On 17/08/2011 12:08, aliko wrote: > Thank you for explanation! The -O0 mode is usefull for me for debuging > purposes because compiller tends to optimize out many things... In > production I use -O2. >
-O0 is a poor choice for most debugging, for a number of reasons. The code it generates is often so long-winded it is hard to understand the generated assembly - there is so much extra code moving data onto and off from the stack that it is hard to see the real action. The code is so much slower than "real" code, that timing differences become relevant. Disabling optimisation disables or limits a number of the static error checking features that you get with gcc - you should compile with /lots/ of warning flags enabled. But the biggest issue is that trying to debug or test code with -O0 hides mistakes that become apparent when you do your real build. Most of your debugging and testing effort should be with the actual code, not a "debug" version. There are some types of debugging, such as stepping through an algorithm, that are easier if the code is not heavily optimised - but using -O1 will not hinder debugging and will make many debugging tasks much easier and much more realistic. ------------------------------------------------------------------------------ Get a FREE DOWNLOAD! and learn more about uberSVN rich system, user administration capabilities and model configuration. Take the hassle out of deploying and managing Subversion and the tools developers use with it. http://p.sf.net/sfu/wandisco-d2d-2 _______________________________________________ Mspgcc-users mailing list Mspgcc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mspgcc-users
