Nilesh M Negi 210f90ae0f [SRC] Enable unroll=1 for gfx950 (#1602) ...

* [SRC] Enable unroll=1 for gfx950

* Fix typo from rebase in generate.py

* Support for unroll=1 and gfx90a when building for all GPU targets

---------

Signed-off-by: nileshnegi <Nilesh.Negi@amd.com>

[ROCm/rccl commit: 307bc10781]

2025-03-27 18:21:35 -05:00

..

network/unpack

Merge remote-tracking branch 'nccl/master' into develop

2025-01-23 12:08:46 -06:00

all_gather.h

removed gfx940 and gfx941 (#1606)

2025-03-20 09:34:53 -06:00

all_reduce.h

Removing the experimental clique kernel files (#1610)

2025-03-20 18:10:01 -06:00

alltoall_pivot.h

removed gfx940 and gfx941 (#1606)

2025-03-20 09:34:53 -06:00

broadcast.h

removed gfx940 and gfx941 (#1606)

2025-03-20 09:34:53 -06:00

common_kernel.h

[SRC] Enable unroll=1 for gfx950 (#1602)

2025-03-27 18:21:35 -05:00

common.cu

[SRC] Enable unroll=1 for gfx950 (#1602)

2025-03-27 18:21:35 -05:00

common.h

[SRC] Enable unroll=1 for gfx950 (#1602)

2025-03-27 18:21:35 -05:00

generate.py

[SRC] Enable unroll=1 for gfx950 (#1602)

2025-03-27 18:21:35 -05:00

msccl_kernel_impl.h

removed gfx940 and gfx941 (#1606)

2025-03-20 09:34:53 -06:00

onerank.cu

[SRC] Enable unroll=1 for gfx950 (#1602)

2025-03-27 18:21:35 -05:00

op128.h

removed gfx940 and gfx941 (#1606)

2025-03-20 09:34:53 -06:00

primitives.h

removed gfx940 and gfx941 (#1606)

2025-03-20 09:34:53 -06:00

prims_ll128.h

Enable LL128 on gfx942 (#1549)

2025-03-16 15:10:05 -07:00

prims_ll.h

Reset barrier and make barrier_next thread local (#1531)

2025-02-05 09:06:48 -08:00

prims_simple.h

Reset barrier and make barrier_next thread local (#1531)

2025-02-05 09:06:48 -08:00

reduce_kernel.h

Merge remote-tracking branch 'nccl/master' into develop

2025-01-23 12:08:46 -06:00

reduce_scatter.h

removed gfx940 and gfx941 (#1606)

2025-03-20 09:34:53 -06:00

reduce.h

removed gfx940 and gfx941 (#1606)

2025-03-20 09:34:53 -06:00

sendrecv.h

removed gfx940 and gfx941 (#1606)

2025-03-20 09:34:53 -06:00