2005-04-16 18:20:36 -04:00
|
|
|
/*
|
|
|
|
* Copyright (C) 1995 Linus Torvalds
|
|
|
|
*
|
|
|
|
* Pentium III FXSR, SSE support
|
|
|
|
* Gareth Hughes <gareth@valinux.com>, May 2000
|
|
|
|
*/
|
|
|
|
|
|
|
|
/*
|
|
|
|
* This file handles the architecture-dependent parts of process handling..
|
|
|
|
*/
|
|
|
|
|
2009-02-11 02:31:00 -05:00
|
|
|
#include <linux/stackprotector.h>
|
2005-06-25 17:54:50 -04:00
|
|
|
#include <linux/cpu.h>
|
2005-04-16 18:20:36 -04:00
|
|
|
#include <linux/errno.h>
|
|
|
|
#include <linux/sched.h>
|
|
|
|
#include <linux/fs.h>
|
|
|
|
#include <linux/kernel.h>
|
|
|
|
#include <linux/mm.h>
|
|
|
|
#include <linux/elfcore.h>
|
|
|
|
#include <linux/smp.h>
|
|
|
|
#include <linux/stddef.h>
|
|
|
|
#include <linux/slab.h>
|
|
|
|
#include <linux/vmalloc.h>
|
|
|
|
#include <linux/user.h>
|
|
|
|
#include <linux/interrupt.h>
|
|
|
|
#include <linux/delay.h>
|
|
|
|
#include <linux/reboot.h>
|
|
|
|
#include <linux/init.h>
|
|
|
|
#include <linux/mc146818rtc.h>
|
|
|
|
#include <linux/module.h>
|
|
|
|
#include <linux/kallsyms.h>
|
|
|
|
#include <linux/ptrace.h>
|
2006-09-26 04:52:28 -04:00
|
|
|
#include <linux/personality.h>
|
2007-02-16 04:28:07 -05:00
|
|
|
#include <linux/tick.h>
|
2007-05-02 13:27:16 -04:00
|
|
|
#include <linux/percpu.h>
|
2008-04-13 18:24:18 -04:00
|
|
|
#include <linux/prctl.h>
|
2008-12-05 21:40:00 -05:00
|
|
|
#include <linux/ftrace.h>
|
2009-01-04 05:48:56 -05:00
|
|
|
#include <linux/uaccess.h>
|
|
|
|
#include <linux/io.h>
|
|
|
|
#include <linux/kdebug.h>
|
2005-04-16 18:20:36 -04:00
|
|
|
|
|
|
|
#include <asm/pgtable.h>
|
|
|
|
#include <asm/system.h>
|
|
|
|
#include <asm/ldt.h>
|
|
|
|
#include <asm/processor.h>
|
|
|
|
#include <asm/i387.h>
|
|
|
|
#include <asm/desc.h>
|
|
|
|
#ifdef CONFIG_MATH_EMULATION
|
|
|
|
#include <asm/math_emu.h>
|
|
|
|
#endif
|
|
|
|
|
|
|
|
#include <linux/err.h>
|
|
|
|
|
2005-06-25 17:54:50 -04:00
|
|
|
#include <asm/tlbflush.h>
|
|
|
|
#include <asm/cpu.h>
|
2008-09-23 22:40:02 -04:00
|
|
|
#include <asm/idle.h>
|
2008-07-21 12:04:13 -04:00
|
|
|
#include <asm/syscalls.h>
|
2008-12-19 09:10:24 -05:00
|
|
|
#include <asm/ds.h>
|
2009-06-01 14:14:55 -04:00
|
|
|
#include <asm/debugreg.h>
|
2005-06-25 17:54:50 -04:00
|
|
|
|
2005-04-16 18:20:36 -04:00
|
|
|
asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Return saved PC of a blocked thread.
