asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
- static int hlt_counter;
-
- unsigned long boot_option_idle_override = 0;
- EXPORT_SYMBOL(boot_option_idle_override);
-
DEFINE_PER_CPU(struct task_struct *, current_task) = &init_task;
EXPORT_PER_CPU_SYMBOL(current_task);
return ((unsigned long *)tsk->thread.sp)[3];
}
- /*
- * Powermanagement idle function, if any..
- */
- void (*pm_idle)(void);
- EXPORT_SYMBOL(pm_idle);
+ #ifdef CONFIG_HOTPLUG_CPU
+ #include <asm/nmi.h>
- void disable_hlt(void)
+ static void cpu_exit_clear(void)
{
- hlt_counter++;
- }
+ int cpu = raw_smp_processor_id();
- EXPORT_SYMBOL(disable_hlt);
+ idle_task_exit();
- void enable_hlt(void)
- {
- hlt_counter--;
- }
+ cpu_uninit();
+ irq_ctx_exit(cpu);
- EXPORT_SYMBOL(enable_hlt);
+ cpu_clear(cpu, cpu_callout_map);
+ cpu_clear(cpu, cpu_callin_map);
- /*
- * We use this if we don't have any better
- * idle routine..
- */
- void default_idle(void)
- {
- if (!hlt_counter && boot_cpu_data.hlt_works_ok) {
- current_thread_info()->status &= ~TS_POLLING;
- /*
- * TS_POLLING-cleared state must be visible before we
- * test NEED_RESCHED:
- */
- smp_mb();
-
- if (!need_resched())
- safe_halt(); /* enables interrupts racelessly */
- else
- local_irq_enable();
- current_thread_info()->status |= TS_POLLING;
- } else {
- local_irq_enable();
- /* loop is done by the caller */
- cpu_relax();
- }
+ numa_remove_cpu(cpu);
}
- #ifdef CONFIG_APM_MODULE
- EXPORT_SYMBOL(default_idle);
- #endif
- #ifdef CONFIG_HOTPLUG_CPU
- #include <asm/nmi.h>
/* We don't actually take CPU down, just spin without interrupts. */
static inline void play_dead(void)
{
/* endless idle loop with no priority at all */
while (1) {
- tick_nohz_stop_sched_tick();
+ tick_nohz_stop_sched_tick(1);
while (!need_resched()) {
- void (*idle)(void);
check_pgt_cache();
rmb();
- idle = pm_idle;
if (rcu_pending(cpu))
rcu_check_callbacks(cpu, 0);
- if (!idle)
- idle = default_idle;
-
if (cpu_is_offline(cpu))
play_dead();
local_irq_disable();
__get_cpu_var(irq_stat).idle_timestamp = jiffies;
- idle();
+ /* Don't trace irqs off for idle */
+ stop_critical_timings();
+ pm_idle();
+ start_critical_timings();
}
tick_nohz_restart_sched_tick();
preempt_enable_no_resched();
tss->x86_tss.io_bitmap_base = INVALID_IO_BITMAP_OFFSET;
put_cpu();
}
+#ifdef CONFIG_X86_DS
+ /* Free any DS contexts that have not been properly released. */
+ if (unlikely(current->thread.ds_ctx)) {
+ /* we clear debugctl to make sure DS is not used. */
+ update_debugctlmsr(0);
+ ds_free(current->thread.ds_ctx);
+ }
+#endif /* CONFIG_X86_DS */
}
void flush_thread(void)
return 0;
}
+#ifdef CONFIG_X86_DS
+static int update_debugctl(struct thread_struct *prev,
+ struct thread_struct *next, unsigned long debugctl)
+{
+ unsigned long ds_prev = 0;
+ unsigned long ds_next = 0;
+
+ if (prev->ds_ctx)
+ ds_prev = (unsigned long)prev->ds_ctx->ds;
+ if (next->ds_ctx)
+ ds_next = (unsigned long)next->ds_ctx->ds;
+
+ if (ds_next != ds_prev) {
+ /* we clear debugctl to make sure DS
+ * is not in use when we change it */
+ debugctl = 0;
+ update_debugctlmsr(0);
+ wrmsr(MSR_IA32_DS_AREA, ds_next, 0);
+ }
+ return debugctl;
+}
+#else
+static int update_debugctl(struct thread_struct *prev,
+ struct thread_struct *next, unsigned long debugctl)
+{
+ return debugctl;
+}
+#endif /* CONFIG_X86_DS */
+
static noinline void
__switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
struct tss_struct *tss)
prev = &prev_p->thread;
next = &next_p->thread;
- debugctl = prev->debugctlmsr;
- if (next->ds_area_msr != prev->ds_area_msr) {
- /* we clear debugctl to make sure DS
- * is not in use when we change it */
- debugctl = 0;
- update_debugctlmsr(0);
- wrmsr(MSR_IA32_DS_AREA, next->ds_area_msr, 0);
- }
+ debugctl = update_debugctl(prev, next, prev->debugctlmsr);
if (next->debugctlmsr != debugctl)
update_debugctlmsr(next->debugctlmsr);
hard_enable_TSC();
}
-#ifdef X86_BTS
+#ifdef CONFIG_X86_PTRACE_BTS
if (test_tsk_thread_flag(prev_p, TIF_BTS_TRACE_TS))
ptrace_bts_take_timestamp(prev_p, BTS_TASK_DEPARTS);
if (test_tsk_thread_flag(next_p, TIF_BTS_TRACE_TS))
ptrace_bts_take_timestamp(next_p, BTS_TASK_ARRIVES);
-#endif
+#endif /* CONFIG_X86_PTRACE_BTS */
if (!test_tsk_thread_flag(next_p, TIF_IO_BITMAP)) {
unsigned long kernel_thread_flags = CLONE_VM | CLONE_UNTRACED;
- unsigned long boot_option_idle_override = 0;
- EXPORT_SYMBOL(boot_option_idle_override);
-
- /*
- * Powermanagement idle function, if any..
