-
公开(公告)号:ZA201604938B
公开(公告)日:2019-02-27
申请号:ZA201604938
申请日:2016-07-15
Applicant: IBM
Inventor: GREINER DAN , FARRELL MARK , OSISEK DAMIAN LEO , SCHMIDT DONALD WILLIAM , BUSABA FADI YUSUF , KUBALA JEFFREY PAUL , BRADBURY JONATHAN DAVID , HELLER LISA CRANTON , SLEGEL TIMOTHY , GAINEY JR CHARLES (DECEASED) , JACOBI CHRISTIAN
Abstract: Embodiments relate to dynamic enablement of multithreading. According to an aspect, a computer-implemented method for dynamic enablement of multithreading in a configuration is provided. The configuration includes a core configurable between a single thread (ST) mode and a multithreading (MT) mode, where the ST mode addresses a primary thread and the MT mode addresses the primary thread and one or more secondary threads on shared resources of the core. The method includes executing in the primary thread in the ST mode, an MT mode setting instruction. A number of threads requested is obtained from a location specified by the MT mode setting instruction. Based on determining that the number of threads requested indicates multiple threads, the MT mode is enabled to execute the multiple threads including the primary thread and the one or more secondary threads.
-
公开(公告)号:ZA201704122B
公开(公告)日:2018-11-28
申请号:ZA201704122
申请日:2017-06-15
Applicant: IBM
Inventor: BRADBURY JONATHAN DAVID , JACOBI CHRISTIAN , SLEGEL TIMOTHY , GSCHWIND MICHAEL KARL
Abstract: Embodiments relate to accessing data in a memory. A method for accessing data in a memory coupled to a processor is provided. The method receives a memory reference instruction for accessing data of a first size at an address in the memory. The method determines an alignment size of the address in the memory. The method accesses the data of the first size in one or more groups of data by accessing each group of data block concurrently. The groups of data have sizes that are multiples of the alignment size.
-
公开(公告)号:PL2769382T3
公开(公告)日:2018-09-28
申请号:PL12871181
申请日:2012-11-15
Applicant: IBM
-
公开(公告)号:HUE037896T2
公开(公告)日:2018-09-28
申请号:HUE15711701
申请日:2015-03-16
Applicant: IBM
-
25.
公开(公告)号:DK2769305T3
公开(公告)日:2018-08-06
申请号:DK12871072
申请日:2012-11-15
Applicant: IBM
Inventor: SLEGEL TIMOTHY , JACOBI CHRISTIAN , BRADBURY JONATHAN DAVID , GSCHWIND MICHAEL KARL , SCHWARZ ERIC MARK
IPC: G06F9/30
-
26.
公开(公告)号:ES2673903T3
公开(公告)日:2018-06-26
申请号:ES15711701
申请日:2015-03-16
Applicant: IBM
Inventor: GREINER DAN , FARRELL MARK , OSISEK DAMIAN LEO , SCHMIDT DONALD WILLIAM , BUSABA FADI YUSUF , KUBALA JEFFREY PAUL , BRADBURY JONATHAN DAVID , HELLER LISA CRANTON , SLEGEL TIMOTHY
Abstract: Un sistema informático, que comprende: una configuración (102) que comprende un núcleo configurable entre un modo de subproceso único (ST) y un modo de subprocesamiento múltiple (MT), el modo ST que se dirige a un subproceso primario y el modo MT que se dirige al subproceso primario y a uno o más subprocesos secundarios sobre recursos compartidos del núcleo; y una facilidad de subprocesamiento múltiple (103) configurada para controlar la utilización de la configuración, en donde la facilidad de subprocesamiento múltiple está adaptada para: acceder al subproceso primario en el modo ST usando un valor de dirección de núcleo; conmutar del modo ST al modo MT; y acceder al subproceso primario o a uno del uno o más subprocesos secundarios en el modo MT, y caracterizado por que la facilidad de subprocesamiento múltiple está adaptada para acceder al subproceso primario o a uno del uno o más subprocesos secundarios en el modo MT usando un valor de dirección expandido, comprendiendo el valor de dirección expandido el valor de dirección de núcleo concatenado con un valor de dirección de subproceso.
-
公开(公告)号:DK3123326T3
公开(公告)日:2018-06-25
申请号:DK15711701
申请日:2015-03-16
Applicant: IBM
Inventor: GREINER DAN , FARRELL MARK , OSISEK DAMIAN LEO , SLEGEL TIMOTHY , SCHMIDT DONALD WILLIAM , BUSABA FADI YUSUF , KUBALA JEFFREY PAUL , BRADBURY JONATHAN DAVID , HELLER LISA CRANTON
IPC: G06F9/50
-
公开(公告)号:AU2015238663B2
公开(公告)日:2017-05-25
申请号:AU2015238663
申请日:2015-03-16
Applicant: IBM
Inventor: GREINER DAN , FARRELL MARK , OSISEK DAMIAN LEO , SCHMIDT DONALD WILLIAM , BUSABA FADI YUSUF , KUBALA JEFFREY PAUL , BRADBURY JONATHAN DAVID , HELLER LISA CRANTON , SLEGEL TIMOTHY , GAINEY JR CHARLES
IPC: G06F9/46
Abstract: THREAD CONTEXT RESTORATION IN A MULTITHREADING COMPUTER SYSTEM Amultithreading computer system includesa configuration with a core configurable between a single thread (ST) mode and a multithreading (MT) mode. The ST mode addresses a primary thread and the MT mode addresses the primary thread and one or more secondary threads on shared resources of the core. A multithreading facility is configured to control utilization of the configuration to perform a method including disabling one or more secondary threads based on switching from MT mode to ST mode. A thread context of secondary threads is made unavailable to programs. Based on a last-set program-specified maximum thread-id indicating MT, the thread context is obtained by a) executing a set MT instruction to resume the MT mode, and b) based on being in the resumed MT mode, accessing the thread context.
-
公开(公告)号:AU2014208864B2
公开(公告)日:2017-04-13
申请号:AU2014208864
申请日:2014-01-07
Applicant: IBM
Inventor: BRADBURY JONATHAN DAVID
IPC: G06F9/30
Abstract: A Vector Galois Field Multiply Sum and Accumulate instruction. Each element of a second operand of the instruction is multiplied in a Galois field with the corresponding element of the third operand to provide one or more products. The one or more products are exclusively ORed with each other and exclusively ORed with a corresponding element of a fourth operand of the instruction. The results are placed in a selected operand.
-
公开(公告)号:MX346496B
公开(公告)日:2017-03-22
申请号:MX2014010947
申请日:2012-11-15
Applicant: IBM
Inventor: SLEGEL TIMOTHY , SCHWARZ ERIC MARK , BRADBURY JONATHAN DAVID , GSCHWIND MICHAEL KARL , JACOBI CHRISTIAN
Abstract: Se provee una instrucción del conteo de carga a frontera de bloque que provee una distancia de una dirección de memoria especificada a una frontera de memoria especificada. La frontera de memoria es una frontera que no es cruzada en la carga de datos. La frontera puede ser especificada de una diversidad de maneras, incluyendo pero limitado a un valor variable en el texto de instrucción, un valor de texto de instrucción fijo codificado en el código de operación o una frontera a base de registro o puede ser determinada dinámicamente.
-
-
-
-
-
-
-
-
-