Messages in this thread | | | Date | Tue, 14 Sep 1999 16:58:29 +0200 | Subject | Re: e2fsprogs-1.15 change for llseek check | From | tytso@mit ... |
| |
Date: Thu, 9 Sep 1999 14:47:04 -0400 From: Roland McGrath <roland@frob.com>
Any objections to this change to use llseek if configure finds it in libc, regardless of OS? We might add llseek to glibc for hurd, and then this would dtrt both before and after without further changes to e2fsprogs.
The reason why I have this there is I've gotten screwed in the past with glibc and its headers --- glibc folks claim that llseek isn't standard, so they reserve the right to jerk things around arbitrarily. People have had filesystems corrupted because of the glibc developer's intransigence, so I don't feel much like depending on glibc keeping llseek()'s behaviour standardized. That's why under Linux systems I don't trust glibc's implementation of llseek at all. Once burned, twice shy....
- Ted
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |