Changes between Version 7 and Version 8 of RTSsummaryEvents

Timestamp:

12/21/11 07:26:41 (18 months ago)

Author:

MikolajKonarski (IP: 95.160.111.162)

Comment:

--

RTSsummaryEvents

@@ -62 +62 @@
 This is quite convoluted: total words copied during parallel GCs divided
 by the average over all parallel GCs of maximal number of words copied
-by any thread in a single par GC.Revents needed: the events added above suffice,
+by any thread in a single par GC. Events needed: the events added above suffice,
 but quite a bit of extra state will have to be maintained when reading
 the events. Perhaps +RTS -s could be modified to make this figure simpler?
@@ -90 +90 @@
 events for that (we calculate this using the SparkCounters events, 
 but we could also use the precise per-spark events).
+
+
+{{{
+  INIT    time    0.00s  (  0.00s elapsed)
+  MUT     time    0.42s  (  0.41s elapsed)
+  GC      time    0.29s  (  0.13s elapsed)
+  EXIT    time    0.00s  (  0.00s elapsed)
+  Total   time    0.71s  (  0.55s elapsed)
+}}}
+
+(Note that there may be more positions here, e.g., for profiling.)
+We can sum up the times from GC events. For other stats, we'd need the MUT
+figure, but it's not obvious if we manage to get it from all the thread (task)
+events that we have and will add above. It's also not clear if adding events
+needed to get the other times is worth it. After all the other events
+are added, let's see if we can get any more of these summary times,
+perhaps by adding a minor event emitted just once. Note that the INIT time is
+necessary for the Productivity figure below (it does not count as "productive
+time" in +RTS -s)..
+
+{{{
+%GC time      89.5%  (75.3% elapsed)
+}}}
+
+Ask JaffaCake why this line is in the docs, but not in the output
+of +RTS -s I've got.
+
+{{{
+  Alloc rate    564,074,971 bytes per MUT second
+}}}
+
+The events added above should suffice, except that we use elapsed time,
+not CPU time.
+
+{{{
+  Productivity  59.0% of total user, 76.6% of total elapsed
+}}}
+
+The events added above should be enough. Again. we only do the elapsed case.