- */
- void (*pm_idle)(void);
- EXPORT_SYMBOL(pm_idle);
-
static ATOMIC_NOTIFIER_HEAD(idle_notifier);
void idle_notifier_register(struct notifier_block *n)
__exit_idle();
}
- /*
- * We use this if we don't have any better
- * idle routine..
- */
- void default_idle(void)
- {
- current_thread_info()->status &= ~TS_POLLING;
- /*
- * TS_POLLING-cleared state must be visible before we
- * test NEED_RESCHED:
- */
- smp_mb();
- if (!need_resched())
- safe_halt(); /* enables interrupts racelessly */
- else
- local_irq_enable();
- current_thread_info()->status |= TS_POLLING;
- }
-
#ifdef CONFIG_HOTPLUG_CPU
DECLARE_PER_CPU(int, cpu_state);
current_thread_info()->status |= TS_POLLING;
/* endless idle loop with no priority at all */
while (1) {
- tick_nohz_stop_sched_tick();
+ tick_nohz_stop_sched_tick(1);
while (!need_resched()) {
- void (*idle)(void);
rmb();
- idle = pm_idle;
- if (!idle)
- idle = default_idle;
+
if (cpu_is_offline(smp_processor_id()))
play_dead();
/*
*/
local_irq_disable();
enter_idle();
- idle();
+ /* Don't trace irqs off for idle */
+ stop_critical_timings();
+ pm_idle();
+ start_critical_timings();
/* In many cases the interrupt that ended idle
has already called exit_idle. But some idle
loops can be woken up without interrupt. */
t->io_bitmap_max = 0;
put_cpu();
}
+#ifdef CONFIG_X86_DS
+ /* Free any DS contexts that have not been properly released. */
+ if (unlikely(t->ds_ctx)) {
+ /* we clear debugctl to make sure DS is not used. */
+ update_debugctlmsr(0);
+ ds_free(t->ds_ctx);
+ }
+#endif /* CONFIG_X86_DS */
}
void flush_thread(void)
p->thread.fs = me->thread.fs;
p->thread.gs = me->thread.gs;
- asm("mov %%gs,%0" : "=m" (p->thread.gsindex));
- asm("mov %%fs,%0" : "=m" (p->thread.fsindex));
- asm("mov %%es,%0" : "=m" (p->thread.es));
- asm("mov %%ds,%0" : "=m" (p->thread.ds));
+ savesegment(gs, p->thread.gsindex);
+ savesegment(fs, p->thread.fsindex);
+ savesegment(es, p->thread.es);
+ savesegment(ds, p->thread.ds);
if (unlikely(test_tsk_thread_flag(me, TIF_IO_BITMAP))) {
p->thread.io_bitmap_ptr = kmalloc(IO_BITMAP_BYTES, GFP_KERNEL);
void
start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp)
{
- asm volatile("movl %0, %%fs; movl %0, %%es; movl %0, %%ds" :: "r"(0));
+ loadsegment(fs, 0);
+ loadsegment(es, 0);
+ loadsegment(ds, 0);
load_gs_index(0);
regs->ip = new_ip;
regs->sp = new_sp;
next = &next_p->thread;
debugctl = prev->debugctlmsr;
- if (next->ds_area_msr != prev->ds_area_msr) {
- /* we clear debugctl to make sure DS
- * is not in use when we change it */
- debugctl = 0;
- update_debugctlmsr(0);
- wrmsrl(MSR_IA32_DS_AREA, next->ds_area_msr);
+
+#ifdef CONFIG_X86_DS
+ {
+ unsigned long ds_prev = 0, ds_next = 0;
+
+ if (prev->ds_ctx)
+ ds_prev = (unsigned long)prev->ds_ctx->ds;
+ if (next->ds_ctx)
+ ds_next = (unsigned long)next->ds_ctx->ds;
+
+ if (ds_next != ds_prev) {
+ /*
+ * We clear debugctl to make sure DS
+ * is not in use when we change it:
+ */
+ debugctl = 0;
+ update_debugctlmsr(0);
+ wrmsrl(MSR_IA32_DS_AREA, ds_next);
+ }
}
+#endif /* CONFIG_X86_DS */
if (next->debugctlmsr != debugctl)
update_debugctlmsr(next->debugctlmsr);
memset(tss->io_bitmap, 0xff, prev->io_bitmap_max);
}
-#ifdef X86_BTS
+#ifdef CONFIG_X86_PTRACE_BTS
if (test_tsk_thread_flag(prev_p, TIF_BTS_TRACE_TS))
ptrace_bts_take_timestamp(prev_p, BTS_TASK_DEPARTS);
if (test_tsk_thread_flag(next_p, TIF_BTS_TRACE_TS))
ptrace_bts_take_timestamp(next_p, BTS_TASK_ARRIVES);
-#endif
+#endif /* CONFIG_X86_PTRACE_BTS */
}
/*
struct task_struct *
__switch_to(struct task_struct *prev_p, struct task_struct *next_p)
{
- struct thread_struct *prev = &prev_p->thread,
- *next = &next_p->thread;
+ struct thread_struct *prev = &prev_p->thread;
+ struct thread_struct *next = &next_p->thread;
int cpu = smp_processor_id();
struct tss_struct *tss = &per_cpu(init_tss, cpu);
+ unsigned fsindex, gsindex;
/* we're going to use this soon, after a few expensive things */
if (next_p->fpu_counter>5)
* Switch DS and ES.
