vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-14 02:49:41 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

temp checkpoint

Browse source
This commit is contained in:
Concedo 2024-11-30 11:59:27 +08:00
commit ec95241e38
23 changed files with 460 additions and 105 deletions
Download patch file Download diff file Expand all files Collapse all files

6
ggml/src/ggml-cpu/ggml-cpu-quants.c
View file

@ -1814,11 +1814,13 @@ void ggml_vec_dot_q4_0_q8_0(int n, float * restrict s, size_t bs, const void * r
sumv0 = vmlaq_f32(sumv0,(vcvtq_f32_s32(vmmlaq_s32((vmmlaq_s32((vmmlaq_s32((vmmlaq_s32(vdupq_n_s32(0), l0, r0)),
l1, r1)), l2, r2)), l3, r3))), scale);
}
float32x4_t sumv1 = vextq_f32(sumv0, sumv0, 2);
float32x4_t sumv1 = vextq_f32 (sumv0, sumv0, 2);
float32x4_t sumv2 = vzip1q_f32(sumv0, sumv1);
vst1_f32(s, vget_low_f32(sumv2));
vst1_f32(s, vget_low_f32 (sumv2));
vst1_f32(s + bs, vget_high_f32(sumv2));
return;
}
#endif