International Business Machines Corporation

Other

Dwain Alan Hicks

Shih-Hsiung Stephen Tung

Brian David Barrick

David Scott Ray

David A Campbell

Takeki Osanai

Michael John Mayfield

Hoichi Cheong

Kimberly Marie Fernsler

Kimming K So

Kin Shing Chan

Jonathan H Raymond

Peichun Peter Liu

Christian Jacobi

George McNeil Lattimore

David James Shippy

George W Rohrbaugh, Iii

Rajinder Paul Singh

Bryan J Lloyd

David Allen Hrusecky

Michael Thomas Dibrino

Pei-Chun Peter Liu

Kerey Michelle Tassin

Avery Cox Topps

Kimberly Marie Fensler

Hanaa Youssef

A computer system includes a translation lookaside buffer (TLB) and a processor. The TLB comprises a first TLB array and a second TLB array, and stores entries comprising virtual address information and corresponding real address information. The processor is configured to receive a first virtual address for translation, and to concurrently determine if the TLB stores a physical address associated with the first virtual address based on a first portion and a second portion of the first virtual address. The first portion is associated with a first page size and the second portion is associated with a second page size (different from the first page size). The first portion is used to perform lookup in either one of the first TLB array and the second TLB array and the second portion is used for performing lookup in other one of the first TLB array and the second TLB array.

Methods and systems for optimized translation of a virtual address having multiple virtual address portions using multiple translation lookaside buffer (TLB) arrays for variable page sizes

A computer system includes a translation lookaside buffer (TLB) data cache and a processor. The TLB data cache includes a hierarchical configuration comprising a first TLB array, a second TLB array, a third TLB array, and a fourth TLB array. The processor is configured to receive a first address for translation to a second address, and determine whether translation should be performed using a hierarchical page table or a hashed page table. The processor also determines (using a first portion of the first address) whether the first array stores a mapping of the first portion of the first address in response to determining that the translation should be performed using the hashed page table, and retrieving the second address from the third TLB array or the fourth TLB array in response to determining that the first TLB array stores the mapping of the first portion of the first address.

Methods and systems for incorporating non-tree based address translation into a hierarchical translation lookaside buffer (TLB)

Operation of a multi-slice processor that includes a plurality of execution slices, a plurality of load/store slices, and one or more page walk caches, where operation includes: receiving, at a load/store slice, an instruction to be issued; determining, at the load/store slice, a process type indicating a source of the instruction to be a host process or a guest process; and determining, in accordance with an allocation policy and in dependence upon the process type, an allocation of an entry of the page walk cache, wherein the page walk cache comprises one or more entries for both host processes and guest processes.

Operation of a multi-slice processor implementing a unified page walk cache

A system and method of stopping program execution includes tagging an entry in a virtual cache with an indicator bit where the virtual address of the entry corresponds to a virtual address range in a break point register, in response to a second virtual cache data access demand matching the entry tagged with the indicator bit, determining whether the second data access demand matches the virtual address range of the breakpoint register, and in response to the second data access demand matching the virtual address range of the break point register, flagging an exception and stopping execution of the program. In an embodiment, the method or system enters a slow-mode in response to the second data access demand matching the virtual cache entry with the indicator bit, and performs a full comparison between the second data access demand and the break point register virtual address range.

Virtual cache mechanism for program break point register exception handling

Operation of a multi-slice processor that includes a plurality of execution slices, a plurality of load/store slices, and one or more translation caches, where operation includes: determining, at the load/store slice, a real address from a cache hit in the translation cache for an effective address for an instruction received at a load/store slice; determining, at the load/store slice, an error condition corresponding to an access of the real address; determining, at the load/store slice, a process type indicating a source of the instruction to be a guest process; and responsive to determining the error condition, initiating, in dependence upon the process type indicating a source of the instruction to be a guest process, an effective address translation corresponding to a cache miss in the translation cache for the effective address for the instruction.

Operation of a multi-slice processor implementing exception handling in a nested translation environment

A method or system of translating a virtualized address to a real address is disclosed that includes receiving a virtualized address for translation; generating a predicted intermediate address translation using a portion of the bit field of the virtualized address; determining a predicted real address using the predicted intermediate address or portion thereof; performing a translation of the virtualized address to an actual intermediate address; determining whether the predicted intermediate address is the same as the actual intermediate address; and in response to the predicted intermediate address being the same as the actual intermediate address, providing the predicted real address as the real address.

Methods and systems for predicting virtual address

A method of optimized congruence class matching for concurrent memory translation requests to avoid memory access conflicts with respect to a virtual memory managed by a processor is provided. The method includes initiating a first table walk by a first memory access of the concurrent memory translation requests and pending a subsequent table walk initiated by a subsequent memory access of the concurrent memory translation requests. Then, the method determines whether the subsequent table walk will cause a memory access conflict with the first table walk based on the optimized congruence class matching. The subsequent memory access is rejected when the subsequent table walk will cause the memory access conflict with the first table walk.

Performance optimized congruence class matching for multiple concurrent radix translations

The present invention provides for a method for a load address dependency mechanism in a high frequency, low power processor. A load instruction corresponding to a memory address is received. At least one unexecuted preceding instruction corresponding to the memory address is identified. The load instruction is stored in a miss queue. And the load instruction is tagged as a local miss.

Load address dependency mechanism system and method in a high frequency, low power processor system

A method is disclosed for executing a load instruction. Address information of the load instruction is used to generate an address of needed data, and the address is used to search a cache memory for the needed data. If the needed data is found in the cache memory, a cache hit signal is generated. At least a portion of the address is used to search a queue for a previous load instruction specifying the same address. If a previous load instruction specifying the same address is found, the cache hit signal is ignored and the load instruction is stored in the queue. A load/store unit, and a processor implementing the method, are also described.

Systems for executing load instructions that achieve sequential load consistency

A method, an apparatus, and a computer program product are provided for detecting load/store dependency in a memory system by dynamically changing the address width for comparison. An incoming load/store operation must be compared to the operations in the pipeline and the queues to avoid address conflicts. Overall, the present invention introduces a cache hit or cache miss input into the load/store dependency logic. If the incoming load operation is a cache hit, then the quadword boundary address value is used for detection. If the incoming load operation is a cache miss, then the cacheline boundary address value is used for detection. This invention enhances the performance of LHS and LHR operations in a memory system.

Growing community of inventors

Pflugerville, TX, United States of America

Dwain Alan Hicks