deraadt [Sat, 10 Oct 2015 14:46:15 +0000 (14:46 +0000)]
I forgot execve would go through the namei codepath, so a program marked
"stdio rpath" this would fail to execve. pre-indicate exec actions to the
namei checker to allow them through.
ok semarie
deraadt [Sat, 10 Oct 2015 14:33:02 +0000 (14:33 +0000)]
must also pledge "getpw", because it will use getpw* and getgr* functions.
discussed with doug and semarie
deraadt [Sat, 10 Oct 2015 14:29:05 +0000 (14:29 +0000)]
pledge "stdio rpath route" seems to be working. route is needed for
pretty printing some addresses.
deraadt [Sat, 10 Oct 2015 14:27:43 +0000 (14:27 +0000)]
pledge "stdio rpath wpath cpath". as a curses program, I expected this
to maybe need "tty", but have not found a path which calls those kind
of curses functions.
ok doug
deraadt [Sat, 10 Oct 2015 14:25:42 +0000 (14:25 +0000)]
pkill has to get all the getopt, getpwuid, libkvm stuff out of the way
first. it can pledge to "stdio" (pgrep case) or "stdio proc" (pkill case)
before parsing and matching the expression.
ok doug
deraadt [Sat, 10 Oct 2015 14:23:46 +0000 (14:23 +0000)]
fairly obvious pledges.
ok doug
deraadt [Sat, 10 Oct 2015 14:23:12 +0000 (14:23 +0000)]
fairly simple pledge to "stdio rpath wpath cpath"
schwarze [Sat, 10 Oct 2015 13:54:22 +0000 (13:54 +0000)]
Code points U+10000 to U+fffff are valid, too.
Fixing a regression in wcrtomb(3) found with the mandoc testsuite
that was caused by the last commit.
OK semarie@ bentley@
schwarze [Sat, 10 Oct 2015 13:20:25 +0000 (13:20 +0000)]
Decide whether to use_pager as early as possible,
in preparation for pledge(2); no functional change intended.
jasper [Sat, 10 Oct 2015 12:20:10 +0000 (12:20 +0000)]
add (currently failing) test for --exists foo,bar.
spotted by aja@
jung [Sat, 10 Oct 2015 12:05:47 +0000 (12:05 +0000)]
add location to the fan description
jung [Sat, 10 Oct 2015 11:57:20 +0000 (11:57 +0000)]
fix wrong brackets in if statement
jung [Sat, 10 Oct 2015 11:42:49 +0000 (11:42 +0000)]
three conversions of fgetln() to getline()
ok eric sunil
ajacoutot [Sat, 10 Oct 2015 09:45:15 +0000 (09:45 +0000)]
Fix /var/spool/smtpd/offline ownership and mode.
ok gilles@
lum [Sat, 10 Oct 2015 09:13:14 +0000 (09:13 +0000)]
Make functions that accept multiple iterations via C-u N, honour 0.
Except C-k which has a defined behaviour. In mg, C-t doesn't complete
n iterations if requested, but probably should, hence it has been
included in this diff.
lum [Sat, 10 Oct 2015 08:35:26 +0000 (08:35 +0000)]
Call onlywind() properly.
stsp [Sat, 10 Oct 2015 07:52:30 +0000 (07:52 +0000)]
Make tcpdump(1) print more information from the HT Capabilities element.
tweak + ok sthen@
stsp [Sat, 10 Oct 2015 07:51:47 +0000 (07:51 +0000)]
Add macros for A-MPDU and MCS data, both found in HT capabilities element.
typo fix + ok sthen@
nicm [Sat, 10 Oct 2015 07:38:18 +0000 (07:38 +0000)]
Unused macros; from Michael McConville.
nicm [Sat, 10 Oct 2015 07:35:16 +0000 (07:35 +0000)]
Move more declarations out of proto.h into better headers, from Michael
McConville. No binary change.
deraadt [Sat, 10 Oct 2015 05:47:54 +0000 (05:47 +0000)]
Some of these large so easy to contain, with "stdio rpath".
ok doug
deraadt [Sat, 10 Oct 2015 05:43:48 +0000 (05:43 +0000)]
simple program using "stdio rpath"
ok doug
doug [Sat, 10 Oct 2015 05:35:22 +0000 (05:35 +0000)]
Add pledge support to cmp(1).
