Encoding:

Format:

FMIN_A.df 

FMIN_A.W wd,ws,wt

FMIN_A.D wd,ws,wt

Purpose:

Vector Floating-Point Minimum Based on Absolute Values

Vector floating-point minimum based on the magnitude, i.e. absolute values.

Description:

wd[i] = absolute_value(ws[i]) < absolute_value(wt[i])? ws[i]: wt[i]

The value with the smallest magnitude, i.e. absol ute value, between corres ponding floating-point elements in vector ws and vector wt are written to vector wd.

The smallest absolute value is defined by the minNumMag operation in the IEEE Standard for Floating-Point Arithmetic 754TM-2008.

The operands and results are values in floating-point data format df.

Restrictions:

Data-dependent exceptions are poss ible as s pecified by the I EEE Standard for Floating-Point Arithmetic 754^TM2008.

Operation:

FMIN_A.W
   for i in 0 .. WRLEN/32-1
      WR[wd]32i+31..32i = MinAbsoluteFP(WR[ws]32i+31..32i, WR[wt]32i+31..32i, 32)
   endfor
FMIN_A.D
   for i in 0 .. WRLEN/64-1
      WR[wd]64i+63..64i = MinAbsoluteFP(WR[ws]64i+63..64i, WR[wt]64i+63..64i, 64)
   endfor
function MinAbsoluteFP(tt, ts, n)
   /* Implementation defined, returns the argument with smallest
         absolute value. For equal absolute values, returns the smallest
         argument.*/
endfunction MinAbsoluteFP

Exceptions:

Reserved Instruction Exception, MSA Disabled Exception, MSA Floating Point Exception.