mailing list of musl libc
 help / color / mirror / code / Atom feed
* [musl] [PATCH v2] x86: Implement arch_prctl(ARCH_VSYSCALL_CONTROL) to disable vsyscall
@ 2021-12-16 18:50 Florian Weimer
  2021-12-27 16:49 ` [musl] " Andrei Vagin
  0 siblings, 1 reply; 3+ messages in thread
From: Florian Weimer @ 2021-12-16 18:50 UTC (permalink / raw)
  To: Andy Lutomirski
  Cc: linux-arch, Linux API, linux-x86_64, kernel-hardening, linux-mm,
	the arch/x86 maintainers, musl, libc-alpha, linux-kernel,
	Dave Hansen, Kees Cook

Distributions struggle with changing the default for vsyscall
emulation because it is a clear break of userspace ABI, something
that should not happen.

The legacy vsyscall interface is supposed to be used by libcs only,
not by applications.  This commit adds a new arch_prctl request,
ARCH_VSYSCALL_CONTROL, with one argument.  If the argument is 0,
executing vsyscalls will cause the process to terminate.  Argument 1
turns vsyscall back on (this is mostly for a largely theoretical
CRIU use case).

Newer libcs can use a zero ARCH_VSYSCALL_CONTROL at startup to disable
vsyscall for the process.  Legacy libcs do not perform this call, so
vsyscall remains enabled for them.  This approach should achieves
backwards compatibility (perfect compatibility if the assumption that
only libcs use vsyscall is accurate), and it provides full hardening
for new binaries.

The chosen value of ARCH_VSYSCALL_CONTROL should avoid conflicts
with other x86-64 arch_prctl requests.  The fact that with
vsyscall=emulate, reading the vsyscall region is still possible
even after a zero ARCH_VSYSCALL_CONTROL is considered limitation
in the current implementation and may change in a future kernel
version.

Future arch_prctls requests commonly used at process startup can imply
ARCH_VSYSCALL_CONTROL with a zero argument, so that a separate system
call for disabling vsyscall is avoided.

Signed-off-by: Florian Weimer <fweimer@redhat.com>

---
v2: ARCH_VSYSCALL_CONTROL instead of ARCH_VSYSCALL_LOCKOUT.  New tests
    for the toggle behavior.  Implement hiding [vsyscall] in
    /proc/PID/maps and test it.  Various other test fixes cleanups
    (e.g., fixed missing second argument to gettimeofday).

 arch/x86/entry/vsyscall/vsyscall_64.c          |  10 +-
 arch/x86/include/asm/mmu.h                     |   6 +
 arch/x86/include/uapi/asm/prctl.h              |   2 +
 arch/x86/kernel/process_64.c                   |   7 +
 tools/arch/x86/include/uapi/asm/prctl.h        |   2 +
 tools/testing/selftests/x86/Makefile           |  13 +-
 tools/testing/selftests/x86/vsyscall_control.c | 891 +++++++++++++++++++++++++
 7 files changed, 927 insertions(+), 4 deletions(-)

diff --git a/arch/x86/entry/vsyscall/vsyscall_64.c b/arch/x86/entry/vsyscall/vsyscall_64.c
index fd2ee9408e91..8eb3bcf2cedf 100644
--- a/arch/x86/entry/vsyscall/vsyscall_64.c
+++ b/arch/x86/entry/vsyscall/vsyscall_64.c
@@ -174,6 +174,12 @@ bool emulate_vsyscall(unsigned long error_code,
 
 	tsk = current;
 
+	if (tsk->mm->context.vsyscall_disabled) {
+		warn_bad_vsyscall(KERN_WARNING, regs,
+				  "vsyscall after lockout (exploit attempt?)");
+		goto sigsegv;
+	}
+
 	/*
 	 * Check for access_ok violations and find the syscall nr.
 	 *
@@ -316,8 +322,10 @@ static struct vm_area_struct gate_vma __ro_after_init = {
 
 struct vm_area_struct *get_gate_vma(struct mm_struct *mm)
 {
+	if (!mm || mm->context.vsyscall_disabled)
+		return NULL;
 #ifdef CONFIG_COMPAT
-	if (!mm || !(mm->context.flags & MM_CONTEXT_HAS_VSYSCALL))
+	if (!(mm->context.flags & MM_CONTEXT_HAS_VSYSCALL))
 		return NULL;
 #endif
 	if (vsyscall_mode == NONE)
diff --git a/arch/x86/include/asm/mmu.h b/arch/x86/include/asm/mmu.h
index 5d7494631ea9..3934d6907910 100644
--- a/arch/x86/include/asm/mmu.h
+++ b/arch/x86/include/asm/mmu.h
@@ -41,6 +41,12 @@ typedef struct {
 #ifdef CONFIG_X86_64
 	unsigned short flags;
 #endif
+#ifdef CONFIG_X86_VSYSCALL_EMULATION
+	/*
+	 * Changed by arch_prctl(ARCH_VSYSCALL_CONTROL).
+	 */
+	bool vsyscall_disabled;
+#endif
 
 	struct mutex lock;
 	void __user *vdso;			/* vdso base address */
diff --git a/arch/x86/include/uapi/asm/prctl.h b/arch/x86/include/uapi/asm/prctl.h
index 754a07856817..aad0bcfbf49f 100644
--- a/arch/x86/include/uapi/asm/prctl.h
+++ b/arch/x86/include/uapi/asm/prctl.h
@@ -18,4 +18,6 @@
 #define ARCH_MAP_VDSO_32	0x2002
 #define ARCH_MAP_VDSO_64	0x2003
 
+#define ARCH_VSYSCALL_CONTROL	0x5001
+
 #endif /* _ASM_X86_PRCTL_H */
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 3402edec236c..834bad068211 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -816,6 +816,13 @@ long do_arch_prctl_64(struct task_struct *task, int option, unsigned long arg2)
 		ret = put_user(base, (unsigned long __user *)arg2);
 		break;
 	}
+#ifdef CONFIG_X86_VSYSCALL_EMULATION
+	case ARCH_VSYSCALL_CONTROL:
+		if (unlikely(arg2 > 1))
+			return -EINVAL;
+		current->mm->context.vsyscall_disabled = !arg2;
+		break;
+#endif
 
