4185 Commits

Author SHA1 Message Date
itojun
45c9804017 avoid memory leak. sync with kame 2001-01-21 07:51:17 +00:00
kleink
f46c3b8662 Teach tcpdump(8) to dump VRRP. 2001-01-19 09:10:13 +00:00
garbled
b0ee440a2f Update the TODO list a little bit. Add a few items that were brought to
my attention.
2001-01-19 07:50:20 +00:00
enami
f93a8a97b8 Parse nfs file handle better on NetBSD. On little endian machine,
it was completely broken.  On big endian machine, it was detected as SUNOS4
and thus 32bit dev_t isn't handled properly.

XXX: The auto detect code should be integrated more cleverly if possible.
XXX: In addition, it is better if there is a way to specify server os type.
2001-01-19 01:28:50 +00:00
lukem
3b32ee2ae2 this is ntpq(8) not pq(8). (conversion problem from the html???) 2001-01-19 01:25:32 +00:00
thorpej
ed7ec7c142 Make this work again for the case where there are devices
configured, but none of those devices happen to have locators.
2001-01-18 07:09:47 +00:00
mycroft
5087e52718 Kill .NOPATH, and fix some DPADDs. 2001-01-18 03:28:19 +00:00
scottr
acef071496 Fix the sorting introduced in rev 1.9 for the netgroup map:
handle entries that span multiple lines correctly.
2001-01-16 07:15:18 +00:00
cgd
d594ce939b comment or delete text after CPP directives. 2001-01-16 02:50:27 +00:00
cgd
bcc8ec9959 include stdlib.h for prototypes (typically for exit()). 2001-01-16 02:44:43 +00:00
cgd
fdfcf1962d labels (even default:) _must_ be followed by statements! usr.bin/fgen/fgen.l 2001-01-16 02:43:37 +00:00
cgd
c338834cd9 use expr to do subtraction. shell math isn't as portable. 2001-01-16 02:42:39 +00:00
cgd
9cfe468c74 avoid C sequence point issues warned about by development version of gcc. 2001-01-16 02:41:17 +00:00
cgd
e209ced86f include string.h for prototypes. 2001-01-16 02:38:36 +00:00
cgd
3e8fb298b1 Avoid accidental use of trigraphs. Pointed out by development version of gcc. 2001-01-16 02:37:03 +00:00
cgd
ede451d831 comment about the backslash used to avoid trigraph 2001-01-16 02:36:29 +00:00
cgd
4edf8bff8e Don't forget the trailing newline. 2001-01-16 02:34:34 +00:00
cgd
f762cf8294 for linking, libraries should be named _after_ the libraries that need them.
In this case, curses should come after cdk & form.
2001-01-16 02:34:14 +00:00
martin
7d77421729 Comparing ch != -1 when ch is of type char is no good idea.
Fix this for ports where char is unsigned by making ch an int.

