happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

opencl: refine condition for q6_K mm

This commit is contained in:
Li He 2026-01-22 22:30:52 -08:00
parent 3228645346
commit b2aacdafd8
1 changed files with 7 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
ggml/src/ggml-opencl/ggml-opencl.cpp
View File

@ -8946,9 +8946,13 @@ static void ggml_cl_mul_mat(ggml_backend_t backend, const ggml_tensor * src0, co
return;
}
case GGML_TYPE_Q6_K: {
//if (ne11 < 32) {
// break;
//}
if (ne11 < 32) {
break;
}
if (!ggml_is_contiguous(src0) || !ggml_is_contiguous(src1)) {
break;
}
kernel = backend_ctx->kernel_mul_mm_q6_k_f32_l4_lm;
nth0 = 128; // calculated as (BM*BN)/(TM*TN)