Executable code for constrained computing environments

Abstract

A system and method for facilitating adjusting computer code running in computing environments where writing of executable memory is restricted. An example method includes using a virtual machine to generate a first executable image from software code, and employing the first executable image to selectively modify the virtual machine and the first executable image, resulting in an updated virtual machine and a second executable image. The second executable image represents an updated version of the first executable image. To transform dynamic code, such as natively produced computer code, into one or more executable images for running on constrained devices, embodiments may leverage existing dynamic code generators (e.g., within Java HotSpot Performance Engine, also simply called HotSpot), selectively modifying them to generate different, i.e., updated code that is usable on subsequent runs of the associated software application, thereby effectively enabling modification of executable code used to run the software.

Description

[0001] Cross References to Related Applications [0002] This application claims priority to U.S. Provisional Patent Application Serial No. 61 / 919,654, filed December 20, 2013, entitled "TRANSFORMINGNATIVEINSTRUCTIONSINTOANIMMUTABLEEXECUTABLEIMAGE," which application is hereby incorporated by reference for all purposes as if in this application The same is fully explained in . technical field [0003] The present application relates to software and more particularly to systems and methods for facilitating dynamic computer code generation and use in a computing environment where writes to executable memory are constrained. Background technique [0004] Software for facilitating dynamic code generation is employed in a variety of demanding applications, including network-distributed enterprise software, general Web-based applications, and the like. Such applications often require portable platform-independent software that can be efficiently adapted to the characteristics of ...

AI Technical Summary

Problems solved by technology

[0008] However, using a remote server for dynamic or adaptive code generation can be extremely slow and inefficient

Furthermore, this approach may require expensive support for potentially time-consuming cross-compilation to account for different target platform characteristics that define or characterize the runtime computing environment, such as central processing unit (CPU) characteristics, operating system endiannes ,and many more

Images