Abstract: An automatic delay detection and receiver adjustment method for a synchronous communications bus system sends a test pattern to the receivers of the system during a detection phase, uses the test pattern to determine a longest delay time for each bus lone, and adjusts a receiver for each bit line to receive incoming signals at a time based on the determined longest delay time.

Abstract: A method and system for a processor to efficiently accesses a remote First-in First-out (FIFO) buffer that is used to record event information. The access involves an interrupt mechanism when the FIFO transitions from the empty state, a mechanism for reading a FIFO entry including FIFO state information, and a mechanism for reading large areas of the FIFO while maintaining the pointers and interrupt protocols.

Abstract: A clocking circuit decreases the load on the local clock signals to save power. The load is decreased by altering the structure of the latches. Typically, a passgate style latch is used where both an NFET and a PFET are used to control dataflow. Here, the PFET has been removed and the load is decreased. However, it is difficult to pass a logical 1 through an NFET and this increases both the rising slew and rising edge delay through the latch. The effect is mitigated, though, by overdriving the local clock block (LCB) local clocks to drive a local clock to the latches by passgates using only NFET transistors in the master latches and slave latches. Overdrivig the NFET gate allows the NFET to pass a full-level logical 1 signal.

Abstract: An apparatus, method and program product for controlling the transfer of data in a data processing system having a processor handling an I/O request in an I/O operation, main storage controlled by the processor for storing data, and one or more I/O devices for sending data to or receiving data from said main storage. The apparatus includes a vector mechanism operable to register I/O requests by the devices to send or receive data from said main storage. A dispatcher is included which is operable to poll the vector mechanism to determine if there is an outstanding I/O request. An override bit has a first condition when an immediate interrupt is to be sent to the processor for handling an I/O request from the I/O device(s), and a second condition when the dispatcher is to poll the vector mechanism to determine if there is an outstanding I/O request. The override bit is set to its first condition or reset to its second condition by the processor.

Abstract: Apparatus and methods implemented in a processor semiconductor logic chip for providing novel “hint instructions” that uniquely preserve and reuse branch predictions replaced in a branch history table (BHT). A branch prediction is lost in the BHT after its associated instruction is replaced in an instruction cache. The unique “hint instructions” are generated and stored in a unique instruction cache which associates each hint instruction with a line of instructions. The hint instructions contains the latest branch history for all branch instructions executed in an associated line of instructions, and they are stored in the instruction cache during instruction cache hits in the associated line. During an instruction cache miss in an instruction line, the associated hint instruction is stored in a second level cache with a copy of the associated instruction line being replaced in the instruction cache.

Abstract: A method and system for an I/O coupling channel to operate in a plurality of modes. The first mode is the new mode providing peer operation with many times more message passing facilities as the old mode. The second mode is used to connect the new channels through a converter to multiple old channels. In this mode, the new channel distributes its message passing resources among the multiple sink ports of the converter that are attached to old channels. The converter keeps no state information and only adjusts line speeds, routs outbound packets, and adds source information to inbound packets. The new channel operating in old compatibility mode gives the illusion to the software of multiple separate channels, one for each converter sink port.

Abstract: Method and apparatus for sending data from one partition to a second partition within a logically partitioned computer. In a data processing system having multiple logical partitions, a send queue is established in the first logical partition, and a receive queue is established in the second logical partition. The send queue is registered in the send queue in a lookup table available to all of the logical partitions. The send queue is registered using as a key the logical partition identification of the first logical partition and the subchannel number (LPAR-ID.SUBCHANNEL#) of the subchannel assigned to the partition. The receive queue is registered in the lookup table using as a key, the internet protocol address of the receive queue in the second partition. A send instruction from the first logical partition is executed which interrogates the lookup table using the LPAR-ID.

Abstract: An SRAM array local clock generator has variable delay settings that are programmable via level scan bits. Program bits from the level scan operation are decoded and used to adjust the number of delay elements in the local clock generator path.

Abstract: An improved method and system is provided for providing the end-user of a mainframe application running a limited, character-oriented transfer protocol like the IBM3270 protocol with a combined rendering of non-character, i.e. new media data and traditional character data preferably on the same end-user computer. It is proposed to install an individually programmed program component (28), called Server Media Resolution Service (SMRS) on the application server (16) site and a matching program component (30), called Client Media Resolution Service (CMRS) which is a universal, standard component without any individual application specific features. The SMRS is told the client computer destination, searches the requested media address and feeds this meta information to the CMRS which in turn manages the start of a client site media renderer in order to render the new media data received from a datastore such as the File System (20).

Abstract: It is one object of the present invention to obtain a clear scaled image, without impairing the rough shape of fonts or the like, on a graphics screen that is handled by an information display device such as an LCD panel or a projector, and that includes a lot of stepped edges like thin lines.

Abstract: LBIST and weighted LBIST tests are performed simultaneously on different portions of the tested object. This new test methodology and design change achieves the same test coverage and test time as the traditional test strategy with dramatic power reduction during test. It can be applied at wafer, chip, MCM, and system levels of test. Most importantly, it does not need new tools for support. Current test software will work as it does with the traditional test strategy. Scheduling the LBIST and weighted LBIST tests in the same test session reduces the overall power consumption because weighted LBIST testing consumes much less power than flat LBIST testing. In the same test session, if some parts of the logic is tested using weighted LBIST while the others were tested using LBIST, the power consumed by the circuit element at any given time is reduced.