This is a simple case of using "stdio rpath" until all files are opened and
then dropping down to "stdio" since it includes "rw" on open fds.
ok deraadt@
deraadt [Sat, 10 Oct 2015 05:32:52 +0000 (05:32 +0000)]
pledge to only use "stdio rpath"; ok doug
doug [Sat, 10 Oct 2015 05:26:57 +0000 (05:26 +0000)]
Add pledge support to getent(1).
This pledges the superset of all requests for the various getent databases
and then drops to the minimum for the chosen database.
ok deraadt@
renato [Sat, 10 Oct 2015 05:12:33 +0000 (05:12 +0000)]
Make use of pledge(2).
ok deraadt
renato [Sat, 10 Oct 2015 05:09:19 +0000 (05:09 +0000)]
Move some interface initialization bits from if_init() to eigrp_if_start()
and call if_init() only during the startup of the eigrpe process.
renato [Sat, 10 Oct 2015 05:07:10 +0000 (05:07 +0000)]
Remove attached neighbors whenever an interface is disabled to speedup
the convergence process.
renato [Sat, 10 Oct 2015 05:06:00 +0000 (05:06 +0000)]
eigrpctl pledges to use stdio and route.
ok deraadt
renato [Sat, 10 Oct 2015 05:03:39 +0000 (05:03 +0000)]
Fix detection of interface up/down events.
benno [Sat, 10 Oct 2015 00:37:40 +0000 (00:37 +0000)]
relayctl pledges to use stdio only
ok reyk@ sure deraadt@
benno [Sat, 10 Oct 2015 00:19:52 +0000 (00:19 +0000)]
pledge("stdio route") needed here, because ipv6
ok deraadt@
benno [Sat, 10 Oct 2015 00:16:23 +0000 (00:16 +0000)]
relayd's ca process pledges to only use stdio.
ok deraadt@
deraadt [Sat, 10 Oct 2015 00:10:07 +0000 (00:10 +0000)]
ksh can run with pledge "stdio rpath wpath cpath getpw fattr proc exec tty"
if the mknod builtin is disabled. It looks like a lot of abilities, but
hey, this is a shell. can't open sockets or do other nasty stuff though.
(we'll leave the mknod builtin enabled on the install media for now; there
is work happening to regain the MAKEDEV performance in a different way)
discussions with otto & millert in particular
deraadt [Fri, 9 Oct 2015 23:55:03 +0000 (23:55 +0000)]
Allow kill(self, sig) in pledge SELF also. the stack protector, abort(),
and readpassphrase() in particular use this.
ok millert tedu semarie
deraadt [Fri, 9 Oct 2015 23:33:54 +0000 (23:33 +0000)]
can use pledge "stdio"; ok benno
tim [Fri, 9 Oct 2015 21:59:34 +0000 (21:59 +0000)]
Remove telnet warnings. Civilization has reached a point where they are no
longer relevant.
OK millert@
tobias [Fri, 9 Oct 2015 21:24:05 +0000 (21:24 +0000)]
The variable errmsg can be static in main.c if code in re.c uses an own
buffer to construct error messages.
with input by and ok millert@
tobias [Fri, 9 Oct 2015 20:27:28 +0000 (20:27 +0000)]
Define functions as static when they are not used outside their own c-files.
ok millert@
tim [Fri, 9 Oct 2015 20:24:37 +0000 (20:24 +0000)]
Use __progname rather than argv[0]; OK millert@
tim [Fri, 9 Oct 2015 20:14:35 +0000 (20:14 +0000)]
Replace readpass(3) with readpassphrase(3). This was the only use of
readpass(3) in base...
OK millert@
millert [Fri, 9 Oct 2015 19:49:08 +0000 (19:49 +0000)]
Lexer states are not needed outside of lex.c.
From mksh via Michael McConville
millert [Fri, 9 Oct 2015 19:47:02 +0000 (19:47 +0000)]
Mark static globals that are only used in their respective .c files.
