International Business Machines Corporation

Qualcomm Incorporated

Motorola Corporation

Barry Joe Wolford

Bernard Charles Drerup

Richard Gerard Hofmann

James Norris Dieffenderfer

Thomas Andrew Sartorius

Thomas Philip Speier

Jaya Prakash Ganasan

Victor Roberts Augsburg

Jaya Prakash Subramaniam Ganasan

James Edward Sullivan, Jr

Richard Nicholas

Gerald Paul Michalak

Krishnamurthy Venkatramani

Richard Siegmund, Jr

Chau-Shing Hui

John Atkinson Fifield

Harm Peter Hofstee

Robert Christopher Dixon

Mark Michael Schaffer

Richard Nicholas Iachetta, Jr

Wendel Glenn Voigt

Charles E Drake

Richard D Wheeler

Alan Grant Singletary

Jimmie Darius Edrington

Sean Eugene Curry

Prudhvi Nadh Nooney

Victor R Angsburg

Victor E Augsburg

Darren P Umstead

Joseph L Van Swearingen

James Edward Sullivan

Jr James Edward Sullivan

Jim Sullivan

Prudhvi N Nooney

A memory system is disclosed. The memory system includes first and second memory devices, and a memory controller configured to selectively enable one of the memory devices, the memory controller having a first line coupled to the first and second memory devices and a second line coupled to the first and second memory devices. The first memory device is configured to provide a notification to the memory controller on the first line and the second memory device is configured to provide a notification to the memory controller on the second line. The first memory device is further configured not to load the first line and the second memory device is further configured not to load the second line when the memory controller is writing to the enabled memory device.

Method and apparatus to enable the cooperative signaling of a shared bus interrupt in a multi-rank memory subsystem

A transaction request passes from an initiator through interconnect paths and a routing ID indicating the interconnect paths is prepended. A temporary ID is assigned to the routing ID, the transaction request with the temporary ID is sent to a target device, and a response having the temporary ID is received. The routing ID is retrieved using the target ID, and the response with the retrieved routing ID is sent to the initiator.

Method and apparatus for scalable network transaction identifier for interconnects

In a meeting or group event, people having a portable device, such as a cell phone or pager, may wish to be discretely notified when an important message is received, an urgent call comes in from a selected person or a selected group of people, or to be alerted to an upcoming important event without any audible alert to disturb the meeting or group event. To convey such a notification, a tactile alert is provided by vibrating the portable device according to a unique vibration pattern associated with the received communication. When a communication is received, a group identification (ID) is assigned based on the communication being a member of a classified group of source addresses. The portable device associates the group ID with a unique vibration pattern. To provide the alert, the portable device is vibrated according to the unique vibration pattern.

Method of associating groups of classified source addresses with vibration patterns

Efficient techniques for controlling synchronization of bus transactions to improve performance and reduce power requirements in a shared memory system are described. Interconnect arrangements in complex processing systems are also described that provide efficient data transfers between bus masters and shared memory devices to improve performance and reduce power use. In one example, a method for controlling synchronization of bus transactions to remote devices is addressed. A device directed memory barrier command is received. The device directed memory barrier command is decoded to determine one or more destination devices. A memory barrier command is selectively routed to the one or more destination devices in response to the decoding. The described techniques combine high speed device directed memory barrier capability, improved bus bandwidth functionality, and power saving features.

Device directed memory barriers

A weakly-ordered processing system implements an execution synchronization bus transaction, or 'memory barrier' bus transaction, to enforce strongly-ordered data transfer bus transactions. A slave device that ensures global observability may 'opt out' of the memory barrier protocol. In various embodiments, the opt-out decision may be made dynamically by each slave device asserting a signal, may be set system-wide during a Power-On Self Test (POST) by polling the slave devices and setting corresponding bits in a global observability register, or it may be hardwired by system designers so that only slave devices capable of performing out-of-order data transfer operations participate in the memory barrier protocol.

Method and apparatus for conditional broadcast of barrier operations

An apparatus is described for data processing in a computer system. The apparatus comprises a data processing device having a data processing master, a functionally coupled data processor core, and a functionally coupled data processor slave. Both the data processing master and the data processing slave are coupled to a common bus or common crossbar switch. The data processing device processes the data associated with transfers to or from the data processor slave. System masters will direct transactions that require data processing to the data processing slave, which will indirectly interact with the target memory slave. System masters will direct transactions that do not require data processing, directly to the target memory slave.

