Abstract: A system includes a PCIe controller coupled to a device through a nontransparent PCIe bridge. The controller is operable to direct I/O operations to the device on behalf of a host system. The system also includes a PCIe driver operable within the host system to generate I/O request descriptors that specify movement of data from the PCIe controller to the host system as well as from the host system to the PCIe controller. The PCIe controller processes the I/O request descriptors and determines which device is involved in the specified movement of data. The PCIe controller generates I/O commands that contain routing information for the device (e.g., memory addresses and steering information) to route the data between the memory address of the host system and the memory address of the device, bypassing a memory of the PCIe controller.

Abstract: A dynamically controllable buffering system includes a data buffer that is communicatively coupled between first and second data interfaces and operable to perform as an elasticity first-in-first-out buffer in a first mode and to perform as a store-and-forward buffer in a second mode. The system also includes a controller that is operable to detect data rates of the first and second data interfaces, to operate the data buffer in the first mode when the first data interface has a data transfer rate that is faster than a data transfer rate of the second data interface, and to operate the data buffer in the second mode when the second data interface has a data transfer rate that is faster than the data transfer rate of the first data interface.

Abstract: A system includes a PCIe controller coupled to a device through a nontransparent PCIe bridge. The controller is operable to direct I/O operations to the device on behalf of a host system. The system also includes a PCIe driver operable within the host system to generate I/O request descriptors that specify movement of data from the PCIe controller to the host system as well as from the host system to the PCIe controller. The PCIe controller processes the I/O request descriptors and determines which device is involved in the specified movement of data. The PCIe controller generates I/O commands that contain routing information for the device (e.g., memory addresses and steering information) to route the data between the memory address of the host system and the memory address of the device, bypassing a memory of the PCIe controller.

Abstract: Methods and structure are provided for trapping incoming requests directed to hardware registers of an electronic device. The electronic device that comprises a set of hardware registers that define a configuration of the electronic device, circuitry that implements programmable logic defining which hardware registers have been flagged for trapping incoming requests, and a shadow memory that includes values corresponding to the flagged hardware registers. The circuitry is further operable to access a value in shadow memory that corresponds to a flagged hardware register, responsive to receiving a request from an external device to access the flagged hardware register.

Abstract: A dynamically controllable buffering system includes a data buffer that is communicatively coupled between first and second data interfaces and operable to perform as an elasticity first-in-first-out buffer in a first mode and to perform as a store-and-forward buffer in a second mode. The system also includes a controller that is operable to detect data rates of the first and second data interfaces, to operate the data buffer in the first mode when the first data interface has a data transfer rate that is faster than a data transfer rate of the second data interface, and to operate the data buffer in the second mode when the second data interface has a data transfer rate that is faster than the data transfer rate of the first data interface.

Abstract: Methods and structure are provided for trapping incoming requests directed to hardware registers of an electronic device. The electronic device that comprises a set of hardware registers that define a configuration of the electronic device, circuitry that implements programmable logic defining which hardware registers have been flagged for trapping incoming requests, and a shadow memory that includes values corresponding to the flagged hardware registers. The circuitry is further operable to access a value in shadow memory that corresponds to a flagged hardware register, responsive to receiving a request from an external device to access the flagged hardware register.

Abstract: Apparatus and methods for translation of data formats between multiple interface types. Translation logic is interposed between a producer circuit and a consumer circuit to translate data formats of data signals generated by the producer for application to the consumer. The translation logic may include multiple translators to provide translations between any of multiple producer data formats and any of multiple consumer data formats. One or more producer circuits may thus be selectively coupled with one or more consumer circuits through the translation logic circuit.

Abstract: Methods and systems for processing a second request before processing of a first request has completed. The first request is associated with a first flow control credit type, and the second request is associated with a second flow control credit type. After a period of time, the second request is selected for processing based on the first flow control credit type and the second flow control credit type.

Abstract: Apparatus methods for capturing flow control errors in FIFO exchanges between producing and consuming circuits operating in different clock domains. Tag information at the start of an exchange is transferred to a synchronizing component before data of a transfer transaction is entered in the FIFO. The tag information is also associated with each unit of data transferred to the FIFO by the producing circuit. The synchronizing component verifies the each unit of data retrieved by the consuming circuit has the expected tag information associated therewith and signals an error is the tag information does not match. Thus an error by the producing circuit in entering too much or too little data for a transfer is detected before erroneous data is retrieved and processed by the consuming circuit.

Abstract: Apparatus methods for capturing flow control errors in FIFO exchanges between producing and consuming circuits operating in different clock domains. Tag information at the start of an exchange is transferred to a synchronizing component before data of a transfer transaction is entered in the FIFO. The tag information is also associated with each unit of data transferred to the FIFO by the producing circuit. The synchronizing component verifies the each unit of data retrieved by the consuming circuit has the expected tag information associated therewith and signals an error is the tag information does not match. Thus an error by the producing circuit in entering too much or too little data for a transfer is detected before erroneous data is retrieved and processed by the consuming circuit.

Abstract: Methods and systems for processing a second request before processing of a first request has completed. The first request is associated with a first flow control credit type, and the second request is associated with a second flow control credit type. After a period of time, the second request is selected for processing based on the first flow control credit type and the second flow control credit type.

