DMARC-Filter: OpenDMARC Filter v1.3.2 mx1.redhat.com 32F168051F
Subject: Re: [PATCH v2 9/9] KVM: mark requests that need synchronization
To: =?UTF-8?B?UmFkaW0gS3LEjW3DocWZ?= <rkrcmar@redhat.com>,
        linux-kernel@vger.kernel.org, kvm@vger.kernel.org
References: <20170426203227.12321-1-rkrcmar@redhat.com>
 <20170426203227.12321-10-rkrcmar@redhat.com>
Cc: Christoffer Dall <cdall@linaro.org>, Andrew Jones <drjones@redhat.com>,
        Marc Zyngier <marc.zyngier@arm.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>
From: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <0099c50a-8270-4162-344c-878625e5e5b8@redhat.com>
Date: Thu, 27 Apr 2017 14:36:04 +0200
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101
 Thunderbird/45.8.0
MIME-Version: 1.0
In-Reply-To: <20170426203227.12321-10-rkrcmar@redhat.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 8bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 802
Lines: 26


On 26/04/2017 22:32, Radim Krčmář wrote:
>  v2: replaces [v1 1/6]
>      Ugh, KVM_ARCH_REQ_WAIT_NO_WAKEUP looks a weird ...

Yeah, let's drop patch 7 and just use bits for now.  I think using
KVM_ARCH_REQ_FLAGS directly should be fine, especially after the default
is flipped from "no wakeup" to "wakeup", but for 4.12 this is the
simplest incremental step.

> +/* TODO: merge with kvm_arch_vcpu_should_kick */
> +static bool kvm_should_kick_request(struct kvm_vcpu *vcpu, unsigned req)

I'm renaming this to kvm_request_needs_ipi; the point of the IPI for
synchronous requests is not the "kick", but the "ack" that comes back.

Paolo

> +{
> +	int mode = kvm_vcpu_exiting_guest_mode(vcpu);
> +
> +	return req & KVM_REQUEST_WAIT ?
> +		mode != OUTSIDE_GUEST_MODE : mode == IN_GUEST_MODE;
> +}
> +