Also make stdin unbuffered since that is the same as using a
single-byte buffer. OK tobias@
tedu [Fri, 9 Oct 2015 19:36:27 +0000 (19:36 +0000)]
remove null check before afree. from Michael McConville
rpe [Fri, 9 Oct 2015 18:30:54 +0000 (18:30 +0000)]
Exit autoinstall in case of an invalid choice.
OK krw@
bluhm [Fri, 9 Oct 2015 17:51:08 +0000 (17:51 +0000)]
Keep relayd test certificate names in sync with syslogd.
gilles [Fri, 9 Oct 2015 17:44:25 +0000 (17:44 +0000)]
upon smtpd restart, when scanning the offline queue, unlink 0-sized offline
messages as they are left-overs from an errored enqueue.
ok millert@, ok eric@
deraadt [Fri, 9 Oct 2015 17:18:20 +0000 (17:18 +0000)]
Have not come up with a great pattern for flock() yet. flock() is permitted
by "getpw" because libc getpw*/getgr* use open() of /var/run/ypbind.lock plus
flock() to detect YP running. The kernel observes this dance to "open up" the
YP door (ugliness should drive us to rewrite this mechanism from SunOS later).
however, flock is also used independently. Current users are
htpasswd mail skeyinit tmux authpf pwd_mkdb ldapd smtpd ypbind
login_token mail.local lockspool
Let's enable flock() for "cpath", and see if that helps these programs,
otherwise we'll try "wpath" next.
deraadt [Fri, 9 Oct 2015 17:09:06 +0000 (17:09 +0000)]
With nfs spool (fork + seteuid/setuid balony) support gone, it becomes
possible to pledge "stdio rpath wpath tty proc"
Noone uses this code anymore. This is a demonstration...
deraadt [Fri, 9 Oct 2015 17:07:21 +0000 (17:07 +0000)]
remove NFS spool support; it stands in the way of pledge(2)
bluhm [Fri, 9 Oct 2015 17:07:06 +0000 (17:07 +0000)]
Add tests for syslogd TLS accept and receive encrypted messages.
bluhm [Fri, 9 Oct 2015 16:58:25 +0000 (16:58 +0000)]
If syslogd is started with -S, it accepts TLS connections to receive
encrypted messages. The server certificates are taken from /etc/ssl
like relayd does.
OK benno@ beck@ deraadt@
sunil [Fri, 9 Oct 2015 16:47:14 +0000 (16:47 +0000)]
Convert fgetln(3) to getline(3).
Ok eric@ todd@ gilles@
bluhm [Fri, 9 Oct 2015 16:44:55 +0000 (16:44 +0000)]
A fork(2) is used in ttymsg() to delay the message to a tty if it
blocks. Fix the potential syslogd's death, add "proc" to pledge.
OK deraadt@
deraadt [Fri, 9 Oct 2015 16:29:17 +0000 (16:29 +0000)]
catch up to tame() -> pledge() rename
tedu [Fri, 9 Oct 2015 16:26:03 +0000 (16:26 +0000)]
pare down the readme so as to not imply we are tracking upstream.
nor do we much care about running this on dec ultrix anymore, etc...
ok deraadt
gilles [Fri, 9 Oct 2015 15:09:09 +0000 (15:09 +0000)]
if an error occurs during offline enqueuing after we've dropped group, then
attempt to ftruncate() the fp back to 0.
suggested and ok millert@, ok eric@
gilles [Fri, 9 Oct 2015 14:37:38 +0000 (14:37 +0000)]
turn our local enqueuer setgid _smtpq and restrict access to offline queue,
the enqueuer will revoke group and regain real gid right after mkstemp.
this would have prevented the symlink/hardlink attacks against offline, and
it will avoid having to deal with new ways users can mess with it.
ok eric@, ok millert@
kettenis [Fri, 9 Oct 2015 13:22:54 +0000 (13:22 +0000)]
Remove evil hack. I've never seen the printf fire, and xenocara no longer
contains any code that can manipulate the affected register directly.
ok jsg@
deraadt [Fri, 9 Oct 2015 12:20:18 +0000 (12:20 +0000)]
this cpp operates file using pledge "stdio rpath wpath cpath"
bluhm [Fri, 9 Oct 2015 12:07:32 +0000 (12:07 +0000)]
Tame syslogd privsep child with "stdio rpath unix inet recvfd".
