eric [Thu, 4 Apr 2019 19:25:45 +0000 (19:25 +0000)]
accept the NULL string in the proc message formatting api and simplify
code accordingly.
jsing [Thu, 4 Apr 2019 16:53:57 +0000 (16:53 +0000)]
Implement legacy fallback for the TLS 1.3 client.
If the Server Hello received indicates that the server did not negotiate
TLS 1.3, fallback to the original TLS client implementation.
ok bcook@, tb@
jsing [Thu, 4 Apr 2019 16:44:24 +0000 (16:44 +0000)]
Clean up the cipher/digest table mess.
The original implementation allows for libcrypto to be compiled without a
given algorithm and libssl then detects that ciphers or digests are
unavailable so that it can disable the associated cipher suites.
This is unnecessary since we do not compile out algorithms.
ok beck@, tb@ (a while back)
jsing [Thu, 4 Apr 2019 15:47:15 +0000 (15:47 +0000)]
Use correct define.
jsing [Thu, 4 Apr 2019 15:10:10 +0000 (15:10 +0000)]
Only assign destlen when src is non-NULL.
This avoids ever having a non-zero len with a NULL pointer.
jsing [Thu, 4 Apr 2019 15:09:09 +0000 (15:09 +0000)]
Switch to pthread_mutex_init().
While PTHREAD_MUTEX_INITIALIZER can be used on OpenBSD, some other
platforms do not like it.
Noted by bcook@
jsing [Thu, 4 Apr 2019 15:04:26 +0000 (15:04 +0000)]
Sync.
jsing [Thu, 4 Apr 2019 15:04:03 +0000 (15:04 +0000)]
Bump libssl/libtls minors due to symbol addition.
jsing [Thu, 4 Apr 2019 15:03:21 +0000 (15:03 +0000)]
Provide SSL chain/cert chain APIs.
These allow for chains to be managed on a per-certificate basis rather than
as a single "extra certificates" list. Note that "chain" in this context
does not actually include the leaf certificate however, unlike
SSL_CTX_use_certificate_chain_{file,mem}().
Thanks to sthen@ for running this through a bulk ports build.
ok beck@ tb@
jsing [Thu, 4 Apr 2019 14:32:49 +0000 (14:32 +0000)]
This case also needs to be fatal.
sthen [Thu, 4 Apr 2019 12:42:01 +0000 (12:42 +0000)]
update root CAs in cert.pem in sync with Mozilla
ok millert@
bluhm [Thu, 4 Apr 2019 12:03:09 +0000 (12:03 +0000)]
Add an update target to regenerate the expected results.
OK krw@
nicm [Thu, 4 Apr 2019 10:25:35 +0000 (10:25 +0000)]
Fix size check for splitw -f and top level pane size for tiled layout,
problems reported by Thomas Sattler.
ratchov [Thu, 4 Apr 2019 09:23:36 +0000 (09:23 +0000)]
Simplify underruns handling: just copy silence produced by upper layer.
This makes the code less bug prone as the "rare" underruns
conditions are handled by the main non-underrun code.
kettenis [Thu, 4 Apr 2019 07:10:05 +0000 (07:10 +0000)]
Avoid potential null-pointer dereference. Found by Kent R. Spillner.
ok deraadt@, jsg@
jmc [Thu, 4 Apr 2019 06:58:56 +0000 (06:58 +0000)]
patrick confirms that this driver does not support powersave;
kettenis [Thu, 4 Apr 2019 06:33:10 +0000 (06:33 +0000)]
The ASUS E200HA reads the register at "address" 0x00 instead of 0x18.
deraadt [Thu, 4 Apr 2019 05:53:27 +0000 (05:53 +0000)]
frame-pointer elimination has encountered some resistance in ports but we
are close to release, so punt the change till after-release
easier for everyoned and especially naddy
ok tedu
deraadt [Thu, 4 Apr 2019 05:35:10 +0000 (05:35 +0000)]
sync
kettenis [Thu, 4 Apr 2019 04:49:16 +0000 (04:49 +0000)]
Add bwfm(4) such that we can use bsd.rd to update over wireless.
bket [Thu, 4 Apr 2019 04:19:54 +0000 (04:19 +0000)]
Add support for not crossing filesystem boundaries (-x) to Option
and behaviour is the same as GPL rsync.
Initial diff received feedback from benno@, schwarze@, deraadt@ and
florian@. Thanks!
OK deraadt@
naddy [Thu, 4 Apr 2019 02:28:23 +0000 (02:28 +0000)]
Remove CONFIGURE_STYLE=autoupdate setting.
