From: "Andi Kleen" <ak@suse.de>
Nop out system call instruction in vsyscall page when not needed. This may
make some attacks more difficult.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
25-akpm/arch/x86_64/kernel/vsyscall.c | 74 +++++++++++++++++++++++++++++++---
1 files changed, 69 insertions(+), 5 deletions(-)
diff -puN arch/x86_64/kernel/vsyscall.c~x86_64-nop-out-system-call-instruction-in-vsyscall-page arch/x86_64/kernel/vsyscall.c
--- 25/arch/x86_64/kernel/vsyscall.c~x86_64-nop-out-system-call-instruction-in-vsyscall-page Wed Mar 23 15:38:30 2005
+++ 25-akpm/arch/x86_64/kernel/vsyscall.c Wed Mar 23 15:38:30 2005
@@ -13,8 +13,8 @@
* vsyscalls. One vsyscall can reserve more than 1 slot to avoid
* jumping out of line if necessary.
*
- * Note: the concept clashes with user mode linux. If you use UML just
- * set the kernel.vsyscall sysctl to 0.
+ * Note: the concept clashes with user mode linux. If you use UML and
+ * want per guest time just set the kernel.vsyscall64 sysctl to 0.
*/
/*
@@ -41,6 +41,7 @@
#include <linux/timer.h>
#include <linux/seqlock.h>
#include <linux/jiffies.h>
+#include <linux/sysctl.h>
#include <asm/vsyscall.h>
#include <asm/pgtable.h>
@@ -107,7 +108,7 @@ static force_inline void do_get_tz(struc
static force_inline int gettimeofday(struct timeval *tv, struct timezone *tz)
{
int ret;
- asm volatile("syscall"
+ asm volatile("vsysc2: syscall"
: "=a" (ret)
: "0" (__NR_gettimeofday),"D" (tv),"S" (tz) : __syscall_clobber );
return ret;
@@ -116,7 +117,7 @@ static force_inline int gettimeofday(str
static force_inline long time_syscall(long *t)
{
long secs;
- asm volatile("syscall"
+ asm volatile("vsysc1: syscall"
: "=a" (secs)
: "0" (__NR_time),"D" (t) : __syscall_clobber);
return secs;
@@ -154,6 +155,69 @@ static long __vsyscall(3) venosys_1(void
return -ENOSYS;
}
+#ifdef CONFIG_SYSCTL
+
+#define SYSCALL 0x050f
+#define NOP2 0x9090
+
+/*
+ * NOP out syscall in vsyscall page when not needed.
+ */
+static int vsyscall_sysctl_change(ctl_table *ctl, int write, struct file * filp,
+ void __user *buffer, size_t *lenp, loff_t *ppos)
+{
+ extern u16 vsysc1, vsysc2;
+ u16 *map1, *map2;
+ int ret = proc_dointvec(ctl, write, filp, buffer, lenp, ppos);
+ if (!write)
+ return ret;
+ /* gcc has some trouble with __va(__pa()), so just do it this
+ way. */
+ map1 = ioremap(__pa_symbol(&vsysc1), 2);
+ if (!map1)
+ return -ENOMEM;
+ map2 = ioremap(__pa_symbol(&vsysc2), 2);
+ if (!map2) {
+ ret = -ENOMEM;
+ goto out;
+ }
+ if (!sysctl_vsyscall) {
+ *map1 = SYSCALL;
+ *map2 = SYSCALL;
+ } else {
+ *map1 = NOP2;
+ *map2 = NOP2;
+ }
+ iounmap(map2);
+out:
+ iounmap(map1);
+ return ret;
+}
+
+static int vsyscall_sysctl_nostrat(ctl_table *t, int __user *name, int nlen,
+ void __user *oldval, size_t __user *oldlenp,
+ void __user *newval, size_t newlen,
+ void **context)
+{
+ return -ENOSYS;
+}
+
+static ctl_table kernel_table2[] = {
+ { .ctl_name = 99, .procname = "vsyscall64",
+ .data = &sysctl_vsyscall, .maxlen = sizeof(int), .mode = 0644,
+ .strategy = vsyscall_sysctl_nostrat,
+ .proc_handler vsyscall_sysctl_change },
+ { 0, }
+};
+
+static ctl_table kernel_root_table2[] = {
+ { .ctl_name = CTL_KERN, .procname = "kernel", .mode = 0555,
+ .child = kernel_table2 },
+ { 0 },
+};
+
+#endif
+
static void __init map_vsyscall(void)
{
extern char __vsyscall_0;
@@ -170,7 +234,7 @@ static int __init vsyscall_init(void)
BUG_ON((VSYSCALL_ADDR(0) != __fix_to_virt(VSYSCALL_FIRST_PAGE)));
map_vsyscall();
sysctl_vsyscall = 1;
-
+ register_sysctl_table(kernel_root_table2, 0);
return 0;
}
_