huggingface/candle
1
0
Fork 0
mirror of https://github.com/huggingface/candle.git synced 2025-06-21 12:20:46 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #88 from LaurentMazare/fix_unsafe_loads

Browse Source 
Fixing unsafe slow load (memcpy).
This commit is contained in:
Nicolas Patry
2023-07-07 00:09:56 +02:00
committed by GitHub
parent 0f679fe42e ce27073feb
commit 666c6f07a1
1 changed files with 3 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
candle-core/src/safetensors.rs
View File

@ -71,7 +71,9 @@ fn convert_<T: WithDType>(view: st::TensorView<'_>, device: &Device) -> Result<T
let data: &[T] = unsafe { std::slice::from_raw_parts(v.as_ptr() as *const T, elem_count) }; let data: &[T] = unsafe { std::slice::from_raw_parts(v.as_ptr() as *const T, elem_count) };
Tensor::from_slice(data, view.shape(), device) Tensor::from_slice(data, view.shape(), device)
} else { } else {
let mut c = Vec::with_capacity(elem_count); // XXX: We need to specify `T` here, otherwise the compiler will infer u8 because of the following cast
// Making this vector too small to fit a full f16/f32/f64 weights, resulting in out-of-bounds access
let mut c: Vec<T> = Vec::with_capacity(elem_count);
// SAFETY: We just created c, so the allocated memory is necessarily // SAFETY: We just created c, so the allocated memory is necessarily
// contiguous and non overlapping with the view's data. // contiguous and non overlapping with the view's data.
// We're downgrading the `c` pointer from T to u8, which removes alignment // We're downgrading the `c` pointer from T to u8, which removes alignment