IP Library Granted Patent US 8,607,035
Granted Patent B2
US 8,607,035 · App. 12/550,951 · Granted Dec 10, 2013

Multi-core processing utilizing prioritized interrupts for optimization

Inventor: Paul Kimelman (Alamo, CA)
Assignee: Texas Instruments Incorporated
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 8,607,035
App. No.
12/550,951
Granted
Dec 10, 2013
Kind
B2
Abstract

This invention relates to multi-core, multi-processing, factory multi-core and DSP multi-core. The nature of the invention is related to more optimal uses of a multi-core system to maximize utilization of the processor cores and minimize power use. The novel and inventive steps are focused on use of interrupts and prioritized interrupts, along with optional in-built methods, to allow systems to run more efficiently and with less effort on the part of the programmer.

Claims (20)

1. A method of performing a data processing operation on a multi-processor system comprising the steps of:

dividing the data processing operation into a plurality of discrete tasks;

linking said tasks in to at least one chain;

assigning priorities to said tasks;

converting each of said tasks into an interrupt service routine; and

performing said interrupt service routines of at least one chain of task in a multi-processor system by responding a received interrupt by assigning a corresponding interrupt service routine to one data processor core of a plurality of data processor cores of the multi-processing system dependent upon respective priority levels of any interrupt service routines operating on said plurality of data processor cores and a priority level of said received interrupt.

2. The method of claim 1 , wherein:

said step of converting each of said tasks into an interrupt service routine includes trigging an end of task interrupt upon completion of said corresponding task; and

said step of performing said interrupt service routines includes

assigning each task at a beginning of a chain to one data processor core of said plurality of data processor cores,

in response to an end of task interrupt, issuing an interrupt for any sequential task in the corresponding chain.

3. The method of claim 1 , wherein:

said step of converting each of said tasks into an interrupt service routine includes converting at least one task into two parallel interrupt service routines and at least one comparison interrupt service routine, said comparison interrupt service routine comparing results of said parallel interrupt service routines to verify correct operation.

4. The method of claim 3 , wherein:

said two parallel interrupt service routines are identical.

5. The method of claim 3 , wherein:

said two parallel interrupt service routines employ different algorithms.

6. The method of claim 3 , wherein:

said step of verifying correct operation notes any data processor core operating incorrectly; and

said step of assigning a corresponding interrupt service routine to one data processor core of a plurality of data processor cores does not assign a interrupt service routine to any data processor core noted as operating incorrectly.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 17, 2010
From: KIMELMAN, PAUL
To: TEXAS INSTRUMENTS INCORPORATED
Reel/Frame 024092/0049 →
Continuity (2)
Provisional Application 61092958 · Aug 29, 2008
Related Publication 20100174886A1 · Jul 8, 2010