diff options
| author | Len Brown <len.brown@intel.com> | 2010-05-24 14:27:44 -0400 |
|---|---|---|
| committer | Len Brown <len.brown@intel.com> | 2010-05-28 13:58:57 -0400 |
| commit | 02cf4f9808382af7265cafc33dc86ec5875526aa (patch) | |
| tree | 6e4b7fc98a125e6af6011ee04e5fbf2d2fe38c4f /drivers/acpi | |
| parent | a7d27c37533524e23bbdc9ba1d78e3241b7483ea (diff) | |
ACPI: acpi_idle: touch TS_POLLING only in the non-MWAIT case
commit d306ebc28649b89877a22158fe0076f06cc46f60
(ACPI: Be in TS_POLLING state during mwait based C-state entry)
fixed an important power & performance issue where ACPI c2 and c3 C-states
were clearing TS_POLLING even when using MWAIT (ACPI_STATE_FFH).
That bug had been causing us to receive redundant scheduling interrups
when we had already been woken up by MONITOR/MWAIT.
Following up on that...
In the MWAIT case, we don't have to subsequently
check need_resched(), as that c heck was there
for the TS_POLLING-clearing case.
Note that not only does the cpuidle calling function
already check need_resched() before calling us, the
low-level entry into monitor/mwait calls it twice --
guaranteeing that a write to the trigger address
can not go un-noticed.
Also, in this case, we don't have to set TS_POLLING
when we wake, because we never cleared it.
Signed-off-by: Len Brown <len.brown@intel.com>
Acked-by: Venkatesh Pallipadi <venki@google.com>
Diffstat (limited to 'drivers/acpi')
| -rw-r--r-- | drivers/acpi/processor_idle.c | 28 |
1 files changed, 16 insertions, 12 deletions
diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index 5939e7f7d8e..a4166e2abb9 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -881,6 +881,7 @@ static int acpi_idle_enter_simple(struct cpuidle_device *dev, return(acpi_idle_enter_c1(dev, state)); local_irq_disable(); + if (cx->entry_method != ACPI_CSTATE_FFH) { current_thread_info()->status &= ~TS_POLLING; /* @@ -888,12 +889,12 @@ static int acpi_idle_enter_simple(struct cpuidle_device *dev, * NEED_RESCHED: */ smp_mb(); - } - if (unlikely(need_resched())) { - current_thread_info()->status |= TS_POLLING; - local_irq_enable(); - return 0; + if (unlikely(need_resched())) { + current_thread_info()->status |= TS_POLLING; + local_irq_enable(); + return 0; + } } /* @@ -918,7 +919,8 @@ static int acpi_idle_enter_simple(struct cpuidle_device *dev, sched_clock_idle_wakeup_event(sleep_ticks*PM_TIMER_TICK_NS); local_irq_enable(); - current_thread_info()->status |= TS_POLLING; + if (cx->entry_method != ACPI_CSTATE_FFH) + current_thread_info()->status |= TS_POLLING; cx->usage++; @@ -968,6 +970,7 @@ static int acpi_idle_enter_bm(struct cpuidle_device *dev, } local_irq_disable(); + if (cx->entry_method != ACPI_CSTATE_FFH) { current_thread_info()->status &= ~TS_POLLING; /* @@ -975,12 +978,12 @@ static int acpi_idle_enter_bm(struct cpuidle_device *dev, * NEED_RESCHED: */ smp_mb(); - } - if (unlikely(need_resched())) { - current_thread_info()->status |= TS_POLLING; - local_irq_enable(); - return 0; + if (unlikely(need_resched())) { + current_thread_info()->status |= TS_POLLING; + local_irq_enable(); + return 0; + } } acpi_unlazy_tlb(smp_processor_id()); @@ -1032,7 +1035,8 @@ static int acpi_idle_enter_bm(struct cpuidle_device *dev, sched_clock_idle_wakeup_event(sleep_ticks*PM_TIMER_TICK_NS); local_irq_enable(); - current_thread_info()->status |= TS_POLLING; + if (cx->entry_method != ACPI_CSTATE_FFH) + current_thread_info()->status |= TS_POLLING; cx->usage++; |