IMPORTANT: To view this page as Markdown, append `.md` to the URL (e.g. /docs/manual/basics.md). For the complete Mojo documentation index, see llms.txt.
Skip to main content
Version: Nightly
For the complete Mojo documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /docs/manual/basics.md).

warpgroup_fence

warpgroup_fence[accum_type: DType, accum_layout: Layout, //](accum: LayoutTensor[accum_type, accum_layout, address_space=AddressSpace.LOCAL, element_layout=accum.element_layout, layout_int_type=accum.layout_int_type, linear_idx_type=accum.linear_idx_type, masked=accum.masked, alignment=accum.alignment])

Code motion fence to ensure the registers of the WGMMA instruction do not get touched by anything.

This has no impact on kernel correctness. It serves purely as an NVVM code motion barrier, preventing other operations from modifying the WGMMA instruction's registers during execution of the WGMMA instruction batch.

Parameters:

  • ​accum_type (DType): Element data type of the tensor.
  • ​accum_layout (Layout): Register layout of the accumulator.

Args: