mpi [Sun, 10 Dec 2017 11:25:18 +0000 (11:25 +0000)]
Remove outdated comment and trailing spaces.
From kshe59@zoho.eu
florian [Sun, 10 Dec 2017 10:07:54 +0000 (10:07 +0000)]
Move privileged initialization from frontend to main process.
Needed for future work where we will spin up children via
fork - privdrop - exec. Child processes will no longer come
up with root privileges.
jmc [Sun, 10 Dec 2017 09:17:24 +0000 (09:17 +0000)]
- add max-count to SYNOPSIS
- list long options with short, where they have an equivalent
- sync usage()
- minor tweaks
jmc [Sun, 10 Dec 2017 07:40:04 +0000 (07:40 +0000)]
add the zte mf831; from christoph r. murauer
while here, a little less capitalisation
jmc [Sun, 10 Dec 2017 07:34:38 +0000 (07:34 +0000)]
sort -r in the description list, and make its grammar match those
of the other options in this page;
dtucker [Sun, 10 Dec 2017 05:55:29 +0000 (05:55 +0000)]
Put remote client info back into the ClientAlive connection termination
message. Based in part on diff from lars.nooden at gmail, ok djm
jsg [Sun, 10 Dec 2017 04:21:55 +0000 (04:21 +0000)]
Enable the phy-supply regulator if present and use the phy id specified
in the device tree.
Patch from Artturi Alm who tested on a Miniand Hackberry.
deraadt [Sun, 10 Dec 2017 01:03:46 +0000 (01:03 +0000)]
-r restricted mode blocks certain operations as ~ operations.
from Jan Klemkow
ok nicm
pirofti [Sat, 9 Dec 2017 18:38:37 +0000 (18:38 +0000)]
Add support for the non-standard grep -m extension.
grep -m num stops after a maximum of num matches are found.
We support -m0 to match GNU behaviour, but we do not allow negative
numbers.
Manpage help from jmc@, OK deraadt@.
jsing [Sat, 9 Dec 2017 16:46:08 +0000 (16:46 +0000)]
Make tls_config_parse_protocols() work correctly when passed a NULL pointer
for a protocol string.
Issue found by semarie@, who also provided the diff.
jsing [Sat, 9 Dec 2017 16:43:09 +0000 (16:43 +0000)]
Add a regress test for tls_config_parse_protocols().
fcambus [Sat, 9 Dec 2017 16:39:54 +0000 (16:39 +0000)]
Add sizes for free() in the AMD PowerNow! K8 driver.
OK deraadt@, sthen@, visa@
krw [Sat, 9 Dec 2017 15:48:04 +0000 (15:48 +0000)]
Add ACTION_IGNORE and nuke ignored_options, ignored_option_count fields
in struct client_config.
jsing [Sat, 9 Dec 2017 14:34:09 +0000 (14:34 +0000)]
Use a test value that fits into a long on both 32-bit and 64-bit
architectures, so that the regress passes on both.
mikeb [Sat, 9 Dec 2017 14:00:21 +0000 (14:00 +0000)]
No need for the txb_used flag anymore
deraadt [Sat, 9 Dec 2017 07:16:51 +0000 (07:16 +0000)]
In the middle of CRYPTO_gcm128_finish() there is a complicated #ifdef
block which defines a variable late, after code. Place this chunk into
a { subblock } to satisfy old compilers and old eyes.
deraadt [Sat, 9 Dec 2017 07:09:25 +0000 (07:09 +0000)]
Please variable decl before code.
deraadt [Sat, 9 Dec 2017 06:50:32 +0000 (06:50 +0000)]
More precision in pledge sysctl report
deraadt [Sat, 9 Dec 2017 06:39:04 +0000 (06:39 +0000)]
recognize .openbsd.randomdata section and indicate it roughly, so
that objects within it are identified as being in read-only space.
ok guenther
deraadt [Sat, 9 Dec 2017 06:35:08 +0000 (06:35 +0000)]
Add a #define for the name of the .openbsd.randomdata section.
ok guenther
bluhm [Fri, 8 Dec 2017 22:10:34 +0000 (22:10 +0000)]
Pullup the mbuf before accessing the version field in the IP header.
Fix the pullup length of the shim header in mpls_do_error().
issue reported by Maxime Villard; OK deraadt@ claudio@
claudio [Fri, 8 Dec 2017 21:59:05 +0000 (21:59 +0000)]
Use m_freem() in error case. Found by Maxime Villard
OK bluhm@
claudio [Fri, 8 Dec 2017 21:56:22 +0000 (21:56 +0000)]
The adjttl functions use m_pullup(). In some cases m_pullup() can return
a new mbuf chain and this chain needs to be returned to the caller else
a use after free may happen.
