CUTLASS 2.4 (Implicit GEMM convolution) (#147)
CUTLASS 2.4 (Implicit GEMM Convolution) Co-authored-by: Manish Gupta <manigupta@nvidia.com>, Haicheng Wu <haichengw@nvidia.com>, Dustyn Blasig <dblasig@nvidia.com>, Andrew Kerr <akerr@nvidia.com>
This commit is contained in:
co-authored by
Manish Gupta <manigupta@nvidia.com>, Haicheng Wu <haichengw@nvidia.com>, Dustyn Blasig <dblasig@nvidia.com>, Andrew Kerr <akerr@nvidia.com>
parent
c2b80ad4e4
commit
6615010cd0
@@ -582,7 +582,7 @@ public:
|
||||
semaphore.fetch();
|
||||
|
||||
// Indicate which position in a serial reduction the output operator is currently updating
|
||||
output_op.set_k_partition(threadblock_tile_offset.k());
|
||||
output_op.set_k_partition(threadblock_tile_offset.k(), params.grid_tiled_shape.k());
|
||||
}
|
||||
}
|
||||
else if (params.mode == GemmUniversalMode::kGemmSplitKParallel) {
|
||||
|
||||
Reference in New Issue
Block a user