On Sun, 2007-08-05 at 22:58 -0400, Fuzzy Logic wrote: > If you know what valgrind is and what it does, I would have expected > you to realize that if valgrind complains, the chances of it being > anything other than a bug are quite close to zero.
Could be. But my point was a different one. It doesn't really help dumping a big valgrind log onto a mailinglist and write "look these are errors" without being sure: a) that they actually really are errors b) that he can prove them to be errors A lot of different factors matter in this view. A compiler used, code generation, compiler version that was used, known bugs in the compiler, optimization, hardware architecture and and and. Then a different factor would be uninitialized variables in the code, variables that are not used at all. Then it's good knowing how the compiler works, what it does in during optimization. You can also make the compiler to keep it's variables in registers rather than some memory area to store data and so on. Yes it makes sense to initialize variables, yes it makes sense to keep a clean code but then it's also the compiler that needs to keep track of such issues and initialize these values or the memory area (with either null) or whatever. Mit freundlichen Grüßen Ali Akcaagac --~--~---------~--~----~------------~-------~--~----~ You received this message from the "vim_dev" maillist. For more information, visit http://www.vim.org/maillist.php -~----------~----~----~----~------~----~------~--~---
