CUDA: Replace init_offsets kernel with iterators in cub-based argsort (#18930)

* CUDA: Replace `init_offsets` with iterators in argsort

This is a QOL improvement, saving us the cost of materializing the
iterator

* Remove unnecessary include from top-k.cu
This commit is contained in:
Oliver Simons
2026-01-20 20:11:01 +08:00
committed by GitHub
parent 08f3f4a8a3
commit d1e3556481
2 changed files with 7 additions and 16 deletions
-1
View File
@@ -4,7 +4,6 @@
#ifdef GGML_CUDA_USE_CUB
# include <cub/cub.cuh>
# if (CCCL_MAJOR_VERSION >= 3 && CCCL_MINOR_VERSION >= 2)
# include <cuda/iterator>
# define CUB_TOP_K_AVAILABLE
using namespace cub;
# endif // CCCL_MAJOR_VERSION >= 3 && CCCL_MINOR_VERSION >= 2