2005-04-16 18:20:36 -04:00
|
|
|
/*
|
|
|
|
* linux/arch/x86_64/kernel/vsyscall.c
|
|
|
|
*
|
|
|
|
* Copyright (C) 2001 Andrea Arcangeli <andrea@suse.de> SuSE
|
|
|
|
* Copyright 2003 Andi Kleen, SuSE Labs.
|
|
|
|
*
|
|
|
|
* Thanks to hpa@transmeta.com for some useful hint.
|
|
|
|
* Special thanks to Ingo Molnar for his early experience with
|
|
|
|
* a different vsyscall implementation for Linux/IA32 and for the name.
|
|
|
|
*
|
|
|
|
* vsyscall 1 is located at -10Mbyte, vsyscall 2 is located
|
|
|
|
* at virtual address -10Mbyte+1024bytes etc... There are at max 4
|
|
|
|
* vsyscalls. One vsyscall can reserve more than 1 slot to avoid
|
|
|
|
* jumping out of line if necessary. We cannot add more with this
|
|
|
|
* mechanism because older kernels won't return -ENOSYS.
|
|
|
|
* If we want more than four we need a vDSO.
|
|
|
|
*
|
|
|
|
* 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.
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <linux/time.h>
|
|
|
|
#include <linux/init.h>
|
|
|
|
#include <linux/kernel.h>
|
|
|
|
#include <linux/timer.h>
|
|
|
|
#include <linux/seqlock.h>
|
|
|
|
#include <linux/jiffies.h>
|
|
|
|
#include <linux/sysctl.h>
|
2006-09-26 04:52:28 -04:00
|
|
|
#include <linux/getcpu.h>
|
2006-11-14 10:57:46 -05:00
|
|
|
#include <linux/cpu.h>
|
|
|
|
#include <linux/smp.h>
|
|
|
|
#include <linux/notifier.h>
|
2005-04-16 18:20:36 -04:00
|
|
|
|
|
|
|
#include <asm/vsyscall.h>
|
|
|
|
#include <asm/pgtable.h>
|
|
|
|
#include <asm/page.h>
|
|
|
|
#include <asm/fixmap.h>
|
|
|
|
#include <asm/errno.h>
|
|
|
|
#include <asm/io.h>
|
2006-09-26 04:52:28 -04:00
|
|
|
#include <asm/segment.h>
|
|
|
|
#include <asm/desc.h>
|
|
|
|
#include <asm/topology.h>
|
2005-04-16 18:20:36 -04:00
|
|
|
|
|
|
|
#define __vsyscall(nr) __attribute__ ((unused,__section__(".vsyscall_" #nr)))
|
2006-12-06 23:37:29 -05:00
|
|
|
#define __syscall_clobber "r11","rcx","memory"
|
2005-04-16 18:20:36 -04:00
|
|
|
|
|
|
|
int __sysctl_vsyscall __section_sysctl_vsyscall = 1;
|
|
|
|
seqlock_t __xtime_lock __section_xtime_lock = SEQLOCK_UNLOCKED;
|
2006-09-26 04:52:28 -04:00
|
|
|
int __vgetcpu_mode __section_vgetcpu_mode;
|
2005-04-16 18:20:36 -04:00
|
|
|
|
|
|
|
#include <asm/unistd.h>
|
|
|
|
|
2006-01-11 16:45:30 -05:00
|
|
|
static __always_inline void timeval_normalize(struct timeval * tv)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
|
|
|
time_t __sec;
|
|
|
|
|
|
|
|
__sec = tv->tv_usec / 1000000;
|
|
|
|
if (__sec) {
|
|
|
|
tv->tv_usec %= 1000000;
|
|
|
|
tv->tv_sec += __sec;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2006-01-11 16:45:30 -05:00
|
|
|
static __always_inline void do_vgettimeofday(struct timeval * tv)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
|
|
|
long sequence, t;
|
|
|
|
unsigned long sec, usec;
|
|
|
|
|
|
|
|
do {
|
|
