International Business Machines Corporation

Kean G Kuiper

Frank Eliot Levine

Enio Manuel Pineda

Robert John Urquhart

Jimmie Earl DeWitt, Jr

Robert Tod Dimpsey

William Preston Alexander, Iii

Jesse Mannes Gordon

Geoffrey Owen Blandy

Scott Thomas Jones

Christopher Michael Richardson

Matthew Francis Peters

William Robert Reynolds

John Day Howard

A process generates a sample based on an event. The process identifies, with a dispatch monitor, a next dispatched monitored thread that is dispatched by a current processor. The process sets a processor affinity of the next dispatched monitored thread such that the next dispatched monitored thread runs only on the current processor without being able to migrate to a different processor. The process also retrieves, with a sampled thread that runs on the current processor, a next dispatched monitored thread call stack after the processor affinity of the next dispatched monitored thread has been set to the processor. The process restores the processor affinity of the next dispatched monitored thread after the next dispatched monitored thread call stack has been obtained.

Idle transitions sampling with execution context

A sample is generated based on an event. Further, an interrupt handler captures information for an interrupted thread on a current processor. In addition, an affinity of the interrupted thread is set such that the interrupted thread runs only on the current processor without being able to migrate to a different processor. A sampler thread that runs on the current processor retrieves a call stack associated with the interrupted thread after the affinity of the interrupted thread has been set to the current processor. The affinity of the interrupted thread is restored after the call stack has been retrieved.

Performing call stack sampling by setting affinity of target thread to a current process to prevent target thread migration

An interrupt is generated based on an event. Further, a thread is selected for monitoring. In addition, an interrupt handler captures information for the monitored thread. An affinity of the monitored thread is set such that the monitored thread runs only on a current processor without being able to migrate to a different processor. A sampler thread that runs on the current processor retrieves a call stack associated with a monitored thread after the affinity of the monitored thread has been set to the current processor.

Scalable system call stack sampling

A computer implemented method, apparatus, and computer usable program code for sampling call stack information. An event is monitored during execution of a plurality of threads executed by a plurality of processors. In response to an occurrence of the event, a thread is identified in the plurality of threads to form an identified thread. A plurality of sampling threads is woken, wherein a sampling thread within the plurality of sampling threads is associated with each processor in the plurality of processors and wherein one sampling thread in the plurality of sampling threads obtains call stack information for the identified thread.

Call stack sampling in a data processing system

A system and method for Java&#x2122; call stack sampling combined with native sampling is presented. A kernel-mode device driver records and stores sampled context information that includes a program counter, a thread identifier, and a process identifier. The sampling thread receives a notification from the kernel-mode device driver, and proceeds to collect call stack information from a Java&#x2122; Virtual Machine. In turn, the sampling thread retrieves the sampled context information and harvests symbols corresponding to a loaded module. Once symbols are harvested, the sampling thread combines returned call stack nodes with native function leaf nodes into a unified output tree that provides the relationship between the two different node types. In one embodiment, the sampling thread may generate native function leaf nodes and the unified output tree in a post-processing manner.

System and method for call stack sampling combined with node and instruction tracing

A computer implemented method, apparatus, and computer usable program code to collect information for a system or processor having a transition between an idle state and a non-idle state to form collected system or processor information. The collected system or processor information is provided for analysis by an application. Idle counts occurring during execution of code are collected. The idle counts are provided to an application for analyzing why a processor becomes idle.

Method and apparatus for analyzing idle states in a data processing system

A method, a system, an apparatus, and a computer program product are presented for loading LDIF ((Lightweight Directory Access Protocol (LDAP) Data Interchange Format) data into a directory. A single reader thread reads entries from an LDIF file; each entry is associated with a priority value, e.g., the priority value for an entry may be based on the number of delimiters in the distinguished name within the entry. Multiple loading threads concurrently add the entries into an LDAP directory in accordance with their associated priority values. The entries are added to the LDAP directory in an order of highest indicated priority to lowest indicated priority; entries that are associated with priority values that indicate equal priority are considered to be equivalent for the purposes of ordering the entries, so entries that have equal priority may be added to the LDAP directory in any order.

Method and system for maintaining consistency during multi-threaded processing of LDIF data

A method and apparatus for identifying differences in runs of a computer program that are due to changes in the code of the computer program. With the apparatus and method, call trees are generated for two or more builds of a computer program. The apparatus and method perform a 'tree-subtract' operation in which the two or more call trees generated during runs of two or more different builds of a computer program are subtracted from one another to identify build to build differences in the execution of the computer program. From the resulting tree, portions of the runs of the different builds of the computer program where the resource utilization of the computer program has not changed are easily identifiable. Moreover, portions of the runs of the different builds of the computer program where there are improvements or regressions in resource utilization of the computer program may be easily identifiable.

Method and apparatus for identifying differences in runs of a computer program due to code changes

A computer implemented method, apparatus, and computer usable program code to collect information for a system or processor having a transition between an idle state and a non-idle state. Idle counts occurring during execution of code are collected to form collected system or processor information. The collected system or processor information is provided to an application for analyzing why a processor becomes idle.

A system and method for incrementally compacting a computer system heap is presented. A heap, such as a garbage-collected heap in a Java environment, is divided into logical regions. When the heap is becoming fragmented, an incremental compaction cycle is commenced. During a first time period, the first region of the heap is compacted, with subsequent regions being compacted during subsequent time periods. A time period commences when a garbage collection event occurs. In a multiprocessor environment the regions can be divided into a number of sections which are each compacted using a different processor. One or more break tables are constructed indicating how far contiguous groups of moveable objects should be moved to better group objects and eliminate interspersed free spaces. References throughout the heap that point to objects within the compacted region are then adjusted so that the references point to the new object locations.

System and method for compacting a computer system heap

A method and system are disclosed for operating an object-oriented environment within a computer system. The execution of the object-oriented environment is initiated. A first memory location is allocated in a memory area within the object-oriented environment for storing an object. A second memory location is allocated in the memory area with the allocation of the first memory location for storing a persistence indicator. The persistence indicator indicates a point during the execution of the object-oriented environment at which the allocation of the first memory location occurred. A total number of bytes of the memory area which had been allocated during the execution of the object-oriented environment when the allocation of the first memory location occurred is stored as the persistence indicator.

Method and system for allocation of a persistence indicator for an object in an object-oriented environment

A threshold size of &ldquo;interesting&rdquo; free space regions (those areas that exceed a minimum size and are therefore worth recovering) is employed during garbage collection when detecting and reclaiming non-allocated memory space. For mark/sweep garbage collection, the threshold is based upon a number of consecutive zero mark bits in a mark bit array. The mark bit array, in which the liveness of each allocated data object is indicated by a logical &ldquo;1&rdquo; at the first memory segment of a live object and a logical &ldquo;0&rdquo; at the first memory segment for dead object, with the remaining memory segments being marked with a logical &ldquo;0&rdquo;, is produced by the mark phase of a mark/sweep garbage collector. During the sweep phase, however, the mark bit array is processed instead of each object in memory. Interesting free space regions, which meet or exceed the threshold size and are targeted for de-allocation and coalescing, are identified from runs of consecutive zeros, with verification that the run corresponds to unused memory. This allows sequences of bits to be inspected concurrently at a high rate, eliminating the need to inspect objects individually and skipping small free space regions intermingled with live objects. By eliminating the need to examine each object for &ldquo;liveness,&rdquo; groups of objects, represented by mark bits, are treated as aggregates which may be processed more quickly.

Growing community of inventors

Round Rock, TX, United States of America

Kean G Kuiper