NonValidatingReload: Monitor reload operations in NavigationControllerImpl

tools/metrics/histograms/histograms.xml

Index: tools/metrics/histograms/histograms.xml
diff --git a/tools/metrics/histograms/histograms.xml b/tools/metrics/histograms/histograms.xml
index 29bd28575de8e128826d0575f666c29494cc543e..3a05e24f4cefe70cbd1f243c1de829933f05da15 100644
--- a/tools/metrics/histograms/histograms.xml
+++ b/tools/metrics/histograms/histograms.xml
@@ -25639,6 +25639,25 @@ http://cs/file:chrome/histograms.xml - but prefer this file for new entries.
   </summary>
 </histogram>
 
+<histogram name="Navigation.Reload.ReloadMainResourceToReloadDuration"
+    units="ms">
+  <owner>toyoshim@chromium.org</owner>
+  <summary>
+    Reported when a user triggers reload without any other navigations after the
+    previous reload in the same page, and the previous reload variant was
+    RELOAD_MAIN_RESOURCE. Duration time between two reloads is reported.
+  </summary>
+</histogram>
+
+<histogram name="Navigation.Reload.ReloadToReloadDuration" units="ms">
+  <owner>toyoshim@chromium.org</owner>
+  <summary>
+    Reported when a user triggers reload without any other navigations after the
+    previous reload in the same page. Duration time between two reloads is
+    reported.
+  </summary>
+</histogram>
+
 <histogram name="Navigation.Scheduled.MaybeCausedAbort"
     enum="ScheduledNavigationType">
   <owner>csharrison@chromium.org</owner>
@@ -53524,7 +53543,7 @@ http://cs/file:chrome/histograms.xml - but prefer this file for new entries.
   <summary>
     The disposition of a stranded Chrome install discovered while updating
     multi-install Chrome Binaries. This metric is only logged in an edge case.
-    The buckets indicate whether or not the &quot;stranded&quot; Chrome install
+    The buckets indicate whether or not the &quot; stranded&quot; Chrome install

[Charlie Reis, 2016/08/08 23:57:54]

Unintentional?

[Takashi Toyoshima, 2016/08/09 12:06:17]

This is not related to my CL, but presubmit check could not pass without this
format fix unfortunately.

[Charlie Reis, 2016/08/10 05:05:13]

Huh.  Seems like a strange requirement from the presubmit check, since it's the
wrong thing to do to the description (i.e., it inserts a typo).

I won't object, but it might be worth looking into whether there's a bug in the
presubmit.

[Takashi Toyoshima, 2016/08/10 06:34:15]

Precisely said, pretty_print.py could not parse if there is no space after
'&NAME;' around the modified code. OK, I'll investigate this in parallel.

[Takashi Toyoshima, 2016/08/10 07:26:46]

I checked if this is sill needed on trunk, and it turned out that we do not need
this workaround I added at Patch Set 4 any more. Let me revert this in the newer
patch set.

     has been used in the last 28 days. If so, it respresents a Chrome that has
     not been updating due to a logic flaw in the installer and will be repaired
     in the current update. If not, it likely represents Chrome Binaries that