1. 15 Aug, 2014 1 commit
    • Jonathan Jenkins's avatar
      scheduler fix for other networks (NOT torus, simplewan) · 0164aa2d
      Jonathan Jenkins authored
      Torus and simplewan each have problems precluding them from the current
      scheduling fix:
      - simplewan - each "device" has N input/output ports. It can't simply tell the
        scheduler when they are (it is?) idle because the scheduler doesn't know which
        packets go to which ports
      - torus - also has N input/output ports (two for each dimension). Also, the same
        routing "queue" (via the "next_link_available_time" var) is used for incoming
        and outgoing messages, so we can't guarantee the scheduler that we'll be
        available at time x (an incoming msg could arrive and then be routed at time
        x-1). This isn't a problem for the dragonfly network as terminals aren't
        intermediate routers. Ideally what needs to happen here is for the
        intermediate packets/chunks to be queued up in the scheduler.
  2. 31 Jul, 2014 3 commits
    • Jonathan Jenkins's avatar
    • Jonathan Jenkins's avatar
      fix - pointer to local buffer escaped · 351a5de3
      Jonathan Jenkins authored
    • Jonathan Jenkins's avatar
      config overhaul + annotation-aware mapping · 4e19d940
      Jonathan Jenkins authored
      All configuration now proceeds on a per-LP level, and requires separate
      registration and configuration calls, as seen in the test programs.
      model_net_set_params is no longer used, and is replaced by model_net_register
      and model_net_configure. The dragonfly network, having two LP types bundled in
      the same code-path, is special-cased in the registration code.
      LP-mapping in model-net now has the following defaults:
      - counts via codes_mapping_get_lp_count are now with respect to the calling
        network LP's annotation.
      - when looking up network LPs via
        codes_mapping_get_lp_info/codes_mapping_get_lp_id, the annotation of the
        calling network LP is used. Hence, routing now occurs only between LPs of the
        same annotation. If the destination LP's group specified by model_net_*event
        does not contain a modelnet LP with the same annotation as the modelnet LP in
        the sender's group, then an error will occur (in codes_mapping).
      Known Issues:
      - modelnet users currently cannot specify which modelnet LP to use in the case
        of multiple modelnet LPs in the sender's group. This will be fixed in future
        commits after a consensus is reached on the best way to expose this
  3. 24 Jul, 2014 1 commit
  4. 18 Jul, 2014 1 commit
  5. 08 Jul, 2014 1 commit
  6. 07 Jul, 2014 2 commits
  7. 06 Jul, 2014 1 commit
  8. 03 Jul, 2014 1 commit
  9. 13 Jun, 2014 1 commit
  10. 11 Jun, 2014 1 commit
  11. 21 May, 2014 1 commit
  12. 14 May, 2014 1 commit
  13. 05 May, 2014 1 commit
  14. 23 Apr, 2014 1 commit
  15. 16 Apr, 2014 1 commit
  16. 08 Apr, 2014 1 commit
  17. 19 Dec, 2013 1 commit
  18. 13 Dec, 2013 1 commit
  19. 29 Aug, 2013 1 commit
  20. 27 Aug, 2013 1 commit
  21. 09 Aug, 2013 1 commit
    • Misbah Mubarak's avatar
      Updates to modelnet (1) Updated the mapping configuration for dragonfly (the... · 7a4ef14f
      Misbah Mubarak authored
      Updates to modelnet (1) Updated the mapping configuration for dragonfly (the mapping now distributes the routers equally on all the PES.  The old dragonly mapping configuration was placing the routers on a single PE which was causing congestion on that PE leading to slow model performance in parallel ) (2) Updated 'make check' to handle dragonfly and torus test cases as well (3) Remove the num_servers argument from model-net test case, the number of servers are now calculated from the config file (4) Placing the ross mapping parameters in the codes mapping file
  22. 08 Aug, 2013 1 commit