Re: [PATCH] Revert "locking/pvqspinlock: Don't wait if vCPU is preempted"

[Paolo Bonzini]

On 09/09/19 12:56, Waiman Long wrote:
> On 9/9/19 2:40 AM, Wanpeng Li wrote:
>> From: Wanpeng Li <wanpengli@xxxxxxxxxxx>
>>
>> This patch reverts commit 75437bb304b20 (locking/pvqspinlock: Don't wait if 
>> vCPU is preempted), we found great regression caused by this commit.
>>
>> Xeon Skylake box, 2 sockets, 40 cores, 80 threads, three VMs, each is 80 vCPUs.
>> The score of ebizzy -M can reduce from 13000-14000 records/s to 1700-1800 
>> records/s with this commit.
>>
>>           Host                       Guest                score
>>
>> vanilla + w/o kvm optimizes     vanilla               1700-1800 records/s
>> vanilla + w/o kvm optimizes     vanilla + revert      13000-14000 records/s
>> vanilla + w/ kvm optimizes      vanilla               4500-5000 records/s
>> vanilla + w/ kvm optimizes      vanilla + revert      14000-15500 records/s
>>
>> Exit from aggressive wait-early mechanism can result in yield premature and 
>> incur extra scheduling latency in over-subscribe scenario.
>>
>> kvm optimizes:
>> [1] commit d73eb57b80b (KVM: Boost vCPUs that are delivering interrupts)
>> [2] commit 266e85a5ec9 (KVM: X86: Boost queue head vCPU to mitigate lock waiter preemption)
>>
>> Tested-by: loobinliu@xxxxxxxxxxx
>> Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
>> Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
>> Cc: Ingo Molnar <mingo@xxxxxxxxxx>
>> Cc: Waiman Long <longman@xxxxxxxxxx>
>> Cc: Paolo Bonzini <pbonzini@xxxxxxxxxx>
>> Cc: Radim KrÄmÃÅ <rkrcmar@xxxxxxxxxx>
>> Cc: loobinliu@xxxxxxxxxxx
>> Cc: stable@xxxxxxxxxxxxxxx 
>> Fixes: 75437bb304b20 (locking/pvqspinlock: Don't wait if vCPU is preempted)
>> Signed-off-by: Wanpeng Li <wanpengli@xxxxxxxxxxx>
>> ---
>>  kernel/locking/qspinlock_paravirt.h | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/kernel/locking/qspinlock_paravirt.h b/kernel/locking/qspinlock_paravirt.h
>> index 89bab07..e84d21a 100644
>> --- a/kernel/locking/qspinlock_paravirt.h
>> +++ b/kernel/locking/qspinlock_paravirt.h
>> @@ -269,7 +269,7 @@ pv_wait_early(struct pv_node *prev, int loop)
>>  	if ((loop & PV_PREV_CHECK_MASK) != 0)
>>  		return false;
>>  
>> -	return READ_ONCE(prev->state) != vcpu_running || vcpu_is_preempted(prev->cpu);
>> +	return READ_ONCE(prev->state) != vcpu_running;
>>  }
>>  
>>  /*
> 
> There are several possibilities for this performance regression:
> 
> 1) Multiple vcpus calling vcpu_is_preempted() repeatedly may cause some
> cacheline contention issue depending on how that callback is implemented.

Unlikely, it is a single percpu read.

> 2) KVM may set the preempt flag for a short period whenver an vmexit
> happens even if a vmenter is executed shortly after. In this case, we
> may want to use a more durable vcpu suspend flag that indicates the vcpu
> won't get a real vcpu back for a longer period of time.

It sets it for exits to userspace, but they shouldn't really happen on a 
properly-configured system.

However, it's easy to test this theory:

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 2e302e977dac..feb6c75a7a88 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -3368,26 +3368,28 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
 {
 	int idx;
 
-	if (vcpu->preempted)
+	if (vcpu->preempted) {
 		vcpu->arch.preempted_in_kernel = !kvm_x86_ops->get_cpl(vcpu);
 
-	/*
-	 * Disable page faults because we're in atomic context here.
-	 * kvm_write_guest_offset_cached() would call might_fault()
-	 * that relies on pagefault_disable() to tell if there's a
-	 * bug. NOTE: the write to guest memory may not go through if
-	 * during postcopy live migration or if there's heavy guest
-	 * paging.
-	 */
-	pagefault_disable();
-	/*
-	 * kvm_memslots() will be called by
-	 * kvm_write_guest_offset_cached() so take the srcu lock.
-	 */
-	idx = srcu_read_lock(&vcpu->kvm->srcu);
-	kvm_steal_time_set_preempted(vcpu);
-	srcu_read_unlock(&vcpu->kvm->srcu, idx);
-	pagefault_enable();
+		/*
+		 * Disable page faults because we're in atomic context here.
+		 * kvm_write_guest_offset_cached() would call might_fault()
+		 * that relies on pagefault_disable() to tell if there's a
+		 * bug. NOTE: the write to guest memory may not go through if
+		 * during postcopy live migration or if there's heavy guest
+		 * paging.
+		 */
+		pagefault_disable();
+		/*
+		 * kvm_memslots() will be called by
+		 * kvm_write_guest_offset_cached() so take the srcu lock.
+		 */
+		idx = srcu_read_lock(&vcpu->kvm->srcu);
+		kvm_steal_time_set_preempted(vcpu);
+		srcu_read_unlock(&vcpu->kvm->srcu, idx);
+		pagefault_enable();
+	}
+
 	kvm_x86_ops->vcpu_put(vcpu);
 	vcpu->arch.last_host_tsc = rdtsc();
 	/*

Wanpeng, can you try?

Paolo

> Perhaps you can add a lock event counter to count the number of
> wait_early events caused by vcpu_is_preempted() being true to see if it
> really cause a lot more wait_early than without the vcpu_is_preempted()
> call.
> 
> I have no objection to this, I just want to find out the root cause of it.
> 
> Cheers,
> Longman
>