eric [Mon, 25 Nov 2019 12:11:26 +0000 (12:11 +0000)]
use crlf line-ending during the internal smtp session
ok gilles@ martijn@
mpi [Mon, 25 Nov 2019 11:33:51 +0000 (11:33 +0000)]
Convert infinite sleeps to tsleep_nsec(9).
ok bluhm@, cheloha@
mpi [Mon, 25 Nov 2019 11:32:17 +0000 (11:32 +0000)]
Convert to tsleep_nsec(9) and use the correct timeout period in msec.
Tested by kevlo@, ok stsp@
djm [Mon, 25 Nov 2019 10:32:35 +0000 (10:32 +0000)]
unbreak tests for recent security key changes
djm [Mon, 25 Nov 2019 10:23:36 +0000 (10:23 +0000)]
redundant test
djm [Mon, 25 Nov 2019 06:53:04 +0000 (06:53 +0000)]
unbreak after security key support landed
djm [Mon, 25 Nov 2019 00:57:51 +0000 (00:57 +0000)]
document the "no-touch-required" certificate extension;
ok markus, feedback deraadt
djm [Mon, 25 Nov 2019 00:57:27 +0000 (00:57 +0000)]
Print a key touch reminder when generating a security key. Most keys
require a touch to authorize the operation.
djm [Mon, 25 Nov 2019 00:55:58 +0000 (00:55 +0000)]
allow "ssh-keygen -x no-touch-required" when generating a security key
keypair to request one that does not require a touch for each
authentication attempt. The default remains to require touch.
feedback deraadt; ok markus@
djm [Mon, 25 Nov 2019 00:54:23 +0000 (00:54 +0000)]
add a "no-touch-required" option for authorized_keys and a similar
extension for certificates. This option disables the default
requirement that security key signatures attest that the user touched
their key to authorize them.
feedback deraadt, ok markus
djm [Mon, 25 Nov 2019 00:52:46 +0000 (00:52 +0000)]
Add a sshd_config PubkeyAuthOptions directive
This directive has a single valid option "no-touch-required" that
causes sshd to skip checking whether user presence was tested before
a security key signature was made (usually by the user touching the
key).
ok markus@
djm [Mon, 25 Nov 2019 00:51:37 +0000 (00:51 +0000)]
Add new structure for signature options
This is populated during signature verification with additional fields
that are present in and covered by the signature. At the moment, it is
only used to record security key-specific options, especially the flags
field.
with and ok markus@
djm [Mon, 25 Nov 2019 00:38:17 +0000 (00:38 +0000)]
memleak in error path
nicm [Sun, 24 Nov 2019 18:37:23 +0000 (18:37 +0000)]
Only substitute patterns starting with ^ once.
florian [Sun, 24 Nov 2019 14:05:39 +0000 (14:05 +0000)]
revert previous, has not been thought through
florian [Sun, 24 Nov 2019 11:15:04 +0000 (11:15 +0000)]
Show the build date of the (about to be) fetched snapshot / release.
OK phessler, pamela, jung, benno
inoguchi [Sun, 24 Nov 2019 07:58:00 +0000 (07:58 +0000)]
Add test for cms operations in appstest.sh
claudio [Sun, 24 Nov 2019 07:56:03 +0000 (07:56 +0000)]
If the RTM_PROPOSAL is a solicitation proposal forward the request to
all interfaces. Most handlers will ignore it but at least umb(4) will
send a response back.
OK florian@
claudio [Sun, 24 Nov 2019 07:54:30 +0000 (07:54 +0000)]
Add umb_rtrequest() to handle RTM_PROPOSAL events and send back the
DNS proposals in that case. Also clear the DNS proposals when the
interface is deconfigured (e.g. when going down).
OK florian@
claudio [Sun, 24 Nov 2019 07:50:55 +0000 (07:50 +0000)]
Make it possible to call ether_rtrequest with a NULL rt argument.
In that case the function can just return. Part of a larger diff
to use the if_rtrequest functions for RTM_PROPOSAL info.
OK florian@
jsg [Sun, 24 Nov 2019 02:29:43 +0000 (02:29 +0000)]
correct return type of sign_extend64()
claudio [Sat, 23 Nov 2019 19:47:04 +0000 (19:47 +0000)]
Move srtdnstosa() static inline function under #ifdef _KERNEL. This way
3rd party software stuck with c90 will still compile. Quick fix since
RTM_PROPOSAL will most porbably change later on.
