<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto"><div><br></div><div>cpu_mode = cpu-passthrough </div><div>cpu_model = none</div><div><br></div><div>Do you think cpu_model make difference ?</div><div><br></div><br><div dir="ltr">Sent from my iPhone</div><div dir="ltr"><br><blockquote type="cite">On Mar 5, 2020, at 7:18 AM, Satish Patel<satish.txt@gmail.com> wrote:<br><br></blockquote></div><blockquote type="cite"><div dir="ltr"><meta http-equiv="content-type" content="text/html; charset=utf-8"><div><br></div>cpu-passthrough <br><br><div dir="ltr">Sent from my iPhone</div><div dir="ltr"><br><blockquote type="cite">On Mar 4, 2020, at 9:24 PM, rui zang <rui.zang@yandex.com> wrote:<br><br></blockquote></div><blockquote type="cite"><div dir="ltr"><div>Hi,</div><div> </div><div>What is the value for the "cpu_mode" configuration option?</div><div><a href="https://docs.openstack.org/mitaka/config-reference/compute/hypervisor-kvm.html">https://docs.openstack.org/mitaka/config-reference/compute/hypervisor-kvm.html</a></div><div> </div><div>Thanks,<br>Zang, Rui</div><div> </div><div> </div><div>05.03.2020, 01:24, "Satish Patel" <satish.txt@gmail.com>:</div><blockquote><p>Folks,<br><br>We are running openstack with KVM and i have noticed kvm presenting<br>wrong CPU Tolopoly to VM and because of that we are seeing bad<br>performance to our application.<br><br>This is openstack compute:<br><br># lstopo-no-graphics --no-io<br>Machine (64GB total)<br> NUMANode L#0 (P#0 32GB) + Package L#0 + L3 L#0 (25MB)<br> L2 L#0 (256KB) + L1d L#0 (32KB) + L1i L#0 (32KB) + Core L#0<br> PU L#0 (P#0)<br> PU L#1 (P#20)<br> L2 L#1 (256KB) + L1d L#1 (32KB) + L1i L#1 (32KB) + Core L#1<br> PU L#2 (P#1)<br> PU L#3 (P#21)<br><br>This is VM running on above compute<br><br># lstopo-no-graphics --no-io<br>Machine (59GB total)<br> NUMANode L#0 (P#0 29GB) + Package L#0 + L3 L#0 (16MB)<br> L2 L#0 (4096KB) + Core L#0<br> L1d L#0 (32KB) + L1i L#0 (32KB) + PU L#0 (P#0)<br> L1d L#1 (32KB) + L1i L#1 (32KB) + PU L#1 (P#1)<br> L2 L#1 (4096KB) + Core L#1<br> L1d L#2 (32KB) + L1i L#2 (32KB) + PU L#2 (P#2)<br> L1d L#3 (32KB) + L1i L#3 (32KB) + PU L#3 (P#3)<br><br>if you noticed P#0 and P#1 has own (32KB) cache per thread that is<br>wrong presentation if you compare with physical CPU.<br><br>This is a screenshot of AWS vs Openstack CPU Topology and looking at<br>openstack its presentation is little odd, is that normal?<br><br><a href="https://imgur.com/a/2sPwJVC">https://imgur.com/a/2sPwJVC</a><br><br>I am running CentOS7.6 with kvm 2.12 version.<br> </p></blockquote></div></blockquote></div></blockquote></body></html>