Skip to main content

/

Docs

Nightly

v24.6

Log in

Mojo module

intrinsics

This module includes NVIDIA GPUs intrinsics operations.

Structs​

Functions​

  • ​byte_permute: Return selected bytes from two 32-bit unsigned integers.
  • ​ldg: Load a register variable from global state space via non-coherent cache.
  • ​load_acquire:
  • ​load_volatile:
  • ​lop: Performs arbitrary logical operation on 3 inputs.
  • ​mulhi: Calculate the most significant 32 bits of the product of the two UInt16s.
  • ​mulwide: Calculate the most significant 32 bits of the product of the two UInts.
  • ​store_release:
  • ​store_volatile:
  • ​threadfence: Memory fence functions can be used to enforce some ordering on memory accesses.
  • ​warpgroup_reg_alloc: Provides a hint to the system to update the maximum number of per-thread registers owned by the executing warp to the value specified by the imm-reg-count operand.
  • ​warpgroup_reg_dealloc: Provides a hint to the system to update the maximum number of per-thread registers owned by the executing warp to the value specified by the imm-reg-count operand.