From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sean Christopherson Date: Sat, 25 Sep 2021 00:55:21 +0000 Subject: [PATCH 07/14] KVM: Don't block+unblock when halt-polling is successful Message-Id: <20210925005528.1145584-8-seanjc@google.com> List-Id: References: <20210925005528.1145584-1-seanjc@google.com> In-Reply-To: <20210925005528.1145584-1-seanjc@google.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Marc Zyngier , Huacai Chen , Aleksandar Markovic , Paul Mackerras , Christian Borntraeger , Janosch Frank , Paolo Bonzini Cc: James Morse , Alexandru Elisei , Suzuki K Poulose , David Hildenbrand , Cornelia Huck , Claudio Imbrenda , Sean Christopherson , Vitaly Kuznetsov , Wanpeng Li , Jim Mattson , Joerg Roedel , linux-arm-kernel@lists.infradead.org, kvmarm@lists.cs.columbia.edu, linux-mips@vger.kernel.org, kvm@vger.kernel.org, kvm-ppc@vger.kernel.org, linux-kernel@vger.kernel.org, David Matlack , Jing Zhang Invoke the arch hooks for block+unblock if and only if KVM actually attempts to block the vCPU. The only non-nop implementation is on arm64, and if halt-polling is successful, there is no need for arm64 to put/load the vGIC as KVM hasn't relinquished control of the vCPU in any way. The primary motivation is to allow future cleanup to split out "block" from "halt", but this is also likely a small performance boost on arm64 when halt-polling is successful. Adjust the post-block path to update "cur" after unblocking, i.e. include vGIC load time in halt_wait_ns and halt_wait_hist, so that the behavior is consistent. Moving just the pre-block arch hook would result in only the vGIC put latency being included in the halt_wait stats. There is no obvious evidence that one way or the other is correct, so just ensure KVM is consistent. Cc: Marc Zyngier Cc: James Morse Cc: Alexandru Elisei Cc: Suzuki K Poulose Signed-off-by: Sean Christopherson --- virt/kvm/kvm_main.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 2015a1f532ce..f96cda8312f3 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -3232,8 +3232,6 @@ void kvm_vcpu_block(struct kvm_vcpu *vcpu) bool waited = false; u64 block_ns; - kvm_arch_vcpu_blocking(vcpu); - start = cur = poll_end = ktime_get(); if (do_halt_poll) { ktime_t stop = ktime_add_ns(ktime_get(), vcpu->halt_poll_ns); @@ -3250,6 +3248,7 @@ void kvm_vcpu_block(struct kvm_vcpu *vcpu) } while (kvm_vcpu_can_poll(cur, stop)); } + kvm_arch_vcpu_blocking(vcpu); prepare_to_rcuwait(&vcpu->wait); for (;;) { @@ -3262,6 +3261,9 @@ void kvm_vcpu_block(struct kvm_vcpu *vcpu) schedule(); } finish_rcuwait(&vcpu->wait); + + kvm_arch_vcpu_unblocking(vcpu); + cur = ktime_get(); if (waited) { vcpu->stat.generic.halt_wait_ns +@@ -3269,8 +3271,8 @@ void kvm_vcpu_block(struct kvm_vcpu *vcpu) KVM_STATS_LOG_HIST_UPDATE(vcpu->stat.generic.halt_wait_hist, ktime_to_ns(cur) - ktime_to_ns(poll_end)); } + out: - kvm_arch_vcpu_unblocking(vcpu); block_ns = ktime_to_ns(cur) - ktime_to_ns(start); /* -- 2.33.0.685.g46640cef36-goog