For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).
Mojo function
naive_gemv
def naive_gemv(c_buf: TileTensor[Engine=c_buf.Engine, address_space=c_buf.address_space, linear_idx_type=c_buf.linear_idx_type], a_buf: TileTensor[Engine=a_buf.Engine, address_space=a_buf.address_space, linear_idx_type=a_buf.linear_idx_type], b_buf: TileTensor[Engine=b_buf.Engine, address_space=b_buf.address_space, linear_idx_type=b_buf.linear_idx_type])
Computes a reference matrix-vector product C = A * b.
Computes each C[m] as a SIMD dot product of row m of A with b. Intended for correctness testing rather than performance-critical paths.
Args:
- โc_buf (
TileTensor[Engine=c_buf.Engine, address_space=c_buf.address_space, linear_idx_type=c_buf.linear_idx_type]): Output TileTensor of length M. - โa_buf (
TileTensor[Engine=a_buf.Engine, address_space=a_buf.address_space, linear_idx_type=a_buf.linear_idx_type]): Input matrix TileTensor of shape (M, K). - โb_buf (
TileTensor[Engine=b_buf.Engine, address_space=b_buf.address_space, linear_idx_type=b_buf.linear_idx_type]): Input vector TileTensor of length K.