llama.cpp
95753896 - metal: add the missing f32 x bf16 mul_mv variants (#28741)

Commit
55 days ago
metal: add the missing f32 x bf16 mul_mv variants (#28741) ggml_conv_1d_dw builds its im2col as f32 when the kernel is bf16, then multiplies the two, so a depthwise convolution over bf16 weights asks for kernel_mul_mv_f32_bf16, which was never instantiated. The base, the _4 and the _short families are filled in next to their bf16 neighbours, inside the same runtime guard, so a device without bf16 support is unaffected.
Parents
Loading