|
|
sequence = read_seqbegin(&__xtime_lock);
|
|
|
|
|
|
|
|
sec = __xtime.tv_sec;
|
2006-10-01 02:28:31 -04:00
|
|
|
usec = __xtime.tv_nsec / 1000;
|
2005-04-16 18:20:36 -04:00
|
|
|
|
2005-05-17 00:53:28 -04:00
|
|
|
if (__vxtime.mode != VXTIME_HPET) {
|
2006-01-11 16:45:24 -05:00
|
|
|
t = get_cycles_sync();
|
2005-04-16 18:20:36 -04:00
|
|
|
if (t < __vxtime.last_tsc)
|
|
|
|
t = __vxtime.last_tsc;
|
|
|
|
usec += ((t - __vxtime.last_tsc) *
|
|
|
|
__vxtime.tsc_quot) >> 32;
|
|
|
|
/* See comment in x86_64 do_gettimeofday. */
|
|
|
|
} else {
|
2006-09-26 04:52:33 -04:00
|
|
|
usec += ((readl((void __iomem *)
|
|
|
|
fix_to_virt(VSYSCALL_HPET) + 0xf0) -
|
2005-04-16 18:20:36 -04:00
|
|
|
__vxtime.last) * __vxtime.quot) >> 32;
|
|
|
|
}
|
|
|
|
} while (read_seqretry(&__xtime_lock, sequence));
|
|
|
|
|
|
|
|
tv->tv_sec = sec + usec / 1000000;
|
|
|
|
tv->tv_usec = usec % 1000000;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* RED-PEN may want to readd seq locking, but then the variable should be write-once. */
|
2006-01-11 16:45:30 -05:00
|
|
|
static __always_inline void do_get_tz(struct timezone * tz)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
|
|
|
*tz = __sys_tz;
|
|
|
|
}
|
|
|
|
|
2006-01-11 16:45:30 -05:00
|
|
|
static __always_inline int gettimeofday(struct timeval *tv, struct timezone *tz)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
|
|
|
int ret;
|
|
|
|
asm volatile("vsysc2: syscall"
|
|
|
|
: "=a" (ret)
|
|
|
|
: "0" (__NR_gettimeofday),"D" (tv),"S" (tz) : __syscall_clobber );
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2006-01-11 16:45:30 -05:00
|
|
|
static __always_inline long time_syscall(long *t)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
|
|
|
long secs;
|
|
|
|
asm volatile("vsysc1: syscall"
|
|
|
|
: "=a" (secs)
|
|
|
|
: "0" (__NR_time),"D" (t) : __syscall_clobber);
|
|
|
|
return secs;
|
|
|
|
}
|
|
|
|
|
2005-09-12 12:49:24 -04:00
|
|
|
int __vsyscall(0) vgettimeofday(struct timeval * tv, struct timezone * tz)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
2006-06-26 07:56:58 -04:00
|
|
|
if (!__sysctl_vsyscall)
|
2005-04-16 18:20:36 -04:00
|
|
|
return gettimeofday(tv,tz);
|
|
|
|
if (tv)
|
|
|
|
do_vgettimeofday(tv);
|
|
|
|
if (tz)
|
|
|
|
do_get_tz(tz);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* This will break when the xtime seconds get inaccurate, but that is
|
|
|
|
* unlikely */
|
2005-09-12 12:49:24 -04:00
|
|
|
time_t __vsyscall(1) vtime(time_t *t)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
2006-06-26 07:56:58 -04:00
|
|
|
if (!__sysctl_vsyscall)
|
2005-04-16 18:20:36 -04:00
|
|
|
return time_syscall(t);
|
|
|
|
else if (t)
|
|
|
|
*t = __xtime.tv_sec;
|
|
|
|
return __xtime.tv_sec;
|
|
|
|
}
|
|
|
|
|
2006-09-26 04:52:28 -04:00
|
|
|
/* Fast way to get current CPU and node.
