src - OpenBSD base system

Age	Commit message (Collapse)	Author
2000-06-30	warnx?/errx? paranoia (use "%s" not a bare string unless it is a	Todd C. Miller
	constant). These are not security holes but it is worth fixing them anyway both for robustness and so folks looking for examples in the tree are not misled into doing something potentially dangerous. Furthermore, it is a bad idea to assume that pathnames will not include '%' in them and that error routines don't return strings with '%' in them (especially in light of the possibility of locales).
1999-04-21	Don't display the string associated with errno for "basedir too long"	Alex Feldman
	diagnostic. Issue a diagnostic message if the master.passwd file isn't specified as an absolute path.
1998-07-15	Fix a u_int vs. off_t issue and provide an estimate of the number	Todd C. Miller
	of elements in the hash based on master.passwd file size, assuming an average 128bytes per entry. This is only an estimate so it doesn't have to be exact.
1998-07-15	Move loop that is executed 6 times into its own function to improve readability.	Todd C. Miller

1998-07-14	sizeof(uid_t) and sizeof(gid_t) not sizeof(int)	Todd C. Miller

1998-07-14	Get rid of unused empty file, pw_scan.h. Nothing uses it anymore.	Todd C. Miller
	When building .db versions of passwd and master.passwd, go split the loops into three (one per key type) so that we get good locality withing the .db file for getnext style operations (getpwent). With this change I see about a 20% speedup of getpwent() on very large passwd files.
1998-06-10	clear flag before pw_scan(); lukem	Theo de Raadt

1998-04-26	use correct db pointer; fair	Theo de Raadt

1998-04-26	indent	Theo de Raadt

1997-12-08	first cut at performance enhancing heuristic	Theo de Raadt

1997-09-15	When the disk gets full, abort properly instead of creating empty	Theo de Raadt
	passwd databases. I found this while analysing netbsd pr#1328 from August 10, 1995 by hag@gnu.ai.mit.edu. A sample fix was supplied on 14, May 96 by greywolf@defender.VAS.viewlogic.com. The PR mentioned about 6 or 7 places where this could happen. Greywolf and I had made all of the fixes ourselves in openbsd a while back (except one subtle one which he pointed out but I had missed), but not a single one of the fixes is found in the netbsd source tree... I wonder if Perry has an exploit for this problem, and perhaps he's using it? Another damn good reason for making /tmp and /var seperate partitions.
1997-01-16	From NetBSD: Warn about potentially unsafe uid/gid overrides.	Todd C. Miller

1997-01-15	getopt(3) returns -1 when out of args, not EOF, whee!	Todd C. Miller

1996-10-16	kill __yp_token.	Jason Downs

1996-09-28	Add -c flag from FreeBSD.	Jason Downs

1996-08-30	limit basedir length	Theo de Raadt

1996-05-22	libutil	Theo de Raadt

1996-05-14	looks like greywolf and i fixed netbsd pr#1328. pwd_mkdb on a filled root	Theo de Raadt
	filesystem would result in gibbled passwd databases.
1996-05-04	strdup re-generated path in pwd_mkdb; "make distribution" passwd files with	Theo de Raadt
	proper arguments
1996-04-23	add "-d basedir" flag to pwd_mkdb. the use of "chroot" is incompatible	Theo de Raadt
	with our cross compilation/installation goals...
1995-10-18	initial import of NetBSD tree	Theo de Raadt