* This won't pick up thread selector changes, but I guess that is ok.
*/
- asm volatile("mov %%es,%0" : "=m" (prev->es));
+ savesegment(es, prev->es);
if (unlikely(next->es | prev->es))
loadsegment(es, next->es);
-
- asm volatile ("mov %%ds,%0" : "=m" (prev->ds));
+
+ savesegment(ds, prev->ds);
if (unlikely(next->ds | prev->ds))
loadsegment(ds, next->ds);
+
+ /* We must save %fs and %gs before load_TLS() because
+ * %fs and %gs may be cleared by load_TLS().
+ *
+ * (e.g. xen_load_tls())
+ */
+ savesegment(fs, fsindex);
+ savesegment(gs, gsindex);
+
load_TLS(next, cpu);
+ /*
+ * 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.
+ */
+ arch_leave_lazy_cpu_mode();
+
/*
* Switch FS and GS.
+ *
+ * Segment register != 0 always requires a reload. Also
+ * reload when it has changed. When prev process used 64bit
+ * base always reload to avoid an information leak.
*/
- {
- unsigned fsindex;
- asm volatile("movl %%fs,%0" : "=r" (fsindex));
- /* segment register != 0 always requires a reload.
- also reload when it has changed.
- when prev process used 64bit base always reload
- to avoid an information leak. */
- if (unlikely(fsindex | next->fsindex | prev->fs)) {
- loadsegment(fs, next->fsindex);
- /* check if the user used a selector != 0
- * if yes clear 64bit base, since overloaded base
- * is always mapped to the Null selector
- */
- if (fsindex)
+ if (unlikely(fsindex | next->fsindex | prev->fs)) {
+ loadsegment(fs, next->fsindex);
+ /*
+ * Check if the user used a selector != 0; if yes
+ * clear 64bit base, since overloaded base is always
+ * mapped to the Null selector
+ */
+ if (fsindex)
prev->fs = 0;
- }
- /* when next process has a 64bit base use it */
- if (next->fs)
- wrmsrl(MSR_FS_BASE, next->fs);
- prev->fsindex = fsindex;
}
- {
- unsigned gsindex;
- asm volatile("movl %%gs,%0" : "=r" (gsindex));
- if (unlikely(gsindex | next->gsindex | prev->gs)) {
- load_gs_index(next->gsindex);
- if (gsindex)
+ /* when next process has a 64bit base use it */
+ if (next->fs)
+ wrmsrl(MSR_FS_BASE, next->fs);
+ prev->fsindex = fsindex;
+
+ if (unlikely(gsindex | next->gsindex | prev->gs)) {
+ load_gs_index(next->gsindex);
+ if (gsindex)
prev->gs = 0;
- }
- if (next->gs)
- wrmsrl(MSR_KERNEL_GS_BASE, next->gs);
- prev->gsindex = gsindex;
}
+ if (next->gs)
+ wrmsrl(MSR_KERNEL_GS_BASE, next->gs);
+ prev->gsindex = gsindex;
/* Must be after DS reload */
unlazy_fpu(prev_p);
write_pda(pcurrent, next_p);
write_pda(kernelstack,
- (unsigned long)task_stack_page(next_p) + THREAD_SIZE - PDA_STACKOFFSET);
+ (unsigned long)task_stack_page(next_p) +
+ THREAD_SIZE - PDA_STACKOFFSET);
#ifdef CONFIG_CC_STACKPROTECTOR
write_pda(stack_canary, next_p->stack_canary);
/*
set_32bit_tls(task, FS_TLS, addr);
if (doit) {
load_TLS(&task->thread, cpu);
- asm volatile("movl %0,%%fs" :: "r"(FS_TLS_SEL));
+ loadsegment(fs, FS_TLS_SEL);
}
task->thread.fsindex = FS_TLS_SEL;
task->thread.fs = 0;
if (doit) {
/* set the selector to 0 to not confuse
__switch_to */
- asm volatile("movl %0,%%fs" :: "r" (0));
+ loadsegment(fs, 0);
ret = checking_wrmsrl(MSR_FS_BASE, addr);
}
}
if (task->thread.gsindex == GS_TLS_SEL)
base = read_32bit_tls(task, GS_TLS);
else if (doit) {
- asm("movl %%gs,%0" : "=r" (gsindex));
+ savesegment(gs, gsindex);
if (gsindex)
rdmsrl(MSR_KERNEL_GS_BASE, base);
else
return 0;
}
-#ifdef X86_BTS
+#ifdef CONFIG_X86_PTRACE_BTS
+/*
+ * The configuration for a particular BTS hardware implementation.
