* addressing hip_fp8 support compatibility issue
* skipping mulsum and avg test for fp8, using hip_fp8 for product
* syncing with nccl-tests
removing the fp8 filter for pre-hopper gpus and resolving the merge conflict
---------
Co-authored-by: Marzieh Berenjkoub <mberenjk@amd.com>
[ROCm/rccl-tests commit: 4b2b635766]
* skipping the prod test for FP8 types in reduce and reduce-scatter
---------
Co-authored-by: Marzieh Berenjkoub <mberenjk@amd.com>
[ROCm/rccl-tests commit: 5e838ad9df]
* 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.
[ROCm/rccl-tests commit: 51af5572bf]
Fix#45 : All_gather and reduce_scatter algorithm bandwidth was
computed as time/count*(nranks-1) which is not consistent with the
way we compute it for other collectives.
This change makes algbw higher; busbw is unchanged.
[ROCm/rccl-tests commit: ec1b5e22e6]
Major rework to merge most of the changes from the NCCL internal
tests into the public ones
Added "-m <agg_iters>" operation aggregation option.
Data integrity checking is now much more performant at scale.
Startup times at scale are improved.
Test latency units are now displayed in usec.
[ROCm/rccl-tests commit: cbe7f65400]
Wenkai Du
Mustafa Abduljabbar
mberenjk
Nilesh M Negi
John Bachan
Bertan Dogancay
Wenkai Du
Edgar Gabriel
Sylvain Jeaugey
David Addison
Stanley Tsang