|
|
|
|
*/
|
|
|
|
unsigned long thread_saved_pc(struct task_struct *tsk)
|
|
|
|
{
|
2008-01-30 07:31:02 -05:00
|
|
|
return ((unsigned long *)tsk->thread.sp)[3];
|
2005-04-16 18:20:36 -04:00
|
|
|
}
|
|
|
|
|
2008-09-03 09:30:23 -04:00
|
|
|
#ifndef CONFIG_SMP
|
|
|
|
static inline void play_dead(void)
|
|
|
|
{
|
|
|
|
BUG();
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2005-04-16 18:20:36 -04:00
|
|
|
/*
|
|
|
|
* The idle thread. There's no useful work to be
|
|
|
|
* done, so just try to conserve power and have a
|
|
|
|
* low exit latency (ie sit in a loop waiting for
|
|
|
|
* somebody to say that they'd like to reschedule)
|
|
|
|
*/
|
2005-06-25 17:54:50 -04:00
|
|
|
void cpu_idle(void)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
2005-11-09 00:39:01 -05:00
|
|
|
int cpu = smp_processor_id();
|
2005-06-25 17:54:50 -04:00
|
|
|
|
2009-02-11 02:31:00 -05:00
|
|
|
/*
|
|
|
|
* If we're the non-boot CPU, nothing set the stack canary up
|
|
|
|
* for us. CPU0 already has it initialized but no harm in
|
|
|
|
* doing it again. This is a good place for updating it, as
|
|
|
|
* we wont ever return from this function (so the invalid
|
|
|
|
* canaries already on the stack wont ever trigger).
|
|
|
|
*/
|
|
|
|
boot_init_stack_canary();
|
|
|
|
|
2006-06-26 07:59:11 -04:00
|
|
|
current_thread_info()->status |= TS_POLLING;
|
[PATCH] sched: resched and cpu_idle rework
Make some changes to the NEED_RESCHED and POLLING_NRFLAG to reduce
confusion, and make their semantics rigid. Improves efficiency of
resched_task and some cpu_idle routines.
* In resched_task:
- TIF_NEED_RESCHED is only cleared with the task's runqueue lock held,
and as we hold it during resched_task, then there is no need for an
atomic test and set there. The only other time this should be set is
when the task's quantum expires, in the timer interrupt - this is
protected against because the rq lock is irq-safe.
- If TIF_NEED_RESCHED is set, then we don't need to do anything. It
won't get unset until the task get's schedule()d off.
- If we are running on the same CPU as the task we resched, then set
TIF_NEED_RESCHED and no further action is required.
- If we are running on another CPU, and TIF_POLLING_NRFLAG is *not* set
after TIF_NEED_RESCHED has been set, then we need to send an IPI.
Using these rules, we are able to remove the test and set operation in
resched_task, and make clear the previously vague semantics of
POLLING_NRFLAG.
* In idle routines:
- Enter cpu_idle with preempt disabled. When the need_resched() condition
becomes true, explicitly call schedule(). This makes things a bit clearer
(IMO), but haven't updated all architectures yet.
- Many do a test and clear of TIF_NEED_RESCHED for some reason. According
to the resched_task rules, this isn't needed (and actually breaks the
assumption that TIF_NEED_RESCHED is only cleared with the runqueue lock
held). So remove that. Generally one less locked memory op when switching
to the idle thread.
- Many idle routines clear TIF_POLLING_NRFLAG, and only set it in the inner
most polling idle loops. The above resched_task semantics allow it to be
set until before the last time need_resched() is checked before going into
a halt requiring interrupt wakeup.
Many idle routines simply never enter such a halt, and so POLLING_NRFLAG
can be always left set, completely eliminating resched IPIs when rescheduling
the idle task.
