On 02/24/13 01:27, Michael Schwendt wrote: > On Sun, 24 Feb 2013 00:54:29 +0100, poma wrote: > >> On 02/23/13 14:03, Michael Schwendt wrote: >> […] >>> Nevertheless, building from source isn't a safe fix either, since it only >>> uses different compiler flags. The crash is due to uninitialized pointers >>> -> bug 914659 >>> >>> So, don't blame the packaging alone. The C source isn't pretty. >>> >> >> So simple tool, and so much trouble with optimization flags(-O1 vs -O2). >> Thanks for jumping in and patching. > > I'm not following you. It's not just a matter of "-O1 vs -O2". It's not as > if the source code were made for -O1 only. Uninitialized variables can > lead to undefined behaviour. With or without optimisation. Already a > single bit decides between 0 and 1, False and True, NULL-ptr and invalid > ptr, for example. Even worse for longer variables. >
Well, as always, defaults matter, not optflags in rpmrc. ;) And within config.mk default is CFLAGS = -O1. Let Hartlich comments, it's his code. Cheers, poma -- users mailing list users@lists.fedoraproject.org To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
