Michael J Sharp

US Patent:

20100042981, Feb 18, 2010

Filed:

Aug 13, 2009

Appl. No.:

12/583188

Inventors:

Robert Scott Dreyer - Menlo Park CA, US
Joel Kevin Jones - Alameda CA, US
Michael Douglas Sharp - Los Gatos CA, US
Ivan Dimitrov Baev - San Jose CA, US

International Classification:

G06F 9/45

US Classification:

717146, 717155, 717149

Abstract:

Generating parallelized executable code from input code includes statically analyzing the input code to determine aspects of data flow and control flow of the input code; dynamically analyzing the input code to determine additional aspects of data flow and control flow of the input code; generating an intermediate representation of the input code based at least in part on the aspects of data flow and control flow of the input code identified by the static analysis and the additional aspects of data and control flow of the input code identified by the dynamic analysis; and processing the intermediate representation to determine portions of the intermediate representation that are eligible for parallel execution; and generating parallelized executable code from the processed intermediate representation

US Patent:

20160335127, Nov 17, 2016

Filed:

May 11, 2015

Appl. No.:

14/709385

Inventors:

- SAN DIEGO CA, US
SUMIT SUR - BOULDER CO, US
ROBERT SCOTT DREYER - MENLO PARK CA, US
MICHAEL DOUGLAS SHARP - LOS GATOS CA, US
JAMES LYALL ESLIGER - RICHMOND HILL, CA

International Classification:

G06F 9/50

Abstract:

Systems and methods for dynamic granularity control of parallelized work in a heterogeneous multi-processor portable computing device (PCD) are provided. During operation a first parallelized portion of an application executing on the PCD is identified. The first parallelized portion comprising a plurality of threads for parallel execution on the PCD. Performance information is obtained about a plurality of processors of the PCD, each of the plurality of processors corresponding to one of the plurality of threads. A number M of workload partition granularities for the plurality of threads is determined, and a total execution cost for each of the M workload partition granularities is determined An optimal granularity comprising a one of the M workload partition granularities with a lowest total execution cost is determined, and the first parallelized portion is partitioned into a plurality of workloads having the optimal granularity.

US Patent:

20160147577, May 26, 2016

Filed:

Nov 25, 2014

Appl. No.:

14/553243

Inventors:

- San Diego CA, US
SUMIT SUR - BOULDER CO, US
ROBERT S. DREYER - MENLO PARK CA, US
MICHAEL D. SHARP - LOS GATOS CA, US
JAMES L. ESLIGER - RICHMOND HILL, CA
WISLON KWAN - TORONTO, CA
CHRISTOS MARGIOLAS - SANTA CLARA CA, US

International Classification:

G06F 9/52

Abstract:

Systems and methods for adaptive thread control in a portable computing device (PCD) are provided. During operation a plurality of parallelized tasks for an application on the PCD are created. The application is executed with at least one processor of the PCD processing at least one main thread of the application. A determination is made whether a portion of the application being executed includes one or more of the parallelized tasks. A determination is made whether to perform the parallelized tasks in parallel. Based on the determination whether to perform the parallelized tasks in parallel, the parallelized tasks are executed with the at least one main thread of the application if the determination is not to perform the parallelized tasks in parallel, or if the determination is to perform the parallelized tasks in parallel, at least one worker thread is activated to execute the parallelized task in parallel with the main thread.

US Patent:

20160139901, May 19, 2016

Filed:

Feb 12, 2015

Appl. No.:

14/620513

Inventors:

- SAN DIEGO CA, US
ROBERT SCOTT DREYER - MENLO PARK CA, US
JASON KIM - SAN JOSE CA, US
MICHAEL DOUGLAS SHARP - LOS GATOS CA, US

International Classification:

G06F 9/45
G06F 9/44

Abstract:

Systems, methods, and computer programs are disclosed for performing runtime auto-parallelization of application code. One embodiment of such a method comprises receiving application code to be executed in a multi-processor system. The application code comprises an injected code cost computation expression for at least one loop in the application code defining a serial workload for processing the loop. A runtime profitability check of the loop is performed based on the injected code cost computation expression to determine whether the serial workload can be profitably parallelized. If the serial workload can be profitably parallelized, the loop is executed in parallel using two or more processors in the multi-processor system.