Issue reported by Maxime Villard
OK bluhm@ deraadt@
claudio [Fri, 8 Dec 2017 21:52:49 +0000 (21:52 +0000)]
mpls_shim_pop() can return NULL. Check it else we end up dereferencing NULL.
Issue reported by Maxime Villard
OK bluhm@ deraadt@
stsp [Fri, 8 Dec 2017 21:16:01 +0000 (21:16 +0000)]
Add support for background scanning to net80211 and iwm(4).
The iwm(4) driver will now roam between access points which share an SSID.
Use 'ifconfig iwm0 debug' and 'tail -f /var/log/messages' to watch it do so.
Tested by several people in various iterations.
As usual, let me know if you run into issues.
ok phessler deraadt
deraadt [Fri, 8 Dec 2017 21:08:35 +0000 (21:08 +0000)]
The per-interface mpls flag should also also be tested on input before
proceeding, as described in ifconfig documentation. Discussion with claudio.
Related to a report from maxime.
ok claudio bluhm
stsp [Fri, 8 Dec 2017 20:55:46 +0000 (20:55 +0000)]
Make iwm(4) restore the ic_bss channel after tweaking this channel as part
of passing the frame to ieee80211_input(). Fixes a race where auth frames
would be sent on the wrong channel during association.
Problem reported by florian@
ok deraadt@
krw [Fri, 8 Dec 2017 20:17:28 +0000 (20:17 +0000)]
When accepting a BOOTP lease, do not leak offered values of lease
time, renewal time, rebinding time. Use the ACTION_DEFAULT mechanism
to set the default expiry time for any lease so gauche as to not
provide one. Use the DHCP default lease times for BOOTP leases instead
of, bizarrely, shorter times.
kettenis [Fri, 8 Dec 2017 19:05:33 +0000 (19:05 +0000)]
Make sure we don't match (and attach) more than the maximum number of
supported CPUs.
martijn [Fri, 8 Dec 2017 18:41:59 +0000 (18:41 +0000)]
Make the r command filename obligatory, similar to what FreeBSD and NetBSD
do for several years.
While here make corresponding error message for missing read and write file
consistent between commands/flag, and shrink the the code of the w flag of
the s command by making it use the same code as the w command.
Prompted by a larger diff by kshe59 <at> zoho <dot> eu
OK millert@
deraadt [Fri, 8 Dec 2017 17:51:26 +0000 (17:51 +0000)]
repair columns; from kshe59@zohu.eu
millert [Fri, 8 Dec 2017 17:26:42 +0000 (17:26 +0000)]
Add missing length checks to make sure we don't dereference a pointer
past the mmap(2)'d buffer. Otherwise, locate will read a single
byte past the end of the buffer. This is often harmless, but if
the length of the buffer is an even multiple of the page size,
locate will crash. OK tb@ espie@ deraadt@
deraadt [Fri, 8 Dec 2017 17:04:14 +0000 (17:04 +0000)]
Convert snprintf+write into dprintf. It is simply easier to read, and
provides retry on short-write file descriptors.
ok florian, previous versions seen by millert
jasper [Fri, 8 Dec 2017 09:45:05 +0000 (09:45 +0000)]
remove description of 'at' field which was removed from vmstat.c -r1.21
mpi [Fri, 8 Dec 2017 08:54:03 +0000 (08:54 +0000)]
Disable SMAP temporarily to display userland traces.
ok visa@, deraadt@, mlarkin@, jasper@
deraadt [Fri, 8 Dec 2017 05:30:16 +0000 (05:30 +0000)]
also cleanout ld.so.a
deraadt [Fri, 8 Dec 2017 05:25:20 +0000 (05:25 +0000)]
Everyone knows this as ld.so, nor by the ancient name rtld.
ok guenther
deraadt [Fri, 8 Dec 2017 03:45:52 +0000 (03:45 +0000)]
time_t printing needs %lld and (long long) casts
ok djm
djm [Fri, 8 Dec 2017 02:14:33 +0000 (02:14 +0000)]
fix ordering in previous to ensure errno isn't clobbered before
logging.
djm [Fri, 8 Dec 2017 02:13:02 +0000 (02:13 +0000)]
for some reason unix_listener() logged most errors twice with each
message containing only some of the useful information; merge these
kevlo [Fri, 8 Dec 2017 01:03:51 +0000 (01:03 +0000)]
Fix the return value of fwscanf(3) when encountering an early matching
failure. This change brings fwscanf(3) back in line with fscanf(3).