POLLING_NRFLAG width can be increased, to reduce the chance of resched IPIs.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Con Kolivas <kernel@kolivas.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-09 00:39:04 -05:00
|
|
|
|
2005-04-16 18:20:36 -04:00
|
|
|
/* endless idle loop with no priority at all */
|
|
|
|
while (1) {
|
2008-07-18 11:27:28 -04:00
|
|
|
tick_nohz_stop_sched_tick(1);
|
2005-04-16 18:20:36 -04:00
|
|
|
while (!need_resched()) {
|
|
|
|
|
2007-05-12 14:15:24 -04:00
|
|
|
check_pgt_cache();
|
2005-04-16 18:20:36 -04:00
|
|
|
rmb();
|
|
|
|
|
2005-06-25 17:54:50 -04:00
|
|
|
if (cpu_is_offline(cpu))
|
|
|
|
play_dead();
|
|
|
|
|
2008-04-25 11:39:01 -04:00
|
|
|
local_irq_disable();
|
2008-05-12 15:20:42 -04:00
|
|
|
/* Don't trace irqs off for idle */
|
|
|
|
stop_critical_timings();
|
2008-06-09 10:59:53 -04:00
|
|
|
pm_idle();
|
2008-05-12 15:20:42 -04:00
|
|
|
start_critical_timings();
|
2005-04-16 18:20:36 -04:00
|
|
|
}
|
2007-02-16 04:28:07 -05:00
|
|
|
tick_nohz_restart_sched_tick();
|
2005-11-09 00:39:01 -05:00
|
|
|
preempt_enable_no_resched();
|
2005-04-16 18:20:36 -04:00
|
|
|
schedule();
|
2005-11-09 00:39:01 -05:00
|
|
|
preempt_disable();
|
2005-04-16 18:20:36 -04:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2008-04-03 09:40:48 -04:00
|
|
|
void __show_regs(struct pt_regs *regs, int all)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
|
|
|
unsigned long cr0 = 0L, cr2 = 0L, cr3 = 0L, cr4 = 0L;
|
2007-07-21 11:10:42 -04:00
|
|
|
unsigned long d0, d1, d2, d3, d6, d7;
|
2008-01-30 07:30:56 -05:00
|
|
|
unsigned long sp;
|
2007-10-19 14:35:03 -04:00
|
|
|
unsigned short ss, gs;
|
|
|
|
|
|
|
|
if (user_mode_vm(regs)) {
|
2008-01-30 07:30:56 -05:00
|
|
|
sp = regs->sp;
|
|
|
|
ss = regs->ss & 0xffff;
|
2009-02-09 08:17:40 -05:00
|
|
|
gs = get_user_gs(regs);
|
2007-10-19 14:35:03 -04:00
|
|
|
} else {
|
2009-10-12 17:09:07 -04:00
|
|
|
sp = kernel_stack_pointer(regs);
|
2007-10-19 14:35:03 -04:00
|
|
|
savesegment(ss, ss);
|
|
|
|
savesegment(gs, gs);
|
|
|
|
}
|
2005-04-16 18:20:36 -04:00
|
|
|
|
2009-12-08 03:29:42 -05:00
|
|
|
show_regs_common();
|
2007-10-19 14:35:03 -04:00
|
|
|
|
2009-12-28 03:26:59 -05:00
|
|
|
printk(KERN_DEFAULT "EIP: %04x:[<%08lx>] EFLAGS: %08lx CPU: %d\n",
|
2008-02-08 15:09:56 -05:00
|
|
|
(u16)regs->cs, regs->ip, regs->flags,
|
2007-10-19 14:35:03 -04:00
|
|
|
smp_processor_id());
|
2008-01-30 07:30:56 -05:00
|
|
|
print_symbol("EIP is at %s\n", regs->ip);
|
2005-04-16 18:20:36 -04:00
|
|
|
|
2009-12-28 03:26:59 -05:00
|
|
|
printk(KERN_DEFAULT "EAX: %08lx EBX: %08lx ECX: %08lx EDX: %08lx\n",
|
2008-01-30 07:30:56 -05:00
|
|
|
regs->ax, regs->bx, regs->cx, regs->dx);
|
2009-12-28 03:26:59 -05:00
|
|
|