+ */
+struct bts_configuration {
+ /* the size of a BTS record in bytes; at most BTS_MAX_RECORD_SIZE */
+ unsigned char sizeof_bts;
+ /* the size of a field in the BTS record in bytes */
+ unsigned char sizeof_field;
+ /* a bitmask to enable/disable BTS in DEBUGCTL MSR */
+ unsigned long debugctl_mask;
+};
+static struct bts_configuration bts_cfg;
+
+#define BTS_MAX_RECORD_SIZE (8 * 3)
+
+
+/*
+ * Branch Trace Store (BTS) uses the following format. Different
+ * architectures vary in the size of those fields.
+ * - source linear address
+ * - destination linear address
+ * - flags
+ *
+ * Later architectures use 64bit pointers throughout, whereas earlier
+ * architectures use 32bit pointers in 32bit mode.
+ *
+ * We compute the base address for the first 8 fields based on:
+ * - the field size stored in the DS configuration
+ * - the relative field position
+ *
+ * In order to store additional information in the BTS buffer, we use
+ * a special source address to indicate that the record requires
+ * special interpretation.
+ *
+ * Netburst indicated via a bit in the flags field whether the branch
+ * was predicted; this is ignored.
+ */
+
+enum bts_field {
+ bts_from = 0,
+ bts_to,
+ bts_flags,
+
+ bts_escape = (unsigned long)-1,
+ bts_qual = bts_to,
+ bts_jiffies = bts_flags
+};
-static int ptrace_bts_get_size(struct task_struct *child)
+static inline unsigned long bts_get(const char *base, enum bts_field field)
{
- if (!child->thread.ds_area_msr)
- return -ENXIO;
+ base += (bts_cfg.sizeof_field * field);
+ return *(unsigned long *)base;
+}
+
+static inline void bts_set(char *base, enum bts_field field, unsigned long val)
+{
+ base += (bts_cfg.sizeof_field * field);;
+ (*(unsigned long *)base) = val;
+}
- return ds_get_bts_index((void *)child->thread.ds_area_msr);
+/*
+ * Translate a BTS record from the raw format into the bts_struct format
+ *
+ * out (out): bts_struct interpretation
+ * raw: raw BTS record
+ */
+static void ptrace_bts_translate_record(struct bts_struct *out, const void *raw)
+{
+ memset(out, 0, sizeof(*out));
+ if (bts_get(raw, bts_from) == bts_escape) {
+ out->qualifier = bts_get(raw, bts_qual);
+ out->variant.jiffies = bts_get(raw, bts_jiffies);
+ } else {
+ out->qualifier = BTS_BRANCH;
+ out->variant.lbr.from_ip = bts_get(raw, bts_from);
+ out->variant.lbr.to_ip = bts_get(raw, bts_to);
+ }
}
-static int ptrace_bts_read_record(struct task_struct *child,
- long index,
+static int ptrace_bts_read_record(struct task_struct *child, size_t index,
struct bts_struct __user *out)
{
struct bts_struct ret;
- int retval;
- int bts_end;
- int bts_index;
+ const void *bts_record;
+ size_t bts_index, bts_end;
+ int error;
- if (!child->thread.ds_area_msr)
- return -ENXIO;
+ error = ds_get_bts_end(child, &bts_end);
+ if (error < 0)
+ return error;
- if (index < 0)
- return -EINVAL;
-
- bts_end = ds_get_bts_end((void *)child->thread.ds_area_msr);
if (bts_end <= index)
return -EINVAL;
+ error = ds_get_bts_index(child, &bts_index);
+ if (error < 0)
+ return error;
+
/* translate the ptrace bts index into the ds bts index */
- bts_index = ds_get_bts_index((void *)child->thread.ds_area_msr);
- bts_index -= (index + 1);
- if (bts_index < 0)
- bts_index += bts_end;
+ bts_index += bts_end - (index + 1);
+ if (bts_end <= bts_index)
+ bts_index -= bts_end;
+
+ error = ds_access_bts(child, bts_index, &bts_record);
+ if (error < 0)
+ return error;
- retval = ds_read_bts((void *)child->thread.ds_area_msr,
- bts_index, &ret);
- if (retval < 0)
- return retval;
+ ptrace_bts_translate_record(&ret, bts_record);
if (copy_to_user(out, &ret, sizeof(ret)))
return -EFAULT;
return sizeof(ret);
}
-static int ptrace_bts_clear(struct task_struct *child)
-{
- if (!child->thread.ds_area_msr)
- return -ENXIO;
-
- return ds_clear((void *)child->thread.ds_area_msr);
-}
-
static int ptrace_bts_drain(struct task_struct *child,
long size,
struct bts_struct __user *out)
{
- int end, i;
- void *ds = (void *)child->thread.ds_area_msr;
-
- if (!ds)
- return -ENXIO;
+ struct bts_struct ret;
+ const unsigned char *raw;
+ size_t end, i;
+ int error;
- end = ds_get_bts_index(ds);
- if (end <= 0)
- return end;
+ error = ds_get_bts_index(child, &end);
