-
公开(公告)号:AU2020214661B2
公开(公告)日:2022-09-22
申请号:AU2020214661
申请日:2020-01-14
Applicant: IBM
Inventor: RAISCH CHRISTOPH , KRAEMER MARCO , LEHNERT FRANK , KLEIN MATTHIAS , BRADBURY JONATHAN , JACOBI CHRISTIAN , BELMAR BRENTON , DRIEVER PETER
Abstract: A data processing system (210) and a method for handling an input/output store instruction (30), comprising a system nest (18) communicatively coupled to at least one input/output bus (22) by an input/output bus controller (20). The data processing system (210) further comprises at least a data processing unit (216) comprising a core (12), a system firmware (10) and an asynchronous core-nest interface (14). The data processing unit (216) is communicatively coupled to the system nest (18) via an aggregation buffer (16). The system nest (18) is configured to asynchronously load from and/or store data to an external device (214) which is communicatively coupled to the input/output bus (22). The data processing unit (216) is configured to complete the input/output store instruction (30) before an execution of the input/output store instruction (30) in the system nest (18) is completed.
-
公开(公告)号:AU2015330266A1
公开(公告)日:2017-03-09
申请号:AU2015330266
申请日:2015-09-14
Applicant: IBM
Inventor: FARRELL MARK , HELLER LISA , KUBALA JEFFREY PAUL , SCHMIDT DONALD WILLIAM , GREINER DAN , SLEGEL TIMOTHY , BUSABA FADI YUSUF , OSISEK DAMIAN , BRADBURY JONATHAN DAVID , LEHNERT FRANK , NERZ BERND , JACOBI CHRISTIAN
Abstract: A system and method of implementing a modified priority routing of an input/output (I/O) interruption. The system and method determines whether the I/O interruption is pending for a core and whether any of a plurality of guest threads of the core is enabled for guest thread processing of the interruption in accordance with the determining that the I/O interruption is pending. Further, the system and method determines whether at least one of the plurality of guest threads enabled for guest thread processing is in a wait state and, in accordance with the determining that the at least one of the plurality of guest threads enabled for guest thread processing is in the wait state, routes the I/O interruption to a guest thread enabled for guest thread processing and in the wait state.
-
公开(公告)号:GB2489307A
公开(公告)日:2012-09-26
申请号:GB201204158
申请日:2012-03-09
Applicant: IBM
Inventor: GERWIG GUENTER , ENGLER EBERHARD , BUTTLAR JOACHIM VON , MEISSNER KLAUS , LEHNERT FRANK
Abstract: A time of day (TOD) system clock associated with a processing core (2) comprises a host clock register (5) incremented by means of a high precision oscillator (3) and a firmware clock register (6), incremented every time the host clock register (5) is incremented. The system monitors for failures of the host clock register (5), and during a failure of the host clock register (5) increments the firmware clock register (6) by means of timing signals of the processing core (2). Upon receiving a clock value read request providing the content of the host clock register (5) if no failure is detected and updating the firmware clock register (6) with the value of the host clock register (5).
-
14.
公开(公告)号:GB2455212B
公开(公告)日:2012-03-21
申请号:GB0823157
申请日:2008-12-19
Applicant: IBM
Inventor: GERWIG GUENTER , MAYER ULRICH , LEHNERT FRANK , SWANEY SCOTT BARNETT , WOERNER ALEXANDER , KRAUCH ULRICH
IPC: G06F11/10
-
公开(公告)号:GB2455009B
公开(公告)日:2012-02-29
申请号:GB0822766
申请日:2008-12-15
Applicant: IBM
Inventor: GERWIG GUENTER , ENGLER EBERHARD , LEHNERT FRANK , MAYER ULRICH
IPC: G06F1/14
-
16.