printk(KERN_DEFAULT "ESI: %08lx EDI: %08lx EBP: %08lx ESP: %08lx\n",
|
2008-01-30 07:30:56 -05:00
|
|
|
regs->si, regs->di, regs->bp, sp);
|
2009-12-28 03:26:59 -05:00
|
|
|
printk(KERN_DEFAULT " DS: %04x ES: %04x FS: %04x GS: %04x SS: %04x\n",
|
2008-02-08 15:09:56 -05:00
|
|
|
(u16)regs->ds, (u16)regs->es, (u16)regs->fs, gs, ss);
|
2007-10-19 14:35:03 -04:00
|
|
|
|
|
|
|
if (!all)
|
|
|
|
return;
|
2005-04-16 18:20:36 -04:00
|
|
|
|
2005-09-03 18:56:36 -04:00
|
|
|
cr0 = read_cr0();
|
|
|
|
cr2 = read_cr2();
|
|
|
|
cr3 = read_cr3();
|
2006-01-06 03:11:50 -05:00
|
|
|
cr4 = read_cr4_safe();
|
2009-12-28 03:26:59 -05:00
|
|
|
printk(KERN_DEFAULT "CR0: %08lx CR2: %08lx CR3: %08lx CR4: %08lx\n",
|
2007-10-19 14:35:03 -04:00
|
|
|
cr0, cr2, cr3, cr4);
|
2007-07-21 11:10:42 -04:00
|
|
|
|
|
|
|
get_debugreg(d0, 0);
|
|
|
|
get_debugreg(d1, 1);
|
|
|
|
get_debugreg(d2, 2);
|
|
|
|
get_debugreg(d3, 3);
|
2009-12-28 03:26:59 -05:00
|
|
|
printk(KERN_DEFAULT "DR0: %08lx DR1: %08lx DR2: %08lx DR3: %08lx\n",
|
2007-07-21 11:10:42 -04:00
|
|
|
d0, d1, d2, d3);
|
2007-10-19 14:35:03 -04:00
|
|
|
|
2007-07-21 11:10:42 -04:00
|
|
|
get_debugreg(d6, 6);
|
|
|
|
get_debugreg(d7, 7);
|
2009-12-28 03:26:59 -05:00
|
|
|
printk(KERN_DEFAULT "DR6: %08lx DR7: %08lx\n",
|
2007-10-19 14:35:03 -04:00
|
|
|
d6, d7);
|
|
|
|
}
|
2007-07-21 11:10:42 -04:00
|
|
|
|
2007-10-19 14:35:03 -04:00
|
|
|
void show_regs(struct pt_regs *regs)
|
|
|
|
{
|
2009-11-02 19:59:15 -05:00
|
|
|
show_registers(regs);
|
2008-01-30 07:33:07 -05:00
|
|
|
show_trace(NULL, regs, ®s->sp, regs->bp);
|
2005-04-16 18:20:36 -04:00
|
|
|
}
|
|
|
|
|
|
|
|
void release_thread(struct task_struct *dead_task)
|
|
|
|
{
|
2006-01-06 03:11:59 -05:00
|
|
|
BUG_ON(dead_task->mm);
|
2005-04-16 18:20:36 -04:00
|
|
|
release_vm86_irqs(dead_task);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* This gets called before we allocate a new thread and copy
|
|
|
|
* the current task into it.
|
|
|
|
*/
|
|
|
|
void prepare_to_copy(struct task_struct *tsk)
|
|
|
|
{
|
|
|
|
unlazy_fpu(tsk);
|
|
|
|
}
|
|
|
|
|
2009-04-02 19:56:59 -04:00
|
|
|
int copy_thread(unsigned long clone_flags, unsigned long sp,
|
2005-04-16 18:20:36 -04:00
|
|
|
unsigned long unused,
|
2009-01-04 05:48:56 -05:00
|
|
|
struct task_struct *p, struct pt_regs *regs)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
2009-01-04 05:48:56 -05:00
|
|
|
struct pt_regs *childregs;
|
2005-04-16 18:20:36 -04:00
|
|
|
struct task_struct *tsk;
|
|
|
|
int err;
|
|
|
|
|
2006-01-12 04:05:41 -05:00
|
|
|
childregs = task_pt_regs(p);
|
2005-05-05 19:15:03 -04:00
|
|
|
*childregs = *regs;
|
2008-01-30 07:30:56 -05:00
|
|
|
childregs->ax = 0;
|
|
|
|
childregs->sp = sp;
|
2005-05-05 19:15:03 -04:00
|
|
|
|
2008-01-30 07:31:02 -05:00
|
|
|
