Inference based condition code generation

A method for generating a condition code that depends upon a numerical output of a floating point operation for a processing pipeline, the numerical output based upon normalization and rounding steps of the processing pipeline, the method comprising:

• determining a first classification for the floating point operation of a received instruction, wherein the determining the first classification includes determining that the floating point operation of the received instruction is one of a subtraction operation and an addition operation and wherein the determining the first classification further includes determining the presence of both operands of the received instruction are zero;
• calculating, in response to the first classification that the presence of both operands of the received instruction are zero and using condition determination logic, a value for the condition code by inferring from data that is available from the processing pipeline before the numerical output is available and before the normalization and rounding steps are completed;
• determining a second classification for the floating point operation of a received instruction, and wherein determining the second classification includes determining that the floating point operation of the received instruction is an effective subtraction operation with a minuend and subtrahend and wherein the data that is available from the processing pipeline includes an indication that the minuend is equal to the subtrahend;
• calculating, in response to the second classification and using condition determination logic, an updated value for the condition code by inferring from data that is available from the processing pipeline before the numerical output is available and before the normalization and rounding steps are completed;
• and
• providing the value for the condition code to branch decision logic of the processing pipeline.