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, 264 insertions, 0 deletions

diff --git a/arch/microblaze/mm/pgtable.c b/arch/microblaze/mm/pgtable.c
new file mode 100644
index 000000000..e96dd1b7a
--- /dev/null
+++ b/arch/microblaze/mm/pgtable.c
@@ -0,0 +1,264 @@
+/*
+ *  This file contains the routines setting up the linux page tables.
+ *
+ * Copyright (C) 2008 Michal Simek
+ * Copyright (C) 2008 PetaLogix
+ *
+ *    Copyright (C) 2007 Xilinx, Inc.  All rights reserved.
+ *
+ *  Derived from arch/ppc/mm/pgtable.c:
+ *    -- paulus
+ *
+ *  Derived from arch/ppc/mm/init.c:
+ *    Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org)
+ *
+ *  Modifications by Paul Mackerras (PowerMac) (paulus@cs.anu.edu.au)
+ *  and Cort Dougan (PReP) (cort@cs.nmt.edu)
+ *    Copyright (C) 1996 Paul Mackerras
+ *  Amiga/APUS changes by Jesper Skov (jskov@cygnus.co.uk).
+ *
+ *  Derived from "arch/i386/mm/init.c"
+ *    Copyright (C) 1991, 1992, 1993, 1994  Linus Torvalds
+ *
+ *  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/export.h>
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/vmalloc.h>
+#include <linux/init.h>
+#include <linux/mm_types.h>
+#include <linux/pgtable.h>
+#include <linux/memblock.h>
+#include <linux/kallsyms.h>
+
+#include <asm/pgalloc.h>
+#include <linux/io.h>
+#include <asm/mmu.h>
+#include <asm/sections.h>
+#include <asm/fixmap.h>
+
+unsigned long ioremap_base;
+unsigned long ioremap_bot;
+EXPORT_SYMBOL(ioremap_bot);
+
+static void __iomem *__ioremap(phys_addr_t addr, unsigned long size,
+		unsigned long flags)
+{
+	unsigned long v, i;
+	phys_addr_t p;
+	int err;
+
+	/*
+	 * Choose an address to map it to.
+	 * Once the vmalloc system is running, we use it.
+	 * Before then, we use space going down from ioremap_base
+	 * (ioremap_bot records where we're up to).
+	 */
+	p = addr & PAGE_MASK;
+	size = PAGE_ALIGN(addr + size) - p;
+
+	/*
+	 * Don't allow anybody to remap normal RAM that we're using.
+	 * mem_init() sets high_memory so only do the check after that.
+	 *
+	 * However, allow remap of rootfs: TBD
+	 */
+
+	if (mem_init_done &&
+		p >= memory_start && p < virt_to_phys(high_memory) &&
+		!(p >= __virt_to_phys((phys_addr_t)__bss_stop) &&
+		p < __virt_to_phys((phys_addr_t)__bss_stop))) {
+		pr_warn("__ioremap(): phys addr "PTE_FMT" is RAM lr %ps\n",
+			(unsigned long)p, __builtin_return_address(0));
+		return NULL;
+	}
+
+	if (size == 0)
+		return NULL;
+
+	/*
+	 * Is it already mapped? If the whole area is mapped then we're
+	 * done, otherwise remap it since we want to keep the virt addrs for
+	 * each request contiguous.
+	 *
+	 * We make the assumption here that if the bottom and top
+	 * of the range we want are mapped then it's mapped to the
+	 * same virt address (and this is contiguous).
