International Business Machines Corporation

Ralf Winkelmann

Gerrit Koch

Christian Habermann

Martin Recktenwald

Hans-Werner Tast

Christian Jacobi

Johannes C Reichart

Carsten Greiner

Michael F Fee

Matthias Klein

Carsten Otte

Edward W Chencinski

Matthias Pflanz

Thomas P Grosser

Hanno Eichelberger

Clinton E Bubb

Diana Lynn Orf

Roopesh Ambalath Matayambath

Chaitanya Kancherla

The present disclosure relates to a method for a computer system comprising a plurality of processor cores including a first processor core and a second processor core, wherein a data item is exclusively assigned to the first processor core, of the plurality of processor cores, for executing an atomic primitive by the first processor core. The method includes receiving by the first processor core, from the second processor core, a request for accessing the data item, and in response to determining by the first processor core that the executing of the atomic primitive is not completed by the first processor core, returning a rejection message to the second processor core.

Method and processor system for executing a TELT instruction to access a data item during execution of an atomic primitive

The present disclosure relates to a method for a computer system comprising a plurality of processor cores, including a first processor core and a second processor core, wherein a cached data item is assigned to a first processor core, of the plurality of processor cores, for exclusively executing an atomic primitive. The method includes receiving, from a second processor core at a cache controller, a request for accessing the data item, and in response to determining that the execution of the atomic primitive is not completed by the first processor core, returning a rejection message to the second processor core.

Executing an atomic primitive in a multi-core processor system

Methods testing a data coherency algorithm via a simulated multi-processor environment are provided, which include implementing: (i) a transactional footprint keeping the address of each cache line used by the processor core, (ii) a reference model operating on and keeping a set of timestamps for a cache line, the set including a construction date representing a global timestamp when new data arrives at a private cache hierarchy and an expiration date representing another global timestamp when a cross-invalidation hits the private cache hierarchy, (iii) a core observed timestamp representing a global timestamp of an oldest construction date of data used before, and (iv) interface events monitoring instruction sequences guaranteed by transactional execution to ensure atomicity of a transaction. Upon detecting a transaction end event and finding a cache line of the transactional footprint having an expiration date older than or equal to a core observed time, an error is reported.

Testing a data coherency algorithm

A method for verification of a design of an electronic circuit is provided. The method includes executing test runs of the design. The method further includes increasing a fail counter if the executing of a test run of the test runs failed. The method further includes increasing a pass counter if the executing of the test run of the test runs passed. The method further includes halting the executing of the test runs of the design if the current ratio of the fail counter versus the pass counter exceeds a predetermined threshold.

System and method for generation of an integrated circuit design

Verification is provided of a functional correctness of a graph-based coherency verification tool for logic designs of arrangements of processors and processor caches, the graph-based coherency verification tool using trace files as input for verifying memory ordering rules of a given processor architecture for accesses to the caches, wherein nodes in a graph represent memory accesses and edges represent dependencies between them. The verifying includes (i) providing a specification of a test case for a self-checking tool, the test case comprising a sequence of statements in a high-level description language format, representing memory access events and system events; and (ii) generating trace files with the self-checking tool for the graph-based coherency verification tool by producing permutations of trace events, which are defined by the sequence of statements of the test case.

Verifying a graph-based coherency verification tool

Testing a data coherency algorithm of a multi-processor environment. The testing includes implementing a global time incremented every processor cycle and used for timestamping; implementing a transactional execution flag representing a processor core guaranteeing the atomicity and coherency of the currently executed instructions; implementing a transactional footprint, which keeps the address of each cache line that was used by the processor core; implementing a reference model, which operates on every cache line and keeps a set of timestamps for every cache line; implementing a core observed timestamp representing a global timestamp, which is the oldest construction date of data used before; implementing interface events; and reporting an error whenever a transaction end event is detected and any cache line is found in the transactional footprint with an expiration date that is older than or equal to the core observed time.

Data caching for use in a computer system including a lower cache memory and a higher cache memory. The higher cache memory receives a fetch request. It is then determined by the higher cache memory the state of the entry to be replaced next. If the state of the entry to be replaced next indicates that the entry is exclusively owned or modified, the state of the entry to be replaced next is changed such that a following cache access is processed at a higher speed compared to an access processed if the state would stay unchanged.

Data caching method

A mechanism for verifying order of entities being processed by a device under test (DUT) is provided. The mechanism includes arranging the entities into a temporal order, and encoding the entities to maintain the temporal order of the entities and produce encoded entities with each being a random value. The encoded entities each have a one-to-one mapping to their corresponding one of the entities in the temporal order. The encoded entities are input into the DUT to verify its output, and responsive to detecting an error in the output corresponding to one encoded entity, the one encoded entity is decoded into a current decoded error entity. It is determined which is lower in the temporal order between the current decoded error entity and a previous decoded error entity. Responsive to the current decoded error entity being lower than the previous decoded error entity, the current decoded error entity is stored.

Growing community of inventors

Holzgerlingen, Germany

Ralf Winkelmann