International Business Machines Corporation

Arnold S Tran

Peter Anthony Sandon

Vijayalakshmi Srinivasan

Hillery C Hunter

Robert Dov Herzl

Philip George Emma

Bartholomew Blaner

Erik L Hedberg

Kenneth Anthony Lauricella

Marc Raymond Faucher

John Edward Barth, Jr

Clarence Rosser Ogilvie

William Robert Reohr

Jason Frederick Cantin

Jack Robert Smith

Shih-Hsiung Stephen Tung

Kenichi Tsuchiya

Jay Gerald Heaslip

Thomas B Chadwick, Jr

Richard J Schmalz

Kevin Arthur Chiarot

John E Barth

Theodore J Schmitt

Henry Harvey Burkhart

Disclosed are a processor and a processing method incorporating an instruction pipeline with direction prediction (i.e., taken or not taken) for conditional branch instructions. In the embodiments, reading of a branch instruction history table (BHT) and a branch instruction target address cache (BTAC) for branch direction prediction occurs in parallel with the current instruction fetch in order to minimize delay in the next instruction fetch. Additionally, direction prediction is performed in the very next clock cycle based either on an initial direction prediction for the specific instruction, as stored in the BHT, or, if applicable, on a prior entry for the specific instruction in the BTAC. An override bit associated with each entry in the BTAC is the determining factor for whether or the BTAC or BHT is controlling. Override bits in the BTAC can be pre-established based on the branch instruction type in order to ensure prediction accuracy.

Processor and data processing method incorporating an instruction pipeline with conditional branch direction prediction for fast access to branch target instructions

A cache for use in a central processing unit (CPU) of a computer includes a data array; a tag array configured to hold a list of addresses corresponding to each data entry held in the data array; a least recently used (LRU) array configured to hold data indicating least recently used data entries in the data array; a line fill buffer configured to receive data from an address in main memory that is located external to the cache in the event of a cache miss; and a shadow register associated with the line fill buffer, wherein the shadow register is configured to hold LRU data indicating a current state of the LRU array.

Shadow registers for least recently used data in cache

A hardware description language (HDL) design structure embodied on a machine-readable data storage medium includes elements that when processed in a computer aided design system generates a machine executable representation of a device for implementing dynamic refresh protocols for DRAM based cache. The HDL design structure further includes a DRAM cache partitioned into a refreshable portion and a non-refreshable portion; and a cache controller configured to assign incoming individual cache lines to one of the refreshable portion and the non-refreshable portion of the cache based on a usage history of the cache lines; wherein cache lines corresponding to data having a usage history below a defined frequency are assigned by the controller to the refreshable portion of the cache, and cache lines corresponding to data having a usage history at or above the defined frequency are assigned to the non-refreshable portion of the cache.

Structure for implementing dynamic refresh protocols for DRAM based cache

A method for implementing dynamic refresh protocols for DRAM based cache includes partitioning a DRAM cache into a refreshable portion and a non-refreshable portion, and assigning incoming individual cache lines to one of the refreshable portion and the non-refreshable portion of the cache based on a usage history of the cache lines. Cache lines corresponding to data having a usage history below a defined frequency are assigned to the refreshable portion of the cache, and cache lines corresponding to data having a usage history at or above the defined frequency are assigned to the non-refreshable portion of the cache.

Method and system for implementing dynamic refresh protocols for DRAM based cache

A method of integrating a hybrid architecture in a set associative cache having a first type of memory structure for one or more ways in each congruence class, and a second type of memory structure for the remaining ways of the congruence class, includes determining whether a memory access request results in a cache hit or a cache miss; in the event of a cache miss, determining whether LRU way of the first type memory structure is also the LRU way of the entire congruence class, and if not, then copying the contents of the LRU way of the first type memory structure into the LRU way of the entire congruence class, and filling the LRU way of the first type memory structure with a new cache line in the event of a cache miss; and updating LRU bits, depending upon the results of the memory access request.

Method and system for integrating SRAM and DRAM architecture in set associative cache

A method for implementing prioritized refresh of a multiple way, set associative DRAM based cache includes identifying, for each of a plurality of sets of the cache, the existence of a most recently used way that has not been accessed during a current assessment period; and for each set, refreshing only the identified most recently used way of the set not accessed during the current assessment period, while ignoring the remaining ways of the set; wherein a complete examination of each set for most recently used ways therein during the current assessment period constitutes a sweep of the cache.

Method and system for implementing prioritized refresh of DRAM based cache

A system and method for selecting a system for initializing a data processing system. A data processing system includes, a processor coupled to a memory via an interconnect and multiple systems for initializing the aforementioned data processing system. A multiplexor is also included to select between multiple initialization systems. In the method of initializing a data processing system, an initialization or reset command is received from an external reset circuit or the processor; at least one selector signal is generated; and a set of initialization data from a selected initialization system is utilized to initialize the data processing system.

System and method for system initializating a data processing system by selecting parameters from one of a user-defined input, a serial non-volatile memory and a parallel non-volatile memory

A memory and memory architecture for use by a processor executing real time code and a system on a chip including the processor and memory containing the code. An effective address is maintained in a cache directory. In the preferred embodiment memory, individual functions are loaded into physical memory at permanently selected locations and selected by the effective address in the cache directory. By preselecting task storage locations, system performance may be tuned or optimized to assure predictable performance or task execution.

Real time invariant behavior cache

A computer system includes a processor and a cache and memory management unit. The processor includes a means for retiring instructions in program order. The cache and memory management unit includes means for detecting when a translation has been evicted from a lookaside buffer and means for communicating eviction information to the means for retiring instructions in program order. The means for retiring instructions in program order includes means for holding a storage related instruction which causes a miss in the lookaside buffer or in the cache in a first pass of execution until the instruction becomes the oldest storage related instruction in program sequence and further includes means responsive to the eviction information for flushing all storage related instructions except the current storage related instruction. The system avoids the occurrence of misses in the buffer late in execution (e.g., PASS 2 or later), thus avoiding a necessity for complex recovery provisions.

Apparatus to guarantee TLB inclusion for store operations

An improved DLAT structure distinguishes between address spaces and data spaces. The DLAT structure classifies data spaces by one or more space identifications which control assignment of virtual page addresses to DLAT rows. In one embodiment, a 'private space bit' is used to select different DLAT addressing algorithms. In another embodiment, data spaces are sub-classified using space identification bits, and for each sub-class, a unique algorithm is selected based on the page address bits. An Exclusive OR function is used to generate the DLAT selection bits. This approach minimizes private space synonyms while maximizing common space synonyms. The result is improved performance since the former minimizes thrashing and the latter maximizes the value of the DLAT common bit.

Growing community of inventors

South Burlington, VT, United States of America

Arnold S Tran