Floating-point division cell

A floating-point division cell for calculating new partial remainder data by subtracting or adding multi-divisor data, which is obtained by multiplying divisor data incorporated in a divisor format and a multiple of 2 together, from/to partial remainder data in which sign data is attached to the highest digit of a remainder format, comprising:

• partial remainder data storing means for registering either parallel-partial-remainder data, in which first partial remainder data is incorporated in a high-order section of the remainder format and second partial remainder data is incorporated in a low-order section of the remainder format, or third partial remainder data is incorporated in an entire section of the remainder format;
• divisor data storing means for storing either parallel-divisor data, in which first divisor data corresponding to the first partial remainder data is incorporated in a high-order section of the divisor format and second divisor data corresponding to the second partial remainder data is incorporated in a low-order section of the divisor format, or third divisor data, corresponding to the third partial remainder, is incorporated in an entire section of the divisor format;
• sign selecting means for selecting the sign data attached to the second partial remainder data in cases where the parallel-partial-remainder data is stored in the partial remainder data storing means and selecting the sign data attached to the third partial remainder data in cases where the third partial remainder data is stored in the partial remainder data storing means, the selection of the sign data being controlled by an external control signal;
• low-order divisor data generating means for receiving both the sign data selected in the sign selecting means and the low-order portion of the divisor data stored in the divisor data storing means and generating low-order divisor data with the sign data;
• low-order partial remainder calculating means for
• (1) obtaining low-order multi-divisor data by multiplying the low-order divisor data with the sign data generated in the low-order divisor data generating means and a multiple of 2 together and
• (2) calculating new low-order partial remainder data by subtracting or adding the low-order multi-divisor data from/to the low-order portion of the partial remainder data stored in the partial remainder data storing means;
• carry data selecting means for
• (1) selecting carry data carried by the calculation performed in the low-order partial remainder calculating means in cases where the third partial remainder data is stored in the partial remainder data storing means and
• (2) selecting a "0" bit in cases where the parallel-partial-remainder data is stored in the partial remainder data storing means, the selection of either the carry data or the "0" bit being controlled by the external control signal;
• high-order divisor data generating means for receiving both the sign data attached to the first or third divisor data stored in the partial remainder data storing means and the high-order portion of the divisor data stored in the divisor data storing means and generating high-order divisor data with the sign data;
• and
• high-order partial remainder calculating means for
• (1) obtaining high-order multi-divisor data by multiplying the high-order divisor data with the sign data generated in the high-order divisor data generating means and a multiple of 2 together,
• (2) calculating new high-order partial remainder data by subtracting or adding the high-order multi-divisor data from/to the high-order portion of the partial remainder data stored in the partial remainder data storing means, and
• (3) adding either the carry data or the "0" bit selected by the carry data selecting means to the lowest digit of the new high-order partial remainder data.