The Divider IP core uses a non-restoring division algorithm to implement the integer division operation. There are N stages of 1-bit division in an integer division operation, where N is the width of the quotient. Each stage generates a 1-bit quotient and partial-remainder. In the last stage, the final quotient and remainder are generated. 1-bit division uses an adder-subtractor to compare the partial remainder and denominator to get a new partial remainder. Quotient-digit selection is based on the sign of the partial remainder. In the last stage, the partial remainder is corrected to get the final remainder.
The Divider IP core supports configurable output latency. The latency can be any number of clock cycles from 1 to N. When latency is set to the value M, M stages of output registers are uniformly distributed into the N stages of 1-bit division operation. The final division stage always has output registers.
Implements Integer Division - The Divider IP core implements integer division with the formula: Numerator = Denominator * Quotient + Remainder
Uses a Non-Restoring Division Algorithm - The Divider IP core uses a non-restoring division algorithm to implement the integer division operation. There are N stages of 1-bit division in an integer division operation, where N is the width of the quotient. In the last stage, the final quotient and remainder are generated. 1-bit division uses an adder-subtractor to compare the partial remainder and denominator to get a new partial remainder. In the last stage, the partial remainder is corrected to get the final remainder.
Supports Configurable Output Latency - The Divider IP core supports configurable output latency. The latency can be any number of clock cycles from 1 to N. When latency is set to the value M, M stages of output registers are uniformly distributed into the N stages of 1-bit division operation. The final division stage always has output registers.