Advanced Micro Devices Corporation

Applied Micro Circuits Corporation

Dell USA L P

Freescale Semiconductor,inc.

Other

Dell Products Inc.

Mercury Computer Systems, Inc.

Mercury Computer Systemc, Inc.

Daniel L Bouvier

Bryan D Marietta

Kenneth L Jeffries

Robert C Frisch

Joseph D Macri

Larry David Hewitt

Michael D Snyder

Gary L Whisenhunt

Thomas James Gibney

Satish Sathe

Russell Conley Smith

Gilberto Hernandez

Wai-Ming Richard Chan

Joseph A Vivio

Mahesh Subramony

Joshua Titus

Sonu Arora

A method of operating a computing system includes storing a memory map identifying a first physical memory address as associated with a high performance memory and identifying a second physical memory address as associated with a low power consumption memory, servicing a first memory access request received from an application by accessing application data at the first physical memory address, in response to a change in one or more operating conditions of the computing system, moving the application data between the first physical memory address and the second physical memory address based on the memory map, and servicing a second memory access request received from the application by accessing the application data at the second physical memory address.

Active hibernate and managed memory cooling in a non-uniform memory access system

Systems, apparatuses, and methods for secure enablement of platform features without user intervention are disclosed. In one embodiment, a system includes at least a motherboard and a processor. The motherboard includes at least a socket and an authentication component. The authentication component can be a chipset, expansion I/O device, or other component. The processor is installed in the socket on the motherboard. During a boot sequence, the processor retrieves a key value from the authentication component and then authenticates the key value. Next, the processor determines which one or more features to enable based on the key value. Then, the processor programs one or more feature control registers to enable the one or more features specified by the key value. Accordingly, during normal operation of the system, the one or more features will be enabled.

Secure enablement of platform features without user intervention

The described embodiments include an apparatus that controls voltages for an integrated circuit chip having a set of circuits. The apparatus includes a switching voltage regulator separate from the integrated circuit chip and two or more low dropout (LDO) regulators fabricated on the integrated circuit chip. During operation, the switching voltage regulator provides an output voltage that is received as an input voltage by each of the two or more LDO regulators, and each of the two or more LDO regulators provides a local output voltage, each local output voltage received as a local input voltage by a different subset of circuits in the set of circuits.

Voltage regulators for an integrated circuit chip

A computer system includes a first processor, a second processor, and a common memory connected to the second processor. The computer system is switched from a high performance mode, in which at least a portion of the first processor and at least a portion of components on the second processor are active, to a low power mode, in which at least a portion of the first processor is active and the components on the second processor are inactive. All central processing unit (CPU) cores on the second processor are quiesced. Traffic from the second processor to the common memory is quiesced. Paths used by the first processor to access the common memory are switched from a first path across the second processor to a second path across the second processor.

Switching a computer system from a high performance mode to a low power mode

A single interconnect is provided between a first processor and a second processor, such that the first processor may access a common memory through the second processor while the second processor can be mostly powered off. The first processor accesses the memory through a memory controller using a standard dynamic random access memory (DRAM) bus protocol. Instead of the memory controller directly connecting to the memory, the access path is through the second processor to the memory. Additionally, a bidirectional communication protocol bus is mapped to the existing DRAM bus signals. When both the first processor and the second processor are active, the bus protocol between the processors switches from the DRAM protocol to the bidirectional communication protocol. This enables the necessary chip-to-chip transaction semantics without requiring the additional cost burden of a dedicated interface for the bidirectional communication protocol.

Polymorphous signal interface between processing units

A system and method are provided for using queue status to manage power in a system-on-chip (SoC). Messages to be processed are accepted in an SoC with a plurality of selectively enabled processors, and queued. The message traffic can be from an external source via an input/output (IO) interface, or intra-SoC messages between processors. The number of queued messages is monitored and, in response to the number of queued messages exceeding a subscription threshold, one or more processors are enabled. Then, the queued messages are distributed to the enabled processors. Enabling a processor is defined by an action such as supplying power to an unpowered processor, increasing the power supply voltage levels to a processor, increasing the operating frequency of a processor, or a combination of the above-mentioned actions. Likewise, processors can be disabled in response to the number of queued messages falling below the subscription threshold.

System-on-chip queue status power management

A multi-processor computer system is provided for managing physical memory domains. The system includes at least one processor having an address interface for sending a memory access message, which includes an address in physical memory and a domain identification (ID). The system also includes a physical memory portioned into a plurality of domains, where each domain includes a plurality of physical addresses. A domain mapping unit (DMU) has an interface to accept the memory access message from the processor. The DMU uses the domain ID to access a permission list, cross-reference the domain ID to a domain including addresses in physical memory, and grant the processor access to the address in response to the address being located in the domain.

