摘要:
A method, apparatus, and system are described. In one embodiment, the method comprises a chipset receiving a plurality of memory requests, wherein each memory request comprises one or more micro-commands that each require one or more memory clock cycles to execute, and scheduling the execution of each of the micro-commands from more than one of the plurality of memory requests in an order to reduce the number of total memory clock cycles required to complete execution of the more than one memory requests.
摘要:
A method, apparatus, and system are disclosed. In one embodiment, the method receiving data from a memory on a first interconnect of at least one interconnect, receiving a source-synchronous data strobe from the memory, creating at least a nominal, an early, and a delayed compensated data strobe from the received data strobe, latching the received data with the nominal, early, or delayed compensated data strobe, outputting the latched data onto one or more of the at least one interconnect.
摘要:
An address translation apparatus and method that can convert a limited-range memory address from a peripheral device to an expanded-range memory address on the fly. The invention can expand the limited address capability of a peripheral bus, such as a PCI bus with a 4 GB address range, to a much larger address capability, such as a 64 GB address range. This conversion can be performed on the fly by hardware, so that no appreciable delay in transfer time is created. The conversion can be performed by adding features to a conventional graphics controller interface, thus minimizing the impact on circuit complexity and system cost.
摘要:
Embodiments of the present invention relate to accessing a first pair of adjacent data blocks using a first channel of a dual channel memory device; and simultaneously accessing a second pair of adjacent data blocks using a second channel of the memory device, the second pair being spaced apart from the first pair by a predetermined interval.
摘要:
The invention comprises an apparatus and method of prefetching from a memory device having interleaved channels. The chipset prefetcher comprises a stride detector to detect a stride in a stream, a prefetch injector to insert prefetches onto the memory device, a channel mapper to map the prefetches to each channel of the memory device, a scheduler to schedule the prefetches onto the memory device in a DRAM-state aware manner, a throttling heuristic to scale the number of prefetches, and a prefetch data buffer to store prefetch data. The method of prefetching comprises tracking the state of streams, detecting a stride on one of the streams, selecting the stream with the stride for prefetch injection, enqueueing prefetches from the selected stream, mapping the prefetches to each of the interleaved channels, injecting the prefetches from the selected stream into each of the interleaved channels, and scheduling the prefetches onto the memory device in a DRAM-state aware manner.
摘要:
An apparatus and method for prefetching cache data in response to data requests. The prefetching uses the memory addresses of requested data to search for other data, from a related address, in a cache. This, or other data, may then be prefetched based on the result of the search.
摘要:
An integrated circuit designed to be coupled to a suspendable memory, the integrated circuit comprising a memory enable deassertion delay (MEDD) logic setting a wait period for the deassertion of a memory enable signal after completion of a memory operation. The wait period is chosen for a preferred latency versus power savings tradeoff.
摘要:
A method and apparatus for splitting a cache operation into multiple phases and multiple clock domains are disclosed. The method according to the present techniques comprises splitting a cache operation into two or more phases and two or more clock domains.
摘要:
A method and apparatus for managing memory access requests have been disclosed. One embodiment of the method includes dynamically modifying attributes of each of a number of requests to access one or more memory devices and arbitrating among the requests to select a request to send to the memory devices in a time slot based on the attributes. Other embodiments are described and claimed.