Add bisection test

[x41lakazam]

In a bisection, each rank is paired with one other rank called 'peer', they both send and receive N messages, this number is bound to the agg_iters parameter.

The selection of this 'peer' rank is defined by the getPeer function:
https://github.com/x41lakazam/nccl-tests/blob/bisection_test/src/bisection.cu#L19

[sjeaugey]

Can't you already do that running the sendrecv_perf test, and setting NCCL_TESTS_SPLIT_MASK=(nranks/2)-1?

Sure, that only works with nranks being a power of two; maybe your code is more generic.

[x41lakazam]

Being able to run this test when nranks is not a power of two is actually important to us

[x41lakazam]

@AddyLaddy @sjeaugey

The code is ready to merge from our side
Please let me know what you think about it, thanks

[sjeaugey]

@x41lakazam Given the new NCCL_TESTS_SPLIT option allowing to split ranks in any way and no longer tied to powers of two, could you describe a use case where this allows to test performance in a new/better way?

[AddyLaddy]

Oh, I wanted the cross rail change removed. Why did you restore it?

[x41lakazam]

@sjeaugey I didn't notice this change, does it permit bisection on any number of ranks ?

[sjeaugey]

Yes, if you set NCCL_TESTS_SPLIT="MOD X" with X=N/2 then it should be a bisection test.