Using domains for physical address management in a multiprocessor system

A system and method are provided for managing cache memory in a computer system. A cache controller portions a cache memory into a plurality of partitions, where each partition includes a plurality of physical cache addresses. Then, the method accepts a memory access message from the processor. The memory access message includes an address in physical memory and a domain identification (ID). A determination is made if the address in physical memory is cacheable. If cacheable, the domain ID is cross-referenced to a cache partition identified by partition bits. An index is derived from the physical memory address, and a partition index is created by combining the partition bits with the index. A processor is granted access (read or write) to an address in cache defined by partition index.

Multi-domain management of a cache in a processor system

A channel-less system and method are provided for multithreaded communications with a direct memory access (DMA) controller. The method accepts a plurality of DMA command messages directed to a fixed port address. The DMA command messages are arranged in a first-in first-out (FIFO) queue, in the order in which they are received. The DMA command messages are supplied to a DMA controller from the FIFO queue, and in response to the DMA command message, data transfer operation are managed by the DMA controller. Following the completion of each data transfer operation, a transfer complete message indicating completion is sent. In one aspect, DMA command messages are arranged in a plurality of parallel FIFO queues, and CD sets are stored in a plurality of context memories, where each context memory is associated with a corresponding FIFO queue.

Multithreaded DMA controller

A data processing system has an interrupt controller which provides an interrupt request along with a corresponding interrupt identifier and a corresponding interrupt vector to a processor. If the processor accepts the interrupt, the processor returns the same interrupt identifier value by way of interrupt identifier, along with interrupt acknowledge, to the interrupt controller. An interrupt taken/not taken indicator may also be provided. The communications interface used to coordinate interrupt processing between the interrupt controller and the processor may be asynchronous.

Interrupt controller for accelerated interrupt handling in a data processing system and method thereof

Channel-less multithreaded DMA controller

A system and method have been provided for pushing cacheable control messages to a processor. The method accepts a first control message, identified as cacheable and addressed to a processor, from a peripheral device. The first control message is allocated into a cache that is associated with the processor, but not associated with the peripheral device. In response to a read-prompt the processor reads the first control message directly from the cache. The read-prompt can be a hardware interrupt generated by the peripheral device referencing the first control message. For example, the peripheral may determine that the first control message has been allocated into the cache and generate a hardware interrupt associated with the first control message. Then, the processor reads the first control message in response to the hardware interrupt read-prompt. Alternately, the read-prompt can be the processor polling the cache for pending control messages.

Cache stashing processor control messages

A digital data system employs multiple error protection mechanisms on messages that pass along a link interconnect fabric from one node or device to another node or device. The nodes may be end points (such as processor or storage units), or may be intermediate devices or branch points (such as routers or switches in the interconnect fabric). The interconnect fabric comprises a set of one or more routers, switches, electrical, optical, electroptical or other links along which messages are passed. Messages are packets having a defined format including, e.g., a header portion, typically with source and target addresses, and codes indicating message-type or other information, followed by one or more data or other fields. A first node ('sending' node) of a digital data system as described sends a data transmission comprising one or more message packets to a second node ('receiving' node) over a link of a fabric as described above. The receiving node returns a control symbol to the sending node for each packet received on the link. The sender uses information in that symbol to control the further transmission of message packets to receiver over the link.

Method and system for link fabric error detection and message flow control

A digital data system comprises a plurality of links for passing messages between nodes, which may be end points such as memory or processing units, or intermediate or branch points such as routers or other devices in the system. A link level flow control is implemented by control symbols passed between adjacent nodes on a link to efficiently regulate message burden on the link. The control symbols may be embedded within in a message packet to quickly effect control on a link&#x2014;such as reducing data flow, requesting retransmission of corrupted data, or other intervention&#x2014;without disruption of the ongoing packet reception. A control symbol may be recognized within the packet by a flag bit, a marker such as a transition in a signal, or a combination of characteristics. The control symbol may be a short word, having a control action identifier code at defined bit positions to indicate the desired link-level control function. A node receiving the control symbol implements the designated control action while maintaining packet order data (such a bit positions and byte counts), and applying message housekeeping processing (such as error checking) as though the control symbol were absent. The control symbol is thus embedded in a manner such that a receiving node need not receive a complete message, or even a complete packet, before acting on the control symbol. The beginning and end of the interrupted packet, meanwhile, are handled as a single message packet, and are processed by the receiving node in properly-aligned bit positions, allowing message flow to continue along the link without interruption. Thus, retransmission of the affected packet or message is not required. The result is that control operations dictated by the embedded control symbol are effected immediately and without slowing down communications.

Growing community of inventors

Austin, TX, United States of America

Daniel L Bouvier