2005-04-16 22:20:36 +00:00
|
|
|
/*
|
2014-10-29 21:33:46 +00:00
|
|
|
* Copyright (c) 2012-2014 Andy Lutomirski <luto@amacapital.net>
|
|
|
|
*
|
|
|
|
* Based on the original implementation which is:
|
2005-04-16 22:20:36 +00:00
|
|
|
* Copyright (C) 2001 Andrea Arcangeli <andrea@suse.de> SuSE
|
|
|
|
* Copyright 2003 Andi Kleen, SuSE Labs.
|
|
|
|
*
|
2014-10-29 21:33:46 +00:00
|
|
|
* Parts of the original code have been moved to arch/x86/vdso/vma.c
|
|
|
|
*
|
|
|
|
* This file implements vsyscall emulation. vsyscalls are a legacy ABI:
|
|
|
|
* Userspace can request certain kernel services by calling fixed
|
|
|
|
* addresses. This concept is problematic:
|
2011-06-05 17:50:24 +00:00
|
|
|
*
|
2014-10-29 21:33:46 +00:00
|
|
|
* - It interferes with ASLR.
|
|
|
|
* - It's awkward to write code that lives in kernel addresses but is
|
|
|
|
* callable by userspace at fixed addresses.
|
|
|
|
* - The whole concept is impossible for 32-bit compat userspace.
|
|
|
|
* - UML cannot easily virtualize a vsyscall.
|
2005-04-16 22:20:36 +00:00
|
|
|
*
|
2014-10-29 21:33:46 +00:00
|
|
|
* As of mid-2014, I believe that there is no new userspace code that
|
|
|
|
* will use a vsyscall if the vDSO is present. I hope that there will
|
|
|
|
* soon be no new userspace code that will ever use a vsyscall.
|
2005-04-16 22:20:36 +00:00
|
|
|
*
|
2014-10-29 21:33:46 +00:00
|
|
|
* The code in this file emulates vsyscalls when notified of a page
|
|
|
|
* fault to a vsyscall address.
|
2005-04-16 22:20:36 +00:00
|
|
|
*/
|
|
|
|
|
|
|
|
#include <linux/kernel.h>
|
|
|
|
#include <linux/timer.h>
|
2011-06-05 17:50:24 +00:00
|
|
|
#include <linux/syscalls.h>
|
|
|
|
#include <linux/ratelimit.h>
|
2005-04-16 22:20:36 +00:00
|
|
|
|
|
|
|
#include <asm/vsyscall.h>
|
2007-02-16 09:28:21 +00:00
|
|
|
#include <asm/unistd.h>
|
2005-04-16 22:20:36 +00:00
|
|
|
#include <asm/fixmap.h>
|
2011-06-05 17:50:24 +00:00
|
|
|
#include <asm/traps.h>
|
2005-04-16 22:20:36 +00:00
|
|
|
|
2011-08-03 13:31:54 +00:00
|
|
|
#define CREATE_TRACE_POINTS
|
|
|
|
#include "vsyscall_trace.h"
|
|
|
|
|
2011-11-08 00:33:41 +00:00
|
|
|
static enum { EMULATE, NATIVE, NONE } vsyscall_mode = EMULATE;
|
2011-08-10 15:15:32 +00:00
|
|
|
|
|
|
|
static int __init vsyscall_setup(char *str)
|
|
|
|
{
|
|
|
|
if (str) {
|
|
|
|
if (!strcmp("emulate", str))
|
|
|
|
vsyscall_mode = EMULATE;
|
|
|
|
else if (!strcmp("native", str))
|
|
|
|
vsyscall_mode = NATIVE;
|
|
|
|
else if (!strcmp("none", str))
|
|
|
|
vsyscall_mode = NONE;
|
|
|
|
else
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
early_param("vsyscall", vsyscall_setup);
|
|
|
|
|
2011-06-05 17:50:24 +00:00
|
|
|
static void warn_bad_vsyscall(const char *level, struct pt_regs *regs,
|
|
|
|
const char *message)
|
2005-04-16 22:20:36 +00:00
|
|
|
{
|
2012-05-22 02:50:07 +00:00
|
|
|
if (!show_unhandled_signals)
|
2011-06-05 17:50:24 +00:00
|
|
|
return;
|
2005-04-16 22:20:36 +00:00
|
|
|
|
2014-07-25 23:30:27 +00:00
|
|
|
printk_ratelimited("%s%s[%d] %s ip:%lx cs:%lx sp:%lx ax:%lx si:%lx di:%lx\n",
|
|
|
|
level, current->comm, task_pid_nr(current),
|
|
|
|
message, regs->ip, regs->cs,
|
|
|
|
regs->sp, regs->ax, regs->si, regs->di);
|
2011-07-13 13:24:09 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static int addr_to_vsyscall_nr(unsigned long addr)
|
|
|
|
{
|
|
|
|
int nr;
|
|
|
|
|
2014-05-05 19:19:36 +00:00
|
|
|
if ((addr & ~0xC00UL) != VSYSCALL_ADDR)
|
2011-07-13 13:24:09 +00:00
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
nr = (addr & 0xC00UL) >> 10;
|
|
|
|
if (nr >= 3)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
return nr;
|
2005-04-16 22:20:36 +00:00
|
|
|
}
|
|
|
|
|
2011-11-08 00:33:40 +00:00
|
|
|
static bool write_ok_or_segv(unsigned long ptr, size_t size)
|
|
|
|
{
|
|
|
|
/*
|
|
|
|
* XXX: if access_ok, get_user, and put_user handled
|
|
|
|
* sig_on_uaccess_error, this could go away.
