8.1 Parallels Between Integer and Floating-Point Instructions

When one learns elementary mechanics in a physics course, certain parallels are often illustrated between the analysis of circular motion and the standard description of linear motion: angular momentum and linear momentum, moment of inertia and mass, etc. Drawing such parallels can assist the learning process both mnemonically through the similarities and conceptually through the contrasts.

Chapters 4 through 6 discussed Itanium integer instructions, several groupings of which have direct analogues in instructions for data in floating-point formats, as depicted in Table 8-1.

Other groupings of Itanium integer instructions with no floating-point counterparts include the shift, byte-manipulation, and data-independent branch instructions. Conversely, several groupings of floating-point instructions discussed in this chapter have no direct integer counterparts.

Table 8-1. Comparisons of Itanium Integer and Floating-Point Instructions

Type of Instruction

Integer

Floating-Point

Load/Store

ld1, ld2, ld4, ld8

ldfs, ldfd, ldfe, ldf8

 

st1, st2, st4, st8

stfs, stfd, stfe, stf8

Arithmetic

add

fadd (pseudo-op)

 

sub

fsub (pseudo-op)

 

xmpy (pseudo-op)

fmpy (pseudo-op)

Logic

and

fand

 

andcm

fandcm

 

or

for

 

xor

fxor

Compare

cmp

fcmp



ItaniumR Architecture for Programmers. Understanding 64-Bit Processors and EPIC Principles
ItaniumR Architecture for Programmers. Understanding 64-Bit Processors and EPIC Principles
ISBN: N/A
EAN: N/A
Year: 2003
Pages: 223

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net