Type tensor_type_f32;
      if (tensor_type.hasRank()) {
        tensor_type_f32 = tensorflow::GetTypeFromTFTensorShape(
            tensor_type.getShape(), FloatType::getF32(context));
      } else {
        tensor_type_f32 = UnrankedTensorType::get(FloatType::getF32(context));
      }
      // Add cast to f32 to conform with element type of result.
      operand = rewriter.create<CastOp>(op.getLoc(), tensor_type_f32, operand);

		for {
			var addr uintptr
			if tp, addr = tp.next(dst + size); addr == 0 {
				break
			}
			dstx := (*uintptr)(unsafe.Pointer(addr))
			srcx := (*uintptr)(unsafe.Pointer(src + (addr - dst)))
			p := buf.get2()
			p[0] = *dstx
			p[1] = *srcx
		}
	}
}

// bulkBarrierPreWriteSrcOnly is like bulkBarrierPreWrite but
// does not execute write barriers for [dst, dst+size).
//