Running autoupdate is not useful in a port context. No port uses this and
it appears none ever has.
ok kn@
benno [Wed, 3 Apr 2019 19:58:04 +0000 (19:58 +0000)]
YUL - Montreal Dorval International has been renamed Montreal-Pierre
Elliott Trudeau International on January 1, 2004.
anton [Wed, 3 Apr 2019 16:23:20 +0000 (16:23 +0000)]
document BIOCSRTIMEOUT errors; ok deraadt@ millert@
anton [Wed, 3 Apr 2019 16:20:23 +0000 (16:20 +0000)]
Reject negative and too large timeouts passed to BIOCSRTIMEOUT. Since
the timeout converted to ticks is later passed timeout_add(), it could
cause a panic if the timeout is negative.
ok deraadt@ millert@
Reported-by: syzbot+82cb4dfe6a1fc3d8b490@syzkaller.appspotmail.com
djm [Wed, 3 Apr 2019 15:48:45 +0000 (15:48 +0000)]
openssh-8.0
tb [Wed, 3 Apr 2019 15:33:37 +0000 (15:33 +0000)]
Avoid some out of bound accesses in aesni_cbc_hmac_sha1_cipher().
The plen variable can be NO_PAYLOAD_LENGTH == (size_t)-1, so doing
tls_aad[plen-4] is no good. Also check that the length of the AAD
set via the control interface is equal to 13 since the whole file
is written with that case in mind.
Note that we no longer use this code in LibreSSL/OpenBSD. We
eliminated the use of these control interfaces and stitched cipher
modes in libssl a while ago.
Problem found by Guido Vranken with his cryptofuzz - thanks!
input & ok beck, jsing
jca [Wed, 3 Apr 2019 14:59:34 +0000 (14:59 +0000)]
Don't test the behavior of ^L as it depends on $TERM
Suggested by anton@
jca [Wed, 3 Apr 2019 14:55:12 +0000 (14:55 +0000)]
Bind ^L (C-l) to clear-screen instead of redraw
Slightly more useful for some, same defaults as bash.
No objection deraadt@ phessler@, ok tb@ kn@ benno@
krw [Wed, 3 Apr 2019 12:57:56 +0000 (12:57 +0000)]
Take the length of an initialized string, not a
soon-but-not-yet-initialized string.
Should fix "Segmentation fault" reported by Raf
Czlonka.
yasuoka [Wed, 3 Apr 2019 11:54:56 +0000 (11:54 +0000)]
Fix the bug that radius module didn't work when the size of radius message
changes.
jmc [Wed, 3 Apr 2019 10:31:18 +0000 (10:31 +0000)]
spelling mistake in previous - sorry, klemens!
jsg [Wed, 3 Apr 2019 10:31:10 +0000 (10:31 +0000)]
Correct access to doorbell. In radeondrm this is only present and used
with gfx7/sea islands hardware. Fixes ring 2 test failure on
carrizo-l/mullins.
Problem found by kettenis@ in a different part of the drm 4.19 tree.
kn [Wed, 3 Apr 2019 08:56:00 +0000 (08:56 +0000)]
Recommend ROUTE_TABLEFILTER over SO_RTABLE
SO_TABLE is not applicable to AF_ROUTE, ROUTE_TABLEFILTER works across all
families and goes well in line with the other ROUTE_* macros.
While here, document RTABLE_ANY and mention rtable(4).
OK claudio jmc
ratchov [Wed, 3 Apr 2019 07:47:20 +0000 (07:47 +0000)]
Always copy data (if any) whenever a new bounce buffer is available.
ratchov [Wed, 3 Apr 2019 07:44:52 +0000 (07:44 +0000)]
Check for available space before copying data to the bounce buffers.
No behavior change as we don't call uaudio_pdata_copy() in situations
when bounce buffers may not be available.
ratchov [Wed, 3 Apr 2019 07:38:12 +0000 (07:38 +0000)]
Make the uaudio_stream->ubuf_xfer relative to current xfer number.
No behavior change. The new representation is equivalent but eases
detection of overflows and underflows.
nicm [Wed, 3 Apr 2019 06:43:19 +0000 (06:43 +0000)]
Do not load /etc/tmux.conf if given -f.
nicm [Wed, 3 Apr 2019 06:43:04 +0000 (06:43 +0000)]
screen_write_fast_copy can no longer assume the target screen is default
(it isn't for the pane status lines).
jmc [Wed, 3 Apr 2019 06:24:07 +0000 (06:24 +0000)]
tweak previous;
deraadt [Wed, 3 Apr 2019 05:48:52 +0000 (05:48 +0000)]
sync
tedu [Wed, 3 Apr 2019 04:52:26 +0000 (04:52 +0000)]
mortimer unlinked libobjc from the build, but didn't know how to cvs rm.
