tvm_tirx.backend.cuda.op.mma_store

tvm_tirx.backend.cuda.op.mma_store(m, n, dst_ptr, src_ptr, src_offset, dst_stride, *, ty)

Store the result of PTX MMA into a destination pointer.