Reported by naddy and aja
cheloha [Sat, 23 Nov 2019 17:22:10 +0000 (17:22 +0000)]
envy(4): *sleep(9) -> *sleep_nsec(9); ok ratchov@
krw [Sat, 23 Nov 2019 17:10:13 +0000 (17:10 +0000)]
Consistently use ISSET() to check for set flags.
remi [Sat, 23 Nov 2019 15:05:21 +0000 (15:05 +0000)]
These dependon related lines fit better a bit further up.
No functional change.
Suggested and OK claudio@
krw [Sat, 23 Nov 2019 12:27:32 +0000 (12:27 +0000)]
Consistently use !ISSET() to check for unset flags.
florian [Sat, 23 Nov 2019 08:57:52 +0000 (08:57 +0000)]
plug memory leak
florian [Sat, 23 Nov 2019 08:17:39 +0000 (08:17 +0000)]
On startup withdraw all proposals for all interfaces by sending an
empty proposal with if_index 0.
It is possible that the set of autoconf interfaces changed between a
stop -> start transition and slaacd would never withdraw nameservers
for interfaces that had the autoconf flag previously but no longer
after the restart.
florian [Sat, 23 Nov 2019 08:17:10 +0000 (08:17 +0000)]
An if_index of zero signals to remove all previous proposals from a
daemon. Soon to be used by slaacd(8) which handles all interfaces but
can't know which interfaces had been set to autoconf before a restart.
krw [Sat, 23 Nov 2019 01:16:05 +0000 (01:16 +0000)]
Consistently use SET() to set bits.
krw [Fri, 22 Nov 2019 22:45:52 +0000 (22:45 +0000)]
Polish logic talking to unwind.
ok florian@
otto [Fri, 22 Nov 2019 20:09:09 +0000 (20:09 +0000)]
Use a table for unbound options; ok florian@
bluhm [Fri, 22 Nov 2019 15:59:53 +0000 (15:59 +0000)]
Sync tests with current NetBSD. Enable t_mkfifo test.
from Moritz Buhl
krw [Fri, 22 Nov 2019 15:34:29 +0000 (15:34 +0000)]
Be consistent and always use CLR() to clear flags.
florian [Fri, 22 Nov 2019 15:32:42 +0000 (15:32 +0000)]
DNS Proposals are treated as replacements in unwind now.
To withdraw a proposal sent an empty list of nameservers.
This one is a bit of a hack by me to keep everything in sync for now.
A better version from krw is coming soon.
florian [Fri, 22 Nov 2019 15:31:25 +0000 (15:31 +0000)]
Treat DNS proposals as replacement of previous proposals from the same
daemon and interface. An empty list is automatically a withdrawl.
prodding deraadt@
florian [Fri, 22 Nov 2019 15:30:00 +0000 (15:30 +0000)]
Merge DNS proposals coming from the same interface.
That way we no longer need to handle withdraws in unwind. A new
proposal from a daemon on an interface is always a replacement.
If nameservers expire or an interface goes away we sent an empty list
of nameservers.
pestering^Wprodding deraadt
florian [Fri, 22 Nov 2019 15:28:05 +0000 (15:28 +0000)]
The DNS proposal list can (soon) be empty to signal a withdraw, relax
the size constraint to allow this to pass through the kernel.
Looks good to deraadt@
dtucker [Fri, 22 Nov 2019 06:50:30 +0000 (06:50 +0000)]
Wait for FD to be readable or writeable during a nonblocking connect,
not just readable. Prevents a timeout when the server doesn't immediately
send a banner (eg multiplexers like sslh) but is also slightly quicker
for other connections since, unlike ssh1, ssh2 doesn't specify that the
client should parse the server banner before sending its own. Patch
from mnissler@chromium.org, ok djm@
claudio [Fri, 22 Nov 2019 06:31:30 +0000 (06:31 +0000)]
sndiod(8) reopens audio interfaces on SIGHUP, which makes a lot of sense
when -F is used. Because of this allow rc.d script to reload sndiod.
