JuliaParallel / MPI.jl / 809 / 2
60%
master: 60%

DEFAULT BRANCH: master
Ran
Files 5
Run time 0s
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst
coverage: 85.063% (+1.3%) from 83.813%
MPI_IMPL=mpich

push

travis-ci

barche 
Implement MPI_IN_PLACE for MPI Operations (#232)

* Extract the value of MPI_IN_PLACE during build phase
* Add non allocating version of MPI_Allreduce, support for MPI_IN_PLACE, and tests.
* Add non allocating version of MPI_Scatter, another method supporting MPI_IN_PLACE, and tests.
* Add non allocating version of MPI_Allgather, support for MPI_IN_PLACE, and tests.
* Add non allocating version of MPI_Allgatherv, support for MPI_IN_PLACE, and tests.
* Add non allocating version of MPI_Alltoall, support for MPI_IN_PLACE, and tests.
* Add non allocating version of Alltoallv and tests
* Modify definition of IN_PLACE and revert function signatures to the old type.
* Make the compiler check if a function can accept ConstantPtr.
* Allreduce:
 - Rename `allreduce(sendbuf, op, comm)` to `Allreduce(sendbuf, op, comm)` for consistency + fix test
 - Move `Allreduce!(send, recv, count op::Function, comm)` converting user provided-functions from `mpi-op.jl` to `mpi-base.jl`
* Reduce:
	- Add nonallocating version `Reduce!`
	- Move and rename `Reduce(send, recv, count op::Function, comm)` converting user provided-functions from `mpi-op.jl` to `mpi-base.jl`
* Add Reduce_in_place! function
* Modify allreduce tests to test N-dimensional tensors
* Gather: Add a non-allocating version
* Add Gather_in_place!
* Add non-allocating Gatherv and support for MPI.IN_PLACE + tests
* Add Gatherv_in_place!
* Add non-allocating Scatterv! and Scatterv_in_place

672 of 790 relevant lines covered (85.06%)

44704.12 hits per line

Source Files on job 809.2 (MPI_IMPL=mpich)