Skip to main content
Log in

Mojo function

tcgen05_cp

tcgen05_cp[*, cta_group: SIMD[int32, 1], datapaths: Int, bits: Int, src_fmt: String = __init__[__mlir_type.!kgen.string](""), dst_fmt: String = __init__[__mlir_type.!kgen.string](""), multicast: String = __init__[__mlir_type.!kgen.string]("")](tmem_addr: SIMD[uint32, 1], s_desc: MMASmemDescriptor)

Copies data from shared memory described by the matrix descriptor s_desc to tensor memory tmem_addr.

Note: This function is only available on NVIDIA Blackwell GPUs (SM 100+).

Parameters:

  • cta_group (SIMD[int32, 1]): The cooperative thread array (CTA) group ID.
  • datapaths (Int): The first dimension of the shape.
  • bits (Int): The second dimension of the shape.
  • src_fmt (String): Source format string.
  • dst_fmt (String): Destination format string.
  • multicast (String): Multicast string.

Args:

  • tmem_addr (SIMD[uint32, 1]): Address of the tensor memory.
  • s_desc (MMASmemDescriptor): Matrix descriptor for the copy operation.

Was this page helpful?