OK kn ratchov aja
claudio [Fri, 22 Nov 2019 06:26:00 +0000 (06:26 +0000)]
Add RTP_PROPOSAL_UMB to the priorities list
claudio [Fri, 22 Nov 2019 06:24:09 +0000 (06:24 +0000)]
Teach route(8) about RTP_PROPOSAL_UMB
claudio [Fri, 22 Nov 2019 06:22:30 +0000 (06:22 +0000)]
Use rtm_proposal() to send out DNS updates. Both when DNS are added and
when the interface is deconfigured (when going down).
claudio [Fri, 22 Nov 2019 06:20:15 +0000 (06:20 +0000)]
Add rtm_proposal, a function to send out RTM_PROPOSAL messages from the
kernel. Will be used to have umb(4) inform unwind(8) about DNS changes.
OK bluhm@ tested by florian@ and deraadt@
krw [Thu, 21 Nov 2019 23:22:14 +0000 (23:22 +0000)]
Be consistent and always use [!]ISSET() to test flags in xs->flags.
krw [Thu, 21 Nov 2019 22:31:26 +0000 (22:31 +0000)]
Be consistent and always use SET() to set flags in xs->flags.
tim [Thu, 21 Nov 2019 20:02:20 +0000 (20:02 +0000)]
A touch of style(9)
OK tb@ tedu@
tim [Thu, 21 Nov 2019 20:01:06 +0000 (20:01 +0000)]
Use explicit_bzero() to clear key material
OK tb@ tedu@
florian [Thu, 21 Nov 2019 19:27:33 +0000 (19:27 +0000)]
We do not send lifetimes or source in DNS proposals.
Do not copy them to the main process.
florian [Thu, 21 Nov 2019 19:26:02 +0000 (19:26 +0000)]
output RTM_PROPOSALs more similar to other route messages
no objection krw@ who has further suggestions for improvement
akoshibe [Thu, 21 Nov 2019 17:24:15 +0000 (17:24 +0000)]
Fix comment typo: specification
tedu [Thu, 21 Nov 2019 16:13:39 +0000 (16:13 +0000)]
zero tmpout too. reminded by related diff from tim
tedu [Thu, 21 Nov 2019 16:07:24 +0000 (16:07 +0000)]
overwrite the key in failure modes in case the caller doesn't check.
ok deraadt
dlg [Thu, 21 Nov 2019 07:30:28 +0000 (07:30 +0000)]
rename struct tun_softc members so they're prefixed with "sc".
this makes tun more consistent with more of our drivers.
akoshibe [Thu, 21 Nov 2019 06:54:25 +0000 (06:54 +0000)]
Remove duplicate #define OFP_ALIGNMENT
akoshibe [Thu, 21 Nov 2019 06:22:57 +0000 (06:22 +0000)]
Remove arguments from ofp13_flowmod() which are unused and unnecessary.
ok guenther@
tb [Thu, 21 Nov 2019 05:18:47 +0000 (05:18 +0000)]
Remove workaround for broken 'openssl rsa -text' output that
was fixed in libcrypto/rsa/rsa_ameth.c r1.24.
ok dtucker inoguchi
florian [Thu, 21 Nov 2019 05:01:22 +0000 (05:01 +0000)]
Track on which interface nameservers have been learned.
prodding deraadt@
florian [Thu, 21 Nov 2019 03:55:22 +0000 (03:55 +0000)]
Configure the default route with RTP_NONE so that the kernel assigns
different priorities for wired vs. wireless interfaces for us.
Gleaned from dhclient.
brynet [Wed, 20 Nov 2019 20:44:22 +0000 (20:44 +0000)]
ksmn(4): Improve the precision of the fractional part of temperatures.
Before we were losing some precision in the raw value sensor conversion.
raw: 478 before: 59.70 after: 59.75 real: 59.750
raw: 479 before: 59.80 after: 59.88 real: 59.875
raw: 480 before: 60.00 after: 60.00 real: 60.000
Tested on Ryzen 7 2700X, Ryzen 5 2500U, and Zen2 by nte.
Diff and feedback from nte <lists at nte.email>
kn [Wed, 20 Nov 2019 18:11:28 +0000 (18:11 +0000)]
Honour DEBUG
OK deraadt
florian [Wed, 20 Nov 2019 18:10:12 +0000 (18:10 +0000)]
Form DNS proposals from router advertisements with a router lifetime
of zero. It just means that the router is not a default router.
This is also true for prefixes, but not yet implemented.
