NVIDIA/cutlass
1
0
Fork 0
mirror of https://github.com/NVIDIA/cutlass.git synced 2026-04-20 06:48:59 +00:00
Code Issues Packages Projects Releases Wiki Activity

remove internal comment (#2841)

Browse Source 
Co-authored-by: bangyus <bangyus@nvidia.com>
This commit is contained in:
bangyu shen
2025-12-04 23:36:21 +08:00
committed by GitHub
parent bc680c7f67
commit 7252a2d17e
2 changed files with 0 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
examples/python/CuTeDSL/distributed/distributed_all_gather_gemm_blackwell.py
View File

@@ -1151,12 +1151,6 @@ class PersistentDenseGemmKernel:
tile_sched_params, cute.arch.block_idx(), cute.arch.grid_dim()
)
if cutlass.const_expr(self.use_tma_store):
# {$nv-internal-release begin}
# (EPI_TILE_M, EPI_TILE_N, STAGE)
# sC = storage.sC.get_tensor(
# c_smem_layout_staged.outer, swizzle=c_smem_layout_staged.inner
# )
# {$nv-internal-release end}
assert tma_atom_c is not None and sC is not None
self.epilogue_tma_store(

6
examples/python/CuTeDSL/distributed/distributed_gemm_all_reduce_blackwell.py
View File

@@ -1056,12 +1056,6 @@ class PersistentDenseGemmKernel:
)
if cutlass.const_expr(self.use_tma_store):
# {$nv-internal-release begin}
# (EPI_TILE_M, EPI_TILE_N, STAGE)
# sC = storage.sC.get_tensor(
# c_smem_layout_staged.outer, swizzle=c_smem_layout_staged.inner
# )
# {$nv-internal-release end}
assert tma_atom_c is not None and sC is not None
self.epilogue_tma_store_release_flag(