p->thread.sp = (unsigned long) childregs;
|
|
|
|
p->thread.sp0 = (unsigned long) (childregs+1);
|
2005-04-16 18:20:36 -04:00
|
|
|
|
2008-01-30 07:31:02 -05:00
|
|
|
p->thread.ip = (unsigned long) ret_from_fork;
|
2005-04-16 18:20:36 -04:00
|
|
|
|
2009-02-09 08:17:40 -05:00
|
|
|
task_user_gs(p) = get_user_gs(regs);
|
2005-04-16 18:20:36 -04:00
|
|
|
|
2009-06-01 14:14:55 -04:00
|
|
|
p->thread.io_bitmap_ptr = NULL;
|
2005-04-16 18:20:36 -04:00
|
|
|
tsk = current;
|
2009-06-01 14:14:55 -04:00
|
|
|
err = -ENOMEM;
|
2009-09-09 13:22:48 -04:00
|
|
|
|
|
|
|
memset(p->thread.ptrace_bps, 0, sizeof(p->thread.ptrace_bps));
|
2009-06-01 14:14:55 -04:00
|
|
|
|
2006-07-09 21:12:39 -04:00
|
|
|
if (unlikely(test_tsk_thread_flag(tsk, TIF_IO_BITMAP))) {
|
2006-10-01 02:27:21 -04:00
|
|
|
p->thread.io_bitmap_ptr = kmemdup(tsk->thread.io_bitmap_ptr,
|
|
|
|
IO_BITMAP_BYTES, GFP_KERNEL);
|
2005-04-16 18:20:36 -04:00
|
|
|
if (!p->thread.io_bitmap_ptr) {
|
|
|
|
p->thread.io_bitmap_max = 0;
|
|
|
|
return -ENOMEM;
|
|
|
|
}
|
2006-07-09 21:12:39 -04:00
|
|
|
set_tsk_thread_flag(p, TIF_IO_BITMAP);
|
2005-04-16 18:20:36 -04:00
|
|
|
}
|
|
|
|
|
2008-01-30 07:30:46 -05:00
|
|
|
err = 0;
|
|
|
|
|
2005-04-16 18:20:36 -04:00
|
|
|
/*
|
|
|
|
* Set a new TLS for the child thread?
|
|
|
|
*/
|
2008-01-30 07:30:46 -05:00
|
|
|
if (clone_flags & CLONE_SETTLS)
|
|
|
|
err = do_set_thread_area(p, -1,
|
2008-01-30 07:30:56 -05:00
|
|
|
(struct user_desc __user *)childregs->si, 0);
|
2005-04-16 18:20:36 -04:00
|
|
|
|
|
|
|
if (err && p->thread.io_bitmap_ptr) {
|
|
|
|
kfree(p->thread.io_bitmap_ptr);
|
|
|
|
p->thread.io_bitmap_max = 0;
|
|
|
|
}
|
2008-12-19 09:10:24 -05:00
|
|
|
|
2009-04-03 10:43:46 -04:00
|
|
|
clear_tsk_thread_flag(p, TIF_DS_AREA_MSR);
|
|
|
|
p->thread.ds_ctx = NULL;
|
2008-12-19 09:10:24 -05:00
|
|
|
|
|
|
|
clear_tsk_thread_flag(p, TIF_DEBUGCTLMSR);
|
|
|
|
p->thread.debugctlmsr = 0;
|
|
|
|
|
2005-04-16 18:20:36 -04:00
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2008-02-20 23:18:40 -05:00
|
|
|
void
|
|
|
|
start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp)
|
|
|
|
{
|
2009-02-09 08:17:40 -05:00
|
|
|
set_user_gs(regs, 0);
|
2008-02-20 23:18:40 -05:00
|
|
|
regs->fs = 0;
|
|
|
|
set_fs(USER_DS);
|
|
|
|
regs->ds = __USER_DS;
|
|
|
|
regs->es = __USER_DS;
|
|
|
|
regs->ss = __USER_DS;
|
|
|
|
regs->cs = __USER_CS;
|
|
|
|
regs->ip = new_ip;
|
|
|
|
regs->sp = new_sp;
|
2008-03-10 18:28:05 -04:00
|
|
|
/*
|
|
|
|
* Free the old FP and other extended state
|
|
|
|
*/
|
|
|
|
free_thread_xstate(current);
|
2008-02-20 23:18:40 -05:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(start_thread);
|
|
|
|
|
2005-04-16 18:20:36 -04:00
|
|
|
|
|
|
|
/*
|
|
|
|
* switch_to(x,yn) should switch tasks from x to y.
