Analog Devices,inc.

Samsung Electronics Co., Ltd.

Other

Intel Corporation

Moinul Syed

Sung-Boem Park

Ju-Hee Choi

David B Witt

Michael J Allen

Lawrence A Booth

William C Anderson

Ravi Kolagotla

Michael S Allen

Gregory T Koker

Hebbalalu S Ramagopal

Richard A Gentile

Michael Allen

Memory management unit (MMU) in an application processor responds to an access request, corresponding to inspection request, including target context and target virtual address and the inspection request is for translating the target virtual address to a first target physical address. The MMU includes context cache, translation cache, invalidation queue and address translation manager (ATM). The context cache stores contexts and context identifiers of the stored contexts, while avoiding duplicating contexts. The translation cache stores first address and first context identifiers second addresses, the first address corresponds to virtual address, the first context identifiers corresponds to first context, and the second addresses corresponds to the first address and the first context. The invalidation queue stores at least one context identifier to be invalidated, of the context identifiers stored in the translation cache. The ATM controls the context cache, the translation cache and the invalidation queue.

Application processor, system-on chip and method of operating memory management unit

A processor includes a page table walk cache that stores address translation information, and a page table walker. The page table walker fetches first output addresses indicated by first indexes of a first input address by looking up the address translation information and at least a part of page tables, and compares a matching level between second indexes of a second input address and the first indexes of the first input address with a walk cache hit level obtained by looking up the page table walk cache using the second indexes.

Processor to detect redundancy of page table walk

A processing system includes a processing circuit having one or more buses, a memory interface unit to control access by the processing circuit to a memory, and a metrics module. The metrics module includes one or more metrics registers and a metrics controller to monitor one or more operations selected from memory interface unit operations and bus operations, and to store metrics information corresponding to the monitored operations in the metrics registers. The monitored operations can include memory access operations, arbitration operations, bus operations, and the like. The metrics information can be analyzed to provide a basis for improving performance of a program that is executed on the processing system.

Metrics modules and methods for monitoring, analyzing and optimizing bus and memory operations in a complex integrated circuit

In parallel with accesses to a cache made by a core processor, a DMA controller is used to pre-load data from a main memory into the cache. In this manner, the pre-load function can make the data available to the processor application before the application references the data, thereby potentially providing a 100% cache hit ratio since the correct data is pre-loaded into the cache. In addition, if a copy-back cache is employed, the cache memory system can also be configured such that processed data can be dynamically unloaded from the cache to the main memory in parallel with accesses to the cache made by the core processor. The pre-loading and/or post unloading of data may be accomplished, for example, by using a DMA controller to burst data into and out of the cache in parallel with accesses to the cache by the core processor. This DMA control function may be integrated into the existing cache control logic so as to reduce the complexity of the cache hardware (e.g., as compared to a multi-port cache), and to alleviate the difficulty associated with addressing the non-contiguous internal address map of the cache. By employing a DMA controller having flexible address generation and transfer control capabilities, data can be transferred from an atypical memory-mapped entity (e.g., a FIFO buffer of a peripheral) directly to the cache.

Cache system with DMA capabilities and method for operating same

Methods and apparatus are provided for clock domain conversion in digital processing systems. The methods include operating a first circuit in a fast clock domain with a fast clock and operating a second circuit in a slow clock domain with a slow clock. To transfer signals from the fast clock domain to the slow clock domain, a first synchronization signal is asserted during each fast clock cycle in which a slow clock edge occurs. A fast signal is transferred from the fast clock domain to the slow clock domain on a fast clock edge when the first synchronization signal is asserted. To transfer signals from the slow clock domain to the fast clock domain, a second synchronization signal is asserted during each fast clock cycle that immediately follows a slow clock edge. A slow signal is transferred from the slow clock domain to the fast clock domain on a fast clock edge when the second synchronization signal is asserted.

Methods and apparatus for clock domain conversion in digital processing systems

An apparatus having a core processor and a memory system is disclosed. The core processor includes at least one data port. The memory system is connected in such a way as to provide substantially simultaneous data accesses through the data port. The memory system can be made user configurable to provide appropriate memory model.

Growing community of inventors

Austin, TX, United States of America

Moinul Syed