mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2026-05-06 17:14:07 +00:00
18ddaea2ae
* vulkan: Optimize GGML_OP_CUMSUM There are two paths: The preexisting one that does a whole row per workgroup in a single shader, and one that splits each row into multiple blocks and does two passes. The first pass computes partials within a block, the second adds the block partials to compute the final result. The multipass shader is used when there are a small number of large rows. In the whole-row shader, handle multiple elements per invocation. * use 2 ELEM_PER_THREAD for AMD/Intel * address feedback