pub struct PrefillSlice<'a> {
pub prompt_tokens: &'a [u32],
pub seq: &'a mut SequenceState,
pub chunk_start: usize,
pub chunk_len: usize,
pub is_last_chunk: bool,
}Expand description
2026-09-25: One prefilling stream’s chunk. prefill_batch_chunk and mixed_forward_batch
take a &mut [PrefillSlice<'_>] and run every stream’s chunk in one forward pass.
Fields§
§prompt_tokens: &'a [u32]2026-09-25: The stream’s full prompt.
seq: &'a mut SequenceState§chunk_start: usize2026-09-25: Offset into prompt_tokens where this chunk starts.
chunk_len: usize§is_last_chunk: bool2026-09-25: Whether this is the stream’s final chunk, which produces its last-token logits.
Auto Trait Implementations§
impl<'a> Freeze for PrefillSlice<'a>
impl<'a> !RefUnwindSafe for PrefillSlice<'a>
impl<'a> Send for PrefillSlice<'a>
impl<'a> Sync for PrefillSlice<'a>
impl<'a> Unpin for PrefillSlice<'a>
impl<'a> !UnwindSafe for PrefillSlice<'a>
Blanket Implementations§
Source§impl<T> BorrowMut<T> for Twhere
T: ?Sized,
impl<T> BorrowMut<T> for Twhere
T: ?Sized,
Source§fn borrow_mut(&mut self) -> &mut T
fn borrow_mut(&mut self) -> &mut T
Mutably borrows from an owned value. Read more