Abstract:
A processing apparatus includes floating point arithmetic circuitry coupled to monitoring circuitry. The monitoring circuitry stores exponent limit data indicating at least one of a maximum exponent value and a minimum exponent value processed when performing the floating point arithmetic operations. The monitoring circuitry may be selectively enabled in dependence upon a virtual machine identifier, an application specific identifier or a program counter value range. Exponent limit data may be gathered in respect of different portions of the floating point arithmetic circuitry and/or may be aggregated to form global exponent limit data for the system.
Abstract:
Embodiments of methods and systems for BMC decoding are described. In an embodiment, a method for BMC decoding involves performing a unit interval estimation of a BMC encoded bit stream, locating a bit boundary of the BMC encoded bit stream based on the unit interval estimation and a known sequence in a preamble of the BMC encoded bit stream, and measuring a time duration across multiple bit transitions from the bit boundary and decoding the BMC encoded bit stream based on the time duration and the unit interval estimation.
Abstract:
A data processing system uses alignment circuitry to align input operands in accordance with a programmable significance parameter to form aligned input operands. The aligned input operands are supplied to arithmetic circuitry, such as an integer adder or an integer multiplier, where a result value is formed. The result value is stored in an output operand storage element, such as a result register. The programmable significance parameter is independent of the result value.
Abstract:
Methods and devices are provided for data compression. Data compression can include receiving a plurality of data chunks, sampling at least some of the plurality of data chunks extracting a common portion from a number of the plurality of data chunks based on the sampling, and storing a remainder of the plurality of data chunks in memory.
Abstract:
A decoding method for biphase-encoded data is provided. The decoding method includes detecting falling-edge transitions in the biphase-encoded data to decode according to a time difference (Δt) between each two adjacent falling-edge transitions and the logic value of previous bit. When Δt is 1 bit period and previous bit is logic 1, it's determined that present bit is logic 1. When Δt is 1 bit period and previous bit is logic 0, it's determined that present bit is logic 0. When Δt is 1.5 bit periods and previous bit is logic 1, it's determined that present and next bits are both logic 0. When Δt is 1.5 bit periods and previous bit is logic 0, it's determined that present bit is logic 1. When Δt is 2 bit periods and previous bit is logic 1, it's determined that present and next bits are logic 0 and 1 respectively.
Abstract:
A gating logic receives a non-return-to-zero (NRZ) input signal and couples the NRZ input signal as an NRZ output signal when operating in a NRZ mode of operation and converts the NRZ input signal to a return-to-zero (RZ) output signal when operating in a RZ mode of operation. A circuit coupled to the gating logic receives a clock signal and couples the clock signal to the gating logic to convert the NRZ input signal to the RZ output signal in the RZ mode of operation. In the NRZ mode of operation, the circuit decouples the clock signal and places a predetermined signal state at the gating logic to pass through the NRZ input signal as the NRZ output signal. The circuit receives a select signal to select between the NRZ and RZ modes of operation and the NRZ and RZ modes are obtained by controlling the clock signal to the gating logic.
Abstract:
A receiving device is provided. A pulse width is measured, it is selected whether a determination process is performed for a pulse width having the length of one bit in accordance with the measured value of the pulse width or a pulse width having the length of ½ bit, it is discriminated whether a current pulse edge is a pulse edge at the center of the bit or a pulse edge at the boundary between bits while considering the bit data right before the determined bit, and when it is determined that the current pulse edge is the pulse edge at the center of the bit, the bit data is determined by the rising edge or the falling edge of the pulse edge.
Abstract:
Passing input strings through an application programming interface between functions that take null byte terminated strings as arguments, where at least some of said input strings contain null bytes internally. This is accomplished by storing the positions of the null bytes relative to the start of the block and storing the non-null bytes in their relative order to prevent said internal null strings from being treated as terminal null strings
Abstract:
When a zero run, which violating G constraint of a run-length-limited (RLL) code, is detected from the data stored in a first input register 1111 and a second input register 1112, bits before and after the zero run is transferred to a temporary register 1150 via a bus for zero run removal 1130 to be combined to each other. Thus, by effectively using the mechanism of bus transfer, a circuit can be simplified, thereby realizing a small circuit.
Abstract:
A fast n-bit to k-bit mapping or translation method and apparatus avoiding the use of content addressable memories (CAMs) is described. It essentially is characterized by using two conventional storage (RAMs). In the first storage (3), the n-bit words are stored preferably in an order determined by the binary search key. The second storage (4) holds the corresponding k-bit translations. Both storages are addressed by essentially the same address, which is established during the (binary) search performed to find a match between an input n-bit word and the contents of the first storage. In variants of the invention, the use of parallel comparisons and of pipelining is demonstrated.