src - OpenBSD base system

Age	Commit message (Collapse)	Author
2013-04-09	do not store a time_t or the result of a time_t division in an int	Gilles Chehade
	prompted by deraadt@, ok claudio@ chl@ guenther@
2013-04-08	division of time_t does not necessarily fit in an int	Gilles Chehade
	spotted by deraadt@
2013-04-06	Turn an error into a warning to be able to start l2tp tunnels even if	Giovanni Bechis
	gre is not allowed. ok yasuoka@
2013-04-05	kill redundant prototype	Florian Obser
	OK dlg@
2013-04-05	send_packet() and writev() return ssize_t, not int. Use correct	Kenneth R Westerback
	type to store the returned value. From dhill.
2013-04-03	Make it possible to explicitly specify the number of vcpus and the amount	Mark Kettenis
	of memory for the primary domain. As a bonus this prevents people from shooting themselves in the foot by creating two domains names "primary".
2013-04-03	do not assume time_t and timeval.tv_sec are the same type	Theo de Raadt
	ok guenther
2013-04-03	Handle big time_t	Philip Guenther
	ok deraadt@
2013-04-02	change log format to fix warning	Gilles Chehade

2013-04-02	Handle big time_t	Philip Guenther
	ok deraadt@
2013-04-02	for parsing time values, fall back to atoll() for time_t	Theo de Raadt
	ok guenther
2013-04-02	tv_sec is not neccessarily the same type (or size) as time_t	Theo de Raadt
	ok guenther
2013-04-02	handle large time_t at least until 2038. then chaos breaks out.	Theo de Raadt
	ok guenther note to tedu: 2038 -> rwhod
2013-04-02	handle long time_t; ok guenther	Theo de Raadt

2013-04-01	Some firmware revisions on UltraSPARC T2 machines apparently insist on having	Mark Kettenis
	"rngs" and "rng" nodes in the Hypervisor machine description.
2013-04-01	'#' starts a comment, not a line continuation.	Tobias Stoeckmann
	ok gilles
2013-03-31	unbreak tree	Theo de Raadt

2013-03-31	ooops, forgot to lower the define after my very last test: it's 10% not 99%	Gilles Chehade

2013-03-31	build this on octeon too	Jasper Lievisse Adriaanse

2013-03-30	have smtpd temporarily refuse new messages if file system holding the queue	Gilles Chehade
	has less than 10% of disk space or inodes left. a warning is logged and the clients will get a temporarily failure asking them to retry later. prompted by deraadt@
2013-03-30	previous had one backslash too many;	Jason McIntyre

2013-03-30	Fix typos.	Lawrence Teo
	From "Mike." the.lists()mgm51.com, thanks! ok jmc
2013-03-29	Prevent race conditions in db file handling.	Tobias Stoeckmann
	ok gilles
2013-03-29	Patch from Ilya Bakulin that allows to put snmpd(8) into read-only mode	Gerhard Roth
	so that all "set" requests will be rejected. ok reyk@
2013-03-29	Enable the response rate limiting code in NSD in the default build,	Stuart Henderson
	see http://www.nlnetlabs.nl/blog/2012/10/11/nsd-ratelimit/ for more information. camield, phessler and deraadt agree this is a sane default.
2013-03-28	shuffle the KERNEL BUILDING text a little; from daniel dickman	Jason McIntyre

2013-03-27	some updates from daniel dickman;	Jason McIntyre

2013-03-27	fix build after asr update	Eric Faurot
	ok gilles@
2013-03-26	Include inttypes.h to get PRIx64; ok gilles@	Todd C. Miller

2013-03-26	for the sake of argument, let's pretend this #if 0 code isn't wanted	Ted Unangst

2013-03-25	the unused palm code goes away.	Theo de Raadt

2013-03-25	the unused palm code goes away.	Theo de Raadt

2013-03-25	remove useless line that breaks with new perl	Stuart Henderson

2013-03-25	sync w/ospfd: improve snapshot handling; ok claudio, sthen	Markus Friedl

2013-03-25	sync w/ospfd: ignore the nbr_adj_timer if we are not forming	Markus Friedl
	adjacencies; ok claudio sthen
2013-03-25	sync w/ospfd: the event NBR_EVT_SEQ_NUM_MIS is not defined in state	Markus Friedl
	NBR_STA_XSTRT; ok claudio@ sthen@
2013-03-25	sync w/ ospfd:	Markus Friedl
	for point-to-point interfaces we need to send lsupdates to the interface address, since there is no DR and multicast messages to the DR will be ignored. ok claudio@, then@
2013-03-25	sync w/ospfd: allow two minutes until neighbor adjacencies are formed;	Markus Friedl
	ok claudio@ sthen@
2013-03-24	Do not leak kernel pointers, unless operating as root. Some other display	Theo de Raadt
	functionality is lost, but one can argue that privacy was being violated there... ok various, including bluhm and guenther
2013-03-23	rework the main loop so we can drop kmem privs a little later,	Ted Unangst
	prepping for a coming kernel change. we need to call sysctl for all the procs to get their vmspace pointer, then we drop, then we go grovelling. ok deraadt
2013-03-22	RFC 2131 4.3.1 seems clear that a server MUST NOT include the	Kenneth R Westerback
	client-identifier option in OFFER or ACK messages. So stop doing so.
2013-03-22	tweak previous; ok sthen	Jason McIntyre

2013-03-22	decl's before code. please push to upstream.	Theo de Raadt

2013-03-22	port from ospfd/lsreq.c r1.20, ok claudio@	Stuart Henderson
	date: 2013/01/17 09:02:22; author: markus; state: Exp; lines: +4 -2 the peer might still send more DD messages, even if the RDE has processed all received DDs (dd_pending == 0), so we need to check for state != NBR_STA_XCHNG; ok claudio@
2013-03-22	Allow use of an alternative control socket, ported from ospfd. ok claudio@	Stuart Henderson

2013-03-22	sync comment typos/whitespace with ospfd	Stuart Henderson

2013-03-22	typo in comment	Stuart Henderson

2013-03-22	gratuitous whitespace diff with ospfd/ospf.h	Stuart Henderson

2013-03-22	whitespace tweak; gratuitous difference with ospf6d	Stuart Henderson

2013-03-22	reinstate code to announce routes to backup carp interfaces. this was removed	Stuart Henderson
	in a cleanup in r1.83, but we do still want to announce these (with a high metric), so that a specific route to the relevant network is maintained during failover. ok florian@ claudio@