llama.cpp

0e76501e - Vulkan: Optimize Matmul parameters for AMD GPUs with Coopmat support (#18749)

Vulkan: Optimize Matmul parameters for AMD GPUs with Coopmat support (#18749) * vulkan: Enable and optimize large matmul parameter combination for AMD * limit tuning to AMD GPUs with coopmat support * use tx_m values instead of _l