 #ifdef CONFIG_CHECKPOINT_RESTORE
 # ifdef CONFIG_X86_X32_ABI
diff --git a/tools/arch/x86/include/uapi/asm/prctl.h b/tools/arch/x86/include/uapi/asm/prctl.h
index 754a07856817..aad0bcfbf49f 100644
--- a/tools/arch/x86/include/uapi/asm/prctl.h
+++ b/tools/arch/x86/include/uapi/asm/prctl.h
@@ -18,4 +18,6 @@
 #define ARCH_MAP_VDSO_32	0x2002
 #define ARCH_MAP_VDSO_64	0x2003
 
+#define ARCH_VSYSCALL_CONTROL	0x5001
+
 #endif /* _ASM_X86_PRCTL_H */
diff --git a/tools/testing/selftests/x86/Makefile b/tools/testing/selftests/x86/Makefile
index 8a1f62ab3c8e..2a7c91ee68e0 100644
--- a/tools/testing/selftests/x86/Makefile
+++ b/tools/testing/selftests/x86/Makefile
@@ -18,7 +18,7 @@ TARGETS_C_32BIT_ONLY := entry_from_vm86 test_syscall_vdso unwind_vdso \
 			test_FCMOV test_FCOMI test_FISTTP \
 			vdso_restorer
 TARGETS_C_64BIT_ONLY := fsgsbase sysret_rip syscall_numbering \
-			corrupt_xstate_header amx
+			corrupt_xstate_header amx vsyscall_control
 # Some selftests require 32bit support enabled also on 64bit systems
 TARGETS_C_32BIT_NEEDED := ldt_gdt ptrace_syscall
 
@@ -72,10 +72,12 @@ all_64: $(BINARIES_64)
 EXTRA_CLEAN := $(BINARIES_32) $(BINARIES_64)
 
 $(BINARIES_32): $(OUTPUT)/%_32: %.c helpers.h
-	$(CC) -m32 -o $@ $(CFLAGS) $(EXTRA_CFLAGS) $^ -lrt -ldl -lm
+	$(CC) -m32 -o $@ $(CFLAGS) $(EXTRA_CFLAGS) $^ \
+		$(or $(LIBS), -lrt -ldl -lm)
 
 $(BINARIES_64): $(OUTPUT)/%_64: %.c helpers.h
-	$(CC) -m64 -o $@ $(CFLAGS) $(EXTRA_CFLAGS) $^ -lrt -ldl
+	$(CC) -m64 -o $@ $(CFLAGS) $(EXTRA_CFLAGS) $^ \
+		$(or $(LIBS), -lrt -ldl -lm)
 
