kanshan/nccl-tests
1
0
Fork 0
mirror of https://github.com/NVIDIA/nccl-tests.git synced 2026-01-13 18:37:16 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity
125 Commits 3 Branches 45 Tags 1.5 MiB
master
Commit Graph

3 Commits

Author SHA1 Message Date
David Addison
1021260ca9 Make verifiable a DSO and add NAME_SUFFIX support 
Build option DSO=1 generates libverifiable.so which can be
used to reduce the combined binary size.

Build option NAME_SUFFIX can be used to a add suffix to all
generated binaries. e.g. NAME_SUFFIX=_mpi

Added new make target: clean_intermediates
 2025-04-23 17:07:24 -07:00
David Addison
501a149d57 Add support for FP8 datatypes 
Added new datatypes: f8e4m3, f8e5m2

Only supported on H100+ architectures and NCCL versions >= 2.24.0
 2025-04-18 19:20:59 -07:00
John Bachan
51af5572bf Resync with NCCL 2.13 
* Added "verifiable", a suite of kernels for generating and verifying reduction
  input and output arrays in a bit-precise way.
* Data corruption errors now reported in number of wrong elements instead of max
  deviation.
* Use ncclGetLastError.
* Don't run hypercube on non-powers of 2 ranks.
* Fix to hypercube data verification.
* Use "thread local" as the defaut CUDA capture mode.
* Replaced pthread_yield -> sched_yield()
* Bugfix to the cpu-side barrier/allreduce implementations.
 2022-08-22 17:51:06 -07:00