Table of Contents
Advertisement
AMD Athlon™ Processor x86 Code Optimization
Declare Local Functions as Static
24
case combine( 1, 1 ):
for( i ... ) {
}
break;
default:
break;
}
The trick here is that there is some up-front work involved in
generating all the combinations for the switch constant and the
total amount of code has doubled. However, it is also clear that
the inner loops are "if()-free". In ideal cases where the
"DoWork*()" functions are inlined, the successive functions
will have greater overlap leading to greater parallelism than
would be possible in the presence of intervening "if()"
statements.
The s a m e i de a c a n b e a p p l i e d t o c o n s t a n t " swi t ch( ) "
statements, or combinations of "switch()" statements and "if()"
statements inside of "for()" loops. The method for combining
the input constants gets more complicated but will be worth it
for the performance benefit.
However, the number of inner loops can also substantially
increase. If the number of inner loops is prohibitively high, then
only the most common cases need to be dealt with directly, and
the remaining cases can fall back to the old code in a "default:"
clause for the "switch()" statement.
This typically comes up when the programmer is considering
runtime generated code. While runtime generated code can
lead to similar levels of performance improvement, it is much
harder to maintain, and the developer must do their own
optimizations for their code generation without the help of an
available compiler.
Functions that are not used outside the file in which they are
defined should always be declared static, which forces internal
linkage. Otherwise, such functions default to external linkage,
DoWork1( i );
DoWork3( i );
Declare Local Functions as Static
22007E/0—November 1999
Advertisement
Table of Contents