+ if (error < 0)
+ return error;
if (size < (end * sizeof(struct bts_struct)))
return -EIO;
- for (i = 0; i < end; i++, out++) {
- struct bts_struct ret;
- int retval;
+ error = ds_access_bts(child, 0, (const void **)&raw);
+ if (error < 0)
+ return error;
- retval = ds_read_bts(ds, i, &ret);
- if (retval < 0)
- return retval;
+ for (i = 0; i < end; i++, out++, raw += bts_cfg.sizeof_bts) {
+ ptrace_bts_translate_record(&ret, raw);
if (copy_to_user(out, &ret, sizeof(ret)))
return -EFAULT;
}
- ds_clear(ds);
+ error = ds_clear_bts(child);
+ if (error < 0)
+ return error;
return end;
}
+static void ptrace_bts_ovfl(struct task_struct *child)
+{
+ send_sig(child->thread.bts_ovfl_signal, child, 0);
+}
+
static int ptrace_bts_config(struct task_struct *child,
long cfg_size,
const struct ptrace_bts_config __user *ucfg)
{
struct ptrace_bts_config cfg;
- int bts_size, ret = 0;
- void *ds;
+ int error = 0;
+
+ error = -EOPNOTSUPP;
+ if (!bts_cfg.sizeof_bts)
+ goto errout;
+ error = -EIO;
if (cfg_size < sizeof(cfg))
- return -EIO;
+ goto errout;
+ error = -EFAULT;
if (copy_from_user(&cfg, ucfg, sizeof(cfg)))
- return -EFAULT;
+ goto errout;
- if ((int)cfg.size < 0)
- return -EINVAL;
+ error = -EINVAL;
+ if ((cfg.flags & PTRACE_BTS_O_SIGNAL) &&
+ !(cfg.flags & PTRACE_BTS_O_ALLOC))
+ goto errout;
- bts_size = 0;
- ds = (void *)child->thread.ds_area_msr;
- if (ds) {
- bts_size = ds_get_bts_size(ds);
- if (bts_size < 0)
- return bts_size;
- }
- cfg.size = PAGE_ALIGN(cfg.size);
+ if (cfg.flags & PTRACE_BTS_O_ALLOC) {
+ ds_ovfl_callback_t ovfl = 0;
+ unsigned int sig = 0;
+
+ /* we ignore the error in case we were not tracing child */
+ (void)ds_release_bts(child);
+
+ if (cfg.flags & PTRACE_BTS_O_SIGNAL) {
+ if (!cfg.signal)
+ goto errout;
+
+ sig = cfg.signal;
+ ovfl = ptrace_bts_ovfl;
+ }
- if (bts_size != cfg.size) {
- ret = ptrace_bts_realloc(child, cfg.size,
- cfg.flags & PTRACE_BTS_O_CUT_SIZE);
- if (ret < 0)
+ error = ds_request_bts(child, /* base = */ 0, cfg.size, ovfl);
+ if (error < 0)
goto errout;
- ds = (void *)child->thread.ds_area_msr;
+ child->thread.bts_ovfl_signal = sig;
}
- if (cfg.flags & PTRACE_BTS_O_SIGNAL)
- ret = ds_set_overflow(ds, DS_O_SIGNAL);
- else
- ret = ds_set_overflow(ds, DS_O_WRAP);
- if (ret < 0)
+ error = -EINVAL;
+ if (!child->thread.ds_ctx && cfg.flags)
goto errout;
if (cfg.flags & PTRACE_BTS_O_TRACE)
- child->thread.debugctlmsr |= ds_debugctl_mask();
+ child->thread.debugctlmsr |= bts_cfg.debugctl_mask;
else
- child->thread.debugctlmsr &= ~ds_debugctl_mask();
+ child->thread.debugctlmsr &= ~bts_cfg.debugctl_mask;
if (cfg.flags & PTRACE_BTS_O_SCHED)
set_tsk_thread_flag(child, TIF_BTS_TRACE_TS);
else
clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS);
- ret = sizeof(cfg);
+ error = sizeof(cfg);
out:
if (child->thread.debugctlmsr)
else
clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
- return ret;
+ return error;
errout:
- child->thread.debugctlmsr &= ~ds_debugctl_mask();
+ child->thread.debugctlmsr &= ~bts_cfg.debugctl_mask;
clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS);
goto out;
}
long cfg_size,
struct ptrace_bts_config __user *ucfg)
{
- void *ds = (void *)child->thread.ds_area_msr;
struct ptrace_bts_config cfg;
+ size_t end;
+ const void *base, *max;
+ int error;
if (cfg_size < sizeof(cfg))
return -EIO;
- memset(&cfg, 0, sizeof(cfg));
+ error = ds_get_bts_end(child, &end);
+ if (error < 0)
+ return error;
- if (ds) {
- cfg.size = ds_get_bts_size(ds);
+ error = ds_access_bts(child, /* index = */ 0, &base);
+ if (error < 0)
+ return error;
- if (ds_get_overflow(ds) == DS_O_SIGNAL)
- cfg.flags |= PTRACE_BTS_O_SIGNAL;
+ error = ds_access_bts(child, /* index = */ end, &max);
+ if (error < 0)
+ return error;
- if (test_tsk_thread_flag(child, TIF_DEBUGCTLMSR) &&
- child->thread.debugctlmsr & ds_debugctl_mask())
- cfg.flags |= PTRACE_BTS_O_TRACE;
+ memset(&cfg, 0, sizeof(cfg));
+ cfg.size = (max - base);
+ cfg.signal = child->thread.bts_ovfl_signal;
+ cfg.bts_size = sizeof(struct bts_struct);
- if (test_tsk_thread_flag(child, TIF_BTS_TRACE_TS))
- cfg.flags |= PTRACE_BTS_O_SCHED;
- }
+ if (cfg.signal)
+ cfg.flags |= PTRACE_BTS_O_SIGNAL;
