x86/entry: Add STACKLEAK erasing the kernel stack at the end of syscalls

The STACKLEAK feature (initially developed by PaX Team) has the following
benefits:

1. Reduces the information that can be revealed through kernel stack leak
   bugs. The idea of erasing the thread stack at the end of syscalls is
   similar to CONFIG_PAGE_POISONING and memzero_explicit() in kernel
   crypto, which all comply with FDP_RIP.2 (Full Residual Information
   Protection) of the Common Criteria standard.

2. Blocks some uninitialized stack variable attacks (e.g. CVE-2017-17712,
   CVE-2010-2963). That kind of bugs should be killed by improving C
   compilers in future, which might take a long time.

This commit introduces the code filling the used part of the kernel
stack with a poison value before returning to userspace. Full
STACKLEAK feature also contains the gcc plugin which comes in a
separate commit.

The STACKLEAK feature is ported from grsecurity/PaX. More information at:
  https://grsecurity.net/
  https://pax.grsecurity.net/

This code is modified from Brad Spengler/PaX Team's code in the last
public patch of grsecurity/PaX based on our understanding of the code.
Changes or omissions from the original code are ours and don't reflect
the original grsecurity/PaX code.

Performance impact:

Hardware: Intel Core i7-4770, 16 GB RAM

Test #1: building the Linux kernel on a single core
        0.91% slowdown

Test #2: hackbench -s 4096 -l 2000 -g 15 -f 25 -P
        4.2% slowdown

So the STACKLEAK description in Kconfig includes: "The tradeoff is the
performance impact: on a single CPU system kernel compilation sees a 1%
slowdown, other systems and workloads may vary and you are advised to
test this feature on your expected workload before deploying it".
Signed-off-by: NAlexander Popov <alex.popov@linux.com>
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NDave Hansen <dave.hansen@linux.intel.com>
Acked-by: NIngo Molnar <mingo@kernel.org>
Signed-off-by: NKees Cook <keescook@chromium.org>

Documentation/x86/x86_64/mm.txt

@@ -24,6 +24,7 @@ ffffffffa0000000 - fffffffffeffffff (1520 MB) module mapping space
 [fixmap start]   - ffffffffff5fffff kernel-internal fixmap range
 ffffffffff600000 - ffffffffff600fff (=4 kB) legacy vsyscall ABI
 ffffffffffe00000 - ffffffffffffffff (=2 MB) unused hole
+STACKLEAK_POISON value in this last hole: ffffffffffff4111
 
 Virtual memory map with 5 level page tables:
 
@@ -50,6 +51,7 @@ ffffffffa0000000 - fffffffffeffffff (1520 MB) module mapping space
 [fixmap start]   - ffffffffff5fffff kernel-internal fixmap range
 ffffffffff600000 - ffffffffff600fff (=4 kB) legacy vsyscall ABI
 ffffffffffe00000 - ffffffffffffffff (=2 MB) unused hole
+STACKLEAK_POISON value in this last hole: ffffffffffff4111
 
 Architecture defines a 64-bit virtual address. Implementations can support
 less. Currently supported are 48- and 57-bit virtual addresses. Bits 63

arch/Kconfig

@@ -419,6 +419,13 @@ config SECCOMP_FILTER
 
 	  See Documentation/userspace-api/seccomp_filter.rst for details.
 
+config HAVE_ARCH_STACKLEAK
+	bool
+	help
+	  An architecture should select this if it has the code which
+	  fills the used part of the kernel stack with the STACKLEAK_POISON
+	  value before returning from system calls.
+
 config HAVE_STACKPROTECTOR
 	bool
 	help

arch/x86/Kconfig

@@ -127,6 +127,7 @@ config X86
 	select HAVE_ARCH_PREL32_RELOCATIONS
 	select HAVE_ARCH_SECCOMP_FILTER
 	select HAVE_ARCH_THREAD_STRUCT_WHITELIST
+	select HAVE_ARCH_STACKLEAK
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_ARCH_TRANSPARENT_HUGEPAGE
 	select HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD if X86_64

arch/x86/entry/calling.h

@@ -329,8 +329,22 @@ For 32-bit we have the following conventions - kernel is built with
 
 #endif
 
+.macro STACKLEAK_ERASE_NOCLOBBER
+#ifdef CONFIG_GCC_PLUGIN_STACKLEAK
+	PUSH_AND_CLEAR_REGS
+	call stackleak_erase
+	POP_REGS
+#endif
+.endm
+
 #endif /* CONFIG_X86_64 */
 
