IMPORTANT: To view this page as Markdown, append `.md` to the URL (e.g. /get-started.md). For the complete documentation index, see llms.txt.
Skip to main content
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

fence_proxy_tensormap_generic_sys_acquire

def fence_proxy_tensormap_generic_sys_acquire[dtype: AnyType](ptr: Pointer[dtype], size: Int32)

Acquires a system-wide memory fence for tensor map operations.

This function establishes a memory fence that ensures proper synchronization between tensor map operations and system memory. It guarantees that all previous memory operations are completed before subsequent tensor map accesses.

Note:

This is a low-level synchronization primitive typically used in conjunction with TMA (Tensor Memory Access) operations on NVIDIA GPUs.

Parameters:

  • โ€‹dtype (AnyType): The data type of the tensor map object being synchronized.

Args:

  • โ€‹ptr (Pointer[dtype]): Pointer to the tensor map object in system memory that needs to be synchronized.
  • โ€‹size (Int32): The size in bytes of the tensor map object being synchronized.