- cfg.bts_size = sizeof(struct bts_struct);
+ if (test_tsk_thread_flag(child, TIF_DEBUGCTLMSR) &&
+ child->thread.debugctlmsr & bts_cfg.debugctl_mask)
+ cfg.flags |= PTRACE_BTS_O_TRACE;
+
+ if (test_tsk_thread_flag(child, TIF_BTS_TRACE_TS))
+ cfg.flags |= PTRACE_BTS_O_SCHED;
if (copy_to_user(ucfg, &cfg, sizeof(cfg)))
return -EFAULT;
return sizeof(cfg);
}
-
static int ptrace_bts_write_record(struct task_struct *child,
const struct bts_struct *in)
{
- int retval;
+ unsigned char bts_record[BTS_MAX_RECORD_SIZE];
- if (!child->thread.ds_area_msr)
- return -ENXIO;
+ BUG_ON(BTS_MAX_RECORD_SIZE < bts_cfg.sizeof_bts);
- retval = ds_write_bts((void *)child->thread.ds_area_msr, in);
- if (retval)
- return retval;
+ memset(bts_record, 0, bts_cfg.sizeof_bts);
+ switch (in->qualifier) {
+ case BTS_INVALID:
+ break;
- return sizeof(*in);
-}
+ case BTS_BRANCH:
+ bts_set(bts_record, bts_from, in->variant.lbr.from_ip);
+ bts_set(bts_record, bts_to, in->variant.lbr.to_ip);
+ break;
-static int ptrace_bts_realloc(struct task_struct *child,
- int size, int reduce_size)
-{
- unsigned long rlim, vm;
- int ret, old_size;
+ case BTS_TASK_ARRIVES:
+ case BTS_TASK_DEPARTS:
+ bts_set(bts_record, bts_from, bts_escape);
+ bts_set(bts_record, bts_qual, in->qualifier);
+ bts_set(bts_record, bts_jiffies, in->variant.jiffies);
+ break;
- if (size < 0)
+ default:
return -EINVAL;
-
- old_size = ds_get_bts_size((void *)child->thread.ds_area_msr);
- if (old_size < 0)
- return old_size;
-
- ret = ds_free((void **)&child->thread.ds_area_msr);
- if (ret < 0)
- goto out;
-
- size >>= PAGE_SHIFT;
- old_size >>= PAGE_SHIFT;
-
- current->mm->total_vm -= old_size;
- current->mm->locked_vm -= old_size;
-
- if (size == 0)
- goto out;
-
- rlim = current->signal->rlim[RLIMIT_AS].rlim_cur >> PAGE_SHIFT;
- vm = current->mm->total_vm + size;
- if (rlim < vm) {
- ret = -ENOMEM;
-
- if (!reduce_size)
- goto out;
-
- size = rlim - current->mm->total_vm;
- if (size <= 0)
- goto out;
}
- rlim = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur >> PAGE_SHIFT;
- vm = current->mm->locked_vm + size;
- if (rlim < vm) {
- ret = -ENOMEM;
-
- if (!reduce_size)
- goto out;
-
- size = rlim - current->mm->locked_vm;
- if (size <= 0)
- goto out;
- }
-
- ret = ds_allocate((void **)&child->thread.ds_area_msr,
- size << PAGE_SHIFT);
- if (ret < 0)
- goto out;
-
- current->mm->total_vm += size;
- current->mm->locked_vm += size;
-
-out:
- if (child->thread.ds_area_msr)
- set_tsk_thread_flag(child, TIF_DS_AREA_MSR);
- else
- clear_tsk_thread_flag(child, TIF_DS_AREA_MSR);
-
- return ret;
+ /* The writing task will be the switched-to task on a context
+ * switch. It needs to write into the switched-from task's BTS
+ * buffer. */
+ return ds_unchecked_write_bts(child, bts_record, bts_cfg.sizeof_bts);
}
void ptrace_bts_take_timestamp(struct task_struct *tsk,
ptrace_bts_write_record(tsk, &rec);
}
-#endif /* X86_BTS */
+
+static const struct bts_configuration bts_cfg_netburst = {
+ .sizeof_bts = sizeof(long) * 3,
+ .sizeof_field = sizeof(long),
+ .debugctl_mask = (1<<2)|(1<<3)|(1<<5)
+};
+
+static const struct bts_configuration bts_cfg_pentium_m = {
+ .sizeof_bts = sizeof(long) * 3,
+ .sizeof_field = sizeof(long),
+ .debugctl_mask = (1<<6)|(1<<7)
+};
+
+static const struct bts_configuration bts_cfg_core2 = {
+ .sizeof_bts = 8 * 3,
+ .sizeof_field = 8,
+ .debugctl_mask = (1<<6)|(1<<7)|(1<<9)
+};
+
+static inline void bts_configure(const struct bts_configuration *cfg)
+{
+ bts_cfg = *cfg;
+}
+
+void __cpuinit ptrace_bts_init_intel(struct cpuinfo_x86 *c)
+{
+ switch (c->x86) {
+ case 0x6:
+ switch (c->x86_model) {
+ case 0xD:
+ case 0xE: /* Pentium M */
+ bts_configure(&bts_cfg_pentium_m);
+ break;
+ case 0xF: /* Core2 */
+ case 0x1C: /* Atom */
+ bts_configure(&bts_cfg_core2);
+ break;
+ default:
+ /* sorry, don't know about them */
+ break;
+ }
+ break;
+ case 0xF:
+ switch (c->x86_model) {
+ case 0x0:
+ case 0x1:
+ case 0x2: /* Netburst */
+ bts_configure(&bts_cfg_netburst);
+ break;
+ default:
+ /* sorry, don't know about them */
+ break;
+ }
+ break;
+ default:
+ /* sorry, don't know about them */
+ break;
+ }
+}
+#endif /* CONFIG_X86_PTRACE_BTS */
/*
* Called by kernel/ptrace.c when detaching..