|
|
|
|
*/
|
|
|
|
|
|
|
|
if (!access_ok(VERIFY_WRITE, (void __user *)ptr, size)) {
|
|
|
|
siginfo_t info;
|
|
|
|
struct thread_struct *thread = ¤t->thread;
|
|
|
|
|
|
|
|
thread->error_code = 6; /* user fault, no page, write */
|
|
|
|
thread->cr2 = ptr;
|
2012-03-12 09:25:55 +00:00
|
|
|
thread->trap_nr = X86_TRAP_PF;
|
2011-11-08 00:33:40 +00:00
|
|
|
|
|
|
|
memset(&info, 0, sizeof(info));
|
|
|
|
info.si_signo = SIGSEGV;
|
|
|
|
info.si_errno = 0;
|
|
|
|
info.si_code = SEGV_MAPERR;
|
|
|
|
info.si_addr = (void __user *)ptr;
|
|
|
|
|
|
|
|
force_sig_info(SIGSEGV, &info, current);
|
|
|
|
return false;
|
|
|
|
} else {
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2011-08-10 15:15:32 +00:00
|
|
|
bool emulate_vsyscall(struct pt_regs *regs, unsigned long address)
|
2005-04-16 22:20:36 +00:00
|
|
|
{
|
2011-06-05 17:50:24 +00:00
|
|
|
struct task_struct *tsk;
|
|
|
|
unsigned long caller;
|
2012-10-01 18:40:45 +00:00
|
|
|
int vsyscall_nr, syscall_nr, tmp;
|
2011-11-08 00:33:40 +00:00
|
|
|
int prev_sig_on_uaccess_error;
|
2011-06-05 17:50:24 +00:00
|
|
|
long ret;
|
|
|
|
|
2011-08-10 15:15:32 +00:00
|
|
|
/*
|
|
|
|
* No point in checking CS -- the only way to get here is a user mode
|
|
|
|
* trap to a high address, which means that we're in 64-bit user code.
