All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=809492

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=7136542ae6e96c38c9c33a7d96575558857e2fd4382443bc31a3db09582c7dfd


Bot(s) for this bug's original alert(s):

chromium-rel-win7-x64-dual

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/12a704b1840000

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/12a704b1840000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14eaf4c2440000

Re-running bisect on a wider range.

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/14eaf4c2440000

NetworkService: Implement URLLoader::OnCertificateRequested by juncai@chromium.org
https://chromium.googlesource.com/chromium/src/+/9409ded1ffb68a4fba35a4a37dc3e39a8f1e5718

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

I took a look at the code in the CL:
https://chromium-review.googlesource.com/c/chromium/src/+/848303
and haven't found obvious change which causes this regression. The only place that may cause this is that we now use ContentBrowserClient::CreateClientCertStore() to create an instance of net::ClientCertStore instead of using ResourceLoaderDelegate::CreateClientCertStore(). Since this change is necessary for the Network Service, I am wondering if this can justify marking this bug as "WontFix".

Most likely it's just misattributed. 
1) Creating a ClientCertStore doesn't actually do any work.
2) I doubt perf tests are using client certs, so they wouldn't be hitting this code at all.