vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

metal : unify mul_mv_id kernels (#6556)

Browse source
This commit is contained in:
slaren 2024-04-12 18:13:20 +02:00 • committed by GitHub
parent 4cc120c744
commit fbbc030ba9
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
4 changed files with 208 additions and 1122 deletions
Download patch file Download diff file Expand all files Collapse all files

5
ggml-metal.m
View file

@ -1926,7 +1926,12 @@ static enum ggml_status ggml_metal_graph_compute(
{
nth0 = 4;
nth1 = 16;
#if QK_K == 64
pipeline = ctx->kernels[GGML_METAL_KERNEL_TYPE_MUL_MV_ID_IQ4_NL_F32].pipeline;
#else
pipeline = ctx->kernels[GGML_METAL_KERNEL_TYPE_MUL_MV_ID_IQ4_XS_F32].pipeline;
#endif
} break;
default:
{