aboutsummaryrefslogtreecommitdiff
path: root/arch/i386/kernel/process.c
diff options
context:
space:
mode:
Diffstat (limited to 'arch/i386/kernel/process.c')
-rw-r--r--arch/i386/kernel/process.c91
1 files changed, 41 insertions, 50 deletions
diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c
index 57d375900af..99308510a17 100644
--- a/arch/i386/kernel/process.c
+++ b/arch/i386/kernel/process.c
@@ -56,6 +56,7 @@
#include <asm/tlbflush.h>
#include <asm/cpu.h>
+#include <asm/pda.h>
asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
@@ -99,22 +100,18 @@ EXPORT_SYMBOL(enable_hlt);
*/
void default_idle(void)
{
- local_irq_enable();
-
if (!hlt_counter && boot_cpu_data.hlt_works_ok) {
current_thread_info()->status &= ~TS_POLLING;
smp_mb__after_clear_bit();
- while (!need_resched()) {
- local_irq_disable();
- if (!need_resched())
- safe_halt();
- else
- local_irq_enable();
- }
+ local_irq_disable();
+ if (!need_resched())
+ safe_halt(); /* enables interrupts racelessly */
+ else
+ local_irq_enable();
current_thread_info()->status |= TS_POLLING;
} else {
- while (!need_resched())
- cpu_relax();
+ /* loop is done by the caller */
+ cpu_relax();
}
}
#ifdef CONFIG_APM_MODULE
@@ -128,14 +125,7 @@ EXPORT_SYMBOL(default_idle);
*/
static void poll_idle (void)
{
- local_irq_enable();
-
- asm volatile(
- "2:"
- "testl %0, %1;"
- "rep; nop;"
- "je 2b;"
- : : "i"(_TIF_NEED_RESCHED), "m" (current_thread_info()->flags));
+ cpu_relax();
}
#ifdef CONFIG_HOTPLUG_CPU
@@ -205,7 +195,7 @@ void cpu_idle(void)
void cpu_idle_wait(void)
{
unsigned int cpu, this_cpu = get_cpu();
- cpumask_t map;
+ cpumask_t map, tmp = current->cpus_allowed;
set_cpus_allowed(current, cpumask_of_cpu(this_cpu));
put_cpu();
@@ -227,6 +217,8 @@ void cpu_idle_wait(void)
}
cpus_and(map, map, cpu_online_map);
} while (!cpus_empty(map));
+
+ set_cpus_allowed(current, tmp);
}
EXPORT_SYMBOL_GPL(cpu_idle_wait);
@@ -254,8 +246,7 @@ void mwait_idle_with_hints(unsigned long eax, unsigned long ecx)
static void mwait_idle(void)
{
local_irq_enable();
- while (!need_resched())
- mwait_idle_with_hints(0, 0);
+ mwait_idle_with_hints(0, 0);
}
void __devinit select_idle_routine(const struct cpuinfo_x86 *c)
@@ -312,8 +303,8 @@ void show_regs(struct pt_regs * regs)
regs->eax,regs->ebx,regs->ecx,regs->edx);
printk("ESI: %08lx EDI: %08lx EBP: %08lx",
regs->esi, regs->edi, regs->ebp);
- printk(" DS: %04x ES: %04x\n",
- 0xffff & regs->xds,0xffff & regs->xes);
+ printk(" DS: %04x ES: %04x GS: %04x\n",
+ 0xffff & regs->xds,0xffff & regs->xes, 0xffff & regs->xgs);
cr0 = read_cr0();
cr2 = read_cr2();
@@ -336,7 +327,6 @@ extern void kernel_thread_helper(void);
int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
{
struct pt_regs regs;
- int err;
memset(&regs, 0, sizeof(regs));
@@ -345,16 +335,14 @@ int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
regs.xds = __USER_DS;
regs.xes = __USER_DS;
+ regs.xgs = __KERNEL_PDA;
regs.orig_eax = -1;
regs.eip = (unsigned long) kernel_thread_helper;
regs.xcs = __KERNEL_CS | get_kernel_rpl();
regs.eflags = X86_EFLAGS_IF | X86_EFLAGS_SF | X86_EFLAGS_PF | 0x2;
/* Ok, create the new process.. */
- err = do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, &regs, 0, NULL, NULL);
- if (err == 0) /* terminate kernel stack */
- task_pt_regs(current)->eip = 0;
- return err;
+ return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, &regs, 0, NULL, NULL);
}
EXPORT_SYMBOL(kernel_thread);
@@ -433,7 +421,6 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long esp,
p->thread.eip = (unsigned long) ret_from_fork;
savesegment(fs,p->thread.fs);
- savesegment(gs,p->thread.gs);
tsk = current;
if (unlikely(test_tsk_thread_flag(tsk, TIF_IO_BITMAP))) {
@@ -510,7 +497,7 @@ void dump_thread(struct pt_regs * regs, struct user * dump)
dump->regs.ds = regs->xds;
dump->regs.es = regs->xes;
savesegment(fs,dump->regs.fs);
- savesegment(gs,dump->regs.gs);
+ dump->regs.gs = regs->xgs;
dump->regs.orig_eax = regs->orig_eax;
dump->regs.eip = regs->eip;
dump->regs.cs = regs->xcs;
@@ -650,22 +637,27 @@ struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct tas
__unlazy_fpu(prev_p);
+
+ /* we're going to use this soon, after a few expensive things */
+ if (next_p->fpu_counter > 5)
+ prefetch(&next->i387.fxsave);
+
/*
* Reload esp0.
*/
load_esp0(tss, next);
/*
- * Save away %fs and %gs. 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.
+ * Save away %fs. No need to save %gs, as it was saved on the
+ * 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.
*/
savesegment(fs, prev->fs);
- savesegment(gs, prev->gs);
/*
* Load the per-thread Thread-Local Storage descriptor.
@@ -673,22 +665,14 @@ struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct tas
load_TLS(next, cpu);
/*
- * Restore %fs and %gs if needed.
+ * Restore %fs if needed.
*
- * Glibc normally makes %fs be zero, and %gs is one of
- * the TLS segments.
+ * Glibc normally makes %fs be zero.
*/
if (unlikely(prev->fs | next->fs))
loadsegment(fs, next->fs);
- if (prev->gs | next->gs)
- loadsegment(gs, next->gs);
-
- /*
- * Restore IOPL if needed.
- */
- if (unlikely(prev->iopl != next->iopl))
- set_iopl_mask(next->iopl);
+ write_pda(pcurrent, next_p);
/*
* Now maybe handle debug registers and/or IO bitmaps
@@ -699,6 +683,13 @@ struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct tas
disable_tsc(prev_p, next_p);
+ /* 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
+ */
+ if (next_p->fpu_counter > 5)
+ math_state_restore();
+
return prev_p;
}