#ifdef TIF_SYSCALL_EMU
clear_tsk_thread_flag(child, TIF_SYSCALL_EMU);
#endif
- if (child->thread.ds_area_msr) {
-#ifdef X86_BTS
- ptrace_bts_realloc(child, 0, 0);
-#endif
- child->thread.debugctlmsr &= ~ds_debugctl_mask();
- if (!child->thread.debugctlmsr)
- clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
- clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS);
- }
+#ifdef CONFIG_X86_PTRACE_BTS
+ (void)ds_release_bts(child);
+
+ child->thread.debugctlmsr &= ~bts_cfg.debugctl_mask;
+ if (!child->thread.debugctlmsr)
+ clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
+
+ clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS);
+#endif /* CONFIG_X86_PTRACE_BTS */
}
#if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION
return copy_regset_to_user(child, &user_x86_32_view,
REGSET_XFP,
0, sizeof(struct user_fxsr_struct),
- datap);
+ datap) ? -EIO : 0;
case PTRACE_SETFPXREGS: /* Set the child extended FPU state. */
return copy_regset_from_user(child, &user_x86_32_view,
REGSET_XFP,
0, sizeof(struct user_fxsr_struct),
- datap);
+ datap) ? -EIO : 0;
#endif
#if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION
/*
* These bits need more cooking - not enabled yet:
*/
-#ifdef X86_BTS
+#ifdef CONFIG_X86_PTRACE_BTS
case PTRACE_BTS_CONFIG:
ret = ptrace_bts_config
(child, data, (struct ptrace_bts_config __user *)addr);
break;
case PTRACE_BTS_SIZE:
- ret = ptrace_bts_get_size(child);
+ ret = ds_get_bts_index(child, /* pos = */ 0);
break;
case PTRACE_BTS_GET:
break;
case PTRACE_BTS_CLEAR:
- ret = ptrace_bts_clear(child);
+ ret = ds_clear_bts(child);
break;
case PTRACE_BTS_DRAIN:
ret = ptrace_bts_drain
(child, data, (struct bts_struct __user *) addr);
break;
-#endif
+#endif /* CONFIG_X86_PTRACE_BTS */
default:
ret = ptrace_request(child, request, addr, data);
#endif
}
- #ifdef CONFIG_X86_32
-
void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code)
{
struct siginfo info;
force_sig_info(SIGTRAP, &info, tsk);
}
- /* notification of system call entry/exit
- * - triggered by current->work.syscall_trace
- */
- int do_syscall_trace(struct pt_regs *regs, int entryexit)
- {
- int is_sysemu = test_thread_flag(TIF_SYSCALL_EMU);
- /*
- * With TIF_SYSCALL_EMU set we want to ignore TIF_SINGLESTEP for syscall
- * interception
- */
- int is_singlestep = !is_sysemu && test_thread_flag(TIF_SINGLESTEP);
- int ret = 0;
-
- /* do the secure computing check first */
- if (!entryexit)
- secure_computing(regs->orig_ax);
-
- if (unlikely(current->audit_context)) {
- if (entryexit)
- audit_syscall_exit(AUDITSC_RESULT(regs->ax),
- regs->ax);
- /* Debug traps, when using PTRACE_SINGLESTEP, must be sent only
- * on the syscall exit path. Normally, when TIF_SYSCALL_AUDIT is
- * not used, entry.S will call us only on syscall exit, not
- * entry; so when TIF_SYSCALL_AUDIT is used we must avoid
- * calling send_sigtrap() on syscall entry.
- *
- * Note that when PTRACE_SYSEMU_SINGLESTEP is used,
- * is_singlestep is false, despite his name, so we will still do
- * the correct thing.
