<div dir="ltr"><br><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Dec 14, 2016 at 4:12 AM, John Ferlan <span dir="ltr"><<a href="mailto:jferlan@redhat.com" target="_blank">jferlan@redhat.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><span class="gmail-"><br>
<br>
On 12/13/2016 11:01 AM, Nitesh Konkar wrote:<br>
</span><span class="gmail-">> Add support and documentation for generic<br>
> perf events like bus_cycles, stalled_cycles_frontend,<br>
> stalled_cycles_backend and ref_cpu_cycles.<br>
><br>
> Nitesh Konkar (4):<br>
> perf: add bus_cycles perf event support<br>
> perf: add stalled_cycles_frontend perf event support<br>
> perf: add stalled_cycles_backend perf event support<br>
> perf: add ref_cpu_cycles perf event support<br>
><br>
> docs/<a href="http://formatdomain.html.in" rel="noreferrer" target="_blank">formatdomain.html.in</a> | 24 +++++++++++++++++<br>
> docs/<a href="http://news.html.in" rel="noreferrer" target="_blank">news.html.in</a> | 6 +++--<br>
> docs/schemas/domaincommon.rng | 4 +++<br>
> include/libvirt/libvirt-<wbr>domain.h | 40 +++++++++++++++++++++++++++++<br>
> src/libvirt-domain.c | 11 ++++++++<br>
> src/qemu/qemu_driver.c | 4 +++<br>
> src/util/virperf.c | 16 +++++++++++-<br>
> src/util/virperf.h | 4 +++<br>
> tests/genericxml2xmlindata/<wbr>generic-perf.xml | 4 +++<br>
> tools/virsh.pod | 16 +++++++++++-<br>
> 10 files changed, 125 insertions(+), 4 deletions(-)<br>
><br>
<br>
</span>As long as you're OK with my word-smithing in patches 2 & 3 regarding a<br>
longer description for those cycles values... Then these are fine and I<br>
can push them once I get the OK. </blockquote><div>Makes sense. I m OK with it. Thanks!<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">If you can provide a better explanation<br>
for ref_cpu_cycles that would be fantastic.<br></blockquote><div><br>Based on the below two links:<br><a href="http://lkml.iu.edu/hypermail/linux/kernel/1112.1/01029.html">http://lkml.iu.edu/hypermail/linux/kernel/1112.1/01029.html</a><br><a href="http://man7.org/linux/man-pages/man2/perf_event_open.2.html">http://man7.org/linux/man-pages/man2/perf_event_open.2.html</a><br><br></div><div>I have come up with the following two explanations.<br><br>The count of total cpu cycles not affected by CPU frequency<br>scaling by applications running on the platform<br><br>OR<br><br>The count of the number of reference core cpu cycles<br>by applications running on the platform<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<br>
Tks -<br>
<br>
John<br></blockquote></div></div></div>