1. 06 Nov, 2014 1 commit
  2. 05 Nov, 2014 2 commits
  3. 04 Nov, 2014 3 commits
  4. 03 Nov, 2014 4 commits
  5. 09 Oct, 2014 2 commits
  6. 29 Aug, 2014 1 commit
  7. 21 Aug, 2014 1 commit
  8. 20 Aug, 2014 1 commit
    • Jonathan Jenkins's avatar
      recv-side queuing support (loggp only) · 6f9bc775
      Jonathan Jenkins authored
      Rather than model-net LPs directly sending messages to other model-net LPs, LPs
      can route the intended message through the scheduler interface to be queued up
      for reception by the receiver (see the diff of loggp.c). This has the benefit of
          enabling things like priority and fairness for N->1 communication patterns.
          Currently, no packetizing is supported, and I haven't yet wrote checks for
          it - beware.
      
      Loggp is currently the only supported model. simplenet could also be supported
      without much trouble, but I doubt there's any demand for it at the moment.  This
      should NOT be used by the dragonfly/torus models, as they have their own routing
      backend.
      6f9bc775
  9. 19 Aug, 2014 2 commits
  10. 18 Aug, 2014 1 commit
  11. 15 Aug, 2014 5 commits
    • 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.
      0164aa2d
    • Jonathan Jenkins's avatar
      scheduler - sched loop is callback-driven now (loggp only) · b3650ede
      Jonathan Jenkins authored
      - previously, packet issues were done without any consideration for device
        availability - within epsilon, preventing any meaningful scheduling
      - enabled for loggp only, other networks will be incorporated shortly
      b3650ede
    • Jonathan Jenkins's avatar
      loggp message logging · 697f5d17
      Jonathan Jenkins authored
      697f5d17
    • Jonathan Jenkins's avatar
      9ab8f039
    • Jonathan Jenkins's avatar
      debug messages for scheduler · 2d9622cf
      Jonathan Jenkins authored
      2d9622cf
  12. 14 Aug, 2014 1 commit
  13. 11 Aug, 2014 1 commit
  14. 08 Aug, 2014 1 commit
  15. 06 Aug, 2014 1 commit
  16. 04 Aug, 2014 1 commit
  17. 31 Jul, 2014 4 commits
    • Jonathan Jenkins's avatar
      11c93b85
    • Jonathan Jenkins's avatar
      d83c4ff2
    • Jonathan Jenkins's avatar
      fix - pointer to local buffer escaped · 351a5de3
      Jonathan Jenkins authored
      351a5de3
    • 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
        information.
      4e19d940
  18. 24 Jul, 2014 2 commits
  19. 18 Jul, 2014 1 commit
  20. 16 Jul, 2014 1 commit
  21. 14 Jul, 2014 1 commit
  22. 08 Jul, 2014 2 commits
  23. 07 Jul, 2014 1 commit