linux-kernel - Re: [External] Re: [RFC] x86/kvm: Use native qspinlock by default when realtime hinted

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <8145bb17-8ba4-4d9d-a995-5f8b09db99c4@google.com>
Date: Wed, 2 Jul 2025 20:23:58 +0800
From: Liangyan <liangyan.peng@...edance.com>
To: Bibo Mao <maobibo@...ngson.cn>, Liangyan <liangyan.peng@...edance.com>,
 pbonzini@...hat.com, vkuznets@...hat.com, tglx@...utronix.de,
 mingo@...hat.com, bp@...en8.de, dave.hansen@...ux.intel.com, hpa@...or.com,
 wanpengli@...cent.com
Cc: linux-kernel@...r.kernel.org, x86@...nel.org, kvm@...r.kernel.org
Subject: Re: [External] Re: [RFC] x86/kvm: Use native qspinlock by default
 when realtime hinted

We test that unixbench spawn has big improvement in Intel 8582c 120-CPU 
guest vm if switch to qspinlock.

Command: ./Run -c 120 spawn

Use virt_spin_lock:
System Benchmarks Partial Index   BASELINE       RESULT  INDEX
Process Creation                     126.0      71878.4   5704.6
                                                         ========
System Benchmarks Index Score (Partial Only)              5704.6


Use qspinlock:
System Benchmarks Partial Index   BASELINE       RESULT    INDEX
Process Creation                     126.0     173566.6  13775.1
                                                         ========
System Benchmarks Index Score (Partial Only              13775.1


Regards,
Liangyan

On 2025/7/2 16:19, Bibo Mao wrote:
> 
> 
> On 2025/7/2 下午2:42, Liangyan wrote:
>> When KVM_HINTS_REALTIME is set and KVM_FEATURE_PV_UNHALT is clear,
>> currently guest will use virt_spin_lock.
>> Since KVM_HINTS_REALTIME is set, use native qspinlock should be safe
>> and have better performance than virt_spin_lock.
> Just be curious, do you have actual data where native qspinlock has 
> better performance than virt_spin_lock()?
> 
> By my understanding, qspinlock is not friendly with VM. When lock is 
> released, it is acquired with one by one order in contending queue. If 
> the first vCPU in contending queue is preempted, the other vCPUs can not 
> get lock. On physical machine it is almost impossible that CPU 
> contending lock is preempted.
> 
> Regards
> Bibo Mao
>>
>> Signed-off-by: Liangyan <liangyan.peng@...edance.com>
>> ---
>>   arch/x86/kernel/kvm.c | 18 +++++++++---------
>>   1 file changed, 9 insertions(+), 9 deletions(-)
>>
>> diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
>> index 921c1c783bc1..9080544a4007 100644
>> --- a/arch/x86/kernel/kvm.c
>> +++ b/arch/x86/kernel/kvm.c
>> @@ -1072,6 +1072,15 @@ static void kvm_wait(u8 *ptr, u8 val)
>>    */
>>   void __init kvm_spinlock_init(void)
>>   {
>> +    /*
>> +     * Disable PV spinlocks and use native qspinlock when dedicated 
>> pCPUs
>> +     * are available.
>> +     */
>> +    if (kvm_para_has_hint(KVM_HINTS_REALTIME)) {
>> +        pr_info("PV spinlocks disabled with KVM_HINTS_REALTIME 
>> hints\n");
>> +        goto out;
>> +    }
>> +
>>       /*
>>        * In case host doesn't support KVM_FEATURE_PV_UNHALT there is 
>> still an
>>        * advantage of keeping virt_spin_lock_key enabled: 
>> virt_spin_lock() is
>> @@ -1082,15 +1091,6 @@ void __init kvm_spinlock_init(void)
>>           return;
>>       }
>> -    /*
>> -     * Disable PV spinlocks and use native qspinlock when dedicated 
>> pCPUs
>> -     * are available.
>> -     */
>> -    if (kvm_para_has_hint(KVM_HINTS_REALTIME)) {
>> -        pr_info("PV spinlocks disabled with KVM_HINTS_REALTIME 
>> hints\n");
>> -        goto out;
>> -    }
>> -
>>       if (num_possible_cpus() == 1) {
>>           pr_info("PV spinlocks disabled, single CPU\n");
>>           goto out;
>>
>