triton.language.atomic_load

triton.language.atomic_load(pointer, mask=None, sem=None, scope=None, _semantic=None)

Atomically loads from the memory locations specified by pointer.

Parameters:

• pointer (Block of dtype=triton.PointerDType) – The memory locations to load from.

• mask (Block of triton.int1, optional) – If mask[idx] is false, the corresponding load is not performed and the result is unspecified.

• sem (str, optional) – Specifies the memory semantics. Acceptable values are “acquire” (default) and “relaxed”.

• scope (str, optional) – Defines the scope of threads that observe the synchronizing effect. Acceptable values are “gpu” (default), “cta”, and “sys”.

This function can also be called as a member function on tensor, as x.atomic_load(...) instead of atomic_load(x, ...).