OSNews: http://www.osnews.com/story/20221/Anatomy_of_Linux_Dynamic_Libraries Exploring the Future of Computing en-us Copyright 2001-2009, David Adams adam+nospam@osnews.com Fri, 11 Dec 2009 04:15:24 GMT http://www.osnews.com/images/osnews.gif OSNews.com http://www.osnews.com Nice Article http://osnews.com/thread?328099 http://osnews.com/thread?328099 I really appreciate articles such as this, since I don't have the shelf-space (or the funds!) to build a comprehensive library on how to program in Linux Wed, 27 Aug 2008 03:59:00 GMT donotreply@osnews.com (StychoKiller) Comments Not a good ELF http://osnews.com/thread?328107 http://osnews.com/thread?328107 I've had a detailed look at 2 binary formats over the years, both ELF and Microsoft's PE.

PE doesn't generate position-independent code, so it's slightly faster than ELF since eg. the EBX register is available to x86 applications to use. However each DLL has to be loaded at a particular location in memory, and that location cannot change after the DLL is loaded, so if that memory region is unavailable in a new process, the DLL has to be "rebased", ie. a new private copy of it loaded somewhere else in memory, which uses extra memory in that process.

ELF does generate position-independent code so it never has the rebasing problem, at the cost of a slight slow down in performance, but in ELF symbol lookup is process scoped, not library scoped like everywhere else. So if 2 libraries define the same symbol, the first library loaded wins. This is a big problem in real world applications, as Autopackage discovered. Wed, 27 Aug 2008 05:24:00 GMT donotreply@osnews.com (draethus) Comments Better use Drepper's http://osnews.com/thread?328164 http://osnews.com/thread?328164 Here is a much much better and in-depth article: http://people.redhat.com/drepper/dsohowto.pdf How to write shared librariesEdited 2008-08-27 14:17 UTC Wed, 27 Aug 2008 14:15:00 GMT donotreply@osnews.com (pgquiles) Comments RE: Not a good ELF http://osnews.com/thread?328173 http://osnews.com/thread?328173
PE doesn't generate position-independent code, so it's slightly faster than ELF since eg. the EBX register is available to x86 applications to use.

Yes. The difference is typically about 5% for processor-bound tasks on x86_32. I doubt the effect is measurable at all on architectures like x86_64 which are not so ridiculously register-starved.

gcc will produce non position independent code if you tell it to. Wed, 27 Aug 2008 15:09:00 GMT donotreply@osnews.com (sbergman27) Comments