Cray Incorporated

Gail A Alverson

Burton Jordan Smith

Charles David Callahan, Ii

Brian D Koblenz

Susan L Coatney

Richard D Korry

Laurence S Kaplan

Simon H Kahan

Allan Porterfield

Mark L Niehaus

Robert L Alverson

Matthew Paul Kelly

Mark Swan

Asaph Zemach

Daniel C Duval

Eric A Hoffman

Kazuya Okubo

An availability system is provided that includes a hierarchy of controllers for providing event notifications relating to availability of components of a scalable MPP system. A controller receives a subscription from a child controller that identifies an event type and a generator. The controller stores in a subscription store an indication that the subscription has been received from the child controller. When a parent controller has not yet been notified of a subscription with a matching event type and generator as indicated by the subscription store, the controller sends the subscription to the parent controller. When the parent controller has already been notified of a subscription with a matching event type and generator as indicated by the subscription store, the controller suppresses the sending of the subscription to the parent controller.

Event notifications relating to system failures in scalable systems

Various techniques for manipulating data using access states of memory, access control fields of pointers and operations, and exception raising and exception trapping in a multithreaded computer system. In particular, the techniques include synchronization support for a thread blocked in a word, demand evaluation of values, parallel access of multiple threads to a list, synchronized and unsynchronized access to a data buffer, use of forwarding to avoid checking for an end of a buffer, use of sentinel word to detect access past a data structure, concurrent access to a word of memory using different synchronization access modes, and use of trapping to detect access to restricted memory.

Synchronization techniques in a multithreaded environment

Various techniques for manipulating data using access states of memory, access control fields of pointers and operations, and exception raising and exception trapping in a multithreaded computer system. In particular, the techniques include synchronization support for a thread blocked in a word, demand evalution of values, parallel access of multiple threads to a list, synchronized and unsynchronized access to a data buffer, use of fowarding to avoid checking for an end of a buffer, use of sentinel work to detect access past a data structure, concurrent access to a word of memory using different synchronization access modes, and use of trapping to detect access to restricted memory.

Accessing a collection of data items in a multithreaded environment

Detecting access to a memory location in a multithreaded environment

A method and system that prepares a task for being swapped out from processor utilization that is executing on a computer with multiple processors that each support multiple streams. The task has one or more teams of threads, where each team represents threads executing on a single processor. The task designates, for each stream that is executing a thread, one stream as a team master stream and one stream as a task master stream. For each team master stream, the task notifies the operating system that the team is ready to be swapped out when each other thread of the team has saved its state and has quit its stream. Finally, for the task master stream, the task notifies the operating system that the task is ready to be swapped when it has saved its state and each other team has notified that it is ready to be swapped out.

Deferred task swapping in a multithreaded environment

A system for debugging targets using various techniques, some of which are particularly useful in a multithread environment. These techniques include implementing breakpoints using out-of-line instruction emulation so that an instruction replaced with a breakpoint instruction does not need to be returned to its original location for single-step execution, executing a debugger nub for each target as part of the target task but using a nub task thread for the nub execution that is separate from the target task threads, providing immunity from breakpoints for specified threads such as the nub thread via specialized breakpoint handlers used by those threads, and virtualizing the debugger nub such that a shared root nub provides a uniform interface between the debugger and the target while specialized nubs provide differing functionality based on the type of target being debugged.

Debugging techniques in a multithreaded environment

Placing a task of a multithreaded environment in a known state

Inter-thread long jumps in a multithreaded environment

Task swap out in a multithreaded environment

Stream management in a multithreaded environment

Restricting access to memory in a multithreaded environment

A method and system in a multithreaded processor for processing events without interrupt notifications. In one aspect of the present invention, an operating system creates a thread to execute on a stream of the processor. During execution of the thread, the thread executes a loop that determines whether an event has occurred and, in response to determining whether an event has occurred, assigns a different thread to process the event so that multiple events can be processed in parallel and so that interrupts are not needed to signal that the event has occurred. Another aspect of the present invention provides a method and system for processing asynchronously occurring events without interrupt notifications. To achieve this processing, a first thread is executed to generate a notification that the event has occurred upon receipt of the asynchronously occurring event. A second thread is also executed that loops determining whether a notification has been generated and, in response to determining that a notification has been generated, performing the processing necessary for the event.

Techniques for reducing the rate of instruction issuance

A method and system that prepares a task for being swapped out from processor utilization that is executing on a computer with multiple processors that each support multiple streams. The task has one or more teams of threads, where each team represents threads executing on a single processor. The task designates, for each stream, one stream that is executing a thread as a team master stream. The task designates one stream that is executing a thread as a task master stream. For each team master stream, the task notifies the operating system that the team is ready to be swapped out when each other thread of the team has saved its state and has quit its stream. Finally, for the task master stream, the task notifies the operating system that the task is ready to be swapped when it has saved its state and each of the other teams have notified the operating system that that team is ready to be swapped out.

User program and operating system interface in a multithreaded environment

Growing community of inventors

Seattle, WA, United States of America

Gail A Alverson