Requalify automatic tesh, as another algorithm is used in init_smp now.

[simgrid.git] / teshsuite / smpi / coll-allreduce / coll-allreduce-automatic.tesh

diff --git a/teshsuite/smpi/coll-allreduce/coll-allreduce-automatic.tesh b/teshsuite/smpi/coll-allreduce/coll-allreduce-automatic.tesh
index b8b7762..2a4d8f9 100644 (file)
--- a/teshsuite/smpi/coll-allreduce/coll-allreduce-automatic.tesh
+++ b/teshsuite/smpi/coll-allreduce/coll-allreduce-automatic.tesh
@@ -19,22 +19,22 @@ $ ${bindir:=.}/../../../smpi_script/bin/smpirun -map -hostfile ../hostfile_coll
 > [rank 13] -> Ginette
 > [rank 14] -> Ginette
 > [rank 15] -> Ginette
-> [  0.427764] (8:7@Jupiter) The quickest allreduce was redbcast on rank 7 and took 0.007546
-> [  0.427764] (5:4@Jupiter) The quickest allreduce was redbcast on rank 4 and took 0.007485
-> [  0.427764] (7:6@Jupiter) The quickest allreduce was redbcast on rank 6 and took 0.007515
-> [  0.427764] (6:5@Jupiter) The quickest allreduce was redbcast on rank 5 and took 0.007515
-> [  0.427976] (14:13@Ginette) The quickest allreduce was mvapich2_two_level on rank 13 and took 0.007278
-> [  0.427976] (13:12@Ginette) The quickest allreduce was mvapich2_two_level on rank 12 and took 0.007247
-> [  0.427976] (16:15@Ginette) The quickest allreduce was ompi on rank 15 and took 0.007263
-> [  0.427976] (15:14@Ginette) The quickest allreduce was mvapich2_two_level on rank 14 and took 0.007278
-> [  0.429367] (2:1@Tremblay) The quickest allreduce was redbcast on rank 1 and took 0.006006
-> [  0.429367] (3:2@Tremblay) The quickest allreduce was redbcast on rank 2 and took 0.006006
-> [  0.429367] (4:3@Tremblay) The quickest allreduce was redbcast on rank 3 and took 0.006037
-> [  0.430519] (12:11@Fafard) The quickest allreduce was mvapich2_two_level on rank 11 and took 0.006523
-> [  0.430519] (10:9@Fafard) The quickest allreduce was mvapich2_two_level on rank 9 and took 0.006492
-> [  0.430519] (9:8@Fafard) The quickest allreduce was mvapich2_two_level on rank 8 and took 0.006462
-> [  0.430519] (11:10@Fafard) The quickest allreduce was mvapich2_two_level on rank 10 and took 0.006492
-> [  0.434504] (1:0@Tremblay) For rank 0, the quickest was redbcast : 0.005991 , but global was mvapich2_two_level : 0.008672 at max
+> [  0.475403] (5:4@Jupiter) The quickest allreduce was redbcast on rank 4 and took 0.007485
+> [  0.475403] (6:5@Jupiter) The quickest allreduce was redbcast on rank 5 and took 0.007515
+> [  0.475403] (7:6@Jupiter) The quickest allreduce was redbcast on rank 6 and took 0.007515
+> [  0.475403] (8:7@Jupiter) The quickest allreduce was redbcast on rank 7 and took 0.007546
+> [  0.475616] (13:12@Ginette) The quickest allreduce was mvapich2_two_level on rank 12 and took 0.007247
+> [  0.475616] (14:13@Ginette) The quickest allreduce was mvapich2_two_level on rank 13 and took 0.007278
+> [  0.475616] (15:14@Ginette) The quickest allreduce was mvapich2_two_level on rank 14 and took 0.007278
+> [  0.475616] (16:15@Ginette) The quickest allreduce was ompi on rank 15 and took 0.007263
+> [  0.477007] (2:1@Tremblay) The quickest allreduce was redbcast on rank 1 and took 0.006006
+> [  0.477007] (3:2@Tremblay) The quickest allreduce was redbcast on rank 2 and took 0.006006
+> [  0.477007] (4:3@Tremblay) The quickest allreduce was redbcast on rank 3 and took 0.006037
+> [  0.478158] (10:9@Fafard) The quickest allreduce was mvapich2_two_level on rank 9 and took 0.006492
+> [  0.478158] (11:10@Fafard) The quickest allreduce was mvapich2_two_level on rank 10 and took 0.006492
+> [  0.478158] (12:11@Fafard) The quickest allreduce was mvapich2_two_level on rank 11 and took 0.006523
+> [  0.478158] (9:8@Fafard) The quickest allreduce was mvapich2_two_level on rank 8 and took 0.006462
+> [  0.482143] (1:0@Tremblay) For rank 0, the quickest was redbcast : 0.005991 , but global was mvapich2_two_level : 0.008672 at max
 > [0] sndbuf=[0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 ]
 > [1] sndbuf=[16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 ]
 > [2] sndbuf=[32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 ]