usercopy: Check valid lifetime via stack depth

One of the things that CONFIG_HARDENED_USERCOPY sanity-checks is whether
an object that is about to be copied to/from userspace is overlapping
the stack at all. If it is, it performs a number of inexpensive
bounds checks. One of the finer-grained checks is whether an object
crosses stack frames within the stack region. Doing this on x86 with
CONFIG_FRAME_POINTER was cheap/easy. Doing it with ORC was deemed too
heavy, and was left out (a while ago), leaving the courser whole-stack
check.

The LKDTM tests USERCOPY_STACK_FRAME_TO and USERCOPY_STACK_FRAME_FROM
try to exercise these cross-frame cases to validate the defense is
working. They have been failing ever since ORC was added (which was
expected). While Muhammad was investigating various LKDTM failures[1],
he asked me for additional details on them, and I realized that when
exact stack frame boundary checking is not available (i.e. everything
except x86 with FRAME_POINTER), it could check if a stack object is at
least "current depth valid", in the sense that any object within the
stack region but not between start-of-stack and current_stack_pointer
should be considered unavailable (i.e. its lifetime is from a call no
longer present on the stack).

Introduce ARCH_HAS_CURRENT_STACK_POINTER to track which architectures
have actually implemented the common global register alias.

Additionally report usercopy bounds checking failures with an offset
from current_stack_pointer, which may assist with diagnosing failures.

The LKDTM USERCOPY_STACK_FRAME_TO and USERCOPY_STACK_FRAME_FROM tests
(once slightly adjusted in a separate patch) pass again with this fixed.

[1] https://github.com/kernelci/kernelci-project/issues/84

Cc: Matthew Wilcox (Oracle) <[e-mail]>
Cc: Josh Poimboeuf <[e-mail]>
Cc: Andrew Morton <[e-mail]>
Cc: [e-mail]
Reported-by: Muhammad Usama Anjum <[e-mail]>
Signed-off-by: Kees Cook <[e-mail]>
---
v1: https://lore.kernel.org/lkml/[e-mail]
v2: https://lore.kernel.org/lkml/[e-mail]
v3: https://lore.kernel.org/lkml/[e-mail]
v4: - improve commit log (akpm)

arch/arm/Kconfig

···
 	select ARCH_32BIT_OFF_T
 	select ARCH_CORRECT_STACKTRACE_ON_KRETPROBE if HAVE_KRETPROBES && FRAME_POINTER && !ARM_UNWIND
 	select ARCH_HAS_BINFMT_FLAT
+	select ARCH_HAS_CURRENT_STACK_POINTER
 	select ARCH_HAS_DEBUG_VIRTUAL if MMU
 	select ARCH_HAS_DMA_WRITE_COMBINE if !ARM_DMA_MEM_BUFFERABLE
 	select ARCH_HAS_ELF_RANDOMIZE

arch/arm64/Kconfig

···
 	select ARCH_ENABLE_SPLIT_PMD_PTLOCK if PGTABLE_LEVELS > 2
 	select ARCH_ENABLE_THP_MIGRATION if TRANSPARENT_HUGEPAGE
 	select ARCH_HAS_CACHE_LINE_SIZE
+	select ARCH_HAS_CURRENT_STACK_POINTER
 	select ARCH_HAS_DEBUG_VIRTUAL
 	select ARCH_HAS_DEBUG_VM_PGTABLE
 	select ARCH_HAS_DMA_PREP_COHERENT

arch/powerpc/Kconfig

···
 	select ARCH_ENABLE_MEMORY_HOTPLUG
 	select ARCH_ENABLE_MEMORY_HOTREMOVE
 	select ARCH_HAS_COPY_MC			if PPC64
+	select ARCH_HAS_CURRENT_STACK_POINTER
 	select ARCH_HAS_DEBUG_VIRTUAL
 	select ARCH_HAS_DEBUG_VM_PGTABLE
 	select ARCH_HAS_DEBUG_WX		if STRICT_KERNEL_RWX