|
|
|
|
*/
|
2011-06-05 17:50:24 +00:00
|
|
|
|
2011-08-10 15:15:32 +00:00
|
|
|
WARN_ON_ONCE(address != regs->ip);
|
2011-07-13 13:24:09 +00:00
|
|
|
|
2011-08-10 15:15:32 +00:00
|
|
|
if (vsyscall_mode == NONE) {
|
|
|
|
warn_bad_vsyscall(KERN_INFO, regs,
|
|
|
|
"vsyscall attempted with vsyscall=none");
|
|
|
|
return false;
|
2011-07-13 13:24:09 +00:00
|
|
|
}
|
|
|
|
|
2011-08-10 15:15:32 +00:00
|
|
|
vsyscall_nr = addr_to_vsyscall_nr(address);
|
2011-08-03 13:31:54 +00:00
|
|
|
|
|
|
|
trace_emulate_vsyscall(vsyscall_nr);
|
|
|
|
|
2011-07-13 13:24:09 +00:00
|
|
|
if (vsyscall_nr < 0) {
|
|
|
|
warn_bad_vsyscall(KERN_WARNING, regs,
|
2011-08-10 15:15:32 +00:00
|
|
|
"misaligned vsyscall (exploit attempt or buggy program) -- look up the vsyscall kernel parameter if you need a workaround");
|
2011-06-05 17:50:24 +00:00
|
|
|
goto sigsegv;
|
|
|
|
}
|
2007-05-21 12:31:52 +00:00
|
|
|
|
2011-06-05 17:50:24 +00:00
|
|
|
if (get_user(caller, (unsigned long __user *)regs->sp) != 0) {
|
2011-08-10 15:15:32 +00:00
|
|
|
warn_bad_vsyscall(KERN_WARNING, regs,
|
|
|
|
"vsyscall with bad stack (exploit attempt?)");
|
2011-06-05 17:50:24 +00:00
|
|
|
goto sigsegv;
|
|
|
|
}
|
2010-07-14 00:56:18 +00:00
|
|
|
|
2011-06-05 17:50:24 +00:00
|
|
|
tsk = current;
|
2011-11-08 00:33:40 +00:00
|
|
|
|
|
|
|
/*
|
2012-10-01 18:40:45 +00:00
|
|
|
* Check for access_ok violations and find the syscall nr.
|
|
|
|
*
|
2012-04-01 18:48:04 +00:00
|
|
|
* NULL is a valid user pointer (in the access_ok sense) on 32-bit and
|
2011-11-08 00:33:40 +00:00
|
|
|
* 64-bit, so we don't need to special-case it here. For all the
|
2012-04-01 18:48:04 +00:00
|
|
|
* vsyscalls, NULL means "don't write anything" not "write it at
|
2011-11-08 00:33:40 +00:00
|
|
|
* address 0".
|
|
|
|
*/
|
2011-06-05 17:50:24 +00:00
|
|
|
switch (vsyscall_nr) {
|
|
|
|
case 0:
|
2011-11-08 00:33:40 +00:00
|
|
|
if (!write_ok_or_segv(regs->di, sizeof(struct timeval)) ||
|
2012-10-01 18:40:45 +00:00
|
|
|
!write_ok_or_segv(regs->si, sizeof(struct timezone))) {
|
|
|
|
ret = -EFAULT;
|
|
|
|
goto check_fault;
|
|
|
|
}
|
2011-11-08 00:33:40 +00:00
|
|
|
|
2012-10-01 18:40:45 +00:00
|
|
|
syscall_nr = __NR_gettimeofday;
|
|
|
|
break;
|
|
|
|
|
|
|
|
case 1:
|
|
|
|
if (!write_ok_or_segv(regs->di, sizeof(time_t))) {
|
|
|
|
ret = -EFAULT;
|
|
|
|
goto check_fault;
|
|
|
|
}
|
|
|
|
|
|
|
|
syscall_nr = __NR_time;
|
|
|
|
break;
|
|
|
|
|
|
|
|
case 2:
|
|
|
|
if (!write_ok_or_segv(regs->di, sizeof(unsigned)) ||
|
|
|
|
!write_ok_or_segv(regs->si, sizeof(unsigned))) {
|
|
|
|
ret = -EFAULT;
|
|
|
|
goto check_fault;
|
|
|
|
}
|
|
|
|
|
|
|
|
syscall_nr = __NR_getcpu;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Handle seccomp. regs->ip must be the original value.
|
|
|
|
* See seccomp_send_sigsys and Documentation/prctl/seccomp_filter.txt.
|
|
|
|
*
|
|
|
|
* We could optimize the seccomp disabled case, but performance
|
|
|
|
* here doesn't matter.
|
|
|
|
*/
|
|
|
|
regs->orig_ax = syscall_nr;
|
|
|
|
regs->ax = -ENOSYS;
|
seccomp,x86,arm,mips,s390: Remove nr parameter from secure_computing
The secure_computing function took a syscall number parameter, but
it only paid any attention to that parameter if seccomp mode 1 was
enabled. Rather than coming up with a kludge to get the parameter
to work in mode 2, just remove the parameter.
To avoid churn in arches that don't have seccomp filters (and may
not even support syscall_get_nr right now), this leaves the
parameter in secure_computing_strict, which is now a real function.
For ARM, this is a bit ugly due to the fact that ARM conditionally
supports seccomp filters. Fixing that would probably only be a
couple of lines of code, but it should be coordinated with the audit
maintainers.