- */
- else if (is_singlestep)
- goto out;
- }
-
- if (!(current->ptrace & PT_PTRACED))
- goto out;
-
- /* If a process stops on the 1st tracepoint with SYSCALL_TRACE
- * and then is resumed with SYSEMU_SINGLESTEP, it will come in
- * here. We have to check this and return */
- if (is_sysemu && entryexit)
- return 0;
-
- /* Fake a debug trap */
- if (is_singlestep)
- send_sigtrap(current, regs, 0);
-
- if (!test_thread_flag(TIF_SYSCALL_TRACE) && !is_sysemu)
- goto out;
-
- /* the 0x80 provides a way for the tracing parent to distinguish
- between a syscall stop and SIGTRAP delivery */
- /* Note that the debugger could change the result of test_thread_flag!*/
- ptrace_notify(SIGTRAP | ((current->ptrace & PT_TRACESYSGOOD) ? 0x80:0));
-
- /*
- * this isn't the same as continuing with a signal, but it will do
- * for normal use. strace only continues with a signal if the
- * stopping signal is not SIGTRAP. -brl
- */
- if (current->exit_code) {
- send_sig(current->exit_code, current, 1);
- current->exit_code = 0;
- }
- ret = is_sysemu;
- out:
- if (unlikely(current->audit_context) && !entryexit)
- audit_syscall_entry(AUDIT_ARCH_I386, regs->orig_ax,
- regs->bx, regs->cx, regs->dx, regs->si);
- if (ret == 0)
- return 0;
-
- regs->orig_ax = -1; /* force skip of syscall restarting */
- if (unlikely(current->audit_context))
- audit_syscall_exit(AUDITSC_RESULT(regs->ax), regs->ax);
- return 1;
- }
-
- #else /* CONFIG_X86_64 */
-
static void syscall_trace(struct pt_regs *regs)
{
+ if (!(current->ptrace & PT_PTRACED))
+ return;
#if 0
printk("trace %s ip %lx sp %lx ax %d origrax %d caller %lx tiflags %x ptrace %x\n",
}
}
- asmlinkage void syscall_trace_enter(struct pt_regs *regs)
+ #ifdef CONFIG_X86_32
+ # define IS_IA32 1
+ #elif defined CONFIG_IA32_EMULATION
+ # define IS_IA32 test_thread_flag(TIF_IA32)
+ #else
+ # define IS_IA32 0
+ #endif
+
+ /*
+ * We must return the syscall number to actually look up in the table.
+ * This can be -1L to skip running any syscall at all.
+ */
+ asmregparm long syscall_trace_enter(struct pt_regs *regs)
{
+ long ret = 0;
+
+ /*
+ * If we stepped into a sysenter/syscall insn, it trapped in
+ * kernel mode; do_debug() cleared TF and set TIF_SINGLESTEP.
+ * If user-mode had set TF itself, then it's still clear from
+ * do_debug() and we need to set it again to restore the user
+ * state. If we entered on the slow path, TF was already set.
+ */
+ if (test_thread_flag(TIF_SINGLESTEP))
+ regs->flags |= X86_EFLAGS_TF;
+
/* do the secure computing check first */
secure_computing(regs->orig_ax);
- if (test_thread_flag(TIF_SYSCALL_TRACE)
- && (current->ptrace & PT_PTRACED))
+ if (unlikely(test_thread_flag(TIF_SYSCALL_EMU)))
+ ret = -1L;
+
+ if (ret || test_thread_flag(TIF_SYSCALL_TRACE))
syscall_trace(regs);
if (unlikely(current->audit_context)) {
- if (test_thread_flag(TIF_IA32)) {
+ if (IS_IA32)
audit_syscall_entry(AUDIT_ARCH_I386,
regs->orig_ax,
regs->bx, regs->cx,
regs->dx, regs->si);
- } else {
+ #ifdef CONFIG_X86_64
+ else
audit_syscall_entry(AUDIT_ARCH_X86_64,
regs->orig_ax,
regs->di, regs->si,
regs->dx, regs->r10);
- }
+ #endif
}
+
+ return ret ?: regs->orig_ax;
}
- asmlinkage void syscall_trace_leave(struct pt_regs *regs)
+ asmregparm void syscall_trace_leave(struct pt_regs *regs)
{
if (unlikely(current->audit_context))
audit_syscall_exit(AUDITSC_RESULT(regs->ax), regs->ax);
- if ((test_thread_flag(TIF_SYSCALL_TRACE)
- || test_thread_flag(TIF_SINGLESTEP))
- && (current->ptrace & PT_PTRACED))
+ if (test_thread_flag(TIF_SYSCALL_TRACE))
syscall_trace(regs);
- }
- #endif /* CONFIG_X86_32 */
+ /*
+ * If TIF_SYSCALL_EMU is set, we only get here because of
+ * TIF_SINGLESTEP (i.e. this is PTRACE_SYSEMU_SINGLESTEP).
+ * We already reported this syscall instruction in
+ * syscall_trace_enter(), so don't do any more now.
+ */
+ if (unlikely(test_thread_flag(TIF_SYSCALL_EMU)))
+ return;
+
+ /*
+ * If we are single-stepping, synthesize a trap to follow the
+ * system call instruction.
+ */
+ if (test_thread_flag(TIF_SINGLESTEP) &&
+ (current->ptrace & PT_PTRACED))
+ send_sigtrap(current, regs, 0);
+ }