Floating-point Reciprocal Approximation InstructionFloating-point Reciprocal Approximation Instructionfrcpa
Instruction Type F
Format
(qp) frcpa.a HREF="6400021.htm">sf f1, p2 = f2, f3
Description
If PR qp is 0, PR p2 is cleared and f1 remains unchanged.
If PR qp is 1, the following will occur:
FR f1 is either set to an approximation (with a relative error < 2-8.886) of the reciprocal of FR f3, or to the IEEE-754 mandated quotient of FR f2/FR f3 — if either FR f2 or FR f3 is in the set {-Infinity, -0, Pseudo-zero, +0, +Infinity, NaN, Unsupported}.
If FR f1 is set to the approximation of the reciprocal of FR f3, then PR p2 is set to 1; otherwise, it is set to 0.
If FR f2 and FR f3 are such that the approximation of FR f3's reciprocal may cause the Newton-Raphson iterations to fail to produce the correct IEEE-754 result of FR f2/FR f3, then a Floating-point Exception fault for Software Assist occurs.
System software is expected to compute the IEEE-754 quotient (FR f2/FR f3), return the result in FR f1, and set PR p2 to 0.
If either FR f2 or FR f3 is a NaTVal, FR f1 is set to NaTVal instead of the computed result, and PR p2 is cleared.
fabs - Floating-point Absolute Value Instruction
fadd - Floating-point Add Instruction
famax - Floating-point Absolute Maximum Instruction
famin - Floating-point Absolute Minimum Instruction
fand - Floating-point AND Instruction
fchkf - Floating-point Check Flags Instruction
fclass - Floating-point Class Instruction
fclrf - Floating-point Clear Flags Instruction
fcmp - Floating-point Compare Instruction
fcvt.xf - Convert Signed Integer to Floating-point Instruction
Floating-Point Arithmetic Instructions
Floating-Point Memory Access Instructions
Floating-Point Non-Arithmetic Instructions
Floating-Point Pseudo-Operations
Floating-Point Register Transfer Instructions
Floating-Point Status Register (FPSR) Status Field Instructions
fma - Floating-point Multiply Add Instruction
fmax - Floating-point Maximum Instruction
fmerge - Floating-point Merge Instruction
fmin - Floating-point Minimum Instruction
fmpy - Floating-point Multiply Instruction
fms - Floating-point Multiply Subtract Instruction
fneg - Floating-point Negate Instruction
fnma - Floating-point Negative Multiply Add Instruction
fnmpy - Floating-point Negative Multiply Instruction
fnorm - Floating-point Normalize Instruction
for - Floating-point OR Instruction
fpabs - Floating-point Parallel Absolute Value Instruction
fpack - Floating-point Pack Instruction
fpamax - Floating-point Parallel Absolute Value Maximum Instruction
fpamin - Floating-point Parallel Absolute Value Minimum Instruction
fpcmp - Floating-point Parallel Compare Instruction
fpcvt.fx - Convert Parallel Floating-point to Integer Instruction
fcvt.xuf - Convert Unsigned Integer to Floating-point Instruction
fpma - Floating-point Parallel Multiply Add Instruction
fpmax - Floating-point Parallel Maximum Instruction
fpmin - Floating-point Parallel Minimum Instruction
fpmerge - Floating-point Parallel Merge Instructions
fpmpy - Floating-point Parallel Multiply Instruction
fpms - Floating-point Parallel Multiply Subtract Instruction
fpneg - Floating-point Parallel Negate Instruction
fpnegabs - Floating-point Parallel Negate Absolute Value Instruction
fpnma - Floating-point Parallel Negative Multiply Add Instruction
fpnmpy - Floating-point Parallel Negative Multiply Instruction
fprcpa - Floating-point Parallel Reciprocal Approximation Instruction
fprsqrta - Floating-point Parallel Reciprocal Square Root Approximation Instruction
frsqrta - Floating-point Reciprocal Square Root Approximation Instruction
fselect - Floating-point Select Instruction
fsetc - Floating-point Set Controls Instruction
fsub - Floating-point Subtract Instruction
fswap - Floating-point Swap Instruction
fsxt - Floating-point Sign Extend Instruction
fwb - Flush Write Buffers Instruction
fxor - Floating-point Exclusive Or Instruction
getf - Get Floating-point Value or Exponent or Significand Instruction
Integer Multiply and Add (in FP Unit) Instructions
ldf - Floating-point Load Instruction
ldfp - Floating-point Load Pair Instruction
mov fr - Move Floating-point Register Instruction
setf - Set Floating-point Value, Exponent, or Significand Instruction