 # x86_64 users should be encouraged to install 32-bit libraries
 ifeq ($(CAN_BUILD_I386)$(CAN_BUILD_X86_64),01)
@@ -105,3 +107,8 @@ $(OUTPUT)/test_syscall_vdso_32: thunks_32.S
 # state.
 $(OUTPUT)/check_initial_reg_state_32: CFLAGS += -Wl,-ereal_start -static
 $(OUTPUT)/check_initial_reg_state_64: CFLAGS += -Wl,-ereal_start -static
+
+# This test does not link against anything (neither libc nor libgcc).
+$(OUTPUT)/vsyscall_control_64: \
+	LIBS := -Wl,-no-pie -static -nostdlib -nostartfiles
+	CFLAGS += -fno-pie -fno-stack-protector -fno-builtin -ffreestanding
diff --git a/tools/testing/selftests/x86/vsyscall_control.c b/tools/testing/selftests/x86/vsyscall_control.c
new file mode 100644
index 000000000000..ee966f936c89
--- /dev/null
+++ b/tools/testing/selftests/x86/vsyscall_control.c
@@ -0,0 +1,891 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * vsyscall_lockout.c - check that disabling vsyscall works
+ * Copyright (C) 2021 Red Hat, Inc.
+ *
+ * This test requires vsyscall=xonly or vsyscall=emulate.  With
+ * vsyscall=emulate, ARCH_VSYSCALL_CONTROL cannot turn off vsyscall
+ * completely (reads still work), but this is not tested here.
+ */
+
+#include <stddef.h>
+
+#include <asm/prctl.h>
+#include <asm/vsyscall.h>
+#include <linux/errno.h>
+#include <linux/fcntl.h>
+#include <linux/signal.h>
+#include <linux/time.h>
+#include <linux/types.h>
+#include <linux/unistd.h>
+
+#ifndef ARCH_VSYSCALL_CONTROL
+#define ARCH_VSYSCALL_CONTROL	0x5001
+#elif ARCH_VSYSCALL_CONTROL != 0x5001
+#error wrong vlaue for ARCH_VSYSCALL_CONTROL
+#endif
+
+
+static inline long syscall0(int nr)
+{
+	unsigned long result;
+
+	__asm__ volatile ("syscall"
+			  : "=a" (result)
+			  : "0" (nr)
+			  : "memory", "cc", "r11", "cx");
+	return result;
+}
+
+static inline long syscall1(int nr, long arg0)
+{
+	register long rdi asm("rdi") = arg0;
+	unsigned long result;
+
+	__asm__ volatile ("syscall"
+			  : "=a" (result)
+			  : "0" (nr), "r" (rdi)
+			  : "memory", "cc", "r11", "cx");
+	return result;
+}
+
+static inline long syscall2(int nr, long arg0, long arg1)
+{
+	register long rdi asm("rdi") = arg0;
+	register long rsi asm("rsi") = arg1;
+	unsigned long result;
+
+	__asm__ volatile ("syscall"
+			  : "=a" (result)
+			  : "0" (nr), "r" (rdi), "r" (rsi)
+			  : "memory", "cc", "r11", "cx");
+	return result;
+}
+
+static inline long syscall3(int nr, long arg0, long arg1, long arg2)
+{
+	register long rdi asm("rdi") = arg0;
+	register long rsi asm("rsi") = arg1;
+	register long rdx asm("rdx") = arg2;
+	unsigned long result;
+
+	__asm__ volatile ("syscall"
+			  : "=a" (result)
+			  : "0" (nr), "r" (rdi), "r" (rsi), "r" (rdx)
+			  : "memory", "cc", "r11", "cx");
+	return result;
+}
+
+static inline long syscall4(int nr, long arg0, long arg1, long arg2, long arg3)
+{
+	register long rdi asm("rdi") = arg0;
+	register long rsi asm("rsi") = arg1;
+	register long rdx asm("rdx") = arg2;
+	register long r10 asm("r10") = arg3;
+	unsigned long result;
+
+	__asm__ volatile ("syscall"
+			  : "=a" (result)
+			  : "0" (nr), "r" (rdi), "r" (rsi), "r" (rdx),
+			    "r" (r10)
+			  : "memory", "cc", "r11", "cx");
+	return result;
+}
+
+static inline long syscall5(int nr, long arg0, long arg1, long arg2, long arg3,
+			    long arg4)
+{
+	register long rdi asm("rdi") = arg0;
+	register long rsi asm("rsi") = arg1;
+	register long rdx asm("rdx") = arg2;
+	register long r10 asm("r10") = arg3;
+	register long r8 asm("r8") = arg4;
+	unsigned long result;
+
+	__asm__ volatile ("syscall"
+			  : "=a" (result)
+			  : "0" (nr), "r" (rdi), "r" (rsi), "r" (rdx),
+			    "r" (r10), "r" (r8)
+			  : "memory", "cc", "r11", "cx");
+	return result;
+}
+
+static inline long vsyscall2(long addr, long arg0, long arg1)
+{
+	register long rdi asm("rdi") = arg0;
+	register long rsi asm("rsi") = arg1;
+	unsigned long result;
+
+	__asm__ volatile ("callq *%%rax"
+			  : "=a" (result)
+			  : "0" (addr), "r" (rdi), "r" (rsi)
+			  : "memory", "cc", "r11", "cx");
+	return result;
+}
+
+static void __attribute__ ((noreturn)) sys_exit(int status)
+{
+	syscall1(__NR_exit, status);
+	__builtin_unreachable();
+}
+
+static int sys_access(const char *pathname, int mode)
+{
+	return syscall2(__NR_access, (long) pathname, mode);
+}
+
+static int sys_mkdir(const char *pathname, __kernel_mode_t mode)
+{
+	return syscall2(__NR_mkdir, (long) pathname, mode);
+}
+
+static int sys_open(const char *pathname, int flags, __kernel_mode_t mode)
+{
+	return syscall3(__NR_open, (long) pathname, flags, mode);
+}
+
+static long sys_read(int fd, void *buffer, size_t length)
+{
+	return syscall3(__NR_read, fd, (long) buffer, length);
+}
+
+static long sys_write(int fd, const void *buffer, size_t length)
+{
+	return syscall3(__NR_write, fd, (long) buffer, length);
+}
+
+static int sys_mount(const char *source, const char *pathname,
+		     const char *fstype, unsigned long flags,
+		     const void *data)
+{
+	return syscall5(__NR_mount, (long) source, (long) pathname,
+			(long) fstype, flags, (long) data);
+}
+
+static void sigabrt(void)
+{
+	syscall2(__NR_kill, syscall0(__NR_getpid), SIGABRT);