+.macro STACKLEAK_ERASE
+#ifdef CONFIG_GCC_PLUGIN_STACKLEAK
+	call stackleak_erase
+#endif
+.endm
+
 /*
  * This does 'call enter_from_user_mode' unless we can avoid it based on
  * kernel config or using the static jump infrastructure.

arch/x86/entry/entry_32.S

@@ -46,6 +46,8 @@
 #include <asm/frame.h>
 #include <asm/nospec-branch.h>
 
+#include "calling.h"
+
 	.section .entry.text, "ax"
 
 /*
@@ -711,6 +713,7 @@ ENTRY(ret_from_fork)
 	/* When we fork, we trace the syscall return in the child, too. */
 	movl    %esp, %eax
 	call    syscall_return_slowpath
+	STACKLEAK_ERASE
 	jmp     restore_all
 
 	/* kernel thread */
@@ -885,6 +888,8 @@ ENTRY(entry_SYSENTER_32)
 	ALTERNATIVE "testl %eax, %eax; jz .Lsyscall_32_done", \
 		    "jmp .Lsyscall_32_done", X86_FEATURE_XENPV
 
+	STACKLEAK_ERASE
+
 /* Opportunistic SYSEXIT */
 	TRACE_IRQS_ON			/* User mode traces as IRQs on. */
 
@@ -996,6 +1001,8 @@ ENTRY(entry_INT80_32)
 	call	do_int80_syscall_32
 .Lsyscall_32_done:
 
+	STACKLEAK_ERASE
+
 restore_all:
 	TRACE_IRQS_IRET
 	SWITCH_TO_ENTRY_STACK

arch/x86/entry/entry_64.S

@@ -329,6 +329,8 @@ syscall_return_via_sysret:
 	 * We are on the trampoline stack.  All regs except RDI are live.
 	 * We can do future final exit work right here.
 	 */
+	STACKLEAK_ERASE_NOCLOBBER
+
 	SWITCH_TO_USER_CR3_STACK scratch_reg=%rdi
 
 	popq	%rdi
@@ -688,6 +690,7 @@ GLOBAL(swapgs_restore_regs_and_return_to_usermode)
 	 * We are on the trampoline stack.  All regs except RDI are live.
 	 * We can do future final exit work right here.
 	 */
+	STACKLEAK_ERASE_NOCLOBBER
 
 	SWITCH_TO_USER_CR3_STACK scratch_reg=%rdi
 

arch/x86/entry/entry_64_compat.S

@@ -261,6 +261,11 @@ GLOBAL(entry_SYSCALL_compat_after_hwframe)
 
 	/* Opportunistic SYSRET */
 sysret32_from_system_call:
+	/*
+	 * We are not going to return to userspace from the trampoline
+	 * stack. So let's erase the thread stack right now.
+	 */
+	STACKLEAK_ERASE
 	TRACE_IRQS_ON			/* User mode traces as IRQs on. */
 	movq	RBX(%rsp), %rbx		/* pt_regs->rbx */
 	movq	RBP(%rsp), %rbp		/* pt_regs->rbp */

include/linux/sched.h

@@ -1192,6 +1192,10 @@ struct task_struct {
 	void				*security;
 #endif
 
+#ifdef CONFIG_GCC_PLUGIN_STACKLEAK
+	unsigned long			lowest_stack;
+#endif
+
 	/*
 	 * New fields for task_struct should be added above here, so that
 	 * they are included in the randomized portion of task_struct.

include/linux/stackleak.h

+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _LINUX_STACKLEAK_H
+#define _LINUX_STACKLEAK_H
+
+#include <linux/sched.h>
+#include <linux/sched/task_stack.h>
+
+/*
+ * Check that the poison value points to the unused hole in the
+ * virtual memory map for your platform.
+ */
+#define STACKLEAK_POISON -0xBEEF
+#define STACKLEAK_SEARCH_DEPTH 128
+
+#ifdef CONFIG_GCC_PLUGIN_STACKLEAK
+#include <asm/stacktrace.h>
+
+static inline void stackleak_task_init(struct task_struct *t)
+{
+	t->lowest_stack = (unsigned long)end_of_stack(t) + sizeof(unsigned long);
+}
+#else /* !CONFIG_GCC_PLUGIN_STACKLEAK */
+static inline void stackleak_task_init(struct task_struct *t) { }
+#endif
+
+#endif

kernel/Makefile

@@ -117,6 +117,10 @@ obj-$(CONFIG_HAS_IOMEM) += iomem.o
 obj-$(CONFIG_ZONE_DEVICE) += memremap.o
 obj-$(CONFIG_RSEQ) += rseq.o
 