公开(公告)号:GB2456891B
公开(公告)日:2012-02-01
申请号:GB0823186
申请日:2008-12-19
Applicant: IBM
Inventor: GERWIG GUENTER , MAYER ULRICH , LEHNERT FRANK , SHUM KEVIN CHUNG-LUNG
-
公开(公告)号:GB2456656A
公开(公告)日:2009-07-29
申请号:GB0822778
申请日:2008-12-15
Applicant: IBM
Inventor: MAYER ULRICH , LEHNERT FRANK , GERWIG GUENTER , SWANEY SCOTT BARNETT
Abstract: The invention relates to apparatus for controlling timeouts and delays of an error recovery procedure in a digital circuit, e.g. a microprocessor. The apparatus comprises a finite state machine (FSM) 10, having a plurality of states 12 and a plurality of transitions 14. Transitions 14 are arranged between two states 12 respectively. States 12 correspond with operation steps (40, 44, 52, 56, 58, 64) of the error recovery procedure, including error classification, a drain operation, a fence operation in which a microprocessor core does not communicate with memory, a reset or refresh operation, and automatic built-in self test (ABIST). Transitions 14 of the FSM 10 depend on conditions (46, 50, 53, 57, 59, 62) for the error recovery procedure. The FSM 10 is coupled with a timeout logic circuit 20 which controls a timer to obtain the timeouts (46, 53, 57, 59) of the error recovery procedure. The FSM is configurable by a data vector which describes states 12 of the FSM for which the timer should be engaged.
-
公开(公告)号:AU2015330266B2
公开(公告)日:2018-10-04
申请号:AU2015330266
申请日:2015-09-14
Applicant: IBM
Inventor: GAINEY CHARLES , FARRELL MARK , HELLER LISA , KUBALA JEFFREY PAUL , SCHMIDT DONALD WILLIAM , GREINER DAN , SLEGEL TIMOTHY , BUSABA FADI YUSUF , OSISEK DAMIAN , BRADBURY JONATHAN DAVID , LEHNERT FRANK , NERZ BERND , JACOBI CHRISTIAN
Abstract: A system and method of implementing a modified priority routing of an input/output (I/O) interruption. The system and method determines whether the I/O interruption is pending for a core and whether any of a plurality of guest threads of the core is enabled for guest thread processing of the interruption in accordance with the determining that the I/O interruption is pending. Further, the system and method determines whether at least one of the plurality of guest threads enabled for guest thread processing is in a wait state and, in accordance with the determining that the at least one of the plurality of guest threads enabled for guest thread processing is in the wait state, routes the I/O interruption to a guest thread enabled for guest thread processing and in the wait state.
-
公开(公告)号:CA2961690A1
公开(公告)日:2016-04-14
申请号:CA2961690
申请日:2015-09-14
Applicant: IBM
Inventor: FARRELL MARK , HELLER LISA , KUBALA JEFFREY PAUL , SCHMIDT DONALD WILLIAM , GREINER DAN , SLEGEL TIMOTHY , BUSABA FADI YUSUF , OSISEK DAMIAN , BRADBURY JONATHAN DAVID , LEHNERT FRANK , NERZ BERND , JACOBI CHRISTIAN , GAINEY CHARLES
IPC: G06F9/48
Abstract: A system and method of implementing a modified priority routing of an input/output (I/O) interruption. The system and method determines whether the I/O interruption is pending for a core and whether any of a plurality of guest threads of the core is enabled for guest thread processing of the interruption in accordance with the determining that the I/O interruption is pending. Further, the system and method determines whether at least one of the plurality of guest threads enabled for guest thread processing is in a wait state and, in accordance with the determining that the at least one of the plurality of guest threads enabled for guest thread processing is in the wait state, routes the I/O interruption to a guest thread enabled for guest thread processing and in the wait state.
-
公开(公告)号:GB2502540A
公开(公告)日:2013-12-04
申请号:GB201209548
申请日:2012-05-30
Applicant: IBM
Inventor: LEHNERT FRANK , JACOBI CHRISTIAN , GERWIG GUENTER , SHUM CHUNG-LUNG K , SLEGEL TIMOTHY J
IPC: G06F1/14
Abstract: Method for generating monotonically increasing time-OF-day (TOD) values in a multiprocessor system, comprising: receiving synchronization impulses 532; and refusing an execution of read instruction of a time-of-day value (STCK B) within processor 504 of the system if said execution is requested after a predefined time after a synchronization impulse, and if trigger signal 502a, indicative of new data received by a related memory 202 external to the processor, has been received after the predefined time. The predefined time is the smallest latency for data value transfer between processors. The memory is a shared cache. The synchronisation pulses are received from PLL 530. A time flag is set after the predetermined time is reached while a reject flag is set if time flag is set and the trigger signal is received. TOD execution is rejected if reject flag is set. Said flags are reset upon reception of synchronisation impulses.
-
-
-
-
-
-
-
-
-