Abstract: A memory interface device uses a driver impedance adjustment engine with state machine for off chip driver (OCD) calibration which is used to set the driver voltage levels of the DRAM memory module or DIMM of the JEDEC DDR-II standard type. By adjusting the pull-up drive strength and pull-down drive strength, the output voltage levels and the rise times can be optimized to find the minimal signal swing that is still immune to noise, while not degrading the data eye significantly. The state machine finds the optimal setting for the DRAM Driver Impedance, using both DC and AC methods adjusting the value of the driver impedance through a master ASIC, and then sampling the known value sent back from the DRAM. The state machine will stop when the optimal value of the driver impedance is found and automates the process of detecting the optimal driver impedance and configuring the DRAM module or DIMM accordingly.

Abstract: A receiving latch with hysteresis circuit for receiving data on cross chip boundaries in a chip to chip interface has a clock section and a feed section and a hysteresis latch section with the feed section receiver enable input pin for a dataline passing through the receiver feed section and hysteresis latch section. The receiver enable input pin D is settable to a high or low voltage level, respectively turning the hysteresis latch section on said dataline ON or OFF. The hysteresis latch pass gate has clock couplings to the pgate and ngate of the PFET and NFET transistors of the pass gate. The drains of said pass gate PFET and NFET are coupled to ground and their sources to a positive potential provided over said data line.

Abstract: A considerable amount of area can be saved according to the present invention by reducing the number of input ports and the number of output ports to the number n of concurrently intended array accesses. This remarkable reduction of ports and thus an extraordinary associated area saving can be achieved when some knowledge about array utilization is exploited: The array accesses are to be performed with concurrent accesses from at most k particular groups. A group is defined by a plurality of array accesses which have at most one access to the same port at a time. Then, for reading the read results are aligned according to a simple re-wiring scheme to the respective read requesters, whereas for writing the accesses are aligned prior to the array access according to the same or a similar scheme.

Abstract: A method and system for a processor to efficiently accesses a remote First-in First-out (FIFO) buffer that is used to record event information. The access involves an interrupt mechanism when the FIFO transitions from the empty state, a mechanism for reading a FIFO entry including FIFO state information, and a mechanism for reading large areas of the FIFO while maintaining the pointers and interrupt protocols.

Abstract: An improved clocking circuit is provided for generating a half swing clock. Previous circuit operations required an additional supply voltage rail (Vdd/2), but the preferred embodiment exploits charge sharing to generate a half swing clock with less power and without the additional supply voltage rail. To drive clock nodes to Vdd/2, a shunt transistor is opened, and the fully charged clock node shares its charge with the fully discharged clock node. When capacitances have been properly matched, both nodes will settle at Vdd/2.

Abstract: A memory subsystem package has a memory controller interface ASIC (application specific integrated circuit) and a plurality of memory modules. The ASIC has a bi-directional serial protocol i2C communication bus to off chip drivers for monitoring temperature and for adjusting the environment surrounding the package by controlling fans using fan switches and variable voltage controls. In addition there is provided an Alternating Current Built in Self Test (AC BIST) with variable data receiver voltage reference for performing high-speed AC memory subsystem self-test for that ASIC enabling writing of pseudo-random patterns to memory, reading them back and comparing the expected results at hardware speeds. Vref can be made to vary across its allowable range during AC self test to provide improved coverage. The system monitors vddq during normal system operation using an ADC. The system varies Vref as a function of Vddq, using a combination of a DAC and ADC.

Abstract: A remote resource management system for managing resources in a symmetrical multiprocessing environment having a plurality of clusters of symmetric multiprocessors each of which provides interfaces between cluster nodes of the symmetric multiprocessor system with a local interface and an interface controller. One or more remote storage controllers each has a local interface controller and a local-to-remote data bus. A remote fetch controller is responsible for processing data accesses across the clusters and a remote store controller is responsible for processing data accesses across the clusters. These controllers work in conjunction to provide a deadlock avoidance system for preventing hangs.

Abstract: A data buffer memory management method and system is provided for increasing the effectiveness and efficiency of buffer replacement selection. Hierarchical Victim Selection (HVS) identifies hot buffer pages, warm buffer pages and cold buffer pages through weights, reference counts, reassignment of levels and ageing of levels, and then explicitly avoids victimizing hot pages while favoring cold pages in the hierarchy. Unlike LRU, pages in the system are identified by both a static manner (through weights) and in a dynamic manner (through reference counts, reassignment of levels and ageing of levels). HVS provides higher concurrency by allowing pages to be victimized from different levels simultaneously. Unlike other approaches, Hierarchical Victim Selection provides the infrastructure for page cleaners to ensure that the next candidate victims will be clean pages by segregating dirty pages in hierarchical levels having multiple separate lists so that the dirty pages may be cleaned asynchronously.

Abstract: A remote resource management system for managing resources in a symmetrical multiprocessing environment having a plurality of clusters of symmetric multiprocessors each of which provides interfaces between cluster nodes of the symmetric multiprocessor system with a local interface and an interface controller. One or more remote storage controllers each has a local interface controller and a local-to-remote data bus. A remote fetch controller is responsible for processing data accesses in accordance with the methods described.