tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-07-26 03:03:25 -04:00
Code Issues Packages Projects Releases Wiki Activity

metal : add kernel_get_rows_i32

Browse Source 
ggml-ci
This commit is contained in:
Georgi Gerganov
2024-01-03 11:35:46 +02:00
parent ab62fc3e55
commit 289313716f
2 changed files with 33 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
ggml-metal.m
View File

@ -87,6 +87,7 @@ struct ggml_metal_context {
GGML_METAL_DECL_KERNEL(get_rows_q4_K);
GGML_METAL_DECL_KERNEL(get_rows_q5_K);
GGML_METAL_DECL_KERNEL(get_rows_q6_K);
GGML_METAL_DECL_KERNEL(get_rows_i32);
GGML_METAL_DECL_KERNEL(rms_norm);
GGML_METAL_DECL_KERNEL(group_norm);
GGML_METAL_DECL_KERNEL(norm);
@ -377,6 +378,7 @@ struct ggml_metal_context * ggml_metal_init(int n_cb) {
GGML_METAL_ADD_KERNEL(get_rows_q4_K);
GGML_METAL_ADD_KERNEL(get_rows_q5_K);
GGML_METAL_ADD_KERNEL(get_rows_q6_K);
GGML_METAL_ADD_KERNEL(get_rows_i32);
GGML_METAL_ADD_KERNEL(rms_norm);
GGML_METAL_ADD_KERNEL(group_norm);
GGML_METAL_ADD_KERNEL(norm);
@ -499,6 +501,7 @@ void ggml_metal_free(struct ggml_metal_context * ctx) {
GGML_METAL_DEL_KERNEL(get_rows_q4_K);
GGML_METAL_DEL_KERNEL(get_rows_q5_K);
GGML_METAL_DEL_KERNEL(get_rows_q6_K);
GGML_METAL_DEL_KERNEL(get_rows_i32);
GGML_METAL_DEL_KERNEL(rms_norm);
GGML_METAL_DEL_KERNEL(group_norm);
GGML_METAL_DEL_KERNEL(norm);
@ -1978,6 +1981,7 @@ void ggml_metal_graph_compute(
case GGML_TYPE_Q4_K: [encoder setComputePipelineState:ctx->pipeline_get_rows_q4_K]; break;
case GGML_TYPE_Q5_K: [encoder setComputePipelineState:ctx->pipeline_get_rows_q5_K]; break;
case GGML_TYPE_Q6_K: [encoder setComputePipelineState:ctx->pipeline_get_rows_q6_K]; break;
case GGML_TYPE_I32: [encoder setComputePipelineState:ctx->pipeline_get_rows_i32]; break;
default: GGML_ASSERT(false && "not implemented");
}