DMARC-Filter: OpenDMARC Filter v1.3.2 mx1.redhat.com 41FC7D1FE2
Date: Fri, 7 Apr 2017 14:29:11 +0200
From: Radim =?utf-8?B?S3LEjW3DocWZ?= <rkrcmar@redhat.com>
To: Marc Zyngier <marc.zyngier@arm.com>
Cc: linux-kernel@vger.kernel.org, kvm@vger.kernel.org,
        Christoffer Dall <cdall@linaro.org>, Andrew Jones <drjones@redhat.com>,
        Paolo Bonzini <pbonzini@redhat.com>,
        Christian Borntraeger <borntraeger@de.ibm.com>,
        Cornelia Huck <cornelia.huck@de.ibm.com>,
        James Hogan <james.hogan@imgtec.com>,
        Paul Mackerras <paulus@ozlabs.org>
Subject: Re: [PATCH RFC 5/6] KVM: mark requests that do not need a wakeup
Message-ID: <20170407122910.GD23559@potion>
References: <20170406202056.18379-1-rkrcmar@redhat.com>
 <20170406202056.18379-6-rkrcmar@redhat.com>
 <eab3a2dd-3ade-57f2-ec2b-1ebd744df0fa@arm.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <eab3a2dd-3ade-57f2-ec2b-1ebd744df0fa@arm.com>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1265
Lines: 32

2017-04-07 09:27+0100, Marc Zyngier:
> On 06/04/17 21:20, Radim Krčmář wrote:
>> Some operations must ensure that the guest is not running with stale
>> data, but if the guest is halted, then the update can wait until another
>> event happens.  kvm_make_all_requests() currently doesn't wake up, so we
>> can mark all requests used with it.
>> 
>> First 8 bits were arbitrarily reserved for request numbers.
>> 
>> Most uses of requests have the request type as a constant, so a compiler
>> will optimize the '&'.
>> 
>> An alternative would be to have an inline function that would return
>> whether the request needs a wake-up or not, but I like this one better
>> even though it might produce worse assembly.
>> 
>> Suggested-by: Christoffer Dall <cdall@linaro.org>
>> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
>> ---
>>   Btw. do you recall which macro allowed to define bitmasks?  (It has
>>   two arguments, FROM and TO.)
> 
> GENMASK (and its _ULL variant), defined in include/linux/bitops.h.

Thank you, it is under BIT() ... I am blind.

> +#define KVM_REQUEST_MASK           0xff

The 0xff should be "GENMASK(7,0)".

First 8 bits is plenty and should be fast even if the compiler doesn't
optimize the masking because request is not constant.