obscura/nccl-tests
1
0
Fork 0
mirror of https://github.com/NVIDIA/nccl-tests.git synced 2026-10-04 07:57:36 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'shsnyder/dev-api-readme' into 'master'

Browse Source 
update README to account for dev API -D,-V options

See merge request nccl/nccl-tests-internal!26
This commit is contained in:
Shane Snyder
2026-01-14 06:38:18 -08:00
parent 056136d232 67d7367734
commit 9590b4f074
1 changed files with 2 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+2
View File
@@ -80,6 +80,8 @@ All tests support the same set of arguments :
* `-G,--cudagraph <num graph launches>` Capture iterations as a CUDA graph and then replay specified number of times. Default : 0.
* `-C,--report_cputime <0/1>` Report CPU time instead of latency. Default : 0.
* `-R,--local_register <0/1/2>` enable local (1) or symmetric (2) buffer registration on send/recv buffers. Default : 0.
* `-D,--device_implementation <implementation number>` use custom device API implementation. Not every collective has a custom device API implementations (currently just all\_reduce and alltoall). Default : 0 (use traditional NCCL host implementation). Note: values > 0 require symmetric memory registration (`-R 2`).
* `-V,--device_cta_count <number>` number of CTAs for device API implementation. Must be positive and less than 128. Default : 16.
* `-S,--report_timestamps <0/1>` Add timestamp (`"%Y-%m-%d %H:%M:%S"`) to each performance report line. Default : 0.
* `-J,--output_file <file>` Write [JSON] output to filepath. Infer type from suffix (only `json` supported presently).
* `-T,--timeout <time in seconds>` timeout each test after specified number of seconds. Default : disabled.