|
|
|
|
*
|
|
|
|
* We fsave/fwait so that an exception goes off at the right time
|
|
|
|
* (as a call from the fsave or fwait in effect) rather than to
|
|
|
|
* the wrong process. Lazy FP saving no longer makes any sense
|
|
|
|
* with modern CPU's, and this simplifies a lot of things (SMP
|
|
|
|
* and UP become the same).
|
|
|
|
*
|
|
|
|
* NOTE! We used to use the x86 hardware context switching. The
|
|
|
|
* reason for not using it any more becomes apparent when you
|
|
|
|
* try to recover gracefully from saved state that is no longer
|
|
|
|
* valid (stale segment register values in particular). With the
|
|
|
|
* hardware task-switch, there is no way to fix up bad state in
|
|
|
|
* a reasonable manner.
|
|
|
|
*
|
|
|
|
* The fact that Intel documents the hardware task-switching to
|
|
|
|
* be slow is a fairly red herring - this code is not noticeably
|
|
|
|
* faster. However, there _is_ some room for improvement here,
|
|
|
|
* so the performance issues may eventually be a valid point.
|
|
|
|
* More important, however, is the fact that this allows us much
|
|
|
|
* more flexibility.
|
|
|
|
*
|
2008-01-30 07:30:56 -05:00
|
|
|
* The return value (in %ax) will be the "prev" task after
|
2005-04-16 18:20:36 -04:00
|
|
|
* the task-switch, and shows up in ret_from_fork in entry.S,
|
|
|
|
* for example.
|
|
|
|
*/
|
2008-12-05 21:40:00 -05:00
|
|
|
__notrace_funcgraph struct task_struct *
|
|
|
|
__switch_to(struct task_struct *prev_p, struct task_struct *next_p)
|
2005-04-16 18:20:36 -04:00
|
|
|
{
|
|
|
|
struct thread_struct *prev = &prev_p->thread,
|
|
|
|
*next = &next_p->thread;
|
|
|
|
int cpu = smp_processor_id();
|
|
|
|
struct tss_struct *tss = &per_cpu(init_tss, cpu);
|
2009-04-24 03:45:26 -04:00
|
|
|
bool preload_fpu;
|
2005-04-16 18:20:36 -04:00
|
|
|
|
|
|
|
/* never put a printk in __switch_to... printk() calls wake_up*() indirectly */
|
|
|
|
|
2009-04-24 03:45:26 -04:00
|
|
|
/*
|
|
|
|
* If the task has used fpu the last 5 timeslices, just do a full
|
|
|
|
* restore of the math state immediately to avoid the trap; the
|
|
|
|
* chances of needing FPU soon are obviously high now
|
|
|
|
*/
|
|
|
|
preload_fpu = tsk_used_math(next_p) && next_p->fpu_counter > 5;
|
2005-04-16 18:20:36 -04:00
|
|
|
|
2009-04-24 03:45:26 -04:00
|
|
|
__unlazy_fpu(prev_p);
|
2006-12-06 20:14:01 -05:00
|
|
|
|
|
|
|
/* we're going to use this soon, after a few expensive things */
|
2009-04-24 03:45:26 -04:00
|
|
|
if (preload_fpu)
|
2008-03-10 18:28:04 -04:00
|
|
|
prefetch(next->xstate);
|
2006-12-06 20:14:01 -05:00
|
|
|
|
2005-04-16 18:20:36 -04:00
|
|
|
/*
|
2005-09-03 18:56:39 -04:00
|
|
|
* Reload esp0.
|
2005-04-16 18:20:36 -04:00
|
|
|
*/
|
2008-01-30 07:31:02 -05:00
|
|
|
load_sp0(tss, next);
|
2005-04-16 18:20:36 -04:00
|
|
|
|
|
|
|
/*
|
2007-02-13 07:26:20 -05:00
|
|
|
* Save away %gs. No need to save %fs, as it was saved on the
|
[PATCH] i386: Use %gs as the PDA base-segment in the kernel
This patch is the meat of the PDA change. This patch makes several related
changes:
1: Most significantly, %gs is now used in the kernel. This means that on
entry, the old value of %gs is saved away, and it is reloaded with
__KERNEL_PDA.
2: entry.S constructs the stack in the shape of struct pt_regs, and this
is passed around the kernel so that the process's saved register
state can be accessed.
