Abstract:
Systems and method are directed to reducing power consumption of data transfer between a processor and a memory. A data to be transferred on a data bus between the processor and the memory is checked for a first data pattern, and if the first data pattern is present, transfer of the first data pattern is suppressed on the data bus. Instead, a first address corresponding to the first data pattern is transferred on a second bus between the processor and the memory. The first address is smaller than the first data pattern. The processor comprises a processor-side first-in-first-out (FIFO) and the memory comprises a memory-side FIFO, wherein the first data pattern is present at the first address in the processor-side FIFO and at the first address in the memory-side FIFO.
Abstract:
Power saving techniques for memory systems are disclosed. In particular, exemplary aspects of the present disclosure contemplate taking advantage of patterns that may exist within memory elements and eliminating duplicative data transfers. Specifically, if data is repetitive, instead of sending the same data repeatedly, the data may be sent only a single time with instructions that cause the data to be replicated at a receiving end to restore the data to its original repeated state. By reducing the amount of data that is transferred between a host and a memory element, power consumption is reduced.
Abstract:
A memory device may include link error correction code (ECC) decoder and correction circuitry. The ECC decoder and correction circuitry may be arranged in a write path and configured for link error detection and correction of write data received over a data link. The memory device may also include memory ECC encoder circuitry. The memory ECC encoder circuitry may be arranged in the write path and configured for memory protection of the write data during storage in a memory array.
Abstract:
A memory sub-system may include a memory controller having error correction code (ECC) encoder/decoder logic. The memory controller may be configured to embed link ECC parity bits in unused data mask bits and/or in a mask write data during a mask write operation. The memory controller may also be configured to protect at least a location of the link ECC parity bits during the mask write operation.
Abstract:
A method of refreshing a dynamic random access memory (DRAM) includes detecting an open page of the DRAM at a row of a DRAM bank within an open sub-array of the DRAM bank. The method also includes delaying issuance of a refresh command to a target refresh row of the DRAM bank when the target refresh row of the DRAM bank is within the open sub-array of the DRAM bank.
Abstract:
Heterogeneous memory systems, and related methods and computer-readable media for supporting heterogeneous memory access requests in processor-based systems are disclosed. A heterogeneous memory system is comprised of a plurality of homogeneous memories that can be accessed for a given memory access request. Each homogeneous memory has particular power and performance characteristics. In this regard, a memory access request can be advantageously routed to one of the homogeneous memories in the heterogeneous memory system based on the memory access request, and power and/or performance considerations. The heterogeneous memory access request policies may be predefined or determined dynamically based on key operational parameters, such as read/write type, frequency of page hits, and memory traffic, as non-limiting examples. In this manner, memory access request times can be optimized to be reduced without the need to make tradeoffs associated with only having one memory type available for storage.
Abstract:
A method and apparatus for providing a clock signal to a charge pump is disclosed. In a particular embodiment, the method includes providing a first clock signal to a first charge pump unit of a charge pump. The method further includes providing a second clock signal to a second charge pump unit of the charge pump. A low-to-high transition of the first clock signal occurs substantially concurrently with a high-to-low transition of the second clock signal. Only one clock signal may be at a logic high voltage level at any given time.
Abstract:
Methods and devices of a capacitor in a semiconductor device having an increased capacitance are disclosed. In a particular embodiment, a method of forming a capacitor is disclosed. A section of a first insulating material between a first metal contact element and a second metal contact element is removed to form a channel. A second insulating material is deposited in the channel between the first metal contact element and the second metal contact element.
Abstract:
Disclosed are techniques for minimizing performance degradation due to refresh operations in a dynamic volatile memory sub-system. In an aspect, a refresh scheduler coupled to the dynamic volatile memory sub-system generates a batch memory refresh command comprising an identification of a plurality of rows of each of one or more banks of the dynamic volatile memory sub-system to refresh, and issues the batch memory refresh command to the dynamic volatile memory sub-system.
Abstract:
Priority adjustment of dynamic random access memory (DRAM) transactions prior to issuing a per-bank refresh for reducing DRAM unavailability is disclosed. In one aspect, DRAM is refreshed on a per-bank basis. If a queued memory transaction corresponds to a memory bank that will soon be refreshed, the transaction may be delayed if a refresh of the corresponding memory bank begins prior to execution of the transaction. To avoid delaying execution of the transaction while waiting for the corresponding memory bank to be refreshed, a priority of the memory transactions may be adjusted based on a memory bank refresh schedule. The priority of the transaction corresponding to the memory bank to be refreshed may be increased, and the priority of other memory transactions may be decreased, if such an adjustment would avoid or reduce delaying execution due to unavailability of the corresponding memory bank.