This will be a slight slowdown on some arches. The right fix is to
pass in all of seccomp_data instead of trying to make just the
syscall nr part be fast.
This is a prerequisite for making two-phase seccomp work cleanly.
Cc: Russell King <linux@arm.linux.org.uk>
Cc: linux-arm-kernel@lists.infradead.org
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: linux-mips@linux-mips.org
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: linux-s390@vger.kernel.org
Cc: x86@kernel.org
Cc: Kees Cook <keescook@chromium.org>
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Kees Cook <keescook@chromium.org>
2014-07-22 01:49:14 +00:00
|
|
|
tmp = secure_computing();
|
2012-10-01 18:40:45 +00:00
|
|
|
if ((!tmp && regs->orig_ax != syscall_nr) || regs->ip != address) {
|
|
|
|
warn_bad_vsyscall(KERN_DEBUG, regs,
|
|
|
|
"seccomp tried to change syscall nr or ip");
|
|
|
|
do_exit(SIGSYS);
|
|
|
|
}
|
2014-11-04 23:36:50 +00:00
|
|
|
regs->orig_ax = -1;
|
2012-10-01 18:40:45 +00:00
|
|
|
if (tmp)
|
|
|
|
goto do_ret; /* skip requested */
|
|
|
|
|
|
|
|
/*
|
|
|
|
* With a real vsyscall, page faults cause SIGSEGV. We want to
|
|
|
|
* preserve that behavior to make writing exploits harder.
|
|
|
|
*/
|
|
|
|
prev_sig_on_uaccess_error = current_thread_info()->sig_on_uaccess_error;
|
|
|
|
current_thread_info()->sig_on_uaccess_error = 1;
|
|
|
|
|
|
|
|
ret = -EFAULT;
|
|
|
|
switch (vsyscall_nr) {
|
|
|
|
case 0:
|
2011-06-05 17:50:24 +00:00
|
|
|
ret = sys_gettimeofday(
|
|
|
|
(struct timeval __user *)regs->di,
|
|
|
|
(struct timezone __user *)regs->si);
|
|
|
|
break;
|
|
|
|
|
|
|
|
case 1:
|
|
|
|
ret = sys_time((time_t __user *)regs->di);
|
|
|
|
break;
|
|
|
|
|
|
|
|
case 2:
|
|
|
|
ret = sys_getcpu((unsigned __user *)regs->di,
|
|
|
|
(unsigned __user *)regs->si,
|
2012-04-01 18:48:04 +00:00
|
|
|
NULL);
|
2011-06-05 17:50:24 +00:00
|
|
|
break;
|
|
|
|
}
|
2010-07-14 00:56:18 +00:00
|
|
|
|
2011-11-08 00:33:40 +00:00
|
|
|
current_thread_info()->sig_on_uaccess_error = prev_sig_on_uaccess_error;
|
|
|
|
|
2012-10-01 18:40:45 +00:00
|
|
|
check_fault:
|
2011-06-05 17:50:24 +00:00
|
|
|
if (ret == -EFAULT) {
|
2011-11-08 00:33:40 +00:00
|
|
|
/* Bad news -- userspace fed a bad pointer to a vsyscall. */
|
2011-06-05 17:50:24 +00:00
|
|
|
warn_bad_vsyscall(KERN_INFO, regs,
|
|
|
|
"vsyscall fault (exploit attempt?)");
|
2011-11-08 00:33:40 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* If we failed to generate a signal for any reason,
|
|
|
|
* generate one here. (This should be impossible.)
