Floating-point multiply-add unit using cascade design

    公开(公告)号:US08892619B2

    公开(公告)日:2014-11-18

    申请号:US13556710

    申请日:2012-07-24

    CPC classification number: G06F7/5443 G06F7/483 G06F7/49947

    Abstract: A floating-point fused multiply-add (FMA) unit embodied in an integrated circuit includes a multiplier circuit cascaded with an adder circuit to produce a result A*C+B. To decrease latency, the FMA includes accumulation bypass circuits forwarding an unrounded result of the adder to inputs of the close path and the far path circuits of the adder, and forwarding an exponent result in carry save format to an input of the exponent difference circuit. Also included in the FMA is a multiply-add bypass circuit forwarding the unrounded result to the inputs of the multiplier circuit. The adder circuit includes an exponent difference circuit implemented in parallel with the multiplier circuit; a close path circuit implemented after the exponent difference circuit; and a far path circuit implemented after the exponent difference circuit.

    Noise-tolerant signaling schemes supporting simplified timing and data recovery

    公开(公告)号:US20070297520A1

    公开(公告)日:2007-12-27

    申请号:US11895415

    申请日:2007-08-23

    CPC classification number: H04L47/10 H04L5/20 H04L25/0276

    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.

    Circuit, apparatus and method for adjusting a duty-cycle of a clock signal in response to incoming serial data
    70.
    发明授权
    Circuit, apparatus and method for adjusting a duty-cycle of a clock signal in response to incoming serial data 失效
    响应于输入的串行数据调整时钟信号的占空比的电路,装置和方法

    公开(公告)号:US07298807B2

    公开(公告)日:2007-11-20

    申请号:US10672853

    申请日:2003-09-26

    Abstract: A circuit, apparatus and method for maximizing system margins by adjusting a duty-cycle of a clock signal in a receive circuit to whatever duty-cycle is optimal for the particular incoming serial data, rather than the typical 50% duty-cycle, is provided in embodiments of the present invention. A receive circuit, including duty-cycle-correction logic, is included in a double-data rate communication apparatus having a transmit circuit transmitting serial data having duty-cycle distortion. A receive circuit includes a first and second sampler to obtain data and edge values of an incoming serial data responsive to a data and edge clock, respectively. A duty-cycle-correction logic generates a duty-cycle-correction signal to a duty-cycle clock integrator that adjusts the edge clock signals while maintaining quadrature to the data clocks. In an embodiment of the present invention, a duty-cycle-correction logic includes an evaluator circuit to generate an up or down signal responsive to the data and/or edge values. In a further embodiment of the present invention, an evaluator circuit is coupled to a counter and a DAC to generate a duty-cycle-correction signal to the duty-cycle clock integrator. A digital filter or coding scheme is also used to reduce the likelihood of misinterpreting malevolent incoming serial data for duty-cycle distortion in an embodiment of the present invention.

    Abstract translation: 通过将接收电路中的时钟信号的占空比调整到任何占空比来最大化系统裕度的电路,装置和方法对于特定的输入串行数据而言不是典型的50%占空比是最佳的 在本发明的实施例中。 包括占空比校正逻辑的接收电路被包括在具有发送电路的双数据速率通信装置中,该发送电路发送具有占空比失真的串行数据。 接收电路包括第一和第二采样器,用于分别响应于数据和边沿时钟获得输入串行数据的数据和边缘值。 占空比校正逻辑产生占空比校正信号到占空比时钟积分器,该占空比校正信号调整边沿时钟信号同时保持与数据时钟的正交。 在本发明的实施例中,占空比校正逻辑包括响应于数据和/或边缘值产生上升或下降信号的评估器电路。 在本发明的另一实施例中,评估器电路耦合到计数器和DAC,以向占空比时钟积分器产生占空比校正信号。 数字滤波器或编码方案也用于在本发明的一个实施例中减少对恶意输入串行数据进行占空比失真的误解的可能性。

Patent Agency Ranking