corey-derochie-amd e056fe8f7e Disable MSCCL for the non-multi-process case by default (#1307) ...

* Added `RCCL_MSCCL_ENABLE_SINGLE_PROCESS` runtime flag to return to the original MSCCL enablement behaviour except when explicitly enabling for multi-thread.

* Added documentation for the new `RCCL_MSCCL_ENABLE_SINGLE_PROCESS` runtime env var.

2024-09-04 11:11:50 -06:00

..

clique

Add support for using contiguous for GPU direct RDMA (#1096)

2024-02-29 10:06:43 -08:00

device

removed hcc mentions (#1291)

2024-08-14 15:04:13 -06:00

graph

graph: fix for MI300X 64 GPU case (#1308)

2024-08-26 18:37:58 -05:00

include

adding all nccl apis to api_support to enable rccl tracing by rocprofv3 (#1297)

2024-08-22 12:36:07 -05:00

misc

Disable MSCCL for the non-multi-process case by default (#1307)

2024-09-04 11:11:50 -06:00

transport

removed hcc mentions (#1291)

2024-08-14 15:04:13 -06:00

bootstrap.cc

Merge remote-tracking branch 'nccl/master' into develop

2024-04-23 13:34:00 -07:00

channel.cc

Fix cudaMemcpyAsync bug

2023-09-20 05:51:14 -07:00

collectives.cc

adding all nccl apis to api_support to enable rccl tracing by rocprofv3 (#1297)

2024-08-22 12:36:07 -05:00

debug.cc

doubling debug buffer size with increased channels

2024-06-03 13:05:05 -05:00

enhcompat.cc

2.12.7-1

2022-03-02 20:48:56 +01:00

enqueue.cc

adding all nccl apis to api_support to enable rccl tracing by rocprofv3 (#1297)

2024-08-22 12:36:07 -05:00

group.cc

adding all nccl apis to api_support to enable rccl tracing by rocprofv3 (#1297)

2024-08-22 12:36:07 -05:00

init_nvtx.cc

Fix undefined symbol when nvtx is not enabled (#1082)

2024-02-13 14:03:43 -08:00

init.cc

adding all nccl apis to api_support to enable rccl tracing by rocprofv3 (#1297)

2024-08-22 12:36:07 -05:00

Makefile

2.20.3-1

2024-02-13 04:22:38 -08:00

msccl.cc

adding all nccl apis to api_support to enable rccl tracing by rocprofv3 (#1297)

2024-08-22 12:36:07 -05:00

nccl.h.in

Enable multi-threading for MSCCL (#1203)

2024-07-04 09:34:38 -06:00

nccl.pc.in

Add pkgconfig file (#190)

2019-04-08 09:16:54 -07:00

net.cc

removed hcc mentions (#1291)

2024-08-14 15:04:13 -06:00

proxy.cc

add 256 channels support

2024-06-03 13:03:18 -05:00

register.cc

adding all nccl apis to api_support to enable rccl tracing by rocprofv3 (#1297)

2024-08-22 12:36:07 -05:00

transport.cc

set MAXCHANNELS to 128

2024-06-03 13:05:05 -05:00