Abstract: Apparatus and methods for translation of data formats between multiple interface types. Translation logic is interposed between a producer circuit and a consumer circuit to translate data formats of data signals generated by the producer for application to the consumer. The translation logic may include multiple translators to provide translations between any of multiple producer data formats and any of multiple consumer data formats. One or more producer circuits may thus be selectively coupled with one or more consumer circuits through the translation logic circuit.

Abstract: Disclosed is a system that generates a write count value for indicating whether data can be read from a FIFO and a system that generates a read count value for indicating whether data can be written to a FIFO. Each of these systems operates in two separate clock domains. In the system that generates a write count value, write strobes are stored in parallel in a register in a first clock domain. The plurality of synchronizers trigger on a rising edge of the write strobe that is stored in the parallel register and generate an increment pulse in a second clock domain. An up/down counter reads the increment pulse in parallel and increments the up/down counter in parallel. A decrement signal from a read strobe decrements the up/down counter. The output of the counter is fed to a register that provides a write counter value to handshake logic that indicates whether data can be read from a FIFO without underflowing the FIFO.

Abstract: Disclosed is a system that generates a write count value for indicating whether data can be read from a FIFO and a system that generates a read count value for indicating whether data can be written to a FIFO. Each of these systems operates in two separate clock domains. In the system that generates a write count value, write strobes are stored in parallel in a register in a first clock domain. The plurality of synchronizers trigger on a rising edge of the write strobe that is stored in the parallel register and generate an increment pulse in a second clock domain. An up/down counter reads the increment pulse in parallel and increments the up/down counter in parallel. A decrement signal from a read strobe decrements the up/down counter. The output of the counter is fed to a register that provides a write counter value to handshake logic that indicates whether data can be read from a FIFO without underflowing the FIFO.

Abstract: Methods and apparatus for selectively allowing and disallowing changes to an impedance control signal applied to bus driver circuits coupling a device or system to a common, shared bus where impedance of the bus may vary over time. Well known impedance sensing circuits may be coupled to a common bus, such as a PCI bus, and may be used to generate an impedance control signal to be applied to well-known bus driver circuits, including, for example, PCI bus driver circuits, to vary the drive level of such bus driver circuits in accordance with the present electrical impedance sensed on the bus. Features and aspects hereof permit selectively allowing and disallowing changes to such impedance control signals as applied to the driver circuits based upon the present state of the bus and/or the present state of signals driven on the bus by the system embodying the features and aspects hereof.

Abstract: Methods and associated structure for providing broadcast of PCI bus transactions using device ID messaging (DIM) features of the PCI bus specifications. A vendor defined class of messages are defined using device ID messaging to provide broadcast of messages across PCI bus bridge devices to multiple PCI bus segments. One aspect hereof provides for using implicitly addressed device ID messaging such that bridge devices, compatible with the vendor defined message classes, will forward the message upstream and downstream. Another feature provides for use of explicitly addressed device ID messaging to effectuate the desired broadcast. Another aspect hereof provides for translation of a received DIM formatted message with broadcast information and applying the broadcast information to a second bus segment as a standard PCI broadcast transaction.

Abstract: A device for variably unweighting or unloading and performing resistance exercises and stretches while performing cardiovascular exercises on a cardiovascular exercise machine or other exercise device, including a plurality of exercise elements at different locations above the machine or device, the exercise elements being graspable individually or in a selected number and resiliently extended in a variety of directions for reducing forces applied against movable elements of the exercise machine or device and for performing a variety of resistance exercises and stretches.

Abstract: A method and apparatus for traversing a queue of commands containing a mixture of read and write commands places a Next Valid Write Address pointer in each queue entry. In this manner, time savings are achieved by allowing preprocessing of the next write command to be executed. The method may be practiced by setting a next valid address pointer in all queue entries. Queue traversal may be forward, backward, or bi-directional.

Abstract: Methods and associated structure for providing broadcast of PCI bus transactions using device ID messaging (DIM) features of the PCI bus specifications. A vendor defined class of messages are defined using device ID messaging to provide broadcast of messages across PCI bus bridge devices to multiple PCI bus segments. One aspect hereof provides for using implicitly addressed device ID messaging such that bridge devices, compatible with the vendor defined message classes, will forward the message upstream and downstream. Another feature provides for use of explicitly addressed device ID messaging to effectuate the desired broadcast. Another aspect hereof provides for translation of a received DIM formatted message with broadcast information and applying the broadcast information to a second bus segment as a standard PCI broadcast transaction.

Abstract: Centrifugation is used to induce and/or enhance binding between macromolecular targets and either small-molecule ligands or larger biomolecules as single entities or mixtures. With the enhanced binding, the method of the invention permits detection of ligands that bind to target substances and improve the design of ligands. The process relies on centrifugal force to establish a differential and selective concentration gradient between macromolecular therapeutic targets and the desired ligands. Once formed, the information about the self-sorting binding events is derived by analyzing the differential gradient of macromolecules and ligands in situ or by fractionating the gradient into individual samples for independent analysis. A variety of methods or combinations thereof, can be used to look for enhanced levels of bound ligands.