llama.cpp
267e9986 - vulkan: Use larger loads in scalar/coopmat1 matmul (#15729)

Commit
32 days ago
vulkan: Use larger loads in scalar/coopmat1 matmul (#15729) I think glslang will translate an access like x[i][1].z to OpAccessChain ... x, i, 1, 2 OpLoad float16_t ... rather than loading all of x[i] in a single OpLoad. Change the code to explicitly load the vector/matrix.
Author
Parents
Loading