IP Library Granted Patent US 8,214,790
Granted Patent B2
US 8,214,790 · App. 12/397,941 · Granted Jul 3, 2012

Low RC global clock distribution

Assignee: Oracle America
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 8,214,790
App. No.
12/397,941
Granted
Jul 3, 2012
Kind
B2
Abstract

A semiconductor die includes: a clock distribution network that distributes a clock signal within the die. The clock distribution network includes: a clock tree corresponding to one or more metal layers of the die, a plurality of clock spines corresponding to a metal layer of the die, a plurality of clock wings corresponding to a metal layer of the die, a plurality of clock grid drivers placed in one or more gaps of a floorplan corresponding to the semiconductor layer of the die, a clock grid placed in the one or more gaps of the floorplan, a plurality of buffers placed in a local gain buffer pair configuration wherein the local gain buffer pair connects the clock grid to a shorting bar, and a plurality of conductors that connect the shorting bar to a plurality of loads.

Claims (76)

1. A computer implemented method of designing a semiconductor die with a flexible floorplan comprising:

arranging a plurality of bumps that connect the die to a mechanical package, the plurality of bumps corresponding to a bump layout of the die;

placing a plurality of blocks in a floorplan corresponding to a semiconductor layer of the die;

placing a clock tree corresponding to one or more metal layers of the die;

placing a plurality of clock spines corresponding to a metal layer of the die; and

placing a plurality of clock wings corresponding to a metal layer of the die;

wherein the clock wings connect one or more of the plurality of clock spines to the clock tree; and

wherein the clock tree is comprised of a plurality of wires arranged at a planar angle of 45 degrees.

2. A computer implemented method of designing a semiconductor die with a flexible floorplan comprising:

arranging a plurality of bumps that connect the die to a mechanical package, the plurality of bumps corresponding to a bump layout of the die;

placing a plurality of blocks in a floorplan corresponding to a semiconductor layer of the die;

placing a clock tree corresponding to one or more metal layers of the die;

placing a plurality of clock spines corresponding and placing a plurality of clock wings corresponding to a metal layer of the die;

wherein the clock wings connect one or more of the plurality of clock spines to the clock tree,

wherein the clock tree is comprised of a plurality of wires arranged at a planar angle of 45 degrees; and

wherein the clock tree provides for diagonal clock routes.

3. A computer implemented method of designing a semiconductor die with a flexible floorplan comprising:

arranging a plurality of bumps that connect the die to a mechanical package, the plurality of bumps corresponding to a bump layout of the die;

placing a plurality of blocks in a floorplan corresponding to a semiconductor layer of the die;

placing a clock tree corresponding to one or more metal layers of the die;

placing a plurality of clock spines corresponding to a metal layer of the die; and

placing a plurality of clock wings corresponding to a metal layer of the die;

wherein the clock wings connect one or more of the plurality of clock spines to the clock tree;

placing a plurality of clock grid drivers in one or more gaps of the floorplan;

placing a clock grid in the one or more gaps of the floorplan, wherein the clock grid drivers connect one or more of the clock spines to the clock grid;

placing a plurality of buffers in a local gain buffer pair configuration, wherein the local gain buffer pair connects the clock grid to a shorting bar; and

placing a plurality of conductors that connect the shorting bar to a plurality of loads.

4. The method of claim 3 , the method further comprising:

clipping an unused portion of each clock wing.

5. The method of claim 3 , wherein the clock tree is comprised of a plurality of wires arranged at a planar angle of 45 degrees.

6. The method of claim 5 , wherein the clock tree provides for diagonal clock routes.

7. The method of claim 3 , wherein the local gain buffer pair corresponds to a second buffer and a third buffer as part of a three-stage FO4 local gain configuration.

8. A semiconductor device comprising:

a mechanical package; and

a semiconductor die comprising:

a semiconductor layer,

a plurality of metal layers,

a clock distribution network that distributes a clock signal within the die, and

a plurality of bumps that connect the die to the mechanical package;

wherein the clock distribution network comprises a low RC global clock distribution, the low RC global clock distribution comprising:

a clock tree corresponding to one or more metal layers of the die,

a plurality of clock spines corresponding to a metal layer of the die,

a plurality of clock wings corresponding to a metal layer of the die, wherein the clock wings connect one or more of the clock spines to the clock tree,

a plurality of clock grid drivers placed in one or more gaps of a floorplan corresponding to the semiconductor layer of the die;

a clock grid placed in the one or more gaps of the floorplan, wherein the clock grid drivers connect one or more of the clock spines to the clock grid,

a plurality of buffers placed in a local gain buffer pair configuration, wherein the local gain buffer pair connects the clock grid to a shorting bar, and

a plurality of conductors that connect the shorting bar to a plurality of loads.

9. The semiconductor device of claim 8 , wherein the unused portion of the clock wings are clipped.

10. The semiconductor device of claim 8 , wherein the clock tree is comprised of a plurality of wires arranged at a planar angle of 45 degrees.

11. The semiconductor device of claim 10 , wherein the clock tree provides for diagonal clock routes.

12. The semiconductor device of claim 10 , wherein the local gain buffer pair corresponds to a second buffer and a third buffer as part of a three-staged FO4 local gain configuration.

13. A system comprising:

an input device;

an output device;

a mechanical chassis;

a printed circuit board; and

a semiconductor device comprising:

a mechanical package, and

a semiconductor die;

wherein the semiconductor die comprises:

a semiconductor layer,

a plurality of metal layers,

a clock distribution network that distributes a clock signal within the die, and

a plurality of bumps that connect the die to the mechanical package; and

wherein the clock distribution network comprises a low RC global clock distribution, the low RC global clock distribution comprising:

a clock tree corresponding to one or more metal layers of the die,

a plurality of clock spines corresponding to a metal layer of the die,

a plurality of clock wings corresponding to a metal layer of the die, wherein the clock wings connect one or more of the clock spines to the clock tree,

a plurality of clock grid drivers placed in one or more gaps of a floorplan corresponding to the semiconductor layer of the die;

a clock grid placed in one or more gaps of the floorplan, wherein the clock grid drivers connect one or more of the clock spines to the clock grid,

a plurality of buffers placed in a local gain buffer pair configuration, wherein the local gain buffer pair connects the clock grid to a shorting bar, and

a plurality of conductors that connect the shorting bar to a plurality of loads.

14. The system of claim 13 , wherein the unused portion of the clock wings are clipped.

15. The system of claim 13 , wherein the clock tree is comprised of a plurality of wires arranged at a planar angle of 45 degrees.

16. The system of claim 15 , wherein the clock tree provides for diagonal clock routes.

17. The system of claim 13 , wherein the local gain buffer pair corresponds to a second buffer and a third buffer as part of a three-stage FO4 local gain configuration.

Assignments (2)
MERGER AND CHANGE OF NAME Recorded Dec 16, 2015
From: ORACLE USA, INC.; SUN MICROSYSTEMS, INC.; ORACLE AMERICA, INC.
To: ORACLE AMERICA, INC.
Reel/Frame 037311/0171 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 9, 2009
From: MASLEID, ROBERT P.; BALLARD, JAMES
To: SUN MICROSYSTEMS, INC.
Reel/Frame 022364/0987 →
Continuity (1)
Related Publication 20100229142A1 · Sep 9, 2010