llama : stage mmap uploads on integrated GPUs - #26023

llama : stage mmap uploads on integrated GPUs#26023

liminfei-amd wants to merge 1 commit into

ggml-org:masterggml-org/llama.cpp:masterfrom

liminfei-amd:amd-rocm/15018-igpu-mmap-stagingliminfei-amd/llama.cpp:amd-rocm/15018-igpu-mmap-stagingCopy head branch name to clipboard