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

Update ggml-quants.c

Browse source
This commit is contained in:
Michael Potter 2023-11-12 15:07:24 -08:00 • committed by GitHub
parent 532dd74e38
commit bf1c8d761a
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 2 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

3
ggml-quants.c
View file

@ -1273,7 +1273,8 @@ static float make_qkx2_quants(int n, int nmax, const float * restrict x, const f
float max = x[0];
float sum_w = weights[0];
float sum_x = sum_w * x[0];
for (int i = 1; i < n; ++i) {
// Mark i as volatile to prevent the -O3 optimizer from unrolling this loop and breaking MacOS Sonoma quantization
for (volatile int i = 1; i < n; ++i) {
if (x[i] < min) min = x[i];
if (x[i] > max) max = x[i];
float w = weights[i];