(The diff looks complicated but it's only a } moved up + reindent.)
patrick [Wed, 20 Nov 2019 16:34:58 +0000 (16:34 +0000)]
Some Intel SD/MMC controllers do not like the voltage being
set to 0V, so add a quirk for Apollo Lake, Gemini Lake and
100 Series.
Tested by and ok bmercer@ and mlarkin@ for 100 Series
Tested by and diff from James Hastings for Apollo and Gemini Lake
patrick [Wed, 20 Nov 2019 16:33:30 +0000 (16:33 +0000)]
regen
patrick [Wed, 20 Nov 2019 16:33:00 +0000 (16:33 +0000)]
Add PCI IDs for Gemini Lake SD/MMC controllers
beck [Wed, 20 Nov 2019 16:21:20 +0000 (16:21 +0000)]
Add accessors to change the buffer in a handshake message.
Needed for doing TLS 1.3 Post Handshake Handshake messages.
ok jsing@
patrick [Wed, 20 Nov 2019 16:14:51 +0000 (16:14 +0000)]
Correctly match another 9560-based device.
From Bryan Vyhmeister
florian [Wed, 20 Nov 2019 15:50:41 +0000 (15:50 +0000)]
Check for a too short answer packet in all callback functions because
otherwise we try to parse an invalid packet.
This can be triggered by captive_portal_resolve_done() when dhcp
provided nameservers do not answer and asr hits a timeout.
answer_packet is NULL and answer_len -1 in that case.
Found the hard way by claudio
visa [Wed, 20 Nov 2019 14:07:25 +0000 (14:07 +0000)]
Ensure that the kernel stack is properly aligned on mips64. This fixes
a panic related to vararg function sppp_auth_send(). The vararg code
generated by clang assumes 16-byte stack alignment.
pppoe(4)-induced panic seen by many on octeon
Fix from miod@
inoguchi [Wed, 20 Nov 2019 12:12:55 +0000 (12:12 +0000)]
Add test for cms -keyopt in appstest.sh
nicm [Wed, 20 Nov 2019 11:42:51 +0000 (11:42 +0000)]
Do not check the client readonly flag when there is no client, GitHub issue 1980.
inoguchi [Wed, 20 Nov 2019 10:46:17 +0000 (10:46 +0000)]
Fix rsa key output format
This fixes openssl(1) rsa -text output format
ok tb@
djm [Tue, 19 Nov 2019 23:32:09 +0000 (23:32 +0000)]
Sync documentation fixes from upstream:
commit
99c4cb00b50b846e4ba2492f67d44171de8c7066
Author: pedro martelletto <pedro@yubico.com>
Date: Mon Nov 18 16:55:37 2019 +0100
documentation fixes
ensure man pages are named after existing functions; pointed out
by Jason McIntyre <jmc@openbsd.org>; thanks!
djm [Tue, 19 Nov 2019 22:23:19 +0000 (22:23 +0000)]
adjust on-wire signature encoding for ecdsa-sk keys to better match
ec25519-sk keys. Discussed with markus@ and Sebastian Kinne
NB. if you are depending on security keys (already?) then make sure you
update both your clients and servers.
djm [Tue, 19 Nov 2019 22:21:15 +0000 (22:21 +0000)]
a little more information from the monitor when signature
verification fails.
gilles [Tue, 19 Nov 2019 22:04:04 +0000 (22:04 +0000)]
fix typos in smtpd.conf.5
from Ryan Kavanagh
bluhm [Tue, 19 Nov 2019 20:01:57 +0000 (20:01 +0000)]
Link system call tests to build.
bluhm [Tue, 19 Nov 2019 19:57:03 +0000 (19:57 +0000)]
Import NetBSD system call regression tests. They were written with
ATF (Automated Testing Framework), so we use a small wrapper to map
it to our bsd.regress.mk framework. Only half of the 80 NetBSD
tests have been taken, the others need more work to adapt. Of them
34 syscall tests pass.
Moritz Buhl ported the tests to OpenBSD.
anton [Tue, 19 Nov 2019 19:19:28 +0000 (19:19 +0000)]
When waiting on pipe I/O, simplify the unlock/relock logic using
rwsleep(). All made possible by the recent switch to using a rwlock as
the exclusive pipe lock.
ok visa@
jmc [Tue, 19 Nov 2019 16:02:32 +0000 (16:02 +0000)]
revert previous: naddy pointed out what's meant to happen. rethink needed...
jmc [Tue, 19 Nov 2019 14:54:47 +0000 (14:54 +0000)]
-c and -s do not make sense with -k; reshuffle -k into the main synopsis/usage;
ok djm
florian [Tue, 19 Nov 2019 14:49:36 +0000 (14:49 +0000)]
Print learned forwarders in unwindctl.
