Re: Efficacy of compiler optimizations

Derk Gwen <derkgwen@HotPOP.com>
10 Aug 2003 10:51:21 -0400

          From comp.compilers
Related articles
Efficacy of compiler optimizations kernel_learner@yahoo.com (2003-08-04)
Re: Efficacy of compiler optimizations derkgwen@HotPOP.com (Derk Gwen) (2003-08-10)
Re: Efficacy of compiler optimizations suresh@sankhya.com (2003-08-10)
Re: Efficacy of compiler optimizations johnfl@cs.uoregon.edu (John Fiskio-Lasseter) (2003-08-10)
| List of all articles for this month |
From: Derk Gwen <derkgwen@HotPOP.com>
Newsgroups: comp.compilers
Date: 10 Aug 2003 10:51:21 -0400
Organization: Quick STOP Groceries
References: 03-08-012
Keywords: optimize, practice
Posted-Date: 10 Aug 2003 10:51:21 EDT

kernel_learner@yahoo.com (Learner) wrote:
# Hi all,
#
# I had a general question about compiler optimizations and believe
# this is a great place to ask!
#
# First of all there are numerous number of compiler optimizations and
# they each have their own characteristics..my questions is two-fold:
#
# 1. What is the cost/benefit ratio of these optimizations.


There is no simple, linear relation. Small changes in the program can have
major effects on the optimisation.


# 2. How do these compiler optimizations interact? i.e. does one
# optimization negate the effect of another optimization if done in a
# particular order?


Optimisation are usually orderred so that one phase sets up conditions
that subsequent phases can assume are true rather repeating
work. Otherwise nonlinearities getting the best orderring and
interleaving of phases is artwork or guesswork.


--
Derk Gwen http://derkgwen.250free.com/html/index.html
The whole world's against us.

Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.