摘要:
A data processor system includes a first data processor unit for transmitting data units to a second data processor unit and a retry buffer for temporarily storing transmitted data units. The second data processor unit receives the transmitted data and includes an error detector for detecting an error in the received data. When an error is detected, the first data processor unit is notified and a controller causes a data selector to select data from a retry buffer. The first data processor unit limits retransmission of a data unit to a predetermined maximum number of times irrespective of whether the data unit is correctly received or not. This allows for an undisturbed flow of streaming data with an increased reliability.
摘要:
A power island for a system-on-a-chip (SoC) includes a first segment, a second segment, and a supply line. The first segment includes a hardware device and operates the hardware device at first power characteristics indicative of at least a first voltage. The second segment includes scalable logic and operates the scalable logic at second power characteristics indicative of at least a second voltage. The second power characteristics of the scalable logic are different from the first power characteristics of the hardware device. The supply line receives an external supply signal (VDD) and directs the external supply signal to both the first segment and the second segment. The second segment changes at least one power characteristic of the external supply signal to operate the scalable logic according to the second power characteristics.
摘要:
A method for transmitting data is described that includes the steps of: Producing a data frame for transmission, the data frame including a sequence number and user data, saving a copy of the data frame in a retransmission buffer, and if said step of saving a copy requires that data already present in the retransmission buffer is overwritten, selecting the one or more oldest data frames in the retransmission buffer to be overwritten, in case an error is determined in the received data frame, communicating an error message to the transmitter of the data frame, which error message at least comprises an indication of the sequence number of the last correctly received data frame—upon receipt of such message and if available, retransmitting one or more data frames from the retransmission buffer having a sequence number higher than the sequence number communicated in the message.
摘要:
Single hardware subsystems that present two software views that appear to be two separate hardware subsystems attached in a hierarchy are implemented with PCI-type arrangements. According to an example embodiment of the present invention, a hardware arrangement is adapted to emulate two virtually separate hierarchical subsystems in a single hardware block. This emulation facilitates the coupling of devices to PCI Express-type communications links while addressing PCI-Express-type linking requirements for such devices.
摘要:
A power management method is disclosed which provides power management for a hardware based Java accelerator. Initially, a Java mode signal is provided from a host processor in response to initiating a Java application. Thereafter, power to the host processor is reduced, and power to a Java processor is increased in response to the Java mode signal. Then, when execution of the Java application halts, a Java completion signal is generated from the Java processor, thus signaling the system to return control back to the host processor.
摘要:
A method and apparatus for managing power consumption in a computer system wherein the method and apparatus is compliant with the proposed Advanced Configuration and Power Interface (ACPI) specification. In one embodiment, a power management processor is sandwiched between platform hardware and the ACPI register layer. The processor processes all operating power management commands and requests while remaining transparent to the user and the operating system. In so doing, routine power management functions, so classified by the operating system, are implemented by the operating system. Sophisticated power management features, on the other hand, are implemented by the present invention independent from operating system control. Accordingly, in the present invention, the operating system need not suspend processing of other threads to process sophisticated power management procedures.
摘要:
The present invention is a desktop personal computer (PC) system having peripheral device bus mastering. The system has four main elements: a Direct Memory Access (DMA) controller, a hardware state machine, a bus controller, and a device controller. The device controller may be an IDE hard disk controller which is able to generate long streams of data in an intermittent fashion wherein any single stream of data is targeted to a number of different host memory locations. The device controller may also be an ECP parallel port controller which interfaces with a number of different peripheral devices over a parallel bus wherein each peripheral device appears to the system as a separate and independent data path.
摘要:
Single hardware subsystems that present two software views that appear to be two separate hardware subsystems attached in a hierarchy are implemented with PCI arrangements. In an embodiment, a hardware arrangement is configured to emulate two virtually separate hierarchical subsystems in a single hardware block. This emulation facilitates the coupling of devices to PCI Express communications links while addressing PCI-Express linking requirements for such devices.
摘要:
A communications port is implemented for configuration in direction and arrangement. According to an example embodiment of the present invention, a communications link, such as a PCI Express type link, is configurable for communicating with devices having different directional and/or polarity configurations. The communications link is configured to match a communications port condition (e.g., a directional and/or polarity condition) of a device coupled to the communications link. In one instance, the communications link is directionally configurable for reassigning input lanes to output lanes and output lanes to input lanes. With this approach, the communications link can be used to communicate with a variety of devices having varied communication characteristics.
摘要:
In one example embodiment, a high-speed parallel-data communication approach transfers digital data in parallel from a first module to a second module over a communication channel including a plurality of parallel data-carrying lines and a clock path. The parallel bus lines are arranged in a plurality of groups, each of the groups including a plurality of data-carrying lines and a clock path adapted to carry a clock signal for synchronizing digital data carried from the first module to the second module. The sets of data are concurrently transferred using the groups of lines of the parallel bus, and at the second module and for each group, the transferred digital data is synchronously collected via the clock signal for the group. At the second module, the data collected for each group is aligned. By grouping the bus lines in groups with each group having its own clock domain, skew across clock-domain groups is tolerated and overcome by processing the data and the skew first within each clock domain group, and then between groups.