[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <CACzj_yVRifkBrMt2q6Vjvr+anXr0n+9J_iERR+UnZVwz95jN5A@mail.gmail.com>
Date: Fri, 16 Jan 2015 13:49:10 +0800
From: Wincy Van <fanwenyi0529@...il.com>
To: Paolo Bonzini <pbonzini@...hat.com>, gleb@...nel.org,
yang.z.zhang@...el.com
Cc: kvm@...r.kernel.org, linux-kernel@...r.kernel.org,
Wanpeng Li <wanpeng.li@...ux.intel.com>,
Jan Kiszka <jan.kiszka@....de>,
θζδΈ <fanwenyi0529@...il.com>
Subject: [PATCH 0/5] KVM: nVMX: Enable nested apicv support.
In nested vmx, the efficiency of interrupt virtualization is
very important, especially in high throughput scenes.
This patch set enables nested apicv support, which makes a
huge improvement in nested interrupt virtualization.
I also have done some simple tests:
L0: Intel Xeon E5-2630 v2
L1: CentOS 6.5 with 3.10.64-1.el6.elrepo.x86_64 kernel
16 vcpus, 32GB memory.
L2: Windows Server 2008 R2 Datacenter
8 vcpus, 16GB memory.
1. Run wprime 32M, 8 threads.
original nested apicv
7.782s 7.172s
Improvement: 7.8%
2. Run iperf -s -w 64k in L1,
iperf -c 10.1.0.2 -p 5001 -i 1 -t 30 -P 8 -w 64k in L2
original nested apicv
2.12 Gbits/s 3.50 Gbits/s
Improvement: 65.0%
_________________________________________________________
L2: CentOS 6.5 with 2.6.32-431 kernel
8 vcpus, 16GB memory.
1. Run iperf -s -w 64k in L1,
iperf -c 10.1.0.2 -p 5001 -i 1 -t 30 -P 8 -w 64k in L2
original nested apicv
6.58 Gbits/s 14.2 Gbits/s
Improvement: 115.8%
Wincy Van (5):
KVM: nVMX: Make nested control MSRs per-cpu.
KVM: nVMX: Enable nested virtualize x2apic mode.
KVM: nVMX: Enable nested apic register virtualization.
KVM: nVMX: Enable nested virtual interrupt delivery.
KVM: nVMX: Enable nested posted interrupt processing.
arch/x86/kvm/vmx.c | 444 +++++++++++++++++++++++++++++++++++++++++-----------
1 files changed, 355 insertions(+), 89 deletions(-)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists