| 摘要 |
A clock distribution network locks a local clock signal to a reference clock signal using a first feedback loop associated with a synchronization circuit (e.g., a PLL or a DLL). The local clock signal can then be selectively distributed to a plurality of clock destination nodes via a clock network. Clock distribution may be disabled as needed to save power. The first feedback loop is active irrespective of whether clock distribution is enabled. The delay through the clock network may drift due to temperature and supply-voltage fluctuations, which introduces phase errors in the distributed clock signals. A second feedback loop is activated when clock distribution is enabled to compensate for this drift.
|