|
|
|
|
This helps to do per node and per CPU caches in user space.
|
|
|
|
The result is not guaranteed without CPU affinity, but usually
|
|
|
|
works out because the scheduler tries to keep a thread on the same
|
|
|
|
CPU.
|
|
|
|
|
|
|
|
tcache must point to a two element sized long array.
|
|
|
|
All arguments can be NULL. */
|
|
|
|
long __vsyscall(2)
|
|
|
|
vgetcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *tcache)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
2006-09-26 04:52:28 -04:00
|
|
|
unsigned int dummy, p;
|
|
|
|
unsigned long j = 0;
|
|
|
|
|
|
|
|
/* Fast cache - only recompute value once per jiffies and avoid
|
|
|
|
relatively costly rdtscp/cpuid otherwise.
|
|
|
|
This works because the scheduler usually keeps the process
|
|
|
|
on the same CPU and this syscall doesn't guarantee its
|
|
|
|
results anyways.
|
|
|
|
We do this here because otherwise user space would do it on
|
|
|
|
its own in a likely inferior way (no access to jiffies).
|
|
|
|
If you don't like it pass NULL. */
|
2006-09-29 19:47:55 -04:00
|
|
|
if (tcache && tcache->blob[0] == (j = __jiffies)) {
|
|
|
|
p = tcache->blob[1];
|
2006-09-26 04:52:28 -04:00
|
|
|
} else if (__vgetcpu_mode == VGETCPU_RDTSCP) {
|
|
|
|
/* Load per CPU data from RDTSCP */
|
|
|
|
rdtscp(dummy, dummy, p);
|
|
|
|
} else {
|
|
|
|
/* Load per CPU data from GDT */
|
|
|
|
asm("lsl %1,%0" : "=r" (p) : "r" (__PER_CPU_SEG));
|
|
|
|
}
|
|
|
|
if (tcache) {
|
2006-09-29 19:47:55 -04:00
|
|
|
tcache->blob[0] = j;
|
|
|
|
tcache->blob[1] = p;
|
2006-09-26 04:52:28 -04:00
|
|
|
}
|
|
|
|
if (cpu)
|
|
|
|
*cpu = p & 0xfff;
|
|
|
|
if (node)
|
|
|
|
*node = p >> 12;
|
|
|
|
return 0;
|
2005-04-16 18:20:36 -04:00
|
|
|
}
|
|
|
|
|
2005-09-12 12:49:24 -04:00
|
|
|
long __vsyscall(3) venosys_1(void)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
|
|
|
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;
|
2006-09-26 04:52:33 -04:00
|
|
|
u16 __iomem *map1;
|
|
|
|
u16 __iomem *map2;
|
2005-04-16 18:20:36 -04:00
|
|
|
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) {
|
2006-09-26 04:52:33 -04:00
|
|
|
writew(SYSCALL, map1);
|
|
|
|
writew(SYSCALL, map2);
|
2005-04-16 18:20:36 -04:00
|
|
|
} else {
|
2006-09-26 04:52:33 -04:00
|
|
|
writew(NOP2, map1);
|
|
|
|
writew(NOP2, map2);
|
2005-04-16 18:20:36 -04:00
|
|
|
}
|
|
|
|
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
|
|
|
|
|
2006-11-14 10:57:46 -05:00
|
|
|
/* Assume __initcall executes before all user space. Hopefully kmod
|
|
|
|
doesn't violate that. We'll find out if it does. */
|
|
|
|
static void __cpuinit vsyscall_set_cpu(int cpu)
|
2006-09-26 04:52:28 -04:00
|
|
|
{
|
|
|
|
unsigned long *d;
|
|
|
|
unsigned long node = 0;
|
|
|
|
#ifdef CONFIG_NUMA
|
|
|
|
node = cpu_to_node[cpu];
|
|
|
|
#endif
|
2006-11-14 10:57:46 -05:00
|
|
|
if (cpu_has(&cpu_data[cpu], X86_FEATURE_RDTSCP))
|
|
|
|
write_rdtscp_aux((node << 12) | cpu);
|
2006-09-26 04:52:28 -04:00
|
|
|
|
|
|
|
/* Store cpu number in limit so that it can be loaded quickly
|
|
|
|
in user space in vgetcpu.
|
|
|
|
12 bits for the CPU and 8 bits for the node. */
|
|
|
|
d = (unsigned long *)(cpu_gdt(cpu) + GDT_ENTRY_PER_CPU);
|
|
|
|
*d = 0x0f40000000000ULL;
|
|
|
|
*d |= cpu;
|
|
|
|
*d |= (node & 0xf) << 12;
|
|
|
|
*d |= (node >> 4) << 48;
|
|
|
|
}
|
|
|
|
|
2006-11-14 10:57:46 -05:00
|
|
|
static void __cpuinit cpu_vsyscall_init(void *arg)
|
|
|
|
{
|
|
|
|
/* preemption should be already off */
|
|
|
|
vsyscall_set_cpu(raw_smp_processor_id());
|
|
|
|
}
|
|
|
|
|
|
|
|
static int __cpuinit
|
|
|
|
cpu_vsyscall_notifier(struct notifier_block *n, unsigned long action, void *arg)
|
|
|
|
{
|
|
|
|
long cpu = (long)arg;
|
|
|
|
if (action == CPU_ONLINE)
|
|
|
|
smp_call_function_single(cpu, cpu_vsyscall_init, NULL, 0, 1);
|
|
|
|
return NOTIFY_DONE;
|
|
|
|
}
|
|
|
|
|
2005-04-16 18:20:36 -04:00
|
|
|
static void __init map_vsyscall(void)
|
|
|
|
{
|
|
|
|
extern char __vsyscall_0;
|
|
|
|
unsigned long physaddr_page0 = __pa_symbol(&__vsyscall_0);
|
|
|
|
|
2006-12-06 20:14:09 -05:00
|
|
|
/* Note that VSYSCALL_MAPPED_PAGES must agree with the code below. */
|
2005-04-16 18:20:36 -04:00
|
|
|
__set_fixmap(VSYSCALL_FIRST_PAGE, physaddr_page0, PAGE_KERNEL_VSYSCALL);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int __init vsyscall_init(void)
|
|
|
|
{
|
|
|
|
BUG_ON(((unsigned long) &vgettimeofday !=
|
|
|
|
VSYSCALL_ADDR(__NR_vgettimeofday)));
|
|
|
|
BUG_ON((unsigned long) &vtime != VSYSCALL_ADDR(__NR_vtime));
|
|
|
|
BUG_ON((VSYSCALL_ADDR(0) != __fix_to_virt(VSYSCALL_FIRST_PAGE)));
|
2006-09-26 04:52:28 -04:00
|
|
|
BUG_ON((unsigned long) &vgetcpu != VSYSCALL_ADDR(__NR_vgetcpu));
|
2005-04-16 18:20:36 -04:00
|
|
|
map_vsyscall();
|
2005-05-17 00:53:33 -04:00
|
|
|
#ifdef CONFIG_SYSCTL
|
2005-04-16 18:20:36 -04:00
|
|
|
register_sysctl_table(kernel_root_table2, 0);
|
2005-05-17 00:53:33 -04:00
|
|
|
#endif
|
2006-11-14 10:57:46 -05:00
|
|
|
on_each_cpu(cpu_vsyscall_init, NULL, 0, 1);
|
|
|
|
hotcpu_notifier(cpu_vsyscall_notifier, 0);
|
2005-04-16 18:20:36 -04:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
__initcall(vsyscall_init);
|