International Business Machines Corporation

Markus Kaltenbach

Christian Jacobi

Ulrich Mayer

Maarten Jakob Boersma

Jentje Leenstra

Anthony Saporito

Simon H Friedmann

Martin Recktenwald

Jens Leenstra

Thomas G Koehler

Johannes C Reichart

Silvia Melitta Mueller

Frank Siegfried Lehnert

Markus Helms

Uwe Brandt

Eyal Naor

Siegmund Schlechter

Girish Gopala Kurup

Willm Hinrichs

Christophe J Layer

Philipp Panitz

Michael K Gschwind

Dung Q Nguyen

Hung Qui Le

Brian William Thompto

Jose Eduardo Moreira

Brett Olsson

Nicolas Maeding

Sam G Chu

David Lang

Christian Zoellin

Wilhelm Ernst Haller

Tim Niggemeier

Ulrich Krauch

Philipp Oehler

Michael Johannes Jaspers

Gregory William Alexander

Jan Van Lunteren

Aaron Tsai

Deanna Postles Dunn Berger

Thomas Pflueger

Rolf Fritz

Thomas Fuchs

Dietmar Schmunkamp

Christoph W Wandel

Richard Joseph Branciforte

Cordt W Starke

Dwifuzi Coe

Chung-Lung K Shum

Michael Andrew Blake

Timothy Carl Bronson

Christopher Michael Abernathy

Michael Klein

Jochen Preiss

Jang-Soo Lee

Joerg Deutschle

Tobias Gemmeke

Ashraf ElSharif

Lior Binyamini

Ulrike Schmidt

Taylor J Pritchard

Maxim Scholl

Ludmila Zernakov

A computer system includes a processor core and a memory system in signal communication with the processor core. The memory system includes a first cache and a second cache. The first cache is arranged at a first level of a hierarchy in the memory system and is configured to store a plurality of first-cache entries. The second cache is arranged at a second level of the hierarchy that is lower than the first level, and stores a plurality of second-cache entries. The first cache maintains a directory that contains information for each of the first-cache entries. The second cache maintains a shadow pointer directory (SPD) that includes one or more SPD entries that maps each of the first-cache entries to a corresponding second cache entry at a lower-level cache location.

Shadow pointer directory in an inclusive hierarchical cache

A machine instruction to find a condition location within registers, such as vector registers. The machine instruction has associated therewith a register to be examined and a result location. The register includes a plurality of elements. In execution, the machine instruction counts a number of contiguous elements of the plurality of elements of the register having a particular value in a selected location within the contiguous elements. Other locations within the contiguous elements are ignored for the counting. The counting provides a count placed in the result location.

Instructions to count a number of contiguous register elements having specific values in a selected location

A computer-implemented method includes assigning a first group of one or more units of an instruction pipeline of a processor as a frontend group and assigning a second group of the one or more units of the instruction pipeline of the processor as a backend group. A frontend logout is performed to transfer one or more trace records from the first group to a trace controller during an in-memory trace of an instruction. A backend logout is performed to transfer one or more trace records from the second group to the trace controller during the in-memory trace of the instruction. A next instruction is started in the first group of the instruction pipeline before the backend logout completes.

In-memory trace with overlapping processing and logout

Driving address translations in a microprocessor system by sending a rejected Lx+1 cache request from a first set of caches of a first level Lx to a central request unit, transferring an Lx+1 cache request having a translation of a virtual address into a physical address stored in a first buffer, from the central request unit to the at least one Lx+1 cache, and keeping an Lx+1 cache request lacking a translation of a virtual address into a physical address stored in the first buffer, pending in the central request unit.

Fetch request arbiter

Embodiments are for special tracking pool enhancement for core L1 address invalidates. An invalidate request is designated to fill an entry in a queue in a local cache of a processor core, the queue including a first allocation associated with processing any type of invalidate request and a second allocation associated with processing an invalidate request not requiring a response in order for a controller to be made available, the entry being in the second allocation. Responsive to designating the invalidate request to fill the entry in the queue in the local cache, a state of the controller that made the invalidate request is changed to available based at least in part on the entry being in the second allocation.

Special tracking pool enhancement for core local cache address invalidates

The present disclosure relates to a method of operating a translation lookaside buffer (TLB) arrangement for a processor supporting virtual addressing, wherein multiple translation engines are used to perform translations on request of one of a plurality of dedicated processor units. The method comprises: maintaining by a cache unit a dependency matrix for the engines to track for each processing unit if an engine is assigned to the each processing unit for a table walk. The cache unit may block a processing unit from allocating an engine to a translation request when the engine is already assigned to the processing unit in the dependency matrix.

Method and apparatus for an efficient TLB lookup

A computer implemented method to operate different processor cache levels of a cache hierarchy for a processor with pipelined execution is suggested. The cache hierarchy comprises at least a lower hierarchy level entity and a higher hierarchy level entity. The method comprises: sending a fetch request to the cache hierarchy; detecting a miss event from a lower hierarchy level entity; sending a fetch request to a higher hierarchy level entity; and scheduling at least one write pass.

Operating different processor cache levels

A computer implemented method for saving cache access power is suggested. The cache is provided with a set predictor logic for providing a generated set selection for selecting a set in the cache, and with a set predictor cache for pre-caching generated set indices of the cache. The method comprises further: receiving a part of a requested memory address; checking, in the set predictor cache, whether the requested memory address is already generated; in the case, that the requested memory address has already been generated: securing that the set predictor cache is switched off; issuing the pre-cached generated set index towards the cache; and securing that only that part of the cache is switched on that is associated with the pre-cached generated set index.

Method and arrangement for saving cache power

Embodiments of the present invention provide systems and methods for mapping the architected state of one or more threads to a set of distributed physical register files to enable independent execution of one or more threads in a multiple slice processor. In one embodiment, a system is disclosed including a plurality of dispatch queues which receive instructions from one or more threads and an even number of parallel execution slices, each parallel execution slice containing a register file. A routing network directs an output from the dispatch queues to the parallel execution slices and the parallel execution slices independently execute the one or more threads.

Independent mapping of threads

A first request is received to access a first set of data in a first cache. A likelihood that a second request to a second cache for the first set of data will be canceled is determined. Access to the first set of data is completed based on the determining the likelihood that the second request to the second cache for the first set of data will be canceled.

Selective downstream cache processing for data access

An address translation facility is provided for multiple virtualization levels, where a guest virtual address may be translated to a guest non-virtual address, the guest non-virtual address corresponding without translation to a host virtual address, and the host virtual address may be translated to a host non-virtual address, where translation within a virtualization level may be specified as a sequence of accesses to address translation tables. The address translation facility may include a first translation engine and a second translation engine, where the first and second translation engines each have capacity to perform address translation within a single virtualization level of the multiple virtualization levels. In operation, based on the first translation engine performing a guest level translation, the second translation engine may perform a host level translation of a resulting guest non-virtual address to a host non-virtual address based on the guest level translation by the first translation engine.

Multi-engine address translation facility

Growing community of inventors

Holzgerlingen, Germany

Markus Kaltenbach