Commit 5fbd036b authored by Peter Zijlstra's avatar Peter Zijlstra Committed by Ingo Molnar
Browse files

sched: Cleanup cpu_active madness 



Stepan found:

CPU0		CPUn

_cpu_up()
  __cpu_up()

		boostrap()
		  notify_cpu_starting()
		  set_cpu_online()
		  while (!cpu_active())
		    cpu_relax()

<PREEMPT-out>

smp_call_function(.wait=1)
  /* we find cpu_online() is true */
  arch_send_call_function_ipi_mask()

  /* wait-forever-more */

<PREEMPT-in>
		  local_irq_enable()

  cpu_notify(CPU_ONLINE)
    sched_cpu_active()
      set_cpu_active()

Now the purpose of cpu_active is mostly with bringing down a cpu, where
we mark it !active to avoid the load-balancer from moving tasks to it
while we tear down the cpu. This is required because we only update the
sched_domain tree after we brought the cpu-down. And this is needed so
that some tasks can still run while we bring it down, we just don't want
new tasks to appear.

On cpu-up however the sched_domain tree doesn't yet include the new cpu,
so its invisible to the load-balancer, regardless of the active state.
So instead of setting the active state after we boot the new cpu (and
consequently having to wait for it before enabling interrupts) set the
cpu active before we set it online and avoid the whole mess.

Reported-by: default avatarStepan Moskovchenko <stepanm@codeaurora.org>
Signed-off-by: default avatarPeter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: default avatarThomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1323965362.18942.71.camel@twins


Signed-off-by: default avatarIngo Molnar <mingo@elte.hu>
parent 5d6523eb

arch/arm/kernel/smp.c

+0 −7
Original line number Diff line number Diff line
@@ -295,13 +295,6 @@ asmlinkage void __cpuinit secondary_start_kernel(void) 
	 */

	percpu_timer_setup();



	while (!cpu_active(cpu))

		cpu_relax();



	/*

	 * cpu_active bit is set, so it's safe to enalbe interrupts

	 * now.

	 */

	local_irq_enable();

	local_fiq_enable();

arch/hexagon/kernel/smp.c

+0 −2
Original line number Diff line number Diff line
@@ -179,8 +179,6 @@ void __cpuinit start_secondary(void) 
	printk(KERN_INFO "%s cpu %d\n", __func__, current_thread_info()->cpu);



	set_cpu_online(cpu, true);

	while (!cpumask_test_cpu(cpu, cpu_active_mask))

		cpu_relax();

	local_irq_enable();



	cpu_idle();

arch/s390/kernel/smp.c

+0 −6
Original line number Diff line number Diff line
@@ -550,12 +550,6 @@ int __cpuinit start_secondary(void *cpuvoid) 
	S390_lowcore.restart_psw.addr =

		PSW_ADDR_AMODE | (unsigned long) psw_restart_int_handler;

	__ctl_set_bit(0, 28); /* Enable lowcore protection */

	/*

	 * Wait until the cpu which brought this one up marked it

	 * active before enabling interrupts.

	 */

	while (!cpumask_test_cpu(smp_processor_id(), cpu_active_mask))

		cpu_relax();

	local_irq_enable();

	/* cpu_idle will call schedule for us */

	cpu_idle();

arch/x86/kernel/smpboot.c

+0 −13
Original line number Diff line number Diff line
@@ -291,19 +291,6 @@ notrace static void __cpuinit start_secondary(void *unused) 
	per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE;

	x86_platform.nmi_init();



	/*

	 * Wait until the cpu which brought this one up marked it

	 * online before enabling interrupts. If we don't do that then

	 * we can end up waking up the softirq thread before this cpu

	 * reached the active state, which makes the scheduler unhappy

	 * and schedule the softirq thread on the wrong cpu. This is

	 * only observable with forced threaded interrupts, but in

	 * theory it could also happen w/o them. It's just way harder

	 * to achieve.

	 */

	while (!cpumask_test_cpu(smp_processor_id(), cpu_active_mask))

		cpu_relax();



	/* enable local interrupts */

	local_irq_enable();

kernel/sched/core.c

+1 −1
Original line number Diff line number Diff line
@@ -5410,7 +5410,7 @@ static int __cpuinit sched_cpu_active(struct notifier_block *nfb, 
				      unsigned long action, void *hcpu)

{

	switch (action & ~CPU_TASKS_FROZEN) {

	case CPU_ONLINE:

	case CPU_STARTING:

	case CPU_DOWN_FAILED:

		set_cpu_active((long)hcpu, true);

		return NOTIFY_OK;

CRA Git | Maintained and supported by SUSTech CRA and CCSE