+}
+
+/*
+ * String buffers.
+ */
+
+struct buffer {
+	char *position;
+	char *limit;
+};
+
+static void buffer_init(struct buffer *b, char *start, size_t length)
+{
+	b->position = start;
+	b->limit = start + length;
+}
+
+static void buffer_append(struct buffer *b, char ch)
+{
+	if (b->position >= b->limit)
+		sigabrt();
+	*b->position = ch;
+	++b->position;
+}
+
+static void buffer_append_string(struct buffer *b, const char *p)
+{
+	while (*p) {
+		buffer_append(b, *p);
+		++p;
+	}
+}
+
+static void buffer_append_dec_1(struct buffer *b, unsigned long val)
+{
+	if (val != 0) {
+		buffer_append_dec_1(b, val / 10);
+		buffer_append(b, '0' + (val % 10));
+	}
+}
+
+static void buffer_append_dec(struct buffer *b, unsigned long val)
+{
+	if (val == 0) {
+		buffer_append(b, '0');
+		return;
+	}
+	buffer_append_dec_1(b, val);
+}
+
+/*
+ * Output to standard output.
+ */
+
+static void print_char(char byte)
+{
+	if (sys_write(1, &byte, 1) < 0)
+		sigabrt();
+}
+
+static void print_string(const char *p)
+{
+	while (*p) {
+		print_char(*p);
+		++p;
+	}
+}
+
+static void print_dec_1(unsigned long val)
+{
+	if (val != 0) {
+		print_dec_1(val / 10);
+		print_char('0' + (val % 10));
+	}
+}
+
+static void print_dec(unsigned long val)
+{
+	if (val == 0)
+		print_char('0');
+	else
+		print_dec_1(val);
+}
+
+static void print_signed_dec(long val)
+{
+	if (val < 0) {
+		print_char('-');
+		print_dec(-(unsigned long)val);
+	} else
+		print_dec(val);
+}
+
+static void print_message(unsigned int lineno, const char *tag, const char *p)
+{
+	print_string(__FILE__);
+	print_char(':');
+	print_dec(lineno);
+	print_char(':');
+	print_char(' ');
+	print_string(tag);
+	print_char(':');
+	print_char(' ');
+	print_string(p);
+}
+
+static void print_info(unsigned int lineno, const char *p)
+{
+	print_message(lineno, "info", p);
+}
+
+static void print_error(unsigned int lineno, const char *p)
+{
+	print_message(lineno, "ERROR", p);
+}
+
+static void print_failure(int lineno, const char *label, long ret)
+{
+	print_error(lineno, label);
+	print_string(" failed: ");
+	print_signed_dec(ret);
+	print_char('\n');
+}
+
+static void print_time(int lineno, const char *label, struct timeval tv)
+{
+	print_info(lineno, label);
+	print_string(": ");
+	print_dec(tv.tv_sec);
+	print_char(' ');
+	print_dec(tv.tv_usec);
+	print_char('\n');
+}
+
+/*
+ * Process-failing (v)syscall wrappers.
+ */
+
+static void xgettimeofday(struct timeval *tv)
+{
+	long ret = syscall2(__NR_gettimeofday, (long) tv, 0);
+
+	if (ret != 0) {
+		print_failure(__LINE__, "gettimeofday", ret);
+		sigabrt();
+	}
+}
+
+static void xvgettimeofday(struct timeval *tv)
+{
+	long ret = vsyscall2(VSYSCALL_ADDR, (long) tv, 0);
+
+	if (ret) {
+		print_failure(__LINE__, "vgettimeofday", ret);
+		sigabrt();
+	}
+}
+
+static int sys_arch_prctl(int code, unsigned long addr)
+{
+	return syscall2(__NR_arch_prctl, code, addr);
+}
+
+static void xclose(int fd)
+{
+	long ret = syscall1(__NR_close, fd);
+
+	if (ret < 0) {
+		print_failure(__LINE__, "close", ret);
+		sigabrt();
+	}
+}
+
+static void xwrite_byte(int fd, char b)
+{
+	long ret = sys_write(fd, &b, 1);
+
+	if (ret != 1) {
+		print_failure(__LINE__, "write", ret);
+		sigabrt();
+	}
+}
+
+static int xread_byte(int fd)
+{
+	char b;
+	long ret = sys_read(fd, &b, 1);
+
+	if (ret != 1) {
+		print_failure(__LINE__, "read", ret);
+		sigabrt();
+	}
+	return b;
+}
+
+static void xpipe(int fds[static 2])
+{
+	long ret = syscall2(__NR_pipe2, (long) fds, O_CLOEXEC);
+
+	if (ret != 0) {
+		print_failure(__LINE__, "pipe2", ret);
+		sigabrt();
+	}
+}
+
+static __kernel_pid_t xfork(void)
+{
+	long ret = syscall0(__NR_fork);
+
+	if (ret < 0) {
+		print_failure(__LINE__, "fork", ret);
+		sigabrt();
+	}
+	return ret;
+}
+
+static void xexecve(const char *pathname, char **argv, char **envp)
+{
+	long ret;
+
+	ret = syscall3(__NR_execve, (long) pathname, (long) argv, (long) envp);
+	print_failure(__LINE__, "execve", ret);
+	sigabrt();
+}
+
+static __kernel_pid_t xwaitpid(__kernel_pid_t pid, int *status, int options)
+{
+	long ret = syscall4(__NR_wait4, pid, (long) status, options, 0);
+
+	if (ret < 0) {
+		print_failure(__LINE__, "wait4", ret);
+		sigabrt();
+	}
+	return ret;
+}
+
+/*
+ * Various helpers.
+ */
+
+static void vsyscall_disable(void)
+{
+	long ret = sys_arch_prctl(ARCH_VSYSCALL_CONTROL, 0);
+
+	if (ret != 0)
+		print_failure(__LINE__, "arch_prctl(ARCH_VSYSCALL_CONTROL, 0)", ret);
+}
+
+static void vsyscall_enable(void)
+{
+	long ret = sys_arch_prctl(ARCH_VSYSCALL_CONTROL, 1);
+
+	if (ret != 0)
+		print_failure(__LINE__, "arch_prctl(ARCH_VSYSCALL_CONTROL, 1)", ret);
+}
+
+static long difftime(struct timeval first, struct timeval second)
+{
+	return second.tv_usec - first.tv_usec +
+		(second.tv_sec - first.tv_sec) * 1000 * 1000;
+}
+
+static void ensure_proc_is_mounted(int *status)
+{
+	int ret;
+
+	if (sys_access("/proc/version", 0) == 0)
+		return;
+
+	ret = sys_mkdir("/proc", 0555);
+	if (ret == EEXIST)
+		return;
+	if (ret != 0) {
+		print_failure(__LINE__, "could not create /proc", ret);
+		*status = 1;
