MPI ops in GPU stream for faster comms (#1356)

This commit is contained in:
Awni Hannun
2024-08-26 15:12:50 -07:00
committed by GitHub
parent 2fdf9eb535
commit 5f7d19d1f5
14 changed files with 220 additions and 26 deletions

View File

@@ -132,6 +132,7 @@ target_sources(
${CMAKE_CURRENT_SOURCE_DIR}/conv.cpp
${CMAKE_CURRENT_SOURCE_DIR}/copy.cpp
${CMAKE_CURRENT_SOURCE_DIR}/custom_kernel.cpp
${CMAKE_CURRENT_SOURCE_DIR}/distributed.cpp
${CMAKE_CURRENT_SOURCE_DIR}/device.cpp
${CMAKE_CURRENT_SOURCE_DIR}/event.cpp
${CMAKE_CURRENT_SOURCE_DIR}/fft.cpp