From FreeBSD; ok deraadt@, millert@
patrick [Thu, 7 Dec 2017 22:47:28 +0000 (22:47 +0000)]
Change the SA payload parser to parse more than the first proposal. This
allows us to select one of the peer's proposals (and not only the first).
ok sthen@ hshoexer@
jmc [Thu, 7 Dec 2017 21:47:22 +0000 (21:47 +0000)]
adjust the "timeout" text, now that leases are no longer
specified in dhclient.conf;
ok krw
deraadt [Thu, 7 Dec 2017 20:38:15 +0000 (20:38 +0000)]
correct indentation
krw [Thu, 7 Dec 2017 19:17:13 +0000 (19:17 +0000)]
Set ifi->offer to NULL after free'ing it.
krw [Thu, 7 Dec 2017 19:03:15 +0000 (19:03 +0000)]
Nuke support for defining last-gasp leases in dhclient.conf.
mikeb [Thu, 7 Dec 2017 16:52:21 +0000 (16:52 +0000)]
Initialize tcp_secret in tcp_init
The initialization of a secret SHA256 context for generating TCP
initial sequence numbers is moved out of tcp_set_iss_tsm used to
set up ISN for new connections and into tcp_init, sparing the
need for a global flag.
OK deraadt, visa, mpi
schwarze [Thu, 7 Dec 2017 15:43:03 +0000 (15:43 +0000)]
* clean up macro usage: use .Ar for command arguments, .Cm for fixed
strings to be used in commands, and .Li for example strings
* clarify what uses the two environment variables, and that only these
two are ignored for issetugid(2) programs
triggered by a question from Jan Stary <hans at stare dot cz>
feedback and OK ratchov@
stsp [Thu, 7 Dec 2017 14:13:05 +0000 (14:13 +0000)]
In iwm(4), keep scanning if net80211 asks for a SCAN->SCAN transition
and the firmware is no longer busy scanning.
Fixes 'ifconfig iwm0 scan' returning no results after resume far away
from the currently configured AP.
ok phessler@
stsp [Thu, 7 Dec 2017 14:12:39 +0000 (14:12 +0000)]
Make iwm_newstate() recover from state transition errors.
On error, we now schedule the init task which will whack the interface when
it gets to run, which prevents the driver from stalling in such situations.
ok phessler@
mpi [Thu, 7 Dec 2017 11:44:02 +0000 (11:44 +0000)]
Set the correct ENCAPSULATION_MODE when doing NAT-T.
Fix at least interoperability with Cisco when isakmpd(8) is initiating
the connections, originally reported by sebastia@ in 2014.
Refreshed diff from and ok hshoexer@, ok sthen@, ok remi@
martijn [Thu, 7 Dec 2017 09:52:26 +0000 (09:52 +0000)]
Make the command formatting more consistent.
s/with/width type-O fix while here.
From kshe59 <at> zoho <dot> eu
OK jmc@
otto [Thu, 7 Dec 2017 06:34:05 +0000 (06:34 +0000)]
zap a few stray backslashes from the time thesse two were #defines; from
Ilya Kaliman
zhuk [Thu, 7 Dec 2017 05:21:57 +0000 (05:21 +0000)]
Now that we have RB_NFIND, the canacar's trick with RB_INSERT+RB_NEXT
is not needed anymore.
okay jmatthew@
zhuk [Thu, 7 Dec 2017 05:09:27 +0000 (05:09 +0000)]
client_addr_init() never fails and its return value is never checked,
so just make it void.
okay jmatthew@
zhuk [Thu, 7 Dec 2017 05:06:08 +0000 (05:06 +0000)]
Fix a potential fd leak in client_aldap_open().
okay jmatthew@
krw [Thu, 7 Dec 2017 02:08:44 +0000 (02:08 +0000)]
Let it compile again.
mlarkin [Thu, 7 Dec 2017 01:54:39 +0000 (01:54 +0000)]
reorder some port numbers, no functional change.
tb [Thu, 7 Dec 2017 01:54:33 +0000 (01:54 +0000)]
Drop the unused second argument from set_prompt(). It used to be used for
early special casing of ! and !! in the PS1 expansion. This was removed
from set_prompt() as part of the implementaion of the character count
toggles \[ and \] back in 2004.
ok jca
espie [Wed, 6 Dec 2017 17:15:43 +0000 (17:15 +0000)]
be consistent in where we call fflush
okay millert@
fcambus [Wed, 6 Dec 2017 16:26:12 +0000 (16:26 +0000)]
Update inaccurate comment: i386_has_xcrypt => amd64_has_xcrypt.
