Filter
Top Products
160 Integer Optimizations Chapter 8
25112 Rev. 3.06 September 2005
Software Optimization Guide for AMD64 Processors
8.1 Replacing Division with Multiplication
Optimization
Replace integer division by constants with multiplication by the reciprocal.
Rationale
Because the AMD Athlon™ 64 and AMD Opteron™ processors have very fast integer multiplication
(3–8 cycles signed, 3–8 cycles unsigned) and the integer division delivers only one bit of quotient per
cycle (22–47 cycles signed, 17–41 cycles unsigned), the equivalent code is much faster. Either follow
the examples in this chapter that illustrate the use of integer division by constants or create the
executables using the code in “Derivation of Algorithm, Multiplier, and Shift Factor for Integer
Division by Constants” on page 186.
Multiplication by Reciprocal (Division) Utility
The code for the utilities is shown in “Derivation of Algorithm, Multiplier, and Shift Factor for
Integer Division by Constants” on page 186. The utilities provided in this document are for reference
only and are not supported by AMD.
Signed Division Utility
The sdiv.exe utility finds the fastest code for signed division by a constant. The utility displays the
code after the user enters a signed constant divisor. To redirect the code to a file, type the following
command:
sdiv > example.out
Unsigned Division Utility
The udiv.exe utility finds the fastest code for unsigned division by a constant. The utility displays
the code after the user enters an unsigned constant divisor. To redirect the code to a file, type the
following command:
udiv > example.out