<div dir="ltr"><div><div><div>Hi Ren, </div>Thank you for clarifying. I really do not have any more comments on the patch. </div>Regards, </div>Prerna </div><div class="gmail_extra"> <div class="gmail_quote">On Thu, Jul 9, 2015 at 12:27 PM, Ren, Qiaowei <<a href="mailto:qiaowei.ren@intel.com" target="_blank">qiaowei.ren@intel.com</a>> wrote: <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="HOEnZb"><div class="h5">On Jul 7, 2015 15:51, Ren, Qiaowei wrote: > > > On Jul 6, 2015 14:49, Prerna wrote: > >> On Sun, Jul 5, 2015 at 5:13 PM, Qiaowei Ren <<a href="mailto:qiaowei.ren@intel.com">qiaowei.ren@intel.com</a> >> <mailto:<a href="mailto:qiaowei.ren@intel.com">qiaowei.ren@intel.com</a>> > wrote: >> >> >> One RFC in >> <a href="https://www.redhat.com/archives/libvir-list/2015-June/msg01509.html" rel="noreferrer" target="_blank">https://www.redhat.com/archives/libvir-list/2015-June/msg01509.html</a> >> >> CMT (Cache Monitoring Technology) can be used to measure the >> usage of cache by VM running on the host. This patch will >> extend the bulk stats API (virDomainListGetStats) to add this >> field. Applications based on libvirt can use this API to achieve >> cache usage of VM. Because CMT implementation in Linux kernel >> is based on perf mechanism, this patch will enable perf event >> for CMT when VM is created and disable it when VM is destroyed. >> >> >> >> >> Hi Ren, >> >> One query wrt this implementation. I see you make a perf ioctl to >> gather CMT stats each time the stats API is invoked. >> >> If the CMT stats are exposed by a hardware counter, then this implies >> logging on a per-cpu (or per-socket ???) basis. >> >> This also implies that the value read will vary as the CPU (or socket) >> on which it is being called changes. >> >> >> Now, with this background, if we need real-world stats on a VM, we need >> this perf ioctl executed on all CPUs/ sockets on which the VM ran. >> Also, once done, we will need to aggregate results from each of these >> sources. >> >> >> In this implementation, I am missing this -- there seems no control >> over which physical CPU the libvirt worker thread will run and collect >> the perf data from. Data collected from this implementation might not >> accurately model the system state. >> >> I _think_ libvirt currently has no way of directing a worker thread to >> collect stats from a given CPU -- if we do, I would be happy to learn >> about it :) >> > > Prerna, thanks for your reply. I checked the CMT implementation in > kernel, and noticed that the series implement new ->count() of pmu > driver which can aggregate the results from each cpu if perf type is > PERF_TYPE_INTEL_CQM . The following is the link for the patch: > <a href="https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?i" rel="noreferrer" target="_blank">https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?i</a> > d=bfe1fc d2688f557a6b6a88f59ea7619228728bd7 > > So I guess that this patch just need to set right perf type and "cpu=-1". Do you > think this is ok? > </div></div>Hi Prerna, Do you have more comments on this patch series? I would be glad to update my implementation. ^-^ Qiaowei </blockquote></div> </div>