You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Added PingHistogram, QualityHistogram, and JitterHistogram structs. Moved
some logic in there as methods, and fixed some duplicaetd code.
Don't assume a PingTracker wants to collect a detailed histogram. (There
is a use case in the code to deal with relays where this is not useful.)
intm_nQualityHistogram100; // This means everything was perfect. If we delivered over 100 packets in the interval and were less than perfect, but greater than 99.5%, we will use 99% instead.
130
-
intm_nQualityHistogram99; // 99%+
131
-
intm_nQualityHistogram97;
132
-
intm_nQualityHistogram95;
133
-
intm_nQualityHistogram90;
134
-
intm_nQualityHistogram75;
135
-
intm_nQualityHistogram50;
136
-
intm_nQualityHistogram1;
137
-
intm_nQualityHistogramDead; // we received nothing during the interval; it looks like the connection dropped
138
-
inlineintQualityHistogramTotalCount() const
139
-
{
140
-
returnm_nQualityHistogram100
141
-
+m_nQualityHistogram99
142
-
+m_nQualityHistogram97
143
-
+m_nQualityHistogram95
144
-
+m_nQualityHistogram90
145
-
+m_nQualityHistogram75
146
-
+m_nQualityHistogram50
147
-
+m_nQualityHistogram1
148
-
+m_nQualityHistogramDead;
149
-
}
212
+
QualityHistogramm_qualityHistogram;
150
213
151
214
// Distribution. Some might be -1, see above for why.
152
215
shortm_nQualityNtile2nd; // 2% of measurement intervals had quality <= N%
0 commit comments