arch/s390/Kconfig

···
 	select ARCH_ENABLE_MEMORY_HOTPLUG if SPARSEMEM
 	select ARCH_ENABLE_MEMORY_HOTREMOVE
 	select ARCH_ENABLE_SPLIT_PMD_PTLOCK if PGTABLE_LEVELS > 2
+	select ARCH_HAS_CURRENT_STACK_POINTER
 	select ARCH_HAS_DEBUG_VM_PGTABLE
 	select ARCH_HAS_DEBUG_WX
 	select ARCH_HAS_DEVMEM_IS_ALLOWED

arch/sh/Kconfig

···
 	select ARCH_HAVE_CUSTOM_GPIO_H
 	select ARCH_HAVE_NMI_SAFE_CMPXCHG if (GUSA_RB || CPU_SH4A)
 	select ARCH_HAS_BINFMT_FLAT if !MMU
+	select ARCH_HAS_CURRENT_STACK_POINTER
 	select ARCH_HAS_GIGANTIC_PAGE
 	select ARCH_HAS_GCOV_PROFILE_ALL
 	select ARCH_HAS_PTE_SPECIAL

arch/x86/Kconfig

···
 	select ARCH_ENABLE_THP_MIGRATION if X86_64 && TRANSPARENT_HUGEPAGE
 	select ARCH_HAS_ACPI_TABLE_UPGRADE	if ACPI
 	select ARCH_HAS_CACHE_LINE_SIZE
+	select ARCH_HAS_CURRENT_STACK_POINTER
 	select ARCH_HAS_DEBUG_VIRTUAL
 	select ARCH_HAS_DEBUG_VM_PGTABLE	if !X86_PAE
 	select ARCH_HAS_DEVMEM_IS_ALLOWED

mm/Kconfig

···
 config ARCH_HAS_CACHE_LINE_SIZE
 	bool
 
+config ARCH_HAS_CURRENT_STACK_POINTER
+	bool
+	help
+	  In support of HARDENED_USERCOPY performing stack variable lifetime
+	  checking, an architecture-agnostic way to find the stack pointer
+	  is needed. Once an architecture defines an unsigned long global
+	  register alias named "current_stack_pointer", this config can be
+	  selected.
+
 config ARCH_HAS_PTE_DEVMAP
 	bool
 

mm/usercopy.c

···
  * Returns:
  *	NOT_STACK: not at all on the stack
  *	GOOD_FRAME: fully within a valid stack frame
- *	GOOD_STACK: fully on the stack (when can't do frame-checking)
+ *	GOOD_STACK: within the current stack (when can't frame-check exactly)
  *	BAD_STACK: error condition (invalid stack position or bad stack frame)
  */
 static noinline int check_stack_object(const void *obj, unsigned long len)
···
 	ret = arch_within_stack_frames(stack, stackend, obj, len);
 	if (ret)
 		return ret;
+
+	/* Finally, check stack depth if possible. */
+#ifdef CONFIG_ARCH_HAS_CURRENT_STACK_POINTER
+	if (IS_ENABLED(CONFIG_STACK_GROWSUP)) {
+		if ((void *)current_stack_pointer < obj + len)
+			return BAD_STACK;
+	} else {
+		if (obj < (void *)current_stack_pointer)
+			return BAD_STACK;
+	}
+#endif
 
 	return GOOD_STACK;
 }
···
 		 */
 		return;
 	default:
-		usercopy_abort("process stack", NULL, to_user, 0, n);
+		usercopy_abort("process stack", NULL, to_user,
+#ifdef CONFIG_ARCH_HAS_CURRENT_STACK_POINTER
+			IS_ENABLED(CONFIG_STACK_GROWSUP) ?
+				ptr - (void *)current_stack_pointer :
+				(void *)current_stack_pointer - ptr,
+#else
+			0,
+#endif
+			n);
 	}
 
 	/* Check for bad heap object. */