ggml

80772659 - CUDA: Improve flash decoding kernel GPU occupancy for BS=1 case (llama/12183)

CUDA: Improve flash decoding kernel GPU occupancy for BS=1 case (llama/12183) - Find out active blocks per SM using cudaOccupancyMaxActiveBlocksPerMultiprocessor API. Use this value to determine the optimal parallel_blocks value. - Prefer vector flash attention kernels over MMA kernel for BS=1 Fixes Issue: #12182 --------- Co-authored-by: Johannes Gäßler <johannesg@5d6.de>