Pointed out by Matthias Pfaller.
2001-01-15 09:26:15 +00:00
itojun
a88307ab86 sync with latest kame tree.
- reduce chances for signal handler rae condition
- decrease chances for misconfiguration
- feedbacks from router renumbering protocol bakeoff
2001-01-15 06:14:05 +00:00
garbled
3c1becf72b Fix a buffer overrun in the parsing of the searchpaths, and check the return
values of some curses functions I overlooked.
Pointed out by Brett Lymn in private email.
2001-01-14 21:23:23 +00:00
christos
d4f4672be8 move target definition after including bsd.prog.mk so that target protection
works.
2001-01-14 10:36:11 +00:00
christos
b4d783f6b6 move target after including bsd.prog.mk so that target protection works. 2001-01-14 10:34:57 +00:00
christos
b420e5eb3e don't include bsd.nls.mk; bsd.prog.mk does it for us. 2001-01-14 06:26:56 +00:00
itojun
48110adac6 correct fd_set allocation. from deraadt 2001-01-12 18:53:20 +00:00
itojun
606c47534d make signo a volatile variable. sync with kame 2001-01-12 17:48:37 +00:00
itojun
5e43f4d020 make signal handler very short, to avoid possibility of race conditions.
commented by deraadt.  sync with kame
2001-01-12 14:55:13 +00:00
cgd
e3e65b0f8d When building gethnamaddr.o and getnetnamadr.o from libc sources,
define _LIBC so that the weak aliases those files normally define if
built in libc get set properly.  Otherwise, this doesn't link statically.
Static linking tested on NetBSD/i386 by sommerfeld, correct runtime
operation not verified (but should be no different than before).
2001-01-12 01:43:48 +00:00
jhawk
0d99070cf2 Change wi(4)'s default from ad-hoc mode to BSS mode. Also
some minor manpage fixes.
2001-01-11 19:33:51 +00:00
enami
30fc09dacc Use PRINTOBJDIR. 2001-01-11 13:15:06 +00:00
enami
5e2a856773 Kill no longer used variable decl. 2001-01-11 13:06:28 +00:00
lukem
e4099568d7 minor knf 2001-01-11 07:58:04 +00:00
lukem
06680ddb2d fix tyop 2001-01-11 03:28:03 +00:00
lukem
30724dba3e LOG_ODELAY is deprecated; it's the default 2001-01-11 03:05:42 +00:00
lukem
cb4cea7120 don't use LOG_CONS 2001-01-11 02:44:03 +00:00
lukem
631bc8fbd3 oops, forgot syslog.h 2001-01-11 01:54:08 +00:00
lukem
27f625d619 use LOG_WARNING for warnings and LOG_ERR for fatal errors 2001-01-11 01:47:55 +00:00
lukem
f0ccfb9aa0 skip past / if we found one in argv[0] 2001-01-11 01:45:07 +00:00
lukem
13220245b2 use explicit name rather than __progname in openlog 2001-01-11 01:43:25 +00:00
lukem
7890e93e47 remove unnecessary ":" in openlog ident 2001-01-11 01:43:10 +00:00
lukem
39c3c68f55 when calling openlog():
- don't use LOG_CONS
- use explicit name rather than __progname
2001-01-11 01:42:50 +00:00
lukem
0645f2f67b use explicit name rather than __progname in openlog 2001-01-11 01:34:28 +00:00
lukem
a78cb15fc9 use "ypxfr" instead of __progname as ident arg to openlog
wrap all debug syslogs in #ifdef DEBUG (rather than just most of them)
2001-01-11 00:35:49 +00:00
lukem
48565a8e05 use "ypserv" instead of __progname as ident arg to openlog 2001-01-11 00:34:51 +00:00
lukem
74590d002a add openlog(...,pid, daemon) 2001-01-11 00:16:11 +00:00
lukem
43b126d3f1 no need for LOG_PERROR if we're about to daemon()ize (and close
stderr in the process)
2001-01-10 23:48:14 +00:00
jdolecek
c7121de1da use getcwd() instead of getwd() - mainly to avoid link warning, this
particular usage is safe
2001-01-10 22:53:22 +00:00
thorpej
e153fd2620 CPPFLAGS is for CPP flags, not COPTS. 2001-01-10 18:11:05 +00:00
garbled
b4aac50fda Changes to sushi to allow users to bind different keys to the various
built-in functions.  This allows users to change F1 to say, ^X, in the case
where a function key might not be available, or perhaps is bound to a window
manager.

problem noted by Nathan Williams
2001-01-10 10:00:29 +00:00
jmc
ca4a8d27dc Move routines which fill in tbuf in addrmerge(). The fixes in 1.4 pick a new
interface but don't fill in tbuf which results in clients getting garbage back
in part of the answer. This makes sure there's something in tbuf.

At some point this whole logic should be looked at, as just picking a random
local up interface for non-local requests doesn't feel right.
2001-01-10 05:23:43 +00:00