Two minor issues are solved:
+ Enclose the kernel function with parenthesis as hip-clang defines
`hipLaunchKernelGGL` as macro.
+ Need to explicitly include <hsa.h> for hip-clang.
[ROCm/rccl commit: 41310144f6]
Current SHM object name would only use pidHash and ranks as
identification, which would collide each other when program runs with
multiple communicators. Here we added commId info into pidHash, it makes
'pidHash'es of different communicators keeping in same process will be
distincted with each other.
Ported from original commit: https://github.com/lowintelligence/nccl/commits/shm
[ROCm/rccl commit: 949d680e49]
* Implement HDP flush when transfer data over PCIe P2P
* Add some descriptions to HDP flushing
* Fix for review comments
[ROCm/rccl commit: b7a6307371]
Improve throughput for about 20%. Also remove P2P over PCIe which was
left enabled at initial release.
Signed-off-by: Wenkai Du <wenkai.du@amd.com>
[ROCm/rccl commit: f45566a8bd]
When initNet fails, we should not print the backtrace as it is
supposed to be normal operation (falling back to sockets)
[ROCm/rccl commit: 57368189e1]
Whether the network supported GPU Direct RDMA or not was ignored,
causing sockets to break when cards were local enough that NCCL
tried to use it.
[ROCm/rccl commit: 4b39a4cf91]
This is to handle EINTR/EGAIN properly (issue #137), and also
make the code consistent with the rest.
Unfortunately posix_fallocate and mmap do not follow the classic
return code/errno pattern, so we need to write wrappers around those
functions.
[ROCm/rccl commit: eed8218e17]
SYSCHECKVAL was not retrying when a retry was needed. Since not all
calls are inside a loop, that means we could silently miss an
EINTR/EAGAIN return code.
Also rework the socket connection code and improve error reporting.
[ROCm/rccl commit: 302d538b73]
Dynamically load external network from libnccl-net.so.
Add init function in networks.
Move PCI scoring to net.cu, only ask transport to provide a path.
Simplify CUDA PCI path detection.
Add dummy external network
[ROCm/rccl commit: 0d3a20f96d]
Improved LL tuning for multi-node jobs.
Improved bootstrap for large job scaling.
Fixed a hang during bootstrap due to socket reuse.
Added operation name to the COLL INFO logging.
[ROCm/rccl commit: b56650c7f5]
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 .
[ROCm/rccl commit: f93fe9bfd9]
Stanley Tsang
Wenkai Du
Wenkai Du
Michael LIAO
gilbertlee-amd
saadrahim
Jeff Daily
Yaxun Sam Liu
Gilbert Lee
Christian Sigg
Sylvain Jeaugey
Christian Sigg
David Addison
Alex Sergeev
Ilya Biryukov
Nathan Luehr
Chad Whipkey