Re: libc-2.10.1 causes previously well-behaved program to abort
Tilman Sauerbeck wrote Thu, 17 Sep 2009 16:49:51 +0200
I'm guessing that the program does something wrong, and that the error was tolerated by earlier glibc versions.
Try running the program in valgrind; that should give you more clues on where the heap corruption happens resp what causes it.
Regards, Tilman
-- You're right. I'm the one at fault here. According to valgrind, there's an illegal write operation which presumably corrupts my memory and (sometimes) causes the next g_malloc call to crash. The malloc thing was just a symptom of the real problem.
The statement that does the illegal write is a call to strcpy, which copies data originally read from the configuration file; that meshes exactly with what I noticed about the effects of changing this file. I think I can see what's illegal about it; it looks like I forgot to convert a void pointer to char. Anyway I'll have a go at fixing it now. Thanks for introducing me to valgrind. It's an awsome program. Hazel -- Hazel Russman <hazel_russman@yahoo.co.uk>
participants (1)
-
Hazel Russman