kettenis [Wed, 6 Dec 2017 16:20:53 +0000 (16:20 +0000)]
Implement support for using interrupt cookies in vbus(4) and vpci(4) as
introduced in version 3.0 of the Interrupt APIs group. This makes it possible
boot OpenBSD on SPARC T7/M7 hardware (although there still may be issues with
the onboard mpii(4) controller).
krw [Wed, 6 Dec 2017 13:57:27 +0000 (13:57 +0000)]
Disentangle dhclient.conf static lease handling from dynamic
lease handling. Simplifies code and makes it easier to
consider excising this 'feature'.
otto [Wed, 6 Dec 2017 13:48:05 +0000 (13:48 +0000)]
strdup -> bstrdup; from Michael W. Bombardieri
abieber [Wed, 6 Dec 2017 13:29:02 +0000 (13:29 +0000)]
Make vmd respect owner when starting non-disabled vms.
OK pd@, benno@
zhuk [Wed, 6 Dec 2017 12:07:08 +0000 (12:07 +0000)]
It's the imsg_compose(3) who accepts 'fd' argument, not imsg_create(3).
djm [Wed, 6 Dec 2017 05:06:21 +0000 (05:06 +0000)]
don't accept junk after "yes" or "no" responses to hostkey prompts.
bz#2803 reported by Maksim Derbasov; ok dtucker@
dtucker [Tue, 5 Dec 2017 23:59:47 +0000 (23:59 +0000)]
Replace atoi and strtol conversions for integer arguments to config
keywords with a checking wrapper around strtonum. This will prevent
and flag invalid and negative arguments to these keywords. ok djm@
dtucker [Tue, 5 Dec 2017 23:56:07 +0000 (23:56 +0000)]
Add missing break for rdomain. Prevents spurious "Deprecated option"
warnings. ok djm@
kettenis [Tue, 5 Dec 2017 22:27:54 +0000 (22:27 +0000)]
regen
kettenis [Tue, 5 Dec 2017 22:26:31 +0000 (22:26 +0000)]
Add SPARC-M7 PCIe; rename existing SPARC PCIe entries.
kettenis [Tue, 5 Dec 2017 21:11:10 +0000 (21:11 +0000)]
Remove DEF_STRONG(__cxa_thread_atexit_impl). This produces an unwanted
_libc___cxa_thread_atexit_impl reference on gcc architectures that breaks
the build.
kettenis [Tue, 5 Dec 2017 21:04:32 +0000 (21:04 +0000)]
Handle 64-bit-address Memory Space.
jca [Tue, 5 Dec 2017 20:31:45 +0000 (20:31 +0000)]
Use clock_gettime(CLOCK_MONOTONIC) to schedule timers
From Scott Cheloha, ok tb@
jmc [Tue, 5 Dec 2017 19:40:16 +0000 (19:40 +0000)]
add missing blank before punctuation;
deraadt [Tue, 5 Dec 2017 18:36:02 +0000 (18:36 +0000)]
sync
espie [Tue, 5 Dec 2017 17:58:10 +0000 (17:58 +0000)]
document PORTS_PRIVSEP now that it's nearing completion
millert [Tue, 5 Dec 2017 17:47:09 +0000 (17:47 +0000)]
Fix a case where we could go off the end of the buffer.
Crash found by Sergey Bronnikov using afl-fuzz.
Based on a diff from and OK by espie@
visa [Tue, 5 Dec 2017 15:39:26 +0000 (15:39 +0000)]
Show board ID and revision in dmesg to ease the identification
of system model. The early boot code already prints them, but
that output is not buffered and tends to be left out from
dmesg submissions.
visa [Tue, 5 Dec 2017 15:26:47 +0000 (15:26 +0000)]
Drop cn30xxpow_intr_establish(), cn30xxpow_intr() and some other
unused code. POW interrupts are now handled in if_cnmac.c.
jca [Tue, 5 Dec 2017 15:02:06 +0000 (15:02 +0000)]
Seperate real and user timer interfaces
Use more descriptive names, and make it clearer that real and user
timers work on different static storage. The end goal is to be able to
reuse those timer functions, instead of inlining other timer
implementations subject to clock jumps.
Discussed with Scott Cheloha
krw [Tue, 5 Dec 2017 14:57:14 +0000 (14:57 +0000)]
When removing duplicate dynamic leases from the cache, compare the
SSID against ifi->ssid, not the SSID of the new lease. They
should be the same, but this makes the intent clearer and removes
an assumption about the contents of ifi->active.
espie [Tue, 5 Dec 2017 14:35:12 +0000 (14:35 +0000)]
strip_comments is also called for dot lines, so sometimes the comment
is all the line.
problem reported by Sergey Bronnikov
otto [Tue, 5 Dec 2017 14:05:22 +0000 (14:05 +0000)]
use a global BN_CTX; from kshe with a twist from myself
kettenis [Tue, 5 Dec 2017 13:45:31 +0000 (13:45 +0000)]
Implement __cxa_thread_atexit to support C++11 thread_local scope. The
interface is also made available as __cxa_thread_atexit_impl to satisfy the
needs of GNU libstdc++.
ok guenther@, millert@
patrick [Tue, 5 Dec 2017 09:06:53 +0000 (09:06 +0000)]
When sending out a proposal we create an SA/SPI for the Child SAs if we
are an initiator and store the information on the proposal, because we
only had one proposal so far. This changes the code to only create one
SA on the first proposal and then apply the SPI to all other proposals
as well.
ok markus@
deraadt [Tue, 5 Dec 2017 06:05:39 +0000 (06:05 +0000)]
adjust for warning: warning repairs
djm [Tue, 5 Dec 2017 01:30:19 +0000 (01:30 +0000)]
include the addr:port in bind/listen failure messages
kettenis [Mon, 4 Dec 2017 22:06:21 +0000 (22:06 +0000)]
Sadly the firmware on SPARC T7 machines omits the executable bit on mappings
that contain code, just like some sun4u machines. So add it for sun4v as
well when entering the mappings into the kernel pmap.
mpi [Mon, 4 Dec 2017 21:12:41 +0000 (21:12 +0000)]
Use _kernel_lock_held() instead of __mp_lock_held(&kernel_lock).
Missed in previous, found by deraadt@, ok visa@
patrick [Mon, 4 Dec 2017 17:22:39 +0000 (17:22 +0000)]
Remove duplicate check that never could execute because the exact same
condition is handled a line before.
patrick [Mon, 4 Dec 2017 17:03:43 +0000 (17:03 +0000)]
Consistently log "malformed payload" instead of "payload malformed", and
replace "minimal" with "minimum".
patrick [Mon, 4 Dec 2017 16:57:40 +0000 (16:57 +0000)]
Remove check that is now a duplicate due to recent refactoring.
patrick [Mon, 4 Dec 2017 16:52:16 +0000 (16:52 +0000)]
The payloads are layered like onions, so you can validate one layer and
then call the next one, which can then validate itself. Thing is, most
layers try to run validations on the upper layer, which is not useful
and rather confusing. This cleans it up.
First change is that the generic payload parser does not anymore pass
the length of the whole datagram, including all remaining payloads, but
passes only the length of the specific payload to the specific payload
parser. Second change is that the payload validators don't check the
length of the upper layer, but only verify their own lengths.
Diff discussed with hshoexer@ and sthen@
Tested by sthen@
bluhm [Mon, 4 Dec 2017 15:13:12 +0000 (15:13 +0000)]
RFC 4861 requires that all neighbor discovery packets have 255 in
their IPv6 header hop limit field. Let pf drop neighbor solicitation,
neighbor advertisement, router solicitation, router advertisement,
and redirect ICMP6 packets that do not comply. This enforces that
bogus packets cannot be routed when pf is enabled.
OK mpi@ sashan@ benno@
tb [Mon, 4 Dec 2017 15:00:03 +0000 (15:00 +0000)]
change "if (lp &&" to "if lp != NULL &&".
pointed out by krw
robert [Mon, 4 Dec 2017 14:56:47 +0000 (14:56 +0000)]
add the alps touchpad that can be found in the ideapad 710s
ok mpi@
tb [Mon, 4 Dec 2017 14:56:07 +0000 (14:56 +0000)]
Avoid a NULL-deref in get_recorded_lease() leading to a segfault seen
by aja and others.
ok krw
patrick [Mon, 4 Dec 2017 14:35:03 +0000 (14:35 +0000)]
Initialize variable, otherwise the pointer might contain stack garbage.
bluhm [Mon, 4 Dec 2017 13:40:34 +0000 (13:40 +0000)]
Make divert lookup similar for all socket types. If PF_TAG_DIVERTED
is set, pf_find_divert() cannot fail so put an assert there.
Explicitly check all possible divert types, panic in the default
case. For raw sockets call pf_find_divert() before of the socket
loop. Divert reply should not match on TCP or UDP listen sockets.
OK sashan@ visa@