From patchwork Thu May 28 18:49:10 2015 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Christoffer Dall X-Patchwork-Id: 6500551 Return-Path: X-Original-To: patchwork-linux-arm@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork1.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.136]) by patchwork1.web.kernel.org (Postfix) with ESMTP id 140D59F38C for ; Thu, 28 May 2015 18:51:22 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id 14FCC2070C for ; Thu, 28 May 2015 18:51:21 +0000 (UTC) Received: from bombadil.infradead.org (bombadil.infradead.org [198.137.202.9]) (using TLSv1.2 with cipher DHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 781282076C for ; Thu, 28 May 2015 18:51:15 +0000 (UTC) Received: from localhost ([127.0.0.1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.80.1 #2 (Red Hat Linux)) id 1Yy2re-00043z-T1; Thu, 28 May 2015 18:49:10 +0000 Received: from mail-la0-f47.google.com ([209.85.215.47]) by bombadil.infradead.org with esmtps (Exim 4.80.1 #2 (Red Hat Linux)) id 1Yy2rY-0003sT-4I for linux-arm-kernel@lists.infradead.org; Thu, 28 May 2015 18:49:05 +0000 Received: by lagv1 with SMTP id v1so39012176lag.3 for ; Thu, 28 May 2015 11:48:41 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:from:to:cc:subject:date:message-id; bh=ruspFXiugKuPFwwoZ21TARv65oRDwbg3ZuluSi+Riro=; b=cBQ3VKuQeRRXrp6OjptLlouczI7QbhsrCnE2il4kqwki3azZ+avjOuLx5+xvr/wyFW OpjH+nzitqigeth7RCq4/k4biWW4eA2i87upmSC78LNmKXXVRC22G0qWrHsiXJAroejV C8PtHAqZx8Mj3jopsb6W7HYd3EpnmjFe+omJu++J2NLA/FZBCEbkog55x2sAZy2yZEQJ MFT12KgSluh/IkbmiVH15qZ4ML8fzBuYYq4kMqZCdKEnU6Ysu9UxD+zxQBadOZuauXFG VgG2+1RmJYhSdBwIjtP7jSJVU/kQRgNw5a4h+h9kFMOtc6sK3WJasvLSrjEQqi4t8jdQ nXrQ== X-Gm-Message-State: ALoCoQlHJywQbjoVbSu1glX/ycWzw6gRSHjYOsJy0gR8b3HGavaLWARsbq0zpWfxk9P7S0cEUs/S X-Received: by 10.112.29.36 with SMTP id g4mr4327509lbh.56.1432838921268; Thu, 28 May 2015 11:48:41 -0700 (PDT) Received: from localhost.localdomain (188-178-240-98-static.dk.customer.tdc.net. [188.178.240.98]) by mx.google.com with ESMTPSA id yq5sm756117lab.32.2015.05.28.11.48.39 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-SHA bits=128/128); Thu, 28 May 2015 11:48:40 -0700 (PDT) From: Christoffer Dall To: kvmarm@lists.cs.columbia.edu, kvm@vger.kernel.org, linux-arm-kernel@lists.infradead.org Subject: [PATCH v2] arm/arm64: KVM: Properly account for guest CPU time Date: Thu, 28 May 2015 20:49:10 +0200 Message-Id: <1432838950-28774-1-git-send-email-christoffer.dall@linaro.org> X-Mailer: git-send-email 2.1.2.330.g565301e.dirty X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20150528_114904_487227_8298F960 X-CRM114-Status: GOOD ( 15.85 ) X-Spam-Score: -0.7 (/) Cc: Marc Zyngier , borntraeger@de.ibm.com, Christoffer Dall , Paolo Bonzini X-BeenThere: linux-arm-kernel@lists.infradead.org X-Mailman-Version: 2.1.18-1 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , MIME-Version: 1.0 Sender: "linux-arm-kernel" Errors-To: linux-arm-kernel-bounces+patchwork-linux-arm=patchwork.kernel.org@lists.infradead.org X-Spam-Status: No, score=-4.2 required=5.0 tests=BAYES_00, RCVD_IN_DNSWL_MED, T_RP_MATCHES_RCVD, UNPARSEABLE_RELAY autolearn=unavailable version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP Until now we have been calling kvm_guest_exit after re-enabling interrupts when we come back from the guest, but this has the unfortunate effect that CPU time accounting done in the context of timer interrupts occurring while the guest is running doesn't properly notice that the time since the last tick was spent in the guest. Inspired by the comment in the x86 code, move the kvm_guest_exit() call below the local_irq_enable() call and change __kvm_guest_exit() to kvm_guest_exit(), because we are now calling this function with interrupts enabled. We have to now explicitly disable preemption and not enable preemption before we've called kvm_guest_exit(), since otherwise we could be preempted and everything happening before we eventually get scheduled again would be accounted for as guest time. At the same time, move the trace_kvm_exit() call outside of the atomic section, since there is no reason for us to do that with interrupts disabled. Signed-off-by: Christoffer Dall --- This patch is based on kvm/queue, because it has the kvm_guest_enter/exit rework recently posted by Christian Borntraeger. I hope I got the logic of this right, there were 2 slightly worrying facts about this: First, we now enable and disable and enable interrupts on each exit path, but I couldn't see any performance overhead on hackbench - yes the only benchmark we care about. Second, looking at the ppc and mips code, they seem to also call kvm_guest_exit() before enabling interrupts, so I don't understand how guest CPU time accounting works on those architectures. Changes since v1: - Tweak comment and commit text based on Marc's feedback. - Explicitly disable preemption and enable it only after kvm_guest_exit(). arch/arm/kvm/arm.c | 21 +++++++++++++++++---- 1 file changed, 17 insertions(+), 4 deletions(-) diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c index e41cb11..fe8028d 100644 --- a/arch/arm/kvm/arm.c +++ b/arch/arm/kvm/arm.c @@ -532,6 +532,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run) kvm_vgic_flush_hwstate(vcpu); kvm_timer_flush_hwstate(vcpu); + preempt_disable(); local_irq_disable(); /* @@ -544,6 +545,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run) if (ret <= 0 || need_new_vmid_gen(vcpu->kvm)) { local_irq_enable(); + preempt_enable(); kvm_timer_sync_hwstate(vcpu); kvm_vgic_sync_hwstate(vcpu); continue; @@ -559,8 +561,10 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run) ret = kvm_call_hyp(__kvm_vcpu_run, vcpu); vcpu->mode = OUTSIDE_GUEST_MODE; - __kvm_guest_exit(); - trace_kvm_exit(kvm_vcpu_trap_get_class(vcpu), *vcpu_pc(vcpu)); + /* + * Back from guest + *************************************************************/ + /* * We may have taken a host interrupt in HYP mode (ie * while executing the guest). This interrupt is still @@ -574,8 +578,17 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run) local_irq_enable(); /* - * Back from guest - *************************************************************/ + * We do local_irq_enable() before calling kvm_guest_exit() so + * that if a timer interrupt hits while running the guest we + * account that tick as being spent in the guest. We enable + * preemption after calling kvm_guest_exit() so that if we get + * preempted we make sure ticks after that is not counted as + * guest time. + */ + kvm_guest_exit(); + trace_kvm_exit(kvm_vcpu_trap_get_class(vcpu), *vcpu_pc(vcpu)); + preempt_enable(); + kvm_timer_sync_hwstate(vcpu); kvm_vgic_sync_hwstate(vcpu);