Unfortunately struct pt_regs doesn't currently have space for %gs
(or %fs). This patch extends pt_regs to add space for gs (no space
is allocated for %fs, since it won't be used, and it would just
complicate the code in entry.S to work around the space).
3: Because %gs is now saved on the stack like %ds, %es and the integer
registers, there are a number of places where it no longer needs to
be handled specially; namely context switch, and saving/restoring the
register state in a signal context.
4: And since kernel threads run in kernel space and call normal kernel
code, they need to be created with their %gs == __KERNEL_PDA.
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Chuck Ebbert <76306.1226@compuserve.com>
Cc: Zachary Amsden <zach@vmware.com>
Cc: Jan Beulich <jbeulich@novell.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
2006-12-06 20:14:02 -05:00
|
|
|
* stack on entry. No need to save %es and %ds, as those are
|
|
|
|
* always kernel segments while inside the kernel. Doing this
|
|
|
|
* before setting the new TLS descriptors avoids the situation
|
|
|
|
* where we temporarily have non-reloadable segments in %fs
|
|
|
|
* and %gs. This could be an issue if the NMI handler ever
|
|
|
|
* used %fs or %gs (it does not today), or if the kernel is
|
|
|
|
* running inside of a hypervisor layer.
|
2005-04-16 18:20:36 -04:00
|
|
|
*/
|
2009-02-09 08:17:40 -05:00
|
|
|
lazy_save_gs(prev->gs);
|
2005-04-16 18:20:36 -04:00
|
|
|
|
|
|
|
/*
|
2005-09-03 18:56:39 -04:00
|
|
|
* Load the per-thread Thread-Local Storage descriptor.
|
2005-04-16 18:20:36 -04:00
|
|
|
*/
|
2005-09-03 18:56:39 -04:00
|
|
|
load_TLS(next, cpu);
|
2005-04-16 18:20:36 -04:00
|
|
|
|
2007-02-13 07:26:21 -05:00
|
|
|
/*
|
|
|
|
* Restore IOPL if needed. In normal use, the flags restore
|
|
|
|
* in the switch assembly will handle this. But if the kernel
|
|
|
|
* is running virtualized at a non-zero CPL, the popf will
|
|
|
|
* not restore flags, so it must be done in a separate step.
|
|
|
|
*/
|
|
|
|
if (get_kernel_rpl() && unlikely(prev->iopl != next->iopl))
|
|
|
|
set_iopl_mask(next->iopl);
|
|
|
|
|
2005-04-16 18:20:36 -04:00
|
|
|
/*
|
2006-07-09 21:12:39 -04:00
|
|
|
* Now maybe handle debug registers and/or IO bitmaps
|
2005-04-16 18:20:36 -04:00
|
|
|
*/
|
2007-07-16 02:41:33 -04:00
|
|
|
if (unlikely(task_thread_info(prev_p)->flags & _TIF_WORK_CTXSW_PREV ||
|
|
|
|
task_thread_info(next_p)->flags & _TIF_WORK_CTXSW_NEXT))
|
|
|
|
__switch_to_xtra(prev_p, next_p, tss);
|
2005-06-27 17:36:36 -04:00
|
|
|
|
2009-04-24 03:45:26 -04:00
|
|
|
/* If we're going to preload the fpu context, make sure clts
|
|
|
|
is run while we're batching the cpu state updates. */
|
|
|
|
if (preload_fpu)
|
|
|
|
clts();
|
|
|
|
|
2007-02-13 07:26:21 -05:00
|
|
|
/*
|
|
|
|
* Leave lazy mode, flushing any hypercalls made here.
|
|
|
|
* This must be done before restoring TLS segments so
|
|
|
|
* the GDT and LDT are properly updated, and must be
|
|
|
|
* done before math_state_restore, so the TS bit is up
|
|
|
|
* to date.
