DMARC-Filter: OpenDMARC Filter v1.3.2 mx1.redhat.com 80C4D3DBDF
Date: Fri, 7 Apr 2017 16:05:41 +0200
From: Radim =?utf-8?B?S3LEjW3DocWZ?= <rkrcmar@redhat.com>
To: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: linux-kernel@vger.kernel.org, kvm@vger.kernel.org,
        Christoffer Dall <cdall@linaro.org>, Andrew Jones <drjones@redhat.com>,
        Marc Zyngier <marc.zyngier@arm.com>,
        Paolo Bonzini <pbonzini@redhat.com>,
        Cornelia Huck <cornelia.huck@de.ibm.com>,
        James Hogan <james.hogan@imgtec.com>,
        Paul Mackerras <paulus@ozlabs.org>
Subject: Re: [PATCH 2/6] KVM: use kvm_{test,clear}_request instead of
 {test,clear}_bit
Message-ID: <20170407140541.GF23559@potion>
References: <20170406202056.18379-1-rkrcmar@redhat.com>
 <20170406202056.18379-3-rkrcmar@redhat.com>
 <e7609664-6d84-9602-1861-9d037e5953d2@de.ibm.com>
 <20170407122413.GC23559@potion>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <20170407122413.GC23559@potion>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1517
Lines: 37

2017-04-07 14:24+0200, Radim Krčmář:
> 2017-04-07 12:55+0200, Christian Borntraeger:
>> On 04/06/2017 10:20 PM, Radim Krčmář wrote:
>>>  static inline bool kvm_check_request(int req, struct kvm_vcpu *vcpu)
>>>  {
>>> -	if (test_bit(req, &vcpu->requests)) {
>>> -		clear_bit(req, &vcpu->requests);
>>> +	if (kvm_test_request(req, vcpu)) {
>>> +		kvm_clear_request(req, vcpu);
>> 
>> This looks fine. I am just asking myself why we do not use
>> test_and_clear_bit? Do we expect gcc to merge all test bits as
>> a fast path? This does not seem to work as far as I can tell and
>> almost everybody does a fast path like in
> 
> test_and_clear_bit() is a slower operation even if the test is false (at
> least on x86), because it needs to be fully atomic.
> 
>> arch/s390/kvm/kvm-s390.c:
>>         if (!vcpu->requests)
>>                 return 0;
>> 
>> arch/x86/kvm/x86.c:
>>     if (vcpu->requests) {
> 
> We'll mostly have only one request set, so splitting the test_and_clear
> improves the performance of many subsequent tests_and_clear()s even if
> the compiler doesn't optimize.
> 
> GCC couldn't even optimize if we used test_and_clear_bit(), because that
> instruction adds barriers, but the forward check for vcpu->requests is
> there because we do not trust the optimizer to do it for us and it would
> make a big difference.

Ugh, I started thinking that bitops are not atomic because I looked at
wrong boot/bitops.h by mistake.  The compiler cannot merge test_bit()s,
but the speed difference holds.