International Business Machines Corporation

David Kevin Siegwart

Frank Eliot Levine

Allan Henry Kielstra

Martin J Hirzel

Matthew Roy Kilner

Reid T Copeland

Enio Manuel Pineda

Toshihiko Koju

Daryl James Maier

Nikola Grcevski

Ryan Andrew Sciampacone

Timothy P Ellison

Iain Alexander Ireland

Rajini Sivaram

Jordan Ryan Zannier

Ramy Shahin

Nikola Groevski

A method, system, and computer-readable medium for binary translation cause a binary translator to combine raw binary code and compiler-produced metadata associated with a compiled program module. The binary translator is caused to further reconcile, using the compiler-produced metadata, original compiler-produced control flow information with how lower-level machine instructions comprise a control flow in the raw binary code, and original compiler-produced aliasing information with how lower-level machine instructions access the memory locations described by the aliasing information according to predetermined criteria. The binary translator further caused to prevent, copy propagation of values in temporary variables for decimal computations beyond offsets in the machine instructions where the temporary variables are killed. The binary translator further caused to remove identified dead store instructions, and to generate a new compiled program module comprising an optimized version of the compiled program module having strict compatibility to an original version of the compiled program module.

Binary translation using raw binary code with compiler produced metadata

A method for automatic verification of optimization of high level constructs includes generating a first executable code by compiling a computer program that includes a high level construct. The compiling includes generating a first set of machine instructions for the high level construct and storing compile-time information for the high level construct. The method further includes optimizing the first executable code which includes converting the first executable code into an intermediate language representation. The optimization further includes generating a second executable code. For this, the method includes generating a second set of machine instructions for the high level construct from the intermediate language representation. If the behavior of the first set of machine instructions and the second set of machine instructions matches, the second set of machine instructions is included in the second executable code, otherwise the first set of machine instructions is included.

Automatic verification of optimization of high level constructs using test vectors

A predicted profile is generated for target code to be executed on a processor of the computing environment. The predicted profile is based on a profile of sampled code. The sampled code is a different version of code than the target code and is a complex build of modules for which it is difficult to determine which versions of the modules have been profiled. Based on the predicted profile for the target code, a determination is made of predicted execution information for the target code. Based on the determining the predicted execution information for the target code, an action is performed to facilitate processing within the computing environment.

Generating a predicted profile for target code based on a profile of sampled code

A method of determining causes of external fragmentation in a memory. The method includes collecting information associated with release of an area of the memory by an application, storing the information in the area of the memory, and analyzing the information to determine why the area of the memory has not been reallocated to any application. In embodiments wherein a first portion of an area of a memory is allocated to an application by an allocator and a second portion of the area of the memory is released by the allocator, the method includes storing in the second portion of the area of the memory an indicator indicating that the second portion is a remaining portion, collecting information associated with release of the second portion, storing the information in the second portion, and analyzing the information to determine why the second portion is not reallocated to any application.

Determining causes of external fragmentation of memory

Aspects of the present invention include a method, system and computer program product that embeds collected profiling test case information into a binary file of a computer program and subsequently recompiles the program. The method includes a processor specifying one or more test cases to be embedded into a compiled binary file, wherein the one or more test cases relate to at least a portion of a computer program representing a compilation unit; executing the one or more embedded test cases under the computer program; gathering performance data associated with the computer program as the one or more embedded test cases are executed; recompiling the compilation unit based on the performance data; and linking the computer program based on the performance data.

Embedding profile tests into profile driven feedback generated binaries

A method of determining causes of external fragmentation in a memory. The method includes collecting information associated with release of an area of the memory by an application, storing the information in the area of the memory, and analyzing the information to determine why the area of the memory has not been reallocated to any application. In embodiments wherein a first portion of an area of a memory is allocated to an application by an allocator and a second portion of the area of the memory is released by the allocator, the method includes storing in the second portion of the area of the memory an indicator indicating that the second portion is a remaining portion, colleting information associated with release of the second portion, storing the information in the second portion, and analyzing the information to determine why the second portion is not reallocated to any application.

