huggingface/candle
1
0
Fork 0
mirror of https://github.com/huggingface/candle.git synced 2025-06-18 19:47:12 +00:00
Code Issues Packages Projects Releases Wiki Activity

simd128 optimized q8_0 vecdot (#972)

Browse Source 
* wasm/simd128 version of the quantized q8_0 vecdot.

* Add the missing conversion.
This commit is contained in:
Laurent Mazare
2023-09-27 11:03:20 +01:00
committed by GitHub
parent 29bd6b2979
commit e59784e353
3 changed files with 54 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
candle-core/src/quantized/k_quants.rs
View File

@ -606,6 +606,9 @@ impl GgmlType for BlockQ8_0 {
#[cfg(target_feature = "neon")]
return super::neon::vec_dot_q8_0_q8_0(n, xs, ys);
#[cfg(target_feature = "simd128")]
return super::simd128::vec_dot_q8_0_q8_0(n, xs, ys);
let qk = QK8_0;
if n % QK8_0 != 0 {
crate::bail!("vec_dot_q8_0_q8_0: {n} is not divisible by {qk}")