Someone please make this pretty.
Tested by deraadt as part of a larger diff.
florian [Tue, 19 Nov 2019 14:47:46 +0000 (14:47 +0000)]
Proposals from dhclient are sometimes replacements.
slaacd should be changed as well.
Tested by deraadt as part of a larger diff.
florian [Tue, 19 Nov 2019 14:46:33 +0000 (14:46 +0000)]
Stop dhcp lease file parsing now that dhclient sends proposals.
Tested by deraadt as part of a larger diff.
krw [Tue, 19 Nov 2019 14:35:07 +0000 (14:35 +0000)]
Inform unwind about DNS servers it finds in leases.
ok florian@
stsp [Tue, 19 Nov 2019 13:47:25 +0000 (13:47 +0000)]
regen
stsp [Tue, 19 Nov 2019 13:47:06 +0000 (13:47 +0000)]
Add PCI ID for an Intel AX 200 device.
ok jsg@
inoguchi [Tue, 19 Nov 2019 10:28:18 +0000 (10:28 +0000)]
More return value check in openssl(1) cms
Checking return value of sk_.*_new_null().
ok beck@ jsing@
inoguchi [Tue, 19 Nov 2019 10:20:10 +0000 (10:20 +0000)]
Add manual descriptions for openssl(1) req -addext
ok jmc@
remi [Tue, 19 Nov 2019 09:55:55 +0000 (09:55 +0000)]
Add point-to-point support for broadcast interfaces.
tested by Kapetanakis Giannis
ok claudio@
deraadt [Tue, 19 Nov 2019 09:46:22 +0000 (09:46 +0000)]
typo (result was bootblocks from / always installed into media); from miod
kevlo [Tue, 19 Nov 2019 06:34:10 +0000 (06:34 +0000)]
Set isr to the correct value (sc->rl_intrs).
ok deraadt@
kevlo [Tue, 19 Nov 2019 06:32:46 +0000 (06:32 +0000)]
Don't assign ifp twice in the re_txeof().
ok deraadt@
otto [Tue, 19 Nov 2019 06:21:26 +0000 (06:21 +0000)]
Update, max /usr has grown
otto [Tue, 19 Nov 2019 06:20:37 +0000 (06:20 +0000)]
Make /usr maximum 6G and grow it faster; ok lots of devs
yasuoka [Tue, 19 Nov 2019 03:33:43 +0000 (03:33 +0000)]
Correct the link speed constant variables. From Shoichi Yamaguchi.
ok jmatthew
kevlo [Tue, 19 Nov 2019 02:01:58 +0000 (02:01 +0000)]
Enable rge(4).
ok deraadt@
dlg [Tue, 19 Nov 2019 00:06:26 +0000 (00:06 +0000)]
take care to avoid a race when creating the same interface.
it was possible for multiple tun0 interfaces to be created concurrently,
which confused the pf_if code. when concurrent tun0 interfaces were
created, the pf_if code tried to add an addrhook for each interface,
but because they shared a name the two hooks ended up on one
interface. if the interface with two addrhooks was destroyed,
KASSERT(TAILQ_EMPTY(&ifp->if_addrhooks)) would trip. before the
KASSERT existed, we'd blindly free a tailq head, which would corrupt
the list, which would cause faults in pfi_detach_ifnet() anyway.
so now we take more care to ensure multiple tun0 interfaces cannot
exist concurrently.
inserting a tun or tap interface into the list of tun or tap
interfaces now checks to ensure that an interface with the same
unit number doesnt already exist. if an existing interface is found,
insert errors with EEXIST and the callers can unwind. the tunopen
and tapopen paths cope with losing the race.
Reported-by: syzbot+2b26012b9ea93834723e@syzkaller.appspotmail.com
sashan@ made a reliable test that could produce the failures
ok sashan@
naddy [Mon, 18 Nov 2019 23:17:48 +0000 (23:17 +0000)]
document '$' environment variable expansion for SecurityKeyProvider; ok djm@
naddy [Mon, 18 Nov 2019 23:16:49 +0000 (23:16 +0000)]
more missing mentions of
ed25519-sk; ok djm@