|
|
|
|
*/
|
|
|
|
if (WARN_ON_ONCE(!sigismember(&tsk->pending.signal, SIGBUS) &&
|
|
|
|
!sigismember(&tsk->pending.signal, SIGSEGV)))
|
|
|
|
goto sigsegv;
|
|
|
|
|
|
|
|
return true; /* Don't emulate the ret. */
|
2011-06-05 17:50:24 +00:00
|
|
|
}
|
2010-07-14 00:56:18 +00:00
|
|
|
|
2011-06-05 17:50:24 +00:00
|
|
|
regs->ax = ret;
|
2005-04-16 22:20:36 +00:00
|
|
|
|
2012-07-13 17:06:35 +00:00
|
|
|
do_ret:
|
2011-06-05 17:50:24 +00:00
|
|
|
/* Emulate a ret instruction. */
|
|
|
|
regs->ip = caller;
|
|
|
|
regs->sp += 8;
|
2011-08-10 15:15:32 +00:00
|
|
|
return true;
|
2011-06-05 17:50:24 +00:00
|
|
|
|
|
|
|
sigsegv:
|
|
|
|
force_sig(SIGSEGV, current);
|
2011-08-10 15:15:32 +00:00
|
|
|
return true;
|
2005-04-16 22:20:36 +00:00
|
|
|
}
|
|
|
|
|
2014-09-23 17:50:51 +00:00
|
|
|
/*
|
|
|
|
* A pseudo VMA to allow ptrace access for the vsyscall page. This only
|
|
|
|
* covers the 64bit vsyscall page now. 32bit has a real VMA now and does
|
|
|
|
* not need special handling anymore:
|
|
|
|
*/
|
|
|
|
static const char *gate_vma_name(struct vm_area_struct *vma)
|
|
|
|
{
|
|
|
|
return "[vsyscall]";
|
|
|
|
}
|
|
|
|
static struct vm_operations_struct gate_vma_ops = {
|
|
|
|
.name = gate_vma_name,
|
|
|
|
};
|
|
|
|
static struct vm_area_struct gate_vma = {
|
|
|
|
.vm_start = VSYSCALL_ADDR,
|
|
|
|
.vm_end = VSYSCALL_ADDR + PAGE_SIZE,
|
|
|
|
.vm_page_prot = PAGE_READONLY_EXEC,
|
|
|
|
.vm_flags = VM_READ | VM_EXEC,
|
|
|
|
.vm_ops = &gate_vma_ops,
|
|
|
|
};
|
|
|
|
|
|
|
|
struct vm_area_struct *get_gate_vma(struct mm_struct *mm)
|
|
|
|
{
|
|
|
|
#ifdef CONFIG_IA32_EMULATION
|
|
|
|
if (!mm || mm->context.ia32_compat)
|
|
|
|
return NULL;
|
|
|
|
#endif
|
2014-10-29 21:33:45 +00:00
|
|
|
if (vsyscall_mode == NONE)
|
|
|
|
return NULL;
|
2014-09-23 17:50:51 +00:00
|
|
|
return &gate_vma;
|
|
|
|
}
|
|
|
|
|
|
|
|
int in_gate_area(struct mm_struct *mm, unsigned long addr)
|
|
|
|
{
|
|
|
|
struct vm_area_struct *vma = get_gate_vma(mm);
|
|
|
|
|
|
|
|
if (!vma)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
return (addr >= vma->vm_start) && (addr < vma->vm_end);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Use this when you have no reliable mm, typically from interrupt
|
|
|
|
* context. It is less reliable than using a task's mm and may give
|
|
|
|
* false positives.
|
|
|
|
*/
|
|
|
|
int in_gate_area_no_mm(unsigned long addr)
|
|
|
|
{
|
2014-10-29 21:33:45 +00:00
|
|
|
return vsyscall_mode != NONE && (addr & PAGE_MASK) == VSYSCALL_ADDR;
|
2014-09-23 17:50:51 +00:00
|
|
|
}
|
|
|
|
|
2008-01-30 12:32:39 +00:00
|
|
|
void __init map_vsyscall(void)
|
2005-04-16 22:20:36 +00:00
|
|
|
{
|
2011-08-10 15:15:32 +00:00
|
|
|
extern char __vsyscall_page;
|
|
|
|
unsigned long physaddr_vsyscall = __pa_symbol(&__vsyscall_page);
|
2005-04-16 22:20:36 +00:00
|
|
|
|
2014-10-29 21:33:45 +00:00
|
|
|
if (vsyscall_mode != NONE)
|
|
|
|
__set_fixmap(VSYSCALL_PAGE, physaddr_vsyscall,
|
|
|
|
vsyscall_mode == NATIVE
|
|
|
|
? PAGE_KERNEL_VSYSCALL
|
|
|
|
: PAGE_KERNEL_VVAR);
|
|
|
|
|
2014-05-05 19:19:36 +00:00
|
|
|
BUILD_BUG_ON((unsigned long)__fix_to_virt(VSYSCALL_PAGE) !=
|
|
|
|
(unsigned long)VSYSCALL_ADDR);
|
2005-04-16 22:20:36 +00:00
|
|
|
}
|