pub struct TensorMap(/* private fields */);Expand description
2026-09-25: A 128-byte TMA descriptor, aligned to 64 bytes.
Implementations§
Source§impl TensorMap
impl TensorMap
Sourcepub fn tiled_2d_bf16(
global: DevicePtr,
rows: u64,
cols: u64,
row_stride_elems: u64,
box_rows: u32,
box_cols: u32,
) -> Result<Self>
pub fn tiled_2d_bf16( global: DevicePtr, rows: u64, cols: u64, row_stride_elems: u64, box_rows: u32, box_cols: u32, ) -> Result<Self>
2026-09-25: Describe a row-major bf16 matrix as tiles of box_rows x box_cols.
rows, cols and row_stride_elems are in elements. row_stride_elems
is the distance between consecutive rows and may exceed cols (a view into
a wider tensor). Out-of-bounds elements of a tile are filled with zero
(CU_TENSOR_MAP_FLOAT_OOB_FILL_NONE).
Errors: a misaligned address or stride, a stride below cols, a zero box
dimension, a libcuda without cuTensorMapEncodeTiled, or a non-zero
CUresult from the encode.
Trait Implementations§
Auto Trait Implementations§
impl Freeze for TensorMap
impl RefUnwindSafe for TensorMap
impl Send for TensorMap
impl Sync for TensorMap
impl Unpin for TensorMap
impl UnwindSafe for TensorMap
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