Script.atomic.global_max

Script.atomic.global_max(dst, values, *, sem='relaxed', scope='gpu', output=None)

Element-wise dst[i] = max(dst[i], values[i]) atomically, on global memory.

See global_add() for the full parameter description.

Notes

• Thread group: Can be executed by any sized thread group.

• Hardware: Requires compute capability 7.0+ (sm_70).

• PTX: atom.{sem}.{scope}.global.max.s32.

Parameters:

• dst (GlobalTensor)

• values (RegisterTensor)

• sem (str)

• scope (str)

• output (RegisterTensor | None)

Return type:

RegisterTensor | None