Make _starpu_predict_transfer_time take into account that each activated GPU eats bandwidth.

doc/starpu.texi

@@ -1358,15 +1358,10 @@ tries to minimize is @code{alpha * T_execution + beta * T_data_transfer}, where
 accurate), and @code{T_data_transfer} is the estimated data transfer time. The
 latter is however estimated based on bus calibration before execution start,
 i.e. with an idle machine. You can force bus re-calibration by running
-@code{starpu_calibrate_bus}. When StarPU manages several GPUs, such estimation
-is not accurate any more. Beta can then be used to correct this by hand. For
-instance, you can use @code{export STARPU_BETA=2} to double the transfer
-time estimation, e.g. because there are two GPUs in the machine. This is of
-course imprecise, but in practice, a rough estimation already gives the good
-results that a precise estimation would give.
-
-Measuring the actual data transfer time is however on our TODO-list to
-accurately estimate data transfer penalty without the need of a hand-tuned beta parameter.
+@code{starpu_calibrate_bus}. The beta parameter defaults to 1, but it can be
+worth trying to tweak it by using @code{export STARPU_BETA=2} for instance.
+This is of course imprecise, but in practice, a rough estimation already gives
+the good results that a precise estimation would give.
 
 @node Power-based scheduling
 @section Power-based scheduling

src/core/perfmodel/perfmodel_bus.c

@@ -1,6 +1,6 @@
 /* StarPU --- Runtime system for heterogeneous multicore architectures.
  *
- * Copyright (C) 2009, 2010  Université de Bordeaux 1
+ * Copyright (C) 2009, 2010-2011  Université de Bordeaux 1
  * Copyright (C) 2010  Centre National de la Recherche Scientifique
  *
  * StarPU is free software; you can redistribute it and/or modify
@@ -1028,6 +1028,7 @@ double _starpu_predict_transfer_time(unsigned src_node, unsigned dst_node, size_
 {
 	double bandwidth = bandwidth_matrix[src_node][dst_node];
 	double latency = latency_matrix[src_node][dst_node];
+	struct starpu_machine_topology_s *topology = &_starpu_get_machine_config()->topology;
 
-	return latency + size/bandwidth;
+	return latency + (size/bandwidth)*2*(topology->ncudagpus+topology->nopenclgpus);
 }