From 1d6715fe2053148908a9c5f59518b8e58c623e2e Mon Sep 17 00:00:00 2001 From: Sylvain Jeaugey Date: Tue, 20 Sep 2016 11:43:31 -0700 Subject: [PATCH] Fix MPI test path --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 05ca7ae49a..5ec3f38d3c 100644 --- a/README.md +++ b/README.md @@ -78,7 +78,7 @@ To install, run `make PREFIX= install` and add `/lib` t ## Usage -NCCL follows the MPI collectives API fairly closely. Before any collectives can be called, a communicator object must be initialized on each GPU. On a single-process machine, all GPUs can be conveniently initialized using `ncclCommInitAll`. For multi-process applications (e.g., with MPI), `ncclCommInitRank` must be called for each GPU. Internally `ncclCommInitRank` invokes a synchronization among all GPUs, so these calls must be invoked in different host threads (or processes) for each GPU. A brief single-process example follows, for an MPI example see src/mpi_test.cu. For details about the API see nccl.h. +NCCL follows the MPI collectives API fairly closely. Before any collectives can be called, a communicator object must be initialized on each GPU. On a single-process machine, all GPUs can be conveniently initialized using `ncclCommInitAll`. For multi-process applications (e.g., with MPI), `ncclCommInitRank` must be called for each GPU. Internally `ncclCommInitRank` invokes a synchronization among all GPUs, so these calls must be invoked in different host threads (or processes) for each GPU. A brief single-process example follows, for an MPI example see test/mpi/mpi_test.cu. For details about the API see nccl.h. ```c #include