With and OK deraadt@
deraadt [Fri, 9 Oct 2015 11:47:30 +0000 (11:47 +0000)]
oops, snuck into a syscalls sync; spotted by sthen
semarie [Fri, 9 Oct 2015 11:42:54 +0000 (11:42 +0000)]
regress pledge
add missing $OpenBSD$ header
semarie [Fri, 9 Oct 2015 11:38:39 +0000 (11:38 +0000)]
regress pledge: remove 'regenerate' target
semarie [Fri, 9 Oct 2015 11:38:05 +0000 (11:38 +0000)]
add "tty" regress for pledge
jmc [Fri, 9 Oct 2015 10:13:48 +0000 (10:13 +0000)]
correct Xr; from theo buehler
gilles [Fri, 9 Oct 2015 09:56:28 +0000 (09:56 +0000)]
if enhanced status class is not set, enhanced status code is never dumped
in disk envelope.
deraadt [Fri, 9 Oct 2015 07:54:28 +0000 (07:54 +0000)]
All commands seem to work fine with pledge "stdio" after the connect(),
direct source and symbol table inspection suggests it is good. The same
principle will likely apply to most of our network daemon *ctl programs,
since many are derived from ospfd. Still, each needs testing.
discussion about network daemons and ctl's has been mostly with renato
deraadt [Fri, 9 Oct 2015 07:39:56 +0000 (07:39 +0000)]
another tame(2), spotted by jmc
lum [Fri, 9 Oct 2015 07:27:56 +0000 (07:27 +0000)]
Fix line number bug when calling onlywind().
semarie [Fri, 9 Oct 2015 06:50:01 +0000 (06:50 +0000)]
hook pledge
semarie [Fri, 9 Oct 2015 06:44:13 +0000 (06:44 +0000)]
follow tame->pledge in regress
deraadt [Fri, 9 Oct 2015 06:10:57 +0000 (06:10 +0000)]
do not use weak; plus this dies next week
deraadt [Fri, 9 Oct 2015 05:55:58 +0000 (05:55 +0000)]
another stray )
deraadt [Fri, 9 Oct 2015 05:30:03 +0000 (05:30 +0000)]
shortcircuit TIOCGETA to directly return ENOTTY for non-ttys. It could
be called against a non-tty fd, so as to test "is this a tty". Discovered
by sthen and rob pierce at the same time.
deraadt [Fri, 9 Oct 2015 04:38:54 +0000 (04:38 +0000)]
oops, typo spotted in temporary .c file, by semarie
deraadt [Fri, 9 Oct 2015 04:13:34 +0000 (04:13 +0000)]
fix a gotcha in the connect refactoring, that could result in dropping
through and trying to bind failed v6 connects.
ok guenther
deraadt [Fri, 9 Oct 2015 03:54:53 +0000 (03:54 +0000)]
the ntp engine can run with "stdio inet proc". For many reasons,
including fork/exec cost, it would be better if constraints were
forked from the master process, which would then tell the ntp
engine. That would increase accuracy and security.
Lots of conversations with reyk and bcook
deraadt [Fri, 9 Oct 2015 03:50:40 +0000 (03:50 +0000)]
Once the constraint engine process is running, it only needs
"stdio inet". It took weeks to get to this point...
deraadt [Fri, 9 Oct 2015 02:44:22 +0000 (02:44 +0000)]
stardate 93370.16: a whitespace appears to have entered our quadrant...
deraadt [Fri, 9 Oct 2015 02:36:46 +0000 (02:36 +0000)]
multicast test backwards; noted by renato
deraadt [Fri, 9 Oct 2015 01:46:27 +0000 (01:46 +0000)]
sync
deraadt [Fri, 9 Oct 2015 01:37:06 +0000 (01:37 +0000)]
Change all tame callers to namechange to pledge(2).
deraadt [Fri, 9 Oct 2015 01:26:40 +0000 (01:26 +0000)]
tame -> pledge.
deraadt [Fri, 9 Oct 2015 01:24:57 +0000 (01:24 +0000)]
tame -> pledge conversion, in libc. I should crank libc, but am cheating
hoping things go well. The old symbol is faked via a stupid stub function,
until next major crank when it can be removed. I am expecting guenther
to scream at me.
