Apple Inc.

Nvidia Corporation

Shane J Keil

Erik P Machnicki

Gregory S Mathews

Brijesh Tripathi

Gilbert H Herbeck

Lakshmi Narasimha Murthy Nukala

Timothy John Millet

Manu Gulati

Sukalpa Biswas

Joshua P De Cesare

Jung Wook Cho

Anand Dalal

John H Edmondson

Kai Lun Hsiung

Yanzhe Liu

Thejasvi Magudilu Vijayaraj

John Stephen Montrym

Gurjeet S Saund

Josh P De Cesare

Rakesh L Notani

Munetoshi Fukami

Fred Gruner

Peter Fu

Michael J Smith

Jason M Kassoff

Derek Fujio Iwamoto

Kevin C Wong

Thejasvi Magudilu Vijavaraj

Chaitanya Kosaraju

Erdem Guleyupoglu

Michael F Culbert

Jonah Matthew Alben

Sean Jeffrey Treichler

Shu-Yi Yu

Kleanthes G Koniaris

Tao Zhang

Dennis K Ma

Ashish Karandikar

Raman S Thiara

David S Warren

Peter Tong

Sonny S Yeoh

Kaustubh Sanghani

Kai Lun Hsuing

Lakshmi Narasimha Nukala

In an embodiment, a system on a chip (SOC) includes a component that remains powered when the remainder of the SOC is powered off. The component may include a sensor capture unit to capture data from various device sensors, and may filter the captured sensor data. Responsive to the filtering, the component may wake up the remainder of the SOC to permit the processing. The component may store programmable configuration data, matching the state at the time the SOC was most recently powered down, for the other components of the SOC, in order to reprogram them after wakeup. In some embodiments, the component may be configured to wake up the memory controller within the SOC and the path to the memory controller, in order to write the data to memory. The remainder of the SOC may remain powered down.

System on a chip with always-on processor

A memory control circuit coupled to multiple memory ranks may receive read and write requests for a different ranks of the multiple memory ranks. The memory control may allocate write requests to different slots based on the write requests target memory rank, and may adjust the number of slots available for a given memory rank during a write turn to improve write efficiency. The memory control circuit may also determine a number of ranks switches within a read turn based on whether a particular quality-of-service requirement associated with the read requests is being satisfied.

Write arbiter circuit with per-rank allocation override mode

Techniques for scheduling memory operations are disclosed in which alternate read/write commands within a multi-bank memory operation are delayed beyond a minimum timing parameter in order to increase memory data bus utilization. The remaining read/write commands are not delayed beyond the minimum timing parameter. Every other clock cycle (e.g., even clock cycles) within the memory operation is reserved for activate commands, while other commands such as sync and read/write are scheduled on the intervening clock cycles (e.g., odd clock cycles). For memory devices for which a sync command (which causes a clock of the memory data bus to start) is to precede a corresponding read/write command by a number of clock cycles that would place it in a cycle reserved for activate commands, a particular operation mode is disclosed in which the memory device internally delays a received sync command.

Memory device bandwidth optimization

Read arbiter circuit with dual memory rank support

Systems, apparatuses, and methods for addressing bank hotspotting are described. A computing system includes a memory controller with an arbiter for determining how to arbitrate access to one or more memory device(s) for received requests. The arbiter categorizes each request in a manner that helps to ensure fair virtual channel distribution across the banks of the memory device(s). The category system includes bank hotspotting functions to give banks that have more requests more chances to go over banks with fewer requests. The category system is implemented proportionally with more category credits given to banks with higher bank depths within the virtual channel.

Memory bank hotspotting

In an embodiment, a system on a chip (SOC) includes a component that remains powered when a central processing unit (CPU) processor and a memory controller of the SOC are powered off. The component may include a sensor capture unit to capture audio samples from an audio detector circuit and write them to a memory of the component. A processor of the component may be configured to search the audio samples for a predetermined pattern during a time when the CPU processor and the memory controller are powered down. In some embodiments, based on the audio samples filling to a threshold level in the memory of the component and a lack of detection of the predetermined pattern, the component is configured to wake up the memory controller and a path to the memory controller in order to write the audio sample to a memory controlled by the memory controller.

System on a chip with always-on component with powered-down configurations to process audio samples

Techniques are disclosed relating to multi-activation techniques for wire operations with multiple partial writes. In some embodiments, a memory controller is configured to access data in a memory device that supports partial writes having a first size using read-modify-write operations and non-partial writes having a second size that is greater than the first size. In some embodiments, the memory controller is configured to queue a first write operation having the second size, where the first write operation includes multiple partial writes. In some embodiments, the memory controller is configured to send separate activate signals to the memory device to activate a bank of the memory device to perform different proper subsets of the multiple partial writes. This may allow interleaving of other accesses to a memory bank and merging of writes while waiting for a current activation, in some embodiments.

Multi-activation techniques for partial write operations

An apparatus includes a memory circuit and a memory controller circuit. The memory controller circuit may include a write request queue. The memory controller circuit may be configured to receive a memory request to access the memory circuit and determine if the memory request includes a read request or a write request. A received read request may be scheduled for execution, while a received write request may be stored in the write request queue. The memory controller circuit may reorder scheduled memory requests based on achieving a specified memory access efficiency and based on a number of write requests stored in the write request queue.

Ordering memory requests based on access efficiency

Techniques relating to arbitration in a memory controller are disclosed. In some embodiments, the memory controller is configured to transition between read turns and writes turn according to a turn schedule. In some embodiments, the memory controller also receives reports from circuitry requesting memory transactions and determines a current latency tolerance value based on the reports. In some embodiments, the memory controller is configured to switch from a write turn to a read turn prior to a scheduled switch based on the current latency tolerance meeting a threshold value.

Write/read turn techniques based on latency tolerance

A memory system includes a memory circuit including a plurality of pages, including a particular page having a page activation time. The memory system also includes a memory controller circuit configured to receive a memory access request corresponding to data of the particular page. The memory controller circuit is also configured to transmit, in response to a determination that the particular page is inactive, an activation command to the memory circuit to activate the particular page, and to schedule a future transmission of an initial memory command for the particular page based on the page activation time.

Scheduling of read and write memory access requests

A memory controller circuit coupled to a memory circuit that includes multiple banks may receive multiple access requests including a particular access request to a particular bank of the plurality of banks. The particular access request is associated with a particular virtual channel of a plurality of virtual channels. The memory controller circuit may select a given access requests of the multiple access requests based on an arbitration category value associated with a virtual channel of the given access request and modify the arbitration category value in response to selecting the given access request.

Memory access scheduling using category arbitration

A memory controller circuit coupled to multiple memory circuits may receive a read request for a particular one of the memory circuits and insert the read request into one of multiple linked lists that includes a linked list whose entries correspond to previously received read requests and are linked according to respective ages of the read requests. The memory controller circuit may schedule the read request using a head pointer of one of the multiple linked lists.

Memory access scheduling using a linked list

A method and apparatus for performing opportunistic refreshes of memory banks is disclosed. Refresh circuitry in a memory controller performs a refresh on each bank of a multi-bank memory at least once during a given refresh interval. At the beginning of an interval, memory banks for which there are no pending transactions (e.g., reads or writes) may be refreshed. During a first portion of the interval, refresh may be skipped for memory banks for which transactions are pending. In a second portion of the interval, refreshes are performed on memory banks that have not been refreshed during the interval, which may cause some memory transactions to be delayed.

System and method for performing per-bank memory refresh

Growing community of inventors

San Jose, CA, United States of America

Shane J Keil