nicolas
/
ik_llama_opt
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Fix compilation error with IQK_FA_ALL_QUANTS enabled (#226) 

Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
This commit is contained in:
Kawrakow 2025-02-23 08:02:16 +02:00 committed by GitHub
parent 299700a4ec
commit 2212c1c636
1 changed files with 2 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
ggml/src/iqk/iqk_mul_mat.cpp
View File

@ -15315,9 +15315,11 @@ struct HelperIQ4nl final : public BaseHelper<step> {
#ifdef __aarch64__
using block_q8 = block_q8_0;
HelperIQ4nl(const char * data, int stride) : Base(data, stride), values(vld1q_s8(iq4k_values)) {}
constexpr static int block_size_q = QK8_0;
#else
HelperIQ4nl(const char * data, int stride) : Base(data, stride) {}
using block_q8 = block_q8_1;
constexpr static int block_size_q = QK8_1;
#endif
// Needed for v * softmax(k * q)