From mboxrd@z Thu Jan 1 00:00:00 1970 From: Yang Zhang Subject: Re: [RFC PATCH v2 0/7] x86/idle: add halt poll support Date: Fri, 1 Sep 2017 14:21:53 +0800 Message-ID: <2c048ae6-329e-1351-2711-e72e31c8554e@gmail.com> References: <1504007201-12904-1-git-send-email-yang.zhang.wz@gmail.com> <6ba7f198-4403-c9d1-f0be-7069cc8cd421@suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 8bit Cc: kvm@vger.kernel.org, wanpeng.li@hotmail.com, mst@redhat.com, pbonzini@redhat.com, tglx@linutronix.de, rkrcmar@redhat.com, dmatlack@google.com, peterz@infradead.org, linux-doc@vger.kernel.org To: Alexander Graf , linux-kernel@vger.kernel.org Return-path: In-Reply-To: <6ba7f198-4403-c9d1-f0be-7069cc8cd421@suse.de> Content-Language: en-US Sender: linux-kernel-owner@vger.kernel.org List-Id: kvm.vger.kernel.org On 2017/8/29 19:58, Alexander Graf wrote: > On 08/29/2017 01:46 PM, Yang Zhang wrote: >> Some latency-intensive workload will see obviously performance >> drop when running inside VM. The main reason is that the overhead >> is amplified when running inside VM. The most cost i have seen is >> inside idle path. >> >> This patch introduces a new mechanism to poll for a while before >> entering idle state. If schedule is needed during poll, then we >> don't need to goes through the heavy overhead path. >> >> Here is the data we get when running benchmark contextswitch to measure >> the latency(lower is better): >> >>     1. w/o patch: >>        2493.14 ns/ctxsw -- 200.3 %CPU >>     2. w/ patch: >>        halt_poll_threshold=10000 -- 1485.96ns/ctxsw -- 201.0 %CPU >>        halt_poll_threshold=20000 -- 1391.26 ns/ctxsw -- 200.7 %CPU >>        halt_poll_threshold=30000 -- 1488.55 ns/ctxsw -- 200.1 %CPU >>        halt_poll_threshold=500000 -- 1159.14 ns/ctxsw -- 201.5 %CPU >>     3. kvm dynamic poll >>        halt_poll_ns=10000 -- 2296.11 ns/ctxsw -- 201.2 %CPU >>        halt_poll_ns=20000 -- 2599.7 ns/ctxsw -- 201.7 %CPU >>        halt_poll_ns=30000 -- 2588.68 ns/ctxsw -- 211.6 %CPU >>        halt_poll_ns=500000 -- 2423.20 ns/ctxsw -- 229.2 %CPU >>     4. idle=poll >>        2050.1 ns/ctxsw -- 1003 %CPU >>     5. idle=mwait >>        2188.06 ns/ctxsw -- 206.3 %CPU > > Could you please try to create another metric for guest initiated, host > aborted mwait? > > For a quick benchmark, reserve 4 registers for a magic value, set them > to the magic value before you enter MWAIT in the guest. Then allow > native MWAIT execution on the host. If you see the guest wants to enter I guess you want to allow native MWAIT execution on the guest not host? > with the 4 registers containing the magic contents and no events are > pending, directly go into the vcpu block function on the host. Mmm..It is not very clear to me. If guest executes MWAIT without vmexit, how to check the register? > > That way any time a guest gets naturally aborted while in mwait, it will > only reenter mwait when an event actually occured. While the guest is > normally running (and nobody else wants to run on the host), we just > stay in guest context, but with a sleeping CPU. > > Overall, that might give us even better performance, as it allows for > turbo boost and HT to work properly. In our testing, we have enough cores(32cores) but only 10VCPUs, so in the best case, we may see the same performance as poll. -- Yang Alibaba Cloud Computing