Imagination Technologies Limited

Mips Tech Limited

Jason Meredith

Hugh Jackson

Methods and systems for improved control of traffic generated by a processor are described. In an embodiment, when a device generates a pre-fetch request for a piece of data or an instruction from a memory hierarchy, the device includes a pre-fetch identifier in the request. This identifier flags the request as a pre-fetch request rather than a non-pre-fetch request, such as a time-critical request. Based on this identifier, the memory hierarchy can then issue an abort response at times of high traffic which suppresses the pre-fetch traffic, as the pre-fetch traffic is not fulfilled by the memory hierarchy. On receipt of an abort response, the device deletes at least a part of any record of the pre-fetch request and if the data/instruction is later required, a new request is issued at a higher priority than the original pre-fetch request.

Control of pre-fetch traffic

An improved mechanism for copying data in memory is described which uses aliasing. In an embodiment, data is accessed from a first location in a memory and stored in a cache line associated with a second, different location in the memory. In response to a subsequent request for data from the second location in the memory, the cache returns the data stored in the cache line associated with the second location in the memory. The method may be implemented using additional hardware logic in the cache which is arranged to receive an aliasing request from a processor which identifies both the first and second locations in memory and triggers the accessing of data from the first location for storing in a cache line associated with the second location.

Mechanism for copying data in memory

Fill partitioning of a shared cache is described. In an embodiment, all threads running in a processor are able to access any data stored in the shared cache; however, in the event of a cache miss, a thread may be restricted such that it can only store data in a portion of the shared cache. The restrictions to storing data may be implemented for all cache miss events or for only a subset of those events. For example, the restrictions may be implemented only when the shared cache is full and/or only for particular threads. The restrictions may also be applied dynamically, for example, based on conditions associated with the cache. Different portions may be defined for different threads (e.g. in a multi-threaded processor) and these different portions may, for example, be separate and non-overlapping. Fill partitioning may be applied to any on-chip cache, for example, a L1 cache.

Fill partitioning of a shared cache

Aspects relate to methods and systems for processing requests and sending data in a bus architecture. At least one master device is connected to at least two slave devices via a bus. An allocator allocates incoming requests from the master device to a target slave device. Incoming requests are buffered for the respective slave device. The master device sends a read request for a first slave device to the bus; the allocator generates a current-state indicator associated with the read request. The allocator generates a priority indicator associated with the read request. If the initial value of the current-state indicator equals the value of the priority indicator, the read request is processed; or if the initial value of the current-state indicator does not equal the value of priority indicator, the read request is deferred until a later time.

Requests and data handling in a bus architecture

The present invention relates to improved methods for processing requests and sending data in a bus architecture. The present invention further relates to an improved bus architecture for processing requests and data. There is provided a method for processing read requests in a bus architecture comprising at least one master device connected to at least two slave devices via a bus. The architecture comprises an allocator for allocating incoming requests from the master device to a target slave device and an optimiser for each slave device. Each optimiser is for buffering incoming requests for the respective slave device. The method comprising the steps of: a) the master device sending a read request for a first slave device to the bus; b) the allocator generating a current-state indicator associated with the read request. The current-state indicator has an initial value, The method further comprises c) the allocator generating a priority indicator associated with the read request; d) the allocator sending the read request, the current-state indicator and the priority indicator to the optimiser of the first slave device; e) the optimiser of the first slave device receiving the read request, the current-state indicator and the priority indicator. Finally, if the initial value of the current-state indicator equals the value of the priority indicator, the method comprises processing the read request; or if the initial value of the current-state indicator does not equal the value of priority indicator, the method comprises deferring processing of the read request until a later time.

Growing community of inventors

Hemel Hempstead, United Kingdom

Jason Meredith