intel_pmu 2.0 - address review comments

Improvements in plugin documentation in collectd.conf.pod
and other small fixes.

Change-Id: Iba5a762afce054b9c0f7adb29120bf37750ec98b
Signed-off-by: Kamil Wiatrowski <kamilx.wiatrowski@intel.com>

diff --git a/src/collectd.conf.in b/src/collectd.conf.in
index 84f38dbe6e3659ce46f369567a8eb978c821b0f4..19dab27087b046d85f3d02a6bafb23136f224a25 100644 (file)
--- a/src/collectd.conf.in
+++ b/src/collectd.conf.in
@@ -828,13 +828,10 @@
 #</Plugin>
 
 #<Plugin intel_pmu>
-##    ReportHardwareCacheEvents true
-##    ReportKernelPMUEvents true
-##    ReportSoftwareEvents true
 #    EventList "/var/cache/pmu/GenuineIntel-6-2D-core.json"
 #    HardwareEvents "L2_RQSTS.CODE_RD_HIT,L2_RQSTS.CODE_RD_MISS" "L2_RQSTS.ALL_CODE_RD"
 #    Cores "[0-3]"
-#    DispatchMultiPmu false
+#    DispatchMultiPmu true
 #</Plugin>
 
 #<Plugin "intel_rdt">

diff --git a/src/collectd.conf.pod b/src/collectd.conf.pod
index 2871d4b3c57c9d1ba718a39abd5d4ee975dbceab..c2e1c42f467e12a2129dd87ee7c587676326efab 100644 (file)
--- a/src/collectd.conf.pod
+++ b/src/collectd.conf.pod
@@ -3878,6 +3878,11 @@ B<except> the ones matched.
 The I<intel_pmu> plugin collects performance counters data on Intel CPUs using
 Linux perf interface. All events are reported on a per core basis.
 
+B<Note:> When using intel_pmu plugin number of reading threads in collectd
+should be adjusted accordingly. The value should be more than a half of
+configured cores, so for 60 monitored cores the recommendation is to set
+B<ReadThreads> > 30.
+
 B<Synopsis:>
 
   <Plugin intel_pmu>
@@ -3886,7 +3891,7 @@ B<Synopsis:>
     Cores "0-3" "4,6" "[12-15]"
     HardwareEvents "L2_RQSTS.PF_MISS"
     Cores "[1,2]"
-    DispatchMultiPmu false
+    DispatchMultiPmu true
   </Plugin>
 
 B<Options:>
@@ -3927,10 +3932,13 @@ This option can be used once for every B<HardwareEvents> set.
 
 =item B<DispatchMultiPmu> B<false>|B<true>
 
-Enable or disable dispatching of cloned multi PMU for uncore events. If
-disabled only total sum is dispatched as single event. If enabled separate
-metric is dispatched for every counter. Name of particular PMU counter is
-added to I<plugin_instance>.
+This option toggles the event value reporting from all the uncore PMUs to
+either dispatch as aggregated value in a single metric or dispatch as
+individual values. If DispatchMultiPmu is set to 'false' uncore events are
+reported as single metric, even for multi PMU. The total value is obtained by
+summing all counters across all the units (e.g. CHAs). If DispatchMultiPmu is
+set to 'true' all cloned PMUs are dispatched separately with 'type' value of
+PMU added to Collectd's I<type_instance>.
 
 =back
 

diff --git a/src/intel_pmu_test.c b/src/intel_pmu_test.c
index 8139a820c9979d7f1b28ccd19851a6ab1f5019ea..e5d696e0baa1b72a583b50b6849fc5806c9d1efd 100644 (file)
--- a/src/intel_pmu_test.c
+++ b/src/intel_pmu_test.c
@@ -1,7 +1,7 @@
 /**
- * collectd - src/logparser_test.c
+ * collectd - src/intel_pmu_test.c
  *
- * Copyright(c) 2018 Intel Corporation. All rights reserved.
+ * Copyright(c) 2020 Intel Corporation. All rights reserved.
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -232,4 +232,4 @@ int main(void) {
   RUN_TEST(config_cores_parse__not_aggregated_groups);
 
   END_TEST;
-}
\ No newline at end of file
+}