+		return;
+	}
+	ret = sys_mount("none", "/proc", "proc", 0, NULL);
+	if (ret != 0) {
+		print_failure(__LINE__, "could not mount /proc", ret);
+		*status = 1;
+		return;
+	}
+	if (sys_access("/proc/version", 0) != 0) {
+		print_error(__LINE__, "no /proc/version after mounting /proc");
+		*status = 1;
+		return;
+	}
+}
+
+/*
+ * Individual subtest functions.
+ */
+
+static void check_time(int *status)
+{
+	struct timeval initial_time = { -1, -1 };
+	struct timeval vsyscall_time = { -1, -1 };
+	struct timeval final_time = { -1, -1 };
+	long vsyscall_diff, final_diff;
+
+	xgettimeofday(&initial_time);
+	xvgettimeofday(&vsyscall_time);
+	xgettimeofday(&final_time);
+	vsyscall_diff = difftime(initial_time, vsyscall_time);
+	final_diff = difftime(vsyscall_time, final_time);
+
+	print_time(__LINE__, "initial gettimeofday", initial_time);
+	print_time(__LINE__, "vsyscall gettimeofday", vsyscall_time);
+	print_time(__LINE__, "final gettimeofday", final_time);
+
+	if (initial_time.tv_sec < 0 || initial_time.tv_usec < 0 ||
+	    vsyscall_time.tv_sec < 0 || vsyscall_time.tv_usec < 0 ||
+	    final_time.tv_sec < 0 || final_time.tv_usec < 0) {
+		print_error(__LINE__, "negative time\n");
+		*status = 1;
+	}
+
+	print_info(__LINE__, "differences: ");
+	print_signed_dec(vsyscall_diff);
+	print_char(' ');
+	print_signed_dec(final_diff);
+	print_char('\n');
+
+	if (vsyscall_diff < 0 || final_diff < 0) {
+		/*
+		 * This may produce false positives if there is an active NTP.
+		 */
+		print_error(__LINE__, "time went backwards\n");
+		*status = 1;
+	}
+}
+
+static void check_lockout_after_fork(int *status, int twice)
+{
+	__kernel_pid_t pid;
+	struct timeval vsyscall_time;
+	int wstatus;
+
+	if (twice) {
+		__kernel_pid_t pid_outer;
+
+		print_info(__LINE__, "checking that lockout is inherited by fork\n");
+
+		pid_outer = xfork();
+		if (pid_outer == 0) {
+			vsyscall_disable();
+			/*
+			 * Logic for the subprocess follows below.
+			 */
+		} else {
+			xwaitpid(pid_outer, &wstatus, 0);
+			if (wstatus != 0) {
+				print_error(__LINE__, "unexpected exit status: ");
+				print_signed_dec(wstatus);
+				print_char('\n');
+				*status = 1;
+			}
+			return;
+		}
+	} else
+		print_info(__LINE__, "checking that lockout works after one fork\n");
+
+	pid = xfork();
+	if (pid == 0) {
+		if (!twice)
+			vsyscall_disable();
+		/*
+		 * This should trigger a fault.
+		 */
+		xvgettimeofday(&vsyscall_time);
+		sys_exit(0);
+	}
+	xwaitpid(pid, &wstatus, 0);
+	switch (wstatus) {
+	case 0:
+		print_error(__LINE__, "no crash after lockout\n");
+		*status = 1;
+		break;
+	case 0x0100:
+		*status = 1;
+		break;
+	case SIGSEGV:
+		print_info(__LINE__, "termination after lockout\n");
+		break;
+	default:
+		print_error(__LINE__, "unexpected exit status: ");
+		print_signed_dec(wstatus);
+		print_char('\n');
+		*status = 1;
+	}
+
+	if (twice)
+		sys_exit(*status);
+
+	/*
+	 * Status in the parent process should be unaffected.
+	 */
+	xvgettimeofday(&vsyscall_time);
+}
+
+static void check_no_lockout_after_enable(int *status)
+{
+	__kernel_pid_t pid;
+	struct timeval vsyscall_time;
+	int wstatus;
+
+	print_info(__LINE__, "checking that vsyscall can be re-enabled\n");
+
+	pid = xfork();
+	if (pid == 0) {
+		vsyscall_disable();
+		vsyscall_enable();
+		xvgettimeofday(&vsyscall_time);
+		print_time(__LINE__, "vsyscall time after re-enable", vsyscall_time);
+		sys_exit(0);
+	}
+
+	xwaitpid(pid, &wstatus, 0);
+	if (wstatus != 0) {
+		print_error(__LINE__, "unexpected exit status: ");
+		print_signed_dec(wstatus);
+		print_char('\n');
+		*status = 1;
+	}
+}
+
+static void check_no_lockout_after_execve(char **argv, int *status)
+{
+	__kernel_pid_t pid;
+	int wstatus;
+
+	print_info(__LINE__, "checking that lockout is not inherited by execve\n");
+	pid = xfork();
+	if (pid == 0) {
+		struct timeval vsyscall_time;
+		char *new_argv[3];
+
+		xvgettimeofday(&vsyscall_time);
+		vsyscall_disable();
+
+		/*
+		 * Re-exec the second stage.  See main_2 below.
+		 */
+		new_argv[0] = argv[0];
+		new_argv[1] = "2";
+		new_argv[2] = NULL;
+		xexecve(argv[0], new_argv, new_argv + 2);
+	}
+
+	xwaitpid(pid, &wstatus, 0);
+	if (wstatus != 0) {
+		print_error(__LINE__, "unexpected exit status: ");
+		print_signed_dec(wstatus);
+		print_char('\n');
+		*status = 1;
+	}
+}
+
+static int has_vsyscall_map(int pid, int *status)
+{
+	int result = 0;
+	char maps_path[50];
+	int fd;
+
+	/*
+	 * Construct /proc/PID/maps path.
+	 */
+	{
+		struct buffer b;
+
+		buffer_init(&b, maps_path, sizeof(maps_path));
+		buffer_append_string(&b, "/proc/");
+		if (pid == 0)
+			buffer_append_string(&b, "self");
+		else
+			buffer_append_dec(&b, pid);
+		buffer_append_string(&b, "/maps");
+		buffer_append(&b, 0);
+	}
+
+	fd = sys_open(maps_path, O_RDONLY, 0);
+	if (fd < 0) {
+		print_error(__LINE__, "maps file ");
+		print_string(maps_path);
+		print_string(": ");
+		print_signed_dec(fd);
+		print_char('\n');
+		*status = 1;
+	}
+
+	/*
+	 * Search for "[vsyscall]\n".
+	 */
+	{
+		char buf[4096];
+		long ret;
+
+		for (size_t i = 0; i < sizeof(buf); ++i)
+			buf[i] = 0;
