* Support random data generation in utilities. Update first 8 byte element in each 4k block on updates to defy dedup.
* Incorporate different packet types into mdtest/md-workbench.
* Integrated utilities memory pattern tools into IOR. Now all tools use the same patterns.
* Added IOR long option for compatibility between IOR and other tools.
* Added new tests for random buffers.
* Basic support for memory allocation on GPU using CUDA unified memory. Partially addressing #284. IOR support completed.
* Support for GPU alloc in MDTest and MD-Workbench
* Option: support repeated parsing of same option (allows option sharing across modules).
* Checks for gpuDirect
* Integrate gpuDirect options and basic hooks, more testing to be done.
* POSIX: basic gpuDirect implementation working with fake-gpudirect library.
* CUDA allow setting of DeviceID for IOR (not yet MDTest).
* CUDA/GPUDirect Support --with-X=<path>
* Bugfix in option parser for flags that are part of an argument for an option, e.g., -O=1, if 1 is a flag it is wrongly assumed to be a flag.
On systems where numTasks is not evenly divisible by 'tasksPerNode' we were
seeing some nodes reading multiple files while others read none after
reordering.
Commonly all nodes have the same number of tasks but there is nothing
requiring that to be the case. Imagine having 64 tasks running against 4
nodes which can run 20 tasks each. Here you get three groups of 20 and one
group of 4. On this sytem nodes running in the group of 4 were previously
getting tasksPerNode of 4 which meant they reordered tasks differently than
the nodes which got tasksPerNode of 20.
The key to fixing this is ensuring that every node reorders tasks the same
way, which means ensuring they all use the same input values. Obviously on
systems where the number of tasks per node is inconsistent the reordering will
also be inconsistent (some tasks may end up on the same node, or not as far
separated as desired, etc.) but at least this way you'll always end up with a
1:1 reordering.
- Renamed nodes/nodeCount to numNodes
- Renamed tasksPerNode to numTasksOnNode0
- Ensured that numTasksOnNode0 will always have the same value regardless of
which node you're on
- Removed inconsistently used globals numTasksWorld and tasksPerNode and
replaced with per-test params equivalents
- Added utility functions for setting these values:
- numNodes -> GetNumNodes
- numTasks -> GetNumTasks
- numTasksOnNode0 -> GetNumNodesOnTask0
- Improved MPI_VERSION < 3 logic for GetNumNodes so it works when numTasks is
not evenly divisible by numTasksOnNode0
- Left 'nodes' and 'tasksPerNode' in output alone to not break compatibility
- Allowed command-line params to override numTasks, numNodes, and
numTasksOnNode0 but default to using the MPI-calculated values
Clean up the header files to only contain those things that
need to be shared between .c files.
Functions that are not shared are now declared static to
make their file scope explicit. Functions that ARE shared
are declared in appropriate headers.
I am not going to claim that I caugh everything, but at
least it is a good start.