+obj-$(CONFIG_GCC_PLUGIN_STACKLEAK) += stackleak.o
+KASAN_SANITIZE_stackleak.o := n
+KCOV_INSTRUMENT_stackleak.o := n
+
 $(obj)/configs.o: $(obj)/config_data.h
 
 targets += config_data.gz

kernel/fork.c

@@ -91,6 +91,7 @@
 #include <linux/kcov.h>
 #include <linux/livepatch.h>
 #include <linux/thread_info.h>
+#include <linux/stackleak.h>
 
 #include <asm/pgtable.h>
 #include <asm/pgalloc.h>
@@ -1880,6 +1881,8 @@ static __latent_entropy struct task_struct *copy_process(
 	if (retval)
 		goto bad_fork_cleanup_io;
 
+	stackleak_task_init(p);
+
 	if (pid != &init_struct_pid) {
 		pid = alloc_pid(p->nsproxy->pid_ns_for_children);
 		if (IS_ERR(pid)) {

kernel/stackleak.c

+// SPDX-License-Identifier: GPL-2.0
+/*
+ * This code fills the used part of the kernel stack with a poison value
+ * before returning to userspace. It's part of the STACKLEAK feature
+ * ported from grsecurity/PaX.
+ *
+ * Author: Alexander Popov <alex.popov@linux.com>
+ *
+ * STACKLEAK reduces the information which kernel stack leak bugs can
+ * reveal and blocks some uninitialized stack variable attacks.
+ */
+
+#include <linux/stackleak.h>
+
+asmlinkage void stackleak_erase(void)
+{
+	/* It would be nice not to have 'kstack_ptr' and 'boundary' on stack */
+	unsigned long kstack_ptr = current->lowest_stack;
+	unsigned long boundary = (unsigned long)end_of_stack(current);
+	unsigned int poison_count = 0;
+	const unsigned int depth = STACKLEAK_SEARCH_DEPTH / sizeof(unsigned long);
+
+	/* Check that 'lowest_stack' value is sane */
+	if (unlikely(kstack_ptr - boundary >= THREAD_SIZE))
+		kstack_ptr = boundary;
+
+	/* Search for the poison value in the kernel stack */
+	while (kstack_ptr > boundary && poison_count <= depth) {
+		if (*(unsigned long *)kstack_ptr == STACKLEAK_POISON)
+			poison_count++;
+		else
+			poison_count = 0;
+
+		kstack_ptr -= sizeof(unsigned long);
+	}
+
+	/*
+	 * One 'long int' at the bottom of the thread stack is reserved and
+	 * should not be poisoned (see CONFIG_SCHED_STACK_END_CHECK=y).
+	 */
+	if (kstack_ptr == boundary)
+		kstack_ptr += sizeof(unsigned long);
+
+	/*
+	 * Now write the poison value to the kernel stack. Start from
+	 * 'kstack_ptr' and move up till the new 'boundary'. We assume that
+	 * the stack pointer doesn't change when we write poison.
+	 */
+	if (on_thread_stack())
+		boundary = current_stack_pointer;
+	else
+		boundary = current_top_of_stack();
+
+	while (kstack_ptr < boundary) {
+		*(unsigned long *)kstack_ptr = STACKLEAK_POISON;
+		kstack_ptr += sizeof(unsigned long);
+	}
+
+	/* Reset the 'lowest_stack' value for the next syscall */
+	current->lowest_stack = current_top_of_stack() - THREAD_SIZE/64;
+}
+

scripts/gcc-plugins/Kconfig

@@ -139,4 +139,23 @@ config GCC_PLUGIN_RANDSTRUCT_PERFORMANCE
 	  in structures.  This reduces the performance hit of RANDSTRUCT
 	  at the cost of weakened randomization.
 
+config GCC_PLUGIN_STACKLEAK
+	bool "Erase the kernel stack before returning from syscalls"
+	depends on GCC_PLUGINS
+	depends on HAVE_ARCH_STACKLEAK
+	help
+	  This option makes the kernel erase the kernel stack before
+	  returning from system calls. That reduces the information which
+	  kernel stack leak bugs can reveal and blocks some uninitialized
+	  stack variable attacks.
+
+	  The tradeoff is the performance impact: on a single CPU system kernel
+	  compilation sees a 1% slowdown, other systems and workloads may vary
+	  and you are advised to test this feature on your expected workload
+	  before deploying it.
+
+	  This plugin was ported from grsecurity/PaX. More information at:
+	   * https://grsecurity.net/
+	   * https://pax.grsecurity.net/
+
 endif