs, address, SIGKILL);
+	else
+		pagefault_out_of_memory();
+	return;
+
+do_sigbus:
+	mmap_read_unlock(mm);
+	if (user_mode(regs)) {
+		force_sig_fault(SIGBUS, BUS_ADRERR, (void __user *)address);
+		return;
+	}
+	bad_page_fault(regs, address, SIGBUS);
+}