US 11,675,574 B2
Framework for user-directed profile-driven optimizations
Teresa Louise Johnson, Los Gatos, CA (US); and Xinliang David Li, Palo Alto, CA (US)
Assigned to Google LLC, Mountain View, CA (US)
Filed by Google LLC, Mountain View, CA (US)
Filed on Mar. 17, 2022, as Appl. No. 17/697,119.
Application 17/697,119 is a continuation of application No. 16/524,929, filed on Jul. 29, 2019, granted, now 11,321,061.
Application 16/524,929 is a continuation of application No. 15/701,296, filed on Sep. 11, 2017, granted, now 10,365,903, issued on Jul. 30, 2019.
Application 15/701,296 is a continuation of application No. 13/855,557, filed on Apr. 2, 2013, granted, now 9,760,351, issued on Sep. 12, 2017.
Prior Publication US 2022/0206769 A1, Jun. 30, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 8/41 (2018.01)
CPC G06F 8/443 (2013.01) [G06F 8/4434 (2013.01); G06F 8/4441 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for using profiling to determine application-specific values for an application, the method comprising:
profiling, by a computer, one or more application-specific parameters for which to determine at least one application-specific value and collecting profile data by performing operations of:
generating, by the computer, an instrumentation binary from an instrumentation build;
executing, by the computer, a training run with one or more representative workloads using the instrumentation binary;
analyzing, by the computer, the collected profile data using a set of standard value profile transformations, wherein the set of standard value profile transformations are provided in one or more compiler support libraries, and wherein the set of standard value profile transformations are specified in user-directed value transformation support interfaces; and
generating, by the computer using the collected profile data, an optimized binary utilizing the at least one application-specific value for the profiled application-specific parameter recorded in the collected profile data.