The present disclosure provides a computer program product and activity recording system for identifying idleness in a processor via a concurrent software environment. A thread state indicator records an indication of a synchronization state of a software thread that is associated with an identification of the software thread. A time profiler identifies a processor of the computer system being idle and records an indication that the processor is idle. A dispatch monitor identifies a dispatch of the software thread to the processor. In response to the dispatch monitor determining the indication identifies that the processor is idle and the indication of a synchronization state of the software thread indicating the software thread ceases to execute in the processor, the dispatch monitor generates a record attributing the idleness of the processor to the software thread and the indicated synchronization state.

Recording activity of software threads in a concurrent software environment

The present disclosure provides a method for identifying idleness in a processor via a concurrent software environment. A thread state indicator records an indication of a synchronization state of a software thread that is associated with an identification of the software thread. A time profiler identifies a processor of the computer system being idle and records an indication that the processor is idle. A dispatch monitor identifies a dispatch of the software thread to the processor. In response to the dispatch monitor determining the indication identifies that the processor is idle and the indication of a synchronization state of the software thread indicating the software thread ceases to execute in the processor, the dispatch monitor generates a record attributing the idleness of the processor to the software thread and the indicated synchronization state.

Synchronization activity recording system for a concurrent software environment

Controlling access to a resource by a plurality of resource requesters is disclosed. The resource lock operates in a contention efficient (heavyweight) operating mode, and in response to a request from a resource requester to acquire the resource lock, a count of a total number of acquisitions of the resource lock in the contention efficient operating mode is incremented. In response to access to the resource not being contended by more than one resource requester, a count of a number of uncontended acquisitions of the resource lock in the contention efficient operating mode is incremented, and a contention rate is calculated as the number of uncontended acquisitions in the contention efficient operating mode divided by the total number of acquisitions in the contention efficient operating mode. In response to the contention rate meeting a threshold contention rate, the resource lock is changed to a non-contention efficient (lightweight) operating mode.

Managing a resource lock

A method, system and program product that in response to a segmented server going offline reallocates only those primary and secondary destinations that need to be reallocated on an evenhanded basis, so there is at most a difference of one in the number of buckets allocated to each segmented server before and after re-allocation.

Reallocating secondary destinations on a network switch

Automatic memory management with a generational copy collector garbage collection algorithm running on parallel threads to perform simultaneous and independent copying and scanning. An object is scanned. It is determined whether an object referenced within the slot of the object being scanned may be copied to the same generation as the scanned object. A mark may be made to indicate that the referenced object may be copied to a different generation than the scanned object. The mark defers the copying of the referenced object.

Defer separating children in parallel copying garbage collection

A computer-implementable method, system and apparatus. The frequency of reference to an object, also known as 'hotness,' can be collected as a part of a profiling phase during a runtime operation of data-processing system by permitting each reference to the object during the runtime operation to randomly guess a predetermined value associated with the object, such that a correct guess thereof implies that the object is referenced frequently. Thereafter, the frequency of reference to the object can be validated by identifying a particular value in a header associated with the object, in response to collecting the frequency of reference to the object during the profiling phase, thereby increasing the scalability and efficiency of the runtime operation while permitting data associated with the frequency of reference to the object to other applications for immediate use.

Efficient object profiling for optimizing object locality

Methods and systems for generating a hashcode for a hash table are provided to improve the hash table performance. A particular method includes receiving an input of a key; applying a hash function to the key to generate an incoming hashcode of the key; encoding a key type into the incoming hashcode to generate a first hashcode; and storing the first hashcode. A particular method includes encoding the key type into at least one bit of a bucket index portion of the incoming hashcode.

Methods and systems for improving hash table performance

A method of operating a resource lock for controlling access to a resource by a plurality of resource requesters, the resource lock operating in a contention efficient (heavyweight) operating mode, and the method being responsive to a request from a resource requester to acquire the resource lock, the method comprising the steps of: incrementing a count of a total number of acquisitions of the resource lock in the contention efficient operating mode; in response to a determination that access to the resource is not contended by more than one resource requester, performing the steps of: a) incrementing a count of a number of uncontended acquisitions of the resource lock in the contention efficient operating mode; b) calculating a contention rate as the number of uncontended acquisitions in the contention efficient operating mode divided by the total number of acquisitions in the contention efficient operating mode; and c) in response to a determination that the contention rate meets a threshold contention rate, causing the resource lock to change to a non-contention efficient (lightweight) operating mode.

Growing community of inventors

Hursley, United Kingdom

David Kevin Siegwart