Actually, that's not true - the 1.5 series technically still supports assignment to physical cpus. However, it is never really tested and very unusual for someone to use, so I suspect it is broken. I very much doubt anyone will fix it.

Also, be aware that physical cpu assignments are not supported in the current developer trunk. This will likely be the case when it is released as the 1.7 series and going forward.


On Feb 2, 2012, at 10:17 AM, teng ma wrote:

Just remove p in your rankfile like

rank 0=host1 slot=0:0
rank 1=host1 slot=0:2
rank 2=host1 slot=0:4
rank 3=host1 slot=0:6
rank 4=host1 slot=1:1
rank 5=host1 slot=1:3
rank 6=host1 slot=1:5
rank 7=host1 slot=1:7

Teng

2012/2/2 François Tessier <francois.tessier@inria.fr>
Hello,

I need to use a rankfile with openMPI 1.5.4 to do some tests on a basic architecture. I'm using a node for which lstopo returns that :

----------------
Machine (24GB)
  NUMANode L#0 (P#0 12GB)
    Socket L#0 + L3 L#0 (8192KB)
      L2 L#0 (256KB) + L1 L#0 (32KB) + Core L#0 + PU L#0 (P#0)
      L2 L#1 (256KB) + L1 L#1 (32KB) + Core L#1 + PU L#1 (P#2)
      L2 L#2 (256KB) + L1 L#2 (32KB) + Core L#2 + PU L#2 (P#4)
      L2 L#3 (256KB) + L1 L#3 (32KB) + Core L#3 + PU L#3 (P#6)
    HostBridge L#0
      PCIBridge
        PCI 8086:10c9
          Net L#0 "eth0"
        PCI 8086:10c9
          Net L#1 "eth1"
      PCIBridge
        PCI 15b3:673c
          Net L#2 "ib0"
          Net L#3 "ib1"
          OpenFabrics L#4 "mlx4_0"
      PCIBridge
        PCI 102b:0522
      PCI 8086:3a22
        Block L#5 "sda"
        Block L#6 "sdb"
        Block L#7 "sdc"
        Block L#8 "sdd"
  NUMANode L#1 (P#1 12GB) + Socket L#1 + L3 L#1 (8192KB)
    L2 L#4 (256KB) + L1 L#4 (32KB) + Core L#4 + PU L#4 (P#1)
    L2 L#5 (256KB) + L1 L#5 (32KB) + Core L#5 + PU L#5 (P#3)
    L2 L#6 (256KB) + L1 L#6 (32KB) + Core L#6 + PU L#6 (P#5)
    L2 L#7 (256KB) + L1 L#7 (32KB) + Core L#7 + PU L#7 (P#7)
----------------

And I would like to use the physical numbering. To do that, I created a rankfile like this :

rank 0=host1 slot=p0:0
rank 1=host1 slot=p0:2
rank 2=host1 slot=p0:4
rank 3=host1 slot=p0:6
rank 4=host1 slot=p1:1
rank 5=host1 slot=p1:3
rank 6=host1 slot=p1:5
rank 7=host1 slot=p1:7

But when I run my job with "mpiexec -np 8 --rankfile rankfile ./foo", I encounter this error :

    Specified slot list: p0:4
    Error: Not found

    This could mean that a non-existent processor was specified, or
    that the specification had improper syntax.



Do you know what I did wrong?

Best regards,

François

-- 
___________________
François TESSIER
PhD Student at University of Bordeaux
Tel : 0033.5.24.57.41.52
francois.tessier@inria.fr


_______________________________________________
users mailing list
users@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users



--
| Teng Ma          Univ. of Tennessee |
| tma@cs.utk.edu        Knoxville, TN |
| http://web.eecs.utk.edu/~tma/       |

_______________________________________________
users mailing list
users@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users