From b401ee0b85a53e89739ff68a5b1a0667d664afc9 Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Tue, 18 Apr 2017 12:41:18 +0200 Subject: [PATCH] KVM: x86: lower default for halt_poll_ns MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In some fio benchmarks, halt_poll_ns=400000 caused CPU utilization to increase heavily even in cases where the performance improvement was small. In particular, bandwidth divided by CPU usage was as much as 60% lower. To some extent this is the expected effect of the patch, and the additional CPU utilization is only visible when running the benchmarks. However, halving the threshold also halves the extra CPU utilization (from +30-130% to +20-70%) and has no negative effect on performance. Signed-off-by: Paolo Bonzini Signed-off-by: Radim Krčmář --- arch/x86/include/asm/kvm_host.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 9c761fea0c98..695605eb1dfb 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -43,7 +43,7 @@ #define KVM_PRIVATE_MEM_SLOTS 3 #define KVM_MEM_SLOTS_NUM (KVM_USER_MEM_SLOTS + KVM_PRIVATE_MEM_SLOTS) -#define KVM_HALT_POLL_NS_DEFAULT 400000 +#define KVM_HALT_POLL_NS_DEFAULT 200000 #define KVM_IRQCHIP_NUM_PINS KVM_IOAPIC_NUM_PINS -- GitLab