Abstract: Margin-testing circuits and methods rely upon the statistics of sampled data to explore the margin characteristics of received data. One margining circuit samples an incoming data stream N times at each of many sample points, each sample point representing a unique sample voltage, unique sample timing, or a unique combination of sample voltage and sample timing. The number of ones sampled at a particular point is a measure of the ones probability for that sample point. The ones probabilities for the collection of unique sample points are then analyzed to measure various aspects of the received data stream, including the data margin.

Abstract: A system for protocol processing in a computer network has an intelligent network interface card (INIC) or communication processing device (CPD) associated with a host computer. The INIC provides a fast-path that avoids protocol processing for most large multi-packet messages, greatly accelerating data communication. The INIC also assists the host for those message packets that are chosen for processing by host software layers. A communication control block for a message is defined that allows DMA controllers of the INIC to move data, free of headers, directly to or from a destination or source in the host. The context is stored in the INIC as a communication control block (CCB) that can be passed back to the host for message processing by the host. The INIC contains specialized hardware circuits that are much faster at their specific tasks than a general purpose CPU.

Abstract: Described are cleaning benches and methods for removing contaminant layers from semiconductor process components using small volumes of hazardous liquids and minimizing cross-contamination between components from different deposition chambers. Components to be cleaned are stored within or supported by a dedicated cassette before they are placed in a receptacle of cleaning liquid. The cassette displaces a significant percentage of the receptacle's volume; consequently, only a relatively small volume of cleaning liquid is needed to fully submerge the component. In typical embodiments, the combined cassette and component displace a volume of liquid that is greater than the volume of liquid used to clean the component. The cleaning bench can include different chemical baths for different components. Cassettes dedicated for use with particular components can be keyed to particular receptacles.

Abstract: A clock recovery circuit samples an incoming data stream that includes sequences of signal transitions. A transition detector categorizes the received signal transitions into various types, such as those associated with 2PAM and 4PAM signaling schemes. Select logic control circuitry analyzes the signal-transition types to determine which of the transition types is best suited for clock recovery. This determination relies upon a number of factors, including for example whether the received signal is a 4PAM signal or a 2PAM signal, the existence of a pattern within the received data, or the relative abundance or scarcity of certain types of transitions.

Abstract: Described are locking clips for mechanically coupling parallel members, such as cables, poles, pipes, and electrical conduits. The clips include retaining elements that accommodate the members in respective retaining bays. The retaining elements are flexible so that each member clips into a respective bay. Locking mechanisms associated with the retaining elements limits the flexibility of the retaining elements once the members are installed, and thus locks the members in place.

Abstract: Described are communication systems that convey differential and common-mode signals over the same differential channel. Noise-tolerant communication schemes use low-amplitude common-mode signals that are easily rejected by differential receivers, thus allowing for very high differential data rates. Some embodiments employ the common-mode signals to transmit backchannel signals for adjusting the characteristics of the differential transmitter. Backchannel control signals are effectively conveyed even if the forward channel transmitter is so maladjusted that the received differential data is unrecognizable. Systems in accordance with the above-described embodiments obtain these advantages without additional pins or communications channels, and are compatible with both AC-coupled and DC-coupled communications channels. Data coding schemes and corresponding data recovery circuits eliminate the need for complex, high-speed CDR circuits.

Abstract: A TCP/IP offload network interface device (NID) receives packets from a plurality of clients and generates, from the socket address of each such packet, a hash value. Each hash value identifies one of a plurality of hash buckets maintained on the NID. In a file server, certain socket address bits of the packets are low entropy bits in that they tend to be the same, regardless of which client sent the packet. Others of the socket address bits are high entropy bits. The hash function employed is such that the hash values resulting from the changing values of the high entropy bits are substantially evenly distributed among the plurality of hash buckets. In a fast-path, the NID uses a first hash function to identify TCBs on the NID. In a slow-path, the NID generates a second hash using a second hash function and a host stack uses the second hash.

Abstract: A system for protocol processing in a computer network has an intelligent network interface card (INIC) or communication processing device (CPD) associated with a host computer. The INIC provides a fast-path that avoids protocol processing for most large multi-packet messages, greatly accelerating data communication. The INIC also assists the host for those message packets that are chosen for processing by host software layers. A communication control block for a message is defined that allows DMA controllers of the INIC to move data, free of headers, directly to or from a destination or source in the host. The context is stored in the INIC as a communication control block (CCB) that can be passed back to the host for message processing by the host. The INIC contains specialized hardware circuits that are much faster at their specific tasks than a general purpose CPU.

Abstract: A tracking switch includes an MOS switching transistor with a control terminal coupled to a CMOS inverter. The relative geometries of the transistors that make up the inverter are significantly imbalanced, resulting is substantially different drive strengths (i.e., substantially different on-resistances). The gate of the switching transistor exhibits parasitic capacitances between its current-handling terminals and its control terminal. When the switching transistor is on, these capacitances shunt a portion of the switched signal to a power-supply node, with the problem increasing with the frequency of the propagated signal. The geometry of the transistor used to turn on the switching transistor is selected to produce a high on-resistance, which introduces a high-impedance path from the control terminal of the switching transistor to ground when the switch is closed.

Abstract: A receiver includes clocked, differential equalization circuitry to compensate for signal attenuation that varies with the frequency of the input signal received over a respective communication channel. The incoming signal is split into filtered and unfiltered signal components. Separate current-steering transistors coupled in parallel amplify the filtered and unfiltered components and sum the results. The filter or filters used to separate the signal components may be tunable, e.g. using voltage-controlled filter components. The ratio of device sizes for the current-steering transistors sets the magnitude of the boost applied to high-frequency components. The embodiments include adjustable or programmable current-steering networks to facilitate adjustments that accommodate the unique characteristics of individual communication channels.