+		ret = sys_read(fd, buf, sizeof(buf));
+
+		if (ret < 0 || ret >= sizeof(buf))
+			print_failure(__LINE__, "read", ret);
+		else {
+			char *bracket = buf;
+
+			while (1) {
+				while (*bracket && *bracket != '[')
+					++bracket;
+				if (*bracket != '[')
+					/*
+					 * End of file has been reached.
+					 */
+					break;
+				++bracket;
+				if (bracket[0] == 'v'
+				    && bracket[1] == 's'
+				    && bracket[2] == 'y'
+				    && bracket[3] == 's'
+				    && bracket[4] == 'c'
+				    && bracket[5] == 'a'
+				    && bracket[6] == 'l'
+				    && bracket[7] == 'l'
+				    && bracket[8] == ']'
+				    && bracket[9] == '\n') {
+					result = 1;
+					break;
+				}
+			}
+		}
+	}
+
+	xclose(fd);
+	return result;
+}
+
+static void check_vsyscall_in_self_maps(int *status)
+{
+	__kernel_pid_t pid;
+	int wstatus;
+
+	print_info(__LINE__, "checking [vsyscall] in /proc/self/maps\n");
+
+	pid = xfork();
+	if (pid == 0) {
+		if (!has_vsyscall_map(0, status)) {
+			print_error(__LINE__, "[vsyscall] missing\n");
+			*status = 1;
+		}
+		vsyscall_disable();
+		if (has_vsyscall_map(0, status)) {
+			print_error(__LINE__, "[vsyscall] present after disable\n");
+			*status = 1;
+		}
+		vsyscall_enable();
+		if (!has_vsyscall_map(0, status)) {
+			print_error(__LINE__, "[vsyscall] missing after enable\n");
+			*status = 1;
+		}
+		sys_exit(*status);
+	}
+
+	xwaitpid(pid, &wstatus, 0);
+	if (wstatus != 0) {
+		print_error(__LINE__, "unexpected exit status: ");
+		print_signed_dec(wstatus);
+		print_char('\n');
+		*status = 1;
+	}
+}
+
+static void check_vsyscall_maps_for_subprocess(int *status)
+{
+	__kernel_pid_t pid;
+	int wstatus;
+	int outer_to_inner[2];
+	int inner_to_outer[2];
+
+	/*
+	 * Create pipes used to synchronize the two processes.
+	 */
+	xpipe(outer_to_inner);
+	xpipe(inner_to_outer);
+
+	print_info(__LINE__, "checking [vsyscall] in /proc/PID/maps\n");
+
+	pid = xfork();
+	if (pid == 0) {
+		xclose(outer_to_inner[1]);
+		xclose(inner_to_outer[0]);
+
+		xread_byte(outer_to_inner[0]);
+		vsyscall_disable();
+		xwrite_byte(inner_to_outer[1], 101);
+
+		xread_byte(outer_to_inner[0]);
+		vsyscall_disable();
+		xwrite_byte(inner_to_outer[1], 102);
+
+		sys_exit(0);
+	}
+
+	xclose(outer_to_inner[0]);
+	xclose(inner_to_outer[1]);
+
+	if (!has_vsyscall_map(pid, status)) {
+		print_error(__LINE__, "subprocess starts without [vsyscall]");
+		*status = 1;
+	}
+
+	xwrite_byte(outer_to_inner[1], 1);
+	xread_byte(inner_to_outer[0]);
+
+	if (has_vsyscall_map(pid, status)) {
+		print_error(__LINE__, "subprocess has [vsyscall] after disable");
+		*status = 1;
+	}
+
+	xwrite_byte(outer_to_inner[1], 2);
+	xread_byte(inner_to_outer[0]);
+
+	if (has_vsyscall_map(pid, status)) {
+		print_error(__LINE__, "subprocess lacks [vsyscall] after enable");
+		*status = 1;
+	}
+
+	xclose(outer_to_inner[1]);
+	xclose(inner_to_outer[0]);
+
+	xwaitpid(pid, &wstatus, 0);
+	if (wstatus != 0) {
+		print_error(__LINE__, "unexpected exit status: ");
+		print_signed_dec(wstatus);
+		print_char('\n');
+		*status = 1;
+	}
+}
+
+static void check_einval(int *status)
+{
+	long ret;
+
+	ret = sys_arch_prctl(ARCH_VSYSCALL_CONTROL, 2);
+	if (ret != -EINVAL) {
+		print_error(__LINE__, "arch_prctl(ARCH_VSYSCALL_CONTROL, 2) returned ");
+		print_signed_dec(ret);
+		print_char('\n');
+		*status = 1;
+	}
+}
+
+/*
+ * Second stage: Check that the lockout is not inherited across execve.
+ * Used from check_no_lockout_after_execve.
+ */
+static int main_2(void)
+{
+	struct timeval vsyscall_time = { -1, -1 };
+	int status = 0;
+
+	xvgettimeofday(&vsyscall_time);
+	print_time(__LINE__, "vsyscall gettimeofday after fork", vsyscall_time);
+	if (vsyscall_time.tv_sec < 0 || vsyscall_time.tv_usec < 0)
+		status = 1;
+
+	return status;
+}
+
+static int main(int argc, char **argv)
+{
+	int status = 0;
+
+	if (argc > 1) {
+		switch (*argv[1]) {
+		case '2':
+			return main_2();
+		default:
+			print_string("usage: ");
+			print_string(argv[0]);
+			print_string("\n");
+			return 1;
+		}
+	}
+
+	ensure_proc_is_mounted(&status);
+
+	if (has_vsyscall_map(0, &status))
+		print_info(__LINE__, "vsyscall active at process start\n");
+	else
+		print_error(__LINE__, "vsyscall inactive at process start\n");
+
+
+	check_time(&status);
+	check_lockout_after_fork(&status, 0);
+	check_lockout_after_fork(&status, 1);
+	check_no_lockout_after_enable(&status);
+	check_no_lockout_after_execve(argv, &status);
+	check_vsyscall_in_self_maps(&status);
+	check_vsyscall_maps_for_subprocess(&status);
+	check_einval(&status);
+
+	print_info(__LINE__, "testing done, exit status: ");
+	print_signed_dec(status);
+	print_char('\n');
+	return status;
+}
+
+static void __attribute__ ((used)) main_trampoline(long *rsp)
+{
+	sys_exit(main(*rsp, (char **) (rsp + 1)));
+}
+
+__asm__(".text\n\t"
+	".globl _start\n"
+	"_start:\n\t"
+	".cfi_startproc\n\t"
+	".cfi_undefined rip\n\t"
+	"movq %rsp, %rdi\n\t"
+	"callq main_trampoline\n\t" /* Results in psABI %rsp alignment.  */
+	".cfi_endproc\n\t"
+	".type _start, @function\n\t"
+	".size _start, . - _start\n\t"
+	".previous");


