Floating point unit for exponential function implementation

A system comprising:

• a processor;
• and
• a floating point unit that is communicatively coupled to the processor, the floating point unit comprises an estimator module circuit that performs a method for fast exponential calculation using only two fully-pipelined instructions, the method comprising:
• computing an intermediate value y′ by multiplying an input operand that is received in floating point representation with a predetermined constant value;
• and
• computing an exponential result for the input operand by concurrently:
• converting the intermediate value y′ to an integer representation z represented by v most significant bits (MSB), and w least significant bits (LSB);
• determining exponent bits of the exponential result based on the v MSB from the integer representation z;
• and
• determining mantissa bits of the exponential result according to a piecewise linear mapping function using a predetermined number of segments based on the w LSB from the integer representation z.

A system comprising:

• a floating point unit configured to compute one or more mathematical operations in response to receipt of one or more requests from one or more processors, the one or more mathematical operations including an exponential function, wherein the exponential function is computed by executing only two fully-pipelined instructions comprising:
• a first instruction for computing an intermediate value y′ by multiplying an input operand that is received in floating point representation with a predetermined constant value;
• and
• a second instruction for computing an exponential result for the input operand, the second instruction being a fused instruction using the intermediate value y′, the fused instruction comprising:
• converting the intermediate value y′ to an integer representation z represented by v most significant bits (MSB), and w least significant bits (LSB);
• determining exponent bits of the exponential result based on the v MSB from the integer representation z;
• and
• determining mantissa bits of the exponential result according to a piecewise linear mapping function using a predetermined number of segments based on the w LSB from the integer representation z;
• wherein, the floating point unit is further configured to return the exponential result in response to the one or more instructions to the floating point unit.