deraadt [Fri, 9 Oct 2015 01:17:18 +0000 (01:17 +0000)]
Rename tame() to pledge(). This fairly interface has evolved to be more
strict than anticipated. It allows a programmer to pledge/promise/covenant
that their program will operate within an easily defined subset of the
Unix environment, or it pays the price.
deraadt [Fri, 9 Oct 2015 01:11:12 +0000 (01:11 +0000)]
sync
deraadt [Fri, 9 Oct 2015 01:10:27 +0000 (01:10 +0000)]
Rename tame() to pledge(). This fairly interface has evolved to be more
strict than anticipated. It allows a programmer to pledge/promise/covenant
that their program will operate within an easily defined subset of the
Unix environment, or it pays the price.
yasuoka [Thu, 8 Oct 2015 22:41:12 +0000 (22:41 +0000)]
After replacement alloca() with alloc(), out-of-heap happened when booting
on a large block size (32K) partition. Increase the HEAP_LIMIT from
0x90000 to 0xA0000.
try this, deraadt
guenther [Thu, 8 Oct 2015 20:13:45 +0000 (20:13 +0000)]
If getaddrinfo() succeeds, then don't try look ups with other flags, even
if the connect()s failed. In concert with some resolver fixes in libc,
this lets ntpd be tame()ed
problem isolated by theo, who had fun untangling the libc and libtls
behaviors to place blame for not being able to tame ntpd
ok beck@ deraadt@ jsing@
deraadt [Thu, 8 Oct 2015 17:29:43 +0000 (17:29 +0000)]
Expose a small set of multicast join operators under the request "mcast".
This will be used by a few daemons. If they lack this feature, then
they would need to operate without tame.
Discussed with renato
tedu [Thu, 8 Oct 2015 16:45:50 +0000 (16:45 +0000)]
add some tame calls. we may need a bunch of permissions to create files
and manipulate the tty for readpassphrase, but once we've parsed options
and have some idea of what's going to happen next, we can reduce down
quite a bit more. particular use case of "signify | patch" is limited to
feeding garbage to patch.
tedu [Thu, 8 Oct 2015 16:41:26 +0000 (16:41 +0000)]
stop trying to gift history files to the original owner. instead, don't
open history files that don't belong to us. probably much safer.
ok deraadt
kettenis [Thu, 8 Oct 2015 15:58:38 +0000 (15:58 +0000)]
Lock the page queues by turning uvm_lock_pageq() and uvm_unlock_pageq() into
mtx_enter() and mtx_leave() operations. Not 100% this won't blow up but
there is only one way to find out, and we need this to make progress on
further unlocking uvm.
prodded by deraadt@
tedu [Thu, 8 Oct 2015 15:54:59 +0000 (15:54 +0000)]
little cleanup from Michael McConville, mostly related to stale comments.
krw [Thu, 8 Oct 2015 14:50:38 +0000 (14:50 +0000)]
Refactor fileprefix() and filecopy() to use warn() instead of err()
to display error message, and to return error indications (NULL and
-1 respectively). Use the error indications in write_efisystem()
to unwind in the face of more error conditions. In other cases just
exit(1) to emulation current behaviour.
ok deraadt@
deraadt [Thu, 8 Oct 2015 14:49:27 +0000 (14:49 +0000)]
tame "stdio rpath wpath cpath proc exec". make is a shell, and appears
to only need these operations. Take note that "exec" is a 2-day old
tame request, so do get a new kernel before you update or risk getting
trapped.
tedu [Thu, 8 Oct 2015 14:46:05 +0000 (14:46 +0000)]
16 years after E801 memprobe was disabled, probably safe to delete it.
ok deraadt jung kettenis ratchov
visa [Thu, 8 Oct 2015 14:24:32 +0000 (14:24 +0000)]
Remove the sc_soft_req_cnt field because the number of tx requests is
already tracked in sc_sendq. Replace the sc_flush logic with a simple
Fetch-and-Add store that avoids an unnecessary IOBDMA transaction.
ok uebayasi@
jmc [Thu, 8 Oct 2015 14:09:34 +0000 (14:09 +0000)]
tweak previous;