^ permalink raw reply	[flat|nested] 3+ messages in thread

* [musl] Re: [PATCH v2] x86: Implement arch_prctl(ARCH_VSYSCALL_CONTROL) to disable vsyscall
  2021-12-16 18:50 [musl] [PATCH v2] x86: Implement arch_prctl(ARCH_VSYSCALL_CONTROL) to disable vsyscall Florian Weimer
@ 2021-12-27 16:49 ` Andrei Vagin
  2021-12-27 17:40   ` Florian Weimer
  0 siblings, 1 reply; 3+ messages in thread
From: Andrei Vagin @ 2021-12-27 16:49 UTC (permalink / raw)
  To: Florian Weimer
  Cc: Andy Lutomirski, linux-arch, Linux API, linux-x86_64,
	kernel-hardening, linux-mm, the arch/x86 maintainers, musl,
	libc-alpha, LKML, Dave Hansen, Kees Cook

On Thu, Dec 16, 2021 at 4:00 PM Florian Weimer <fweimer@redhat.com> wrote:
>
> Distributions struggle with changing the default for vsyscall
> emulation because it is a clear break of userspace ABI, something
> that should not happen.
>
> The legacy vsyscall interface is supposed to be used by libcs only,
> not by applications.  This commit adds a new arch_prctl request,
> ARCH_VSYSCALL_CONTROL, with one argument.  If the argument is 0,
> executing vsyscalls will cause the process to terminate.  Argument 1
> turns vsyscall back on (this is mostly for a largely theoretical
> CRIU use case).
>
> Newer libcs can use a zero ARCH_VSYSCALL_CONTROL at startup to disable
> vsyscall for the process.  Legacy libcs do not perform this call, so
> vsyscall remains enabled for them.  This approach should achieves
> backwards compatibility (perfect compatibility if the assumption that
> only libcs use vsyscall is accurate), and it provides full hardening
> for new binaries.
>
> The chosen value of ARCH_VSYSCALL_CONTROL should avoid conflicts
> with other x86-64 arch_prctl requests.  The fact that with
> vsyscall=emulate, reading the vsyscall region is still possible
> even after a zero ARCH_VSYSCALL_CONTROL is considered limitation
> in the current implementation and may change in a future kernel
> version.
>
> Future arch_prctls requests commonly used at process startup can imply
> ARCH_VSYSCALL_CONTROL with a zero argument, so that a separate system
> call for disabling vsyscall is avoided.
>
> Signed-off-by: Florian Weimer <fweimer@redhat.com>

Acked-by: Andrei Vagin <avagin@gmail.com>

pls read inline comments.

>
> ---
> v2: ARCH_VSYSCALL_CONTROL instead of ARCH_VSYSCALL_LOCKOUT.  New tests
>     for the toggle behavior.  Implement hiding [vsyscall] in
>     /proc/PID/maps and test it.  Various other test fixes cleanups
>     (e.g., fixed missing second argument to gettimeofday).
>
>  arch/x86/entry/vsyscall/vsyscall_64.c          |  10 +-
>  arch/x86/include/asm/mmu.h                     |   6 +
>  arch/x86/include/uapi/asm/prctl.h              |   2 +
>  arch/x86/kernel/process_64.c                   |   7 +
>  tools/arch/x86/include/uapi/asm/prctl.h        |   2 +
>  tools/testing/selftests/x86/Makefile           |  13 +-
>  tools/testing/selftests/x86/vsyscall_control.c | 891 +++++++++++++++++++++++++
>  7 files changed, 927 insertions(+), 4 deletions(-)
>
> diff --git a/arch/x86/entry/vsyscall/vsyscall_64.c b/arch/x86/entry/vsyscall/vsyscall_64.c
> index fd2ee9408e91..8eb3bcf2cedf 100644
> --- a/arch/x86/entry/vsyscall/vsyscall_64.c
> +++ b/arch/x86/entry/vsyscall/vsyscall_64.c
> @@ -174,6 +174,12 @@ bool emulate_vsyscall(unsigned long error_code,
>
>         tsk = current;
>
> +       if (tsk->mm->context.vsyscall_disabled) {
> +               warn_bad_vsyscall(KERN_WARNING, regs,
> +                                 "vsyscall after lockout (exploit attempt?)");

I don't think that we need this warning message. If we disable
vsyscall, its address range is
not differ from other addresses around and has to be handled the same
way. For example,
gVisor or any other sandbox engines may want to emulate vsyscall, but
the kernel log will
be full of such messages.