I don't know much about compilers, but what I do have are a very particular
set of skills. Skills I have acquired over a very long career.
florian [Wed, 3 Apr 2019 03:48:45 +0000 (03:48 +0000)]
Default to port 853 if DoT is used; pointed out by tedu
naddy [Wed, 3 Apr 2019 02:18:03 +0000 (02:18 +0000)]
Remove the CONFIGURE_STYLE=automake setting. Its semantics have been a
long-standing source of confusion and it is no longer used by any ports.
ok sthen@ kn@ ajacoutot@
krw [Wed, 3 Apr 2019 01:17:53 +0000 (01:17 +0000)]
Update *.ok files to reflect recent modifications to
the auto allocation tables.
krw [Wed, 3 Apr 2019 01:10:30 +0000 (01:10 +0000)]
Add comment to prod updates to regress/sbin/disklabel
*.ok files whenever space_allocation tables are
modified.
dlg [Wed, 3 Apr 2019 00:02:02 +0000 (00:02 +0000)]
actually use the words "split-horizon" when talking about vpls bridge setup
dlg [Tue, 2 Apr 2019 23:50:14 +0000 (23:50 +0000)]
document how to do split-horizon setups with mpw
this is needed^Wuseful since the kernel doesn't implicitly implement
split horizon in bridge for mpw anymore. it does allow the joining
of different VPLS meshes on the same bridge now though.
based on discussion with Lee Nelson and Mitchell Krome
thfr [Tue, 2 Apr 2019 20:24:32 +0000 (20:24 +0000)]
disable MSI for AMD Summit Ridge/Raven Ridge HD Audio as workaround for audio stopping after varying amounts of time. ok brynet@, deraadt@
jmc [Tue, 2 Apr 2019 19:45:23 +0000 (19:45 +0000)]
add acpipci;
jmc [Tue, 2 Apr 2019 19:41:53 +0000 (19:41 +0000)]
kettenis has written an mi page for this, so remove this one; ok kettenis (i think)
jmc [Tue, 2 Apr 2019 19:39:32 +0000 (19:39 +0000)]
spelling;
jmc [Tue, 2 Apr 2019 19:37:04 +0000 (19:37 +0000)]
be consistent with "DoT"; from raf czlonka
nicm [Tue, 2 Apr 2019 18:41:24 +0000 (18:41 +0000)]
Silence flag should use the same option as activity, reported by Thomas
Sattler.
kn [Tue, 2 Apr 2019 14:12:09 +0000 (14:12 +0000)]
Clarify preposition
Any messages sent to the kernel are returned, and copies are sent to all
interested listeners. The kernel will provide the process ID for the
sender [...]
Change the last "for" to "of", making it even clearer that the kernel tells
listeners about the sender's PID; it does not provide anything back to the
sender in particular.
"Seems right" deraadt
visa [Tue, 2 Apr 2019 13:07:28 +0000 (13:07 +0000)]
Restrict which filesystems are available for swap. This rules out
obvious misconfigurations that cannot work.
OK mpi@ tedu@
mortimer [Tue, 2 Apr 2019 13:01:52 +0000 (13:01 +0000)]
Enable frame pointer elimination on i386, amd64 and mips64 when given
optimization flags.
ok kettenis@ visa@
jsg [Tue, 2 Apr 2019 12:59:56 +0000 (12:59 +0000)]
libLLVM.a is no longer installed
deraadt [Tue, 2 Apr 2019 12:59:34 +0000 (12:59 +0000)]
sync
deraadt [Tue, 2 Apr 2019 12:57:00 +0000 (12:57 +0000)]
Do not install libLLVM.a. At present the clang tools are either
static-linked against private copy, or dynamic tools against the *.so,
or ports use independent componented versions. Saves ~85MB in /usr.
ok jsg
deraadt [Tue, 2 Apr 2019 12:55:05 +0000 (12:55 +0000)]
new variable NOLIBSTATIC= prevents installation of a static library,
in cases where you don't want it installed, or it isn't built.
ok jsg
sthen [Tue, 2 Apr 2019 12:30:20 +0000 (12:30 +0000)]
fix broken comment
deraadt [Tue, 2 Apr 2019 11:30:07 +0000 (11:30 +0000)]
Remove commented-out historical ranlib operation, which would be
duplicate work.
deraadt [Tue, 2 Apr 2019 11:29:34 +0000 (11:29 +0000)]
only static libraries undergo ranlink
dlg [Tue, 2 Apr 2019 11:10:54 +0000 (11:10 +0000)]
print MPLS_MCAST as mpls, rather than unknown
The caveat with this is that MPLS and MPLS_MCAST look exactly the
same. I could prefix the MCAST line with "multicast" or something,
but like everything else in MPