|
@@ -390,6 +390,21 @@ $ starpu_paje_state_stats -v native.trace simgrid.trace
|
|
|
$ starpu_paje_state_stats native.trace simgrid.trace
|
|
|
\endverbatim
|
|
|
|
|
|
+An other way to get statistics of StarPU states (without installing R and
|
|
|
+pj_dump) is to use the starpu_trace_state_stats.py script which parses the
|
|
|
+generated trace.rec file instead of the paje.trace file. The output is similar
|
|
|
+to the previous script but it doesn't need any dependencies.
|
|
|
+
|
|
|
+Here's an example how to use it:
|
|
|
+
|
|
|
+\verbatim
|
|
|
+$ python starpu_trace_state_stats.py trace.rec | column -t -s ","
|
|
|
+"Callback" 220 0.075978
|
|
|
+"chol_model_11" 10 565.176
|
|
|
+"chol_model_21" 45 9184.828
|
|
|
+"chol_model_22" 165 64712.07
|
|
|
+\endverbatim
|
|
|
+
|
|
|
And one can plot histograms of execution times, of several states for instance:
|
|
|
\verbatim
|
|
|
$ starpu_paje_draw_histogram -n chol_model_11,chol_model_21,chol_model_22 native.trace simgrid.trace
|