> +               goto sigsegv;
> +       }
> +
>         /*
>          * Check for access_ok violations and find the syscall nr.
>          *
> @@ -316,8 +322,10 @@ static struct vm_area_struct gate_vma __ro_after_init = {
>
>  struct vm_area_struct *get_gate_vma(struct mm_struct *mm)
>  {
> +       if (!mm || mm->context.vsyscall_disabled)
> +               return NULL;
>  #ifdef CONFIG_COMPAT
> -       if (!mm || !(mm->context.flags & MM_CONTEXT_HAS_VSYSCALL))
> +       if (!(mm->context.flags & MM_CONTEXT_HAS_VSYSCALL))
>                 return NULL;
>  #endif
>         if (vsyscall_mode == NONE)
> diff --git a/arch/x86/include/asm/mmu.h b/arch/x86/include/asm/mmu.h
> index 5d7494631ea9..3934d6907910 100644
> --- a/arch/x86/include/asm/mmu.h
> +++ b/arch/x86/include/asm/mmu.h
> @@ -41,6 +41,12 @@ typedef struct {
>  #ifdef CONFIG_X86_64
>         unsigned short flags;
>  #endif
> +#ifdef CONFIG_X86_VSYSCALL_EMULATION
> +       /*
> +        * Changed by arch_prctl(ARCH_VSYSCALL_CONTROL).
> +        */
> +       bool vsyscall_disabled;
> +#endif
>
>         struct mutex lock;
>         void __user *vdso;                      /* vdso base address */
> diff --git a/arch/x86/include/uapi/asm/prctl.h b/arch/x86/include/uapi/asm/prctl.h
> index 754a07856817..aad0bcfbf49f 100644
> --- a/arch/x86/include/uapi/asm/prctl.h
> +++ b/arch/x86/include/uapi/asm/prctl.h
> @@ -18,4 +18,6 @@
>  #define ARCH_MAP_VDSO_32       0x2002
>  #define ARCH_MAP_VDSO_64       0x2003
>
> +#define ARCH_VSYSCALL_CONTROL  0x5001
> +
>  #endif /* _ASM_X86_PRCTL_H */
> diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
> index 3402edec236c..834bad068211 100644
> --- a/arch/x86/kernel/process_64.c
> +++ b/arch/x86/kernel/process_64.c
> @@ -816,6 +816,13 @@ long do_arch_prctl_64(struct task_struct *task, int option, unsigned long arg2)
>                 ret = put_user(base, (unsigned long __user *)arg2);
>                 break;
>         }
> +#ifdef CONFIG_X86_VSYSCALL_EMULATION
> +       case ARCH_VSYSCALL_CONTROL:
> +               if (unlikely(arg2 > 1))
> +                       return -EINVAL;
> +               current->mm->context.vsyscall_disabled = !arg2;
> +               break;
> +#endif
>
>  #ifdef CONFIG_CHECKPOINT_RESTORE
>  # ifdef CONFIG_X86_X32_ABI
> diff --git a/tools/arch/x86/include/uapi/asm/prctl.h b/tools/arch/x86/include/uapi/asm/prctl.h
> index 754a07856817..aad0bcfbf49f 100644
> --- a/tools/arch/x86/include/uapi/asm/prctl.h
> +++ b/tools/arch/x86/include/uapi/asm/prctl.h
> @@ -18,4 +18,6 @@
>  #define ARCH_MAP_VDSO_32       0x2002
>  #define ARCH_MAP_VDSO_64       0x2003
>
> +#define ARCH_VSYSCALL_CONTROL  0x5001
> +
>  #endif /* _ASM_X86_PRCTL_H */
> diff --git a/tools/testing/selftests/x86/Makefile b/tools/testing/selftests/x86/Makefile
> index 8a1f62ab3c8e..2a7c91ee68e0 100644
> --- a/tools/testing/selftests/x86/Makefile
> +++ b/tools/testing/selftests/x86/Makefile
> @@ -18,7 +18,7 @@ TARGETS_C_32BIT_ONLY := entry_from_vm86 test_syscall_vdso unwind_vdso \
>                         test_FCMOV test_FCOMI test_FISTTP \
>                         vdso_restorer
>  TARGETS_C_64BIT_ONLY := fsgsbase sysret_rip syscall_numbering \
> -                       corrupt_xstate_header amx
> +                       corrupt_xstate_header amx vsyscall_control
>  # Some selftests require 32bit support enabled also on 64bit systems
>  TARGETS_C_32BIT_NEEDED := ldt_gdt ptrace_syscall
>
> @@ -72,10 +72,12 @@ all_64: $(BINARIES_64)
>  EXTRA_CLEAN := $(BINARIES_32) $(BINARIES_64)
>
>  $(BINARIES_32): $(OUTPUT)/%_32: %.c helpers.h
> -       $(CC) -m32 -o $@ $(CFLAGS) $(EXTRA_CFLAGS) $^ -lrt -ldl -lm
> +       $(CC) -m32 -o $@ $(CFLAGS) $(EXTRA_CFLAGS) $^ \
> +               $(or $(LIBS), -lrt -ldl -lm)
>
>  $(BINARIES_64): $(OUTPUT)/%_64: %.c helpers.h
> -       $(CC) -m64 -o $@ $(CFLAGS) $(EXTRA_CFLAGS) $^ -lrt -ldl
> +       $(CC) -m64 -o $@ $(CFLAGS) $(EXTRA_CFLAGS) $^ \
> +               $(or $(LIBS), -lrt -ldl -lm)
>
>  # x86_64 users should be encouraged to install 32-bit libraries
>  ifeq ($(CAN_BUILD_I386)$(CAN_BUILD_X86_64),01)
> @@ -105,3 +107,8 @@ $(OUTPUT)/test_syscall_vdso_32: thunks_32.S
>  # state.
>  $(OUTPUT)/check_initial_reg_state_32: CFLAGS += -Wl,-ereal_start -static
>  $(OUTPUT)/check_initial_reg_state_64: CFLAGS += -Wl,-ereal_start -static
> +
> +# This test does not link against anything (neither libc nor libgcc).
> +$(OUTPUT)/vsyscall_control_64: \
> +       LIBS := -Wl,-no-pie -static -nostdlib -nostartfiles
> +       CFLAGS += -fno-pie -fno-stack-protector -fno-builtin -ffreestanding
> diff --git a/tools/testing/selftests/x86/vsyscall_control.c b/tools/testing/selftests/x86/vsyscall_control.c
> new file mode 100644
> index 000000000000..ee966f936c89
> --- /dev/null
> +++ b/tools/testing/selftests/x86/vsyscall_control.c

I would move the test in a separate patch...

Thanks,
Andrei

^ permalink raw reply	[flat|nested] 3+ messages in thread

* [musl] Re: [PATCH v2] x86: Implement arch_prctl(ARCH_VSYSCALL_CONTROL) to disable vsyscall
  2021-12-27 16:49 ` [musl] " Andrei Vagin
@ 2021-12-27 17:40   ` Florian Weimer
  0 siblings, 0 replies; 3+ messages in thread
From: Florian Weimer @ 2021-12-27 17:40 UTC (permalink / raw)
  To: Andrei Vagin
  Cc: Andy Lutomirski, linux-arch, Linux API, linux-x86_64,
	kernel-hardening, linux-mm, the arch/x86 maintainers, musl,
	libc-alpha, LKML, Dave Hansen, Kees Cook

* Andrei Vagin:

>> diff --git a/arch/x86/entry/vsyscall/vsyscall_64.c b/arch/x86/entry/vsyscall/vsyscall_64.c
>> index fd2ee9408e91..8eb3bcf2cedf 100644
>> --- a/arch/x86/entry/vsyscall/vsyscall_64.c
>> +++ b/arch/x86/entry/vsyscall/vsyscall_64.c
>> @@ -174,6 +174,12 @@ bool emulate_vsyscall(unsigned long error_code,
>>
>>         tsk = current;
>>
>> +       if (tsk->mm->context.vsyscall_disabled) {
>> +               warn_bad_vsyscall(KERN_WARNING, regs,
>> +                                 "vsyscall after lockout (exploit attempt?)");
>
> I don't think that we need this warning message. If we disable
> vsyscall, its address range is not differ from other addresses around
> and has to be handled the same way. For example, gVisor or any other
> sandbox engines may want to emulate vsyscall, but the kernel log will
> be full of such messages.

But with vsyscall=none, such messages are already printed.  That's why I
added the warning for the lockout case as well.

>> diff --git a/tools/testing/selftests/x86/vsyscall_control.c b/tools/testing/selftests/x86/vsyscall_control.c
>> new file mode 100644
>> index 000000000000..ee966f936c89
>> --- /dev/null
>> +++ b/tools/testing/selftests/x86/vsyscall_control.c
>
> I would move the test in a separate patch...

I can do that if it simplifies matters.

Thanks,
Florian


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2021-12-27 17:41 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-12-16 18:50 [musl] [PATCH v2] x86: Implement arch_prctl(ARCH_VSYSCALL_CONTROL) to disable vsyscall Florian Weimer
2021-12-27 16:49 ` [musl] " Andrei Vagin
2021-12-27 17:40   ` Florian Weimer

Code repositories for project(s) associated with this inbox:

	https://git.vuxu.org/mirror/musl/

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).