* Initial commit of all_reduce_only support
* Working AllReduce only build
* Removing printfs and restoring release build
* Restore P2P index
* Updates to build_allreduce_only mode.
* cleaning up macro ifdefs
* Select sendrecv path based on collective data size
* Add comments on packing and unpacking group field
* Toggling RCCL_P2P_NET_DISABLE in combined calls unit tests
* Re-enabling mp unit tests
* Fixing shared memory leak and other bugs related to shared mem for MP unit tests
* Revert 43bfbfc97bf9edbae1f386d461439091618ff8ed
* Further tightening up unlinks
* Moving test check macros to separate header file
* Tightening up shared memory unlinking for clique kernels, add munmap for host barrier for MP unit tests
* Updating new MP unit test
* Fixing mqueue bug
* Fixing memory leak in MP unit tests
* Enabling clique for any XGMI-connected topology, adding tuning
* Updating CHANGELOG for clique tuning
* Re-working clique barrier system to work on multi-process / multi-gpu
* Fixing temp file creation/deletion for Clique kernel mode.
* Refactoring of MP unit tests; include bugfixes and general support for any number of GPUs
* GroupCall MP UT properly quits when too many devices specified
* MP UT will programmatically set NCCL_COMM_ID if not specified; updated install script
* gtest: add scatter to combined calls and use loops
* gtest: run validation inside loop
* gtest: revert small element count to 2520
* gtest: fix memory leak in validation
(cherry picked from commit b0853ccd51)
* Fix combined call UT
* Fix memory leak
* Fix alltoallv test
* gtest: add scatter to combined calls and use loops
* gtest: run validation inside loop
* gtest: revert small element count to 2520
* gtest: fix memory leak in validation
* Add alltoallv API and implementation
* Extend Rome P2P channel limit to multinode and alltoall kernels
* topo_expl: fix compilation and sync up with main
* gtest: use RCCL alltoallv API
* Code review changes
RCCL uses -fgpu-rdc to compile its source objects. When linking
the RCCL static library, the link and archive step must do through
hipcc and uses the flag --emit-static-lib. When compiling
UnitTests, the librccl.a must be consumed through -l and -L.
1. Fix RCCL unit test
2. Add ROME detection and tuning
3. Change default P2P level
4. Fix search algorithm for XGMI
5. Remove explicit channel duplication with implicit by using half of link speed
6. Add collective trace support
7. Correct Intel Skylake CPU detection and bandwidth
8. Fix topo connect function
9. Disable GDR read and remove unreachable code
10. Disable LL128 kernels
11. Add tuning parameters
12. Use original clock64() implementation which returns RTC counter value
13. Print out timestamp of collective trace
14. Do not use struct ncclColl in kernel launch parameter
15. Fix abort handling and add tracing
17. Add __launch_bounds__ to kernel functions
18. Remove unused abortCount
19. Unset default MIN_NRINGS and MIN_NCHANNELS
20. Do not allocate shared memory when not using LL128 kernels
21. Correct time print out in tuning log
Add support for inter-node communication using sockets and InfiniBand/RoCE.
Improve latency.
Add support for aggregation.
Improve LL/regular tuning.
Remove tests as those are now at github.com/nvidia/nccl-tests .
Wenkai Du
Stanley Tsang
Wenkai Du
Eiden Yoshida
gilbertlee-amd
Gilbert Lee
Aaron Enye Shi
saadrahim
Akila Premachandra
Sylvain Jeaugey
sclarkson