摘要:
A high speed DRAM cache architecture. One disclosed embodiment includes a multiplexed bus interface to interface with a multiplexed bus. A cache control circuit drives a row address portion of an address on the multiplexed bus interface and a command to open a memory page containing data for a plurality of ways. The cache control circuit subsequently drives a column address including at least a way indicator to the multiplexed bus interface.
摘要:
A high speed DRAM cache architecture. One disclosed embodiment includes a multiplexed bus interface to interface with a multiplexed bus. A cache control circuit drives a row address portion of an address on the multiplexed bus interface and a command to open a memory page containing data for a plurality of ways. The cache control circuit subsequently drives a column address including at least a way indicator to the multiplexed bus interface.
摘要:
Methods and apparatuses for mapping cache contents to memory arrays. In one embodiment, an apparatus includes a processor portion and a cache controller that maps the cache ways to memory banks. In one embodiment, each bank includes data from one cache way. In another embodiment, each bank includes data from each way. In another embodiment, memory array banks contain data corresponding to sequential cache lines.
摘要:
In some embodiments, a chip includes at least four groups of memory banks and at least four groups of output conductors wherein each group of output conductors corresponds to a different one of the groups of memory banks. The chip also includes circuitry to perform a read operation by providing read data from at least one of the banks of each of the groups of memory banks to its corresponding group of output conductors. Other embodiments are described.
摘要:
Embodiments of the invention are generally directed to systems, methods, and apparatuses to save dynamic random access memory (DRAM) self-refresh power. In some embodiments, the refresh frequency of a DRAM is reduced and errors are allowed to occur. In error check mode, the DRAM stores data and corresponding error check bits. The error check bits may be used to scrub the memory and fix the errors.
摘要:
Embodiments of the invention are generally directed to systems, methods, and apparatuses for a common memory device for variable device width and scalable pre-fetch and page size. In some embodiments, a common memory device (such as a DRAM) can operate in any of a number of modes including, for example, a x4 mode, a x8 mode, and a x16 mode. The page size provided by the DRAM may vary depending on the mode of the DRAM. In some embodiments, the amount of data pre-fetched by the DRAM also varies depending on the mode of the DRAM.
摘要:
Embodiments of the invention are generally directed to systems, methods, and apparatuses for a common memory device for variable device width and scalable pre-fetch and page size. In some embodiments, a common memory device (such as a DRAM) can operate in any of a number of modes including, for example, a ×4 mode, a ×8 mode, and a ×16 mode. The page size provided by the DRAM may vary depending on the mode of the DRAM. In some embodiments, the amount of data pre-fetched by the DRAM also varies depending on the mode of the DRAM.
摘要:
Embodiments of the invention are generally directed to systems, methods, and apparatuses for a common memory device for variable device width and scalable pre-fetch and page size. In some embodiments, a common memory device (such as a DRAM) can operate in any of a number of modes including, for example, a ×4 mode, a ×8 mode, and a ×16 mode. The page size provided by the DRAM may vary depending on the mode of the DRAM. In some embodiments, the amount of data pre-fetched by the DRAM also varies depending on the mode of the DRAM.
摘要:
Embodiments of the invention are generally directed to systems, methods, and apparatuses to save dynamic random access memory (DRAM) self-refresh power. In some embodiments, the refresh frequency of a DRAM is reduced and errors are allowed to occur. In error check mode, the DRAM stores data and corresponding error check bits. The error check bits may be used to scrub the memory and fix the errors.