Abstract: Described are methods, systems, and chemistries for removing layers of stubborn silicon and silicon-nitride contamination layers from the inside surfaces of such articles as deposition tubes. In such embodiments, a tube to be cleaned is gently rolled on it side while a portion the tube's interior surface is exposed to an etchant. The tube is only partially filled with etchant to reduce the requisite etchant volume, and the rolling motion evenly exposes the contaminated inner surface to the etchant.

Abstract: Low-latency equalization mechanisms for multi-PAM communication systems are disclosed that reduce delay and complexity in signal correction mechanisms. The equalization mechanisms tap into input signals for a multi-PAM signal driver, and compensate for attenuation along a signal transmission line, crosstalk between adjacent lines, and signal reflections due to impedance discontinuities along the line.

Abstract: Described are approaches to routing buffered reference clock signals to a plurality of input/output (I/O) cell instances on an integrated circuit (IC) die. All or a subset of the I/O cell instances include clock routing resources optimized to deliver high-speed, low jitter clock signals within and through the particular instance. The clock routing resources in physically adjacent instances of the input/output cells for a given IC die automatically interconnect, collectively forming clock routing infrastructure optimized for groups of cell instances. This modular approach to clock routing simplifies the task of combining I/O cell instances with other I/O cell instances and with other types of circuitry.

Abstract: A functional-level instruction-set computing (FLIC) architecture executes higher-level functional instructions such as lookups and bit-compares of variable-length operands. Each FLIC processing-engine slice has specialized processing units including a lookup unit that searches for a matching entry in a lookup cache. Variable-length operands are stored in execution buffers. The operand length and location in the execution buffer are stored in fixed-length general-purpose registers (GPRs) that also store fixed-length operands. A copy/move unit moves data between input and output buffers and one or more FLIC processing-engine slices. Multiple contexts can each have a set of GPRs and execution buffers. An expansion buffer in a FLIC slice can be allocated to a context to expand that context's execution buffer for storing longer operands.

Abstract: A loop filter of a compensating phase-locked loop contains capacitors formed from transistors with thin gate oxide dielectric layers. Leakage current leaks through the capacitors. To avoid jitter in the output signal of the phase-locked loop that would otherwise be caused by the leakage current, a leakage compensation circuit is provided. The leakage compensation circuit of a first embodiment replicates the leakage current using a replication capacitor and a current mirror. The voltage across the replication capacitor is proportional to the control voltage of a voltage-controlled oscillator of the compensating phase-locked loop. A second embodiment generates the compensation current by controlling the voltage on the gate of a transistor. The gate voltage depends on charge added and subtracted by a charge pump in addition to the charge pumps in the loop filter. A third embodiment applies a leakage compensation circuit to a delay locked loop.

Abstract: A system for protocol processing in a computer network has an intelligent network interface card (INIC) or communication processing device (CPD) associated with a host computer. The INIC provides a fast-path that avoids protocol processing for most large multi-packet messages, greatly accelerating data communication. The INIC also assists the host for those message packets that are chosen for processing by host software layers. A communication control block for a message is defined that allows DMA controllers of the INIC to move data, free of headers, directly to or from a destination or source in the host. The context is stored in the INIC as a communication control block (CCB) that can be passed back to the host for message processing by the host. The INIC contains specialized hardware circuits that are much faster at their specific tasks than a general purpose CPU.

Abstract: An apparatus is described having a feedback loop. The feedback loop has an output that approaches a steady state as a data line voltage approaches a reference voltage. The apparatus also includes a driving transistor that drives the data line. The driving transistor has an output impedance that is controlled by the feedback loop output, the feedback loop output keeps the driving transistor output impedance within a high output impedance region when the feedback loop output reaches the steady state.

Abstract: Described are controllable termination impedances that may be adjusted collectively by a combination of digital and analog signals. Each adjustable impedance, responsive to the digital signals, establishes a gross termination resistance for one of a plurality of communication channels by enabling one or more of a plurality of parallel-coupled impedance legs. Each leg includes at least one transistor for controlling the impedance of the leg over a continuous range. An analog compensation voltage is level shifted and the resulting level-shifted signal is applied to the control terminals of the transistors of the selected impedance legs. The compensation voltage, and consequently the level-shifted signal, varies with supply-voltage and temperature fluctuations in a manner that causes the collective impedance of the selected legs for each channel to remain stable despite the fluctuations.

Abstract: A copy instruction executed by a functional-level instruction-set computing (FLIC) processor copies a variable-length data block from one resource to another resource through a cross-bar switch. Resources include general-purpose registers, input, output, and execution buffers, DRAM, SRAM, and other memory. A copy-with-validate instruction has an operand pointing to a first rule in an immediate rule table. The first rule controls validation of a first data-item in the data being copied. Validation includes range and equality checking of the data-item. The value of the data-item or the current offset can be written to a register. A format field in the rule indicates the size of the data-item, or the size is read from the data-item for variable-size formats. The current offset is incremented by the size. The next data-item is validated by a next rule, and other rules in the immediate table control validation of other data-items in the data block.

Abstract: A system for protocol processing in a computer network has an intelligent network interface card (INIC) or communication processing device (CPD) associated with a host computer. The INIC provides a fast-path that avoids protocol processing for most large multi-packet messages, greatly accelerating data communication. The INIC also assists the host for those message packets that are chosen for processing by host software layers. A communication control block for a message is defined that allows DMA controllers of the INIC to move data, free of headers, directly to or from a destination or source in the host. The context is stored in the INIC as a communication control block (CCB) that can be passed back to the host for message processing by the host. The INIC contains specialized hardware circuits that are much faster at their specific tasks than a general purpose CPU.