# US 11,113,225 B2 · Extending multichip package link off package

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 11,113,225 B2
- Document ID: US11113225B2
- Status: Granted
- Kind: B2
- Application number: 16/946,109
- Filed: 2020-06-05
- Published: 2020-12-31
- Granted: 2021-09-07
- Earliest claimed: 2020-06-05
- Assignees: Intel Corporation
- Applicants: Intel Corporation
- Inventors: Debendra Das Sharma; Zuoguo Wu; Mahesh Wagh; Mohiuddin M. Mazumder; Venkatraman Iyer; Jeff C. Morriss
- Web page: https://patentgrep.com/patent/US11113225B2

## CPC

- G06F 13/4027: {using bus bridges (G06F13/4022 takes precedence)}
- G06F 13/4022: {using switching circuits, e.g. switching matrix, connection or expansion network (G06F13/4009 takes precedence)}
- G06F 13/405: {where the bridge performs a synchronising function}
- G06F 13/4265: {on a point to point bus (G06F13/4247, G06F13/4282 take precedence)}
- H10W 70/611: {for connecting multiple chips together}
- H10W 70/65: Shapes or dispositions of interconnections
- H10W 90/00: Package configurations
- H10W 70/63: Vias, e.g. via plugs
- H10W 90/724: {between a chip and a stacked insulating package substrate, interposer or RDL}
- Y02D 10/00: Energy efficient computing, e.g. low power processors, power management or thermal management
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- H10W: GENERIC PACKAGES, INTERCONNECTIONS, CONNECTORS OR OTHER CONSTRUCTIONAL DETAILS OF DEVICES COVERED BY CLASS H10
- Y02D: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE

## Abstract

An interconnect interface is provided to enable communication with an off-package device over a link including a plurality of lanes. Logic of the interconnect interface includes receiver logic to receive a valid signal from the off-package device on a dedicated valid lane of the link indicating that data is to arrive on a plurality of dedicated data lanes in the plurality of lanes, receive the data on the data lanes from the off-package device sampled based on arrival of the valid signal, and receive a stream signal from the off-package device on a dedicated stream lane in the plurality of lanes. The stream signal corresponds to the data and indicates a particular data type of the data. The particular data type can be one of a plurality of different data types capable of being received on the plurality of data lanes of the link.

## Claims

### Claim 1

A multi-chip package comprising:
a substrate;
a first integrated circuit (IC) die mounted on the substrate; and
a second IC die mounted on the substrate and coupled to the first IC die,
the first IC die comprising:
a host processor, and
a multi-chip package link (MCPL) interface coupled to the host processor, the MCPL interface including a first physical layer (PHY) block to physically transmit data to the second IC die and receive data from the second IC die over a first plurality of data lanes,
the first PHY block comprising:
a first plurality of transmitters to transmit data to the second IC die, each transmitter associated with one of the first plurality of data lanes;
a first plurality of receivers to receive data from the second IC die, each receiver associated with one of the first plurality of data lanes;
a clock recovery circuit to ensure that a receiver clock signal is centered with respect to phase, the clock recovery circuit to receive a forwarded clock signal from the second IC die over a strobe lane, the clock recovery circuit including a phase interpolator to determine a phase by which to adjust the forwarded clock signal to generate a receiver clock signal;
a reference voltage generator to add an offset to an input voltage to generate a reference voltage; and
one or more samplers of the first plurality of receivers to sample data signals received over one or more of the first plurality of data lanes based on the reference voltage and the receiver clock signal;
the second IC die comprising:
an off-package interface to couple the multi-chip package to another multi-chip package and at least one input/output (IO) device, the off-package interface to implement a plurality of layered protocol stacks, including a first layered protocol stack to communicate with the other multi-chip package and a second layered protocol stack to communicate with the at least one 10 device,
the off-package interface comprising:
a second PHY block comprising a plurality of modular, protocol-agnostic electrical sub-blocks, each electrical sub-block to be associated with either the first layered protocol stack or the second layered protocol stack, the second PHY block further comprising:
a second plurality of transmitters and receivers of a first electrical sub-block associated with a second plurality of data lanes, the second plurality of data lanes to couple the multi-chip package to the other multi-chip package,
a third plurality of transmitters and receivers of a second electrical sub-block associated with a third plurality of data lanes, the third plurality of data lanes comprising 10 lanes to couple the multi-chip package to the at least one 10 device,
a first upper layer logic block to communicate data in accordance with the first layered protocol stack,
a second upper layer logic block to communicate data in accordance with the second layered protocol stack, and
routing logic to selectively route the first upper layer logic block to the second plurality of data lanes and to selectively route the second upper layer logic block to the third plurality of data lanes.

### Claim 2 (depends on 1)

The multi-chip package of claim 1, wherein the other multi-chip package comprises a graphics processor.

### Claim 3 (depends on 1)

The multi-chip package of claim 1, wherein the second layered protocol stack comprises a serial interconnect logic block to implement a serial interconnect protocol.

### Claim 4 (depends on 1)

The multi-chip package of claim 1, wherein the first layered protocol stack is to implement a memory access protocol.

### Claim 5 (depends on 1)

The multi-chip package of claim 1, wherein the routing logic comprises a multiplexer.

### Claim 6 (depends on 1)

The multi-chip package of claim 1, wherein the host processor comprises a multiprocessor.

### Claim 7 (depends on 6)

The multi-chip package of claim 6, wherein the host processor comprises a processing core.

### Claim 8 (depends on 1)

The multi-chip package of claim 1, wherein the other multi-chip package comprises a system memory device.

### Claim 9 (depends on 1)

The multi-chip package of claim 1, wherein at least the first PHY block is to perform single-ended signaling.