56fd68e7dd
CCCL (CUDA C++ Core Libraries) provides: - CUB: device/block/warp-level GPU primitives (reduce, scan, sort, topk) - Thrust: high-level parallel algorithms (transform_reduce, sort, scan) - libcudacxx: CUDA C++ standard library (atomics, barriers, memory) - cudax: experimental features (memory resources, allocators) - Tuning policies: per-SM hardware-specific algorithm parameters Competition optimization vectors mapped to CCCL: - Output TPS (83% weight): warp_reduce, block_reduce, device_topk - Input TPS (14% weight): device_scan, block_load, prefetch - Cache TPS (3% weight): prefix caching strategy patterns - Memory (0.9 util): pooled/cached/buddy allocators Source: https://github.com/NVIDIA/cccl (shallow clone, HEAD only) License: Apache-2.0
5 lines
220 B
Plaintext
5 lines
220 B
Plaintext
CHECK: Computing C = A + B for sparse vectors A and B
|
|
CHECK-NEXT: A (2,10) (3,60) (5,20) (8,40)
|
|
CHECK-NEXT: B (1,50) (2,30) (4,80) (5,30) (7,90) (8,10)
|
|
CHECK-NEXT: C (1,50) (2,40) (3,60) (4,80) (5,50) (7,90) (8,50)
|