documentation improvements

doc/doxygen/chapters/210_check_list_performance.doxy

@@ -13,6 +13,9 @@ TODO: improve!
 To achieve good
 performance, we give below a list of features which should be checked.
 
+For a start, you can use \ref OfflinePerformanceTools to get a Gantt chart which
+will show roughly where time is spent, and focus correspondingly.
+
 \section ConfigurationImprovePerformance Configuration That May Improve Performance
 
 The \ref enable-fast "--enable-fast" configuration option disables all

doc/doxygen/chapters/380_offline_performance_tools.doxy

@@ -123,6 +123,9 @@ $ vite paje.trace
 To get names of tasks instead of "unknown", fill the optional
 starpu_codelet::name, or use a performance model for them.
 
+One can also introduce user-defined events in the diagram thanks to the
+starpu_fxt_trace_user_event_string() function.
+
 In the MPI execution case, \ref STARPU_GENERATE_TRACE will not work as expected
 (each node will try to generate paje.trace, thus mixing outputs...), you have to
 collect the trace files from the MPI nodes, and

doc/doxygen/chapters/410_mpi_support.doxy

@@ -190,7 +190,9 @@ int main(int argc, char **argv)
 
 We have here replaced <c>MPI_Recv()</c> and <c>MPI_Send()</c> with starpu_mpi_irecv_detached()
 and starpu_mpi_isend_detached(), which just submit the communication to be
-performed. The only remaining synchronization with starpu_data_acquire() is at
+performed. The implicit sequential consistency dependencies provide
+synchronization between mpi reception and emission and the corresponding tasks.
+The only remaining synchronization with starpu_data_acquire() is at
 the beginning and the end.
 
 \section MPIInitialization How to Initialize StarPU-MPI