Intel Corporation

Nadav Bonen

Kuljit Singh Bains

Alexey Kostinsky

Zeev Sperber

Naveen Kumar

John B Halbert

Ron Gabor

Shantanu R Gupta

Zvika Greenfield

Tomer Levy

Vjekoslav Svilan

David N Mackintosh

Jose A Baiocchi Paredes

Siddhartha Chhabra

Shay Gueron

Christopher E Cox

Barnes Cooper

Randy B Osborne

Sridhar Muthrasanallur

Chia-Hung Sophia Kuo

Anant V Nori

Julius Yuli Mandelblat

Jaya L Jeyaseelan

Toby Opferman

Todd M Witter

Israel Diamand

Eran Shifer

Vijay Bahirji

Virendra Vikramsinh Adsure

Vishwanath Somayaji

Srinivas Pandruvada

Nir Sucher

Prashant Kodali

Nivedha Krishnakumar

James Mccall

A mechanism where the locked pages are saved and restored by a hardware accelerator which is transparent to the OS. Prior to standby entry, the OS puts all DMA capable devices in the lowest-powered device low-power state after disabling bus mastering. The OS flushes all pageable memory to an NVM (in segments that are kept in self-refresh) and provides a list of pinned and locked pages in the DRAM to a power management controller (p-unit). The p-unit checks for all Bus Mastering DMA to be turned off and checks if a next OS timer wake event (TNTE) is greater than a threshold, to decide whether to enable or disable PASR or MPSM in Standby. If the conditions are met, the p-unit triggers a hardware accelerator to consolidate the pinned and locked pages in the DRAM to certain segment(s) of the DRAM during standby states, making it transparent to the OS.

Memory power management method and apparatus

System and techniques for multi-tenant cryptographic memory isolation are described herein. A multiple key total memory encryption (MKTME) circuitry may receive a read request for encrypted memory. Here, the read request may include an encrypted memory address that itself includes a sequence of keyid bits and physical address bits. The MKTME circuitry may retrieve a keyid-nonce from a key table using the keyid bits. The MKTME circuitry may construct a tweak from the keyid-nonce, the keyid bits, and the physical address bits. The MKTME circuitry may then decrypt data specified by the read request using the tweak and a common key.

Multi-tenant cryptographic memory isolation

Logical memory is divided into two regions. Data in the first region is always retained. The first region of memory is designated online (or powered on) and is not offlined during standby or low power mode. The second region is the rest of the memory which can be potentially placed in non-self-refresh mode during standby by offlining the memory region. Content in the second region can be moved to the first region or can be flushed to another memory managed by the operating system. When the first region does not have enough space to accommodate data from the second region, the operating system can increase the logical size of the first region. Retaining the content of the first region by putting that region in self-refresh and saving power in the second region by not putting it in self-refresh is performed by an improved Partial Array Self Refresh scheme.

Memory management to improve power performance

One embodiment provides an apparatus. The apparatus includes first memory controller circuitry to control read and/or write access to first memory circuitry via a first conductive bus. The apparatus includes second memory controller circuitry to control read and/or write access to second memory circuitry via a second conductive bus. The apparatus includes power control circuitry coupled to the first memory controller circuitry and the second memory controller circuitry. The power control circuitry transfers data from the second memory circuitry with the second memory controller circuitry via the second conductive bus to the first memory circuitry with the first memory controller circuitry via the first conductive bus. The power control circuitry powers down the second memory circuitry after the transfer of the data from the second memory circuitry to the first memory circuitry. The power control circuitry decreases power consumption of the apparatus and may increase batter life of the apparatus.

Architecture for dynamic transformation of memory configuration

In an embodiment, the present invention includes an execution unit to execute instructions of a first type, a local power gate circuit coupled to the execution unit to power gate the execution unit while a second execution unit is to execute instructions of a second type, and a controller coupled to the local power gate circuit to cause it to power gate the execution unit when an instruction stream does not include the first type of instructions. Other embodiments are described and claimed.

Performing local power gating in a processor

On-die termination (ODT) control enables programmable ODT latency settings. A memory device can couple to an associated memory controller via one or more buses shared by multiple memory devices organized ranks of memory. The memory controller generates a memory access command for a target rank. In response to the command, memory devices can selectively engage ODT for the memory access operation based on being in the target rank or a non-target rank, and based on whether the access command includes a Read or a Write. The memory device can engage ODT in accordance with a programmable ODT latency setting. The programmable ODT latency setting can set different ODT timing values for Read and Write transactions.

Programmable on-die termination timing in a multi-rank system

Described herein are embodiments of asymmetric memory management to enable high bandwidth accesses. In embodiments, a high bandwidth cache or high bandwidth region can be synthesized using the bandwidth capabilities of more than one memory source. In one embodiment, memory management circuitry includes input/output (I/O) circuitry coupled with a first memory and a second memory. The I/O circuitry is to receive memory access requests. The memory management circuitry also includes logic to determine if the memory access requests are for data in a first region of system memory or a second region of system memory, and in response to a determination that one of the memory access requests is to the first region and a second of the memory access requests is to the second region, access data in the first region from the cache of the first memory and concurrently access data in the second region from the second memory.

Concurrent accesses of asymmetrical memory sources

Memory subsystem refresh management enables commands to access one or more identified banks across different bank groups with a single command. Instead of sending commands identifying a bank or banks in separate bank groups by each bank group individually, the command can cause the memory device to access banks in different bank groups. The command can be a refresh command. The command can be a precharge command.

Precharging and refreshing banks in memory device with bank group architecture

Apparatus, systems, and methods to implement polarity based data transfer function for volatile memory power reduction are described. The transfer function take into account certain data values, all zeroes in particular, that are common and transforms them to predetermined values that consume less power and are less common. Similarly, these predetermined values are transformed to the common values.

Polarity based data transfer function for volatile memory

A memory subsystem includes a multi-device package including multiple memory devices organized as multiple ranks of memory. A control unit for the memory subsystem sends a memory access command concurrently to some or all of the ranks of memory, and triggers some of all of the memory ranks that receive the memory access command to change on-die termination (ODT) settings. One of the ranks is selected to execute the memory access command, and executes the command while all ranks triggered to change the ODT setting have the changed ODT setting.

On-die termination control without a dedicated pin in a multi-rank system

Growing community of inventors

Ofer, Israel

Nadav Bonen