+	 *  -- Cort
+	 */
+
+	if (mem_init_done) {
+		struct vm_struct *area;
+		area = get_vm_area(size, VM_IOREMAP);
+		if (area == NULL)
+			return NULL;
+		v = (unsigned long) area->addr;
+	} else {
+		v = (ioremap_bot -= size);
+	}
+
+	if ((flags & _PAGE_PRESENT) == 0)
+		flags |= _PAGE_KERNEL;
+	if (flags & _PAGE_NO_CACHE)
+		flags |= _PAGE_GUARDED;
+
+	err = 0;
+	for (i = 0; i < size && err == 0; i += PAGE_SIZE)
+		err = map_page(v + i, p + i, flags);
+	if (err) {
+		if (mem_init_done)
+			vfree((void *)v);
+		return NULL;
+	}
+
+	return (void __iomem *) (v + ((unsigned long)addr & ~PAGE_MASK));
+}
+
+void __iomem *ioremap(phys_addr_t addr, unsigned long size)
+{
+	return __ioremap(addr, size, _PAGE_NO_CACHE);
+}
+EXPORT_SYMBOL(ioremap);
+
+void iounmap(volatile void __iomem *addr)
+{
+	if ((__force void *)addr > high_memory &&
+					(unsigned long) addr < ioremap_bot)
+		vfree((void *) (PAGE_MASK & (unsigned long) addr));
+}
+EXPORT_SYMBOL(iounmap);
+
+
+int map_page(unsigned long va, phys_addr_t pa, int flags)
+{
+	p4d_t *p4d;
+	pud_t *pud;
+	pmd_t *pd;
+	pte_t *pg;
+	int err = -ENOMEM;
+
+	/* Use upper 10 bits of VA to index the first level map */
+	p4d = p4d_offset(pgd_offset_k(va), va);
+	pud = pud_offset(p4d, va);
+	pd = pmd_offset(pud, va);
+	/* Use middle 10 bits of VA to index the second-level map */
+	pg = pte_alloc_kernel(pd, va); /* from powerpc - pgtable.c */
+	/* pg = pte_alloc_kernel(&init_mm, pd, va); */
+
+	if (pg != NULL) {
+		err = 0;
+		set_pte_at(&init_mm, va, pg, pfn_pte(pa >> PAGE_SHIFT,
+				__pgprot(flags)));
+		if (unlikely(mem_init_done))
+			_tlbie(va);
+	}
+	return err;
+}
+
+/*
+ * Map in all of physical memory starting at CONFIG_KERNEL_START.
+ */
+void __init mapin_ram(void)
+{
+	unsigned long v, p, s, f;
+
+	v = CONFIG_KERNEL_START;
+	p = memory_start;
+	for (s = 0; s < lowmem_size; s += PAGE_SIZE) {
+		f = _PAGE_PRESENT | _PAGE_ACCESSED |
+				_PAGE_SHARED | _PAGE_HWEXEC;
+		if (!is_kernel_text(v))
+			f |= _PAGE_WRENABLE;
+		else
+			/* On the MicroBlaze, no user access
+			   forces R/W kernel access */
+			f |= _PAGE_USER;
+		map_page(v, p, f);
+		v += PAGE_SIZE;
+		p += PAGE_SIZE;
+	}
+}
+
+/* is x a power of 2? */
+#define is_power_of_2(x)	((x) != 0 && (((x) & ((x) - 1)) == 0))
+
+/* Scan the real Linux page tables and return a PTE pointer for
+ * a virtual address in a context.
+ * Returns true (1) if PTE was found, zero otherwise.  The pointer to
+ * the PTE pointer is unmodified if PTE is not found.
+ */
+static int get_pteptr(struct mm_struct *mm, unsigned long addr, pte_t **ptep)
+{
+	pgd_t	*pgd;
+	p4d_t	*p4d;
+	pud_t	*pud;
+	pmd_t	*pmd;
+	pte_t	*pte;
+	int     retval = 0;
+
+	pgd = pgd_offset(mm, addr & PAGE_MASK);
+	if (pgd) {
+		p4d = p4d_offset(pgd, addr & PAGE_MASK);
+		pud = pud_offset(p4d, addr & PAGE_MASK);
+		pmd = pmd_offset(pud, addr & PAGE_MASK);
+		if (pmd_present(*pmd)) {
+			pte = pte_offset_kernel(pmd, addr & PAGE_MASK);
+			if (pte) {
+				retval = 1;
+				*ptep = pte;
+			}
+		}
+	}
+	return retval;
+}
+
+/* Find physical address for this virtual address.  Normally used by
+ * I/O functions, but anyone can call it.
+ */
+unsigned long iopa(unsigned long addr)
+{
+	unsigned long pa;
+
+	pte_t *pte;
+	struct mm_struct *mm;
+
+	/* Allow mapping of user addresses (within the thread)
+	 * for DMA if necessary.
+	 */
+	if (addr < TASK_SIZE)
+		mm = current->mm;
+	else
+		mm = &init_mm;
+
+	pa = 0;
+	if (get_pteptr(mm, addr, &pte))
+		pa = (pte_val(*pte) & PAGE_MASK) | (addr & ~PAGE_MASK);
+
+	return pa;
+}
+
+__ref pte_t *pte_alloc_one_kernel(struct mm_struct *mm)
+{
+	if (mem_init_done)
+		return __pte_alloc_one_kernel(mm);
+	else
+		return memblock_alloc_try_nid(PAGE_SIZE, PAGE_SIZE,
+					      MEMBLOCK_LOW_LIMIT,
+					      memory_start + kernel_tlb,
+					      NUMA_NO_NODE);
+}
+
+void __set_fixmap(enum fixed_addresses idx, phys_addr_t phys, pgprot_t flags)
+{
+	unsigned long address = __fix_to_virt(idx);
+
+	if (idx >= __end_of_fixed_addresses)
+		BUG();
+
+	map_page(address, phys, pgprot_val(flags));
+}