Data path master/slave data processing device apparatus

A system and structure for snooping cache memories of several snooping masters connected to a bus macro, wherein each non-originating snooping master has a cache memory, and wherein some, but less than all the cache memories, may have the data requested by an originating snooping master and wherein the needed data in an non-originating snooping master is marked as updated, and wherein a main memory having addresses for all data is connected to the bus macro. Only those non-originating snooping masters which may have the requested data are queried. All the non-originating snooping masters that have been queried reply. If a non-originating snooping master has the requested data marked as updated, that non-originating snooping master returns the updated data to the originating snooping master and possibly to the main memory. If none of the non-originating snooping masters has the requested data marked as updated, then the requested data is read from main memory.

Selective snooping by snoop masters to locate updated data

In one embodiment taught herein, a memory module selectively uses its write data mask input as a status output on which it provides status signaling to an associated memory controller. The memory module configures its data mask input as a status output at one or more times not conflicting with write operations. Correspondingly, the memory controller configures its write data mask output as a status input at such times, for receipt of status signaling from the memory module. In one embodiment, the memory module maintains a status register related to one or more operating conditions of the module, such as temperature, and signals status information changes to the memory controller by driving the module's data mask input. In response to such signaling, the memory controller initiates a read of the module's status register to obtain updated status information, and takes appropriate action, such as by changing the module's refresh rate.

Method and apparatus for obtaining memory status information cross-reference to related applications

Status information comprising data not stored in a memory array is efficiently read from a plurality of parallel memory devices sharing an N-bit data bus by configuring each memory device to drive the status information on a different subset M of the N bits, and tri-state the remaining N-M bits. Each memory device is additionally configured to drive zero, one or more data strobes associated with the subset M, and tri-state the remaining data strobes. A memory controller may simultaneously read status information from two or more memory devices in parallel, with each memory device driving a separate subset M of the N-bit bus. Each memory device may serialize the status information, and drive it on the subset M of the bus in burst form. Each memory device may include a configuration register initialized by a memory controller to define its subset M.

Concurrent status register read

A DRAM or SDRAM component maintains an indicator that indicates whether or not an independently refreshable memory unit of a DRAM array, such as a row, contains valid data. When a refresh operation is directed to the associated memory, the refresh operation is suppressed if the memory does not contain valid data. Significant power savings may be realized by suppressing refresh operations directed to invalid data.

System and method to reduce dynamic RAM power consumption via the use of valid data indicators

An apparatus and method is described for data processing in a computer system. The apparatus comprises a data processing device having a data processing master, a functionally coupled data processor core, and a functionally coupled data processor slave. Both the data processing master and the data processing slave are coupled to a common bus or common crossbar switch. The data processing device processes the data associated with transfers to or from the data processor slave. System masters will direct transactions that require data processing to the data processing slave, which will indirectly interact with the target memory slave. System masters will direct transactions that do not require data processing, directly to the target memory slave.

Data path master/slave data processing device apparatus and method

A method, and bus prefetching mechanism are provided for implementing enhanced buffer control. A computer system includes a plurality of masters and at least one slave exchanging data over a system bus and the slave prefetches read data under control of a master. The master generates a continue bus signal that indicates a new or a continued request. The master generates a prefetch bus signal that indicates an amount to prefetch including no prefetching. The master includes a mechanism for continuing a sequence of reads allowing prefetching until a request is made indicating a prefetch amount of zero.

Method and bus prefetching mechanism for implementing enhanced buffer control

A method and structure for snooping cache memories of several snooping masters connected to a bus macro, wherein each non-originating snooping master has cache memory, and wherein some, but less than all the cache memories, may have the data requested by an originating snooping master and wherein the needed data in a non-originating snooping master is marked as updated, and wherein a main memory having addresses for all data is connected to the bus macro. Only those non-originating snooping masters which may have the requested data are queried. All the non-originating snooping masters that have been queried reply. If a non-originating snooping master has the requested data marked as updated, that non-originating snooping master returns the updated data to the originating snooping master and possibly to the main memory. If none of the non-originating snooping masters has the requested data marked as updated, then the requested data is read from main memory.

A method of designing a system on a chip (SoC) to operate with varying latencies and frequencies. A layout of the chip is designed with specific placement of devices, including a bus controller, initiator, and target devices. The time for a signal to propagate from a source device to a destination device is determined relative to a default propagation time. A pipeline stage is then inserted into a bus path between said source device and destination device for each additional time the signal takes to propagate. Each device (i.e., initiators, targets, and bus controller) is designed with logic to control a protocol that functions with a variety of response latencies. With the additional logic, the devices do not need to be changed when pipeline stages are inserted in the various paths. Registers are utilized as the pipeline stages that are inserted within the paths.

Growing community of inventors

Austin, TX, United States of America

Barry Joe Wolford