|
|
|
|
*/
|
2009-02-18 14:18:57 -05:00
|
|
|
arch_end_context_switch(next_p);
|
2007-02-13 07:26:21 -05:00
|
|
|
|
2009-04-24 03:45:26 -04:00
|
|
|
if (preload_fpu)
|
|
|
|
__math_state_restore();
|
2006-12-06 20:14:01 -05:00
|
|
|
|
2007-02-13 07:26:21 -05:00
|
|
|
/*
|
|
|
|
* Restore %gs if needed (which is common)
|
|
|
|
*/
|
|
|
|
if (prev->gs | next->gs)
|
2009-02-09 08:17:40 -05:00
|
|
|
lazy_load_gs(next->gs);
|
2007-02-13 07:26:21 -05:00
|
|
|
|
percpu: add optimized generic percpu accessors
It is an optimization and a cleanup, and adds the following new
generic percpu methods:
percpu_read()
percpu_write()
percpu_add()
percpu_sub()
percpu_and()
percpu_or()
percpu_xor()
and implements support for them on x86. (other architectures will fall
back to a default implementation)
The advantage is that for example to read a local percpu variable,
instead of this sequence:
return __get_cpu_var(var);
ffffffff8102ca2b: 48 8b 14 fd 80 09 74 mov -0x7e8bf680(,%rdi,8),%rdx
ffffffff8102ca32: 81
ffffffff8102ca33: 48 c7 c0 d8 59 00 00 mov $0x59d8,%rax
ffffffff8102ca3a: 48 8b 04 10 mov (%rax,%rdx,1),%rax
We can get a single instruction by using the optimized variants:
return percpu_read(var);
ffffffff8102ca3f: 65 48 8b 05 91 8f fd mov %gs:0x7efd8f91(%rip),%rax
I also cleaned up the x86-specific APIs and made the x86 code use
these new generic percpu primitives.
tj: * fixed generic percpu_sub() definition as Roel Kluin pointed out
* added percpu_and() for completeness's sake
* made generic percpu ops atomic against preemption
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Tejun Heo <tj@kernel.org>
2009-01-15 08:15:53 -05:00
|
|
|
percpu_write(current_task, next_p);
|
2007-02-13 07:26:21 -05:00
|
|
|
|
2005-04-16 18:20:36 -04:00
|
|
|
return prev_p;
|
|
|
|
}
|
|
|
|
|
|
|
|
#define top_esp (THREAD_SIZE - sizeof(unsigned long))
|
|
|
|
#define top_ebp (THREAD_SIZE - 2*sizeof(unsigned long))
|
|
|
|
|
|
|
|
unsigned long get_wchan(struct task_struct *p)
|
|
|
|
{
|
2008-01-30 07:30:56 -05:00
|
|
|
unsigned long bp, sp, ip;
|
2005-04-16 18:20:36 -04:00
|
|
|
unsigned long stack_page;
|
|
|
|
int count = 0;
|
|
|
|
if (!p || p == current || p->state == TASK_RUNNING)
|
|
|
|
return 0;
|
2006-01-12 04:05:41 -05:00
|
|
|
stack_page = (unsigned long)task_stack_page(p);
|
2008-01-30 07:31:02 -05:00
|
|
|
sp = p->thread.sp;
|
2008-01-30 07:30:56 -05:00
|
|
|
if (!stack_page || sp < stack_page || sp > top_esp+stack_page)
|
2005-04-16 18:20:36 -04:00
|
|
|
return 0;
|
2008-01-30 07:30:56 -05:00
|
|
|
/* include/asm-i386/system.h:switch_to() pushes bp last. */
|
|
|
|
bp = *(unsigned long *) sp;
|
2005-04-16 18:20:36 -04:00
|
|
|
do {
|
2008-01-30 07:30:56 -05:00
|
|
|
if (bp < stack_page || bp > top_ebp+stack_page)
|
2005-04-16 18:20:36 -04:00
|
|
|
return 0;
|
2008-01-30 07:30:56 -05:00
|
|
|
ip = *(unsigned long *) (bp+4);
|
|
|
|
if (!in_sched_functions(ip))
|
|
|
|
return ip;
|
|
|
|
bp = *(unsigned long *) bp;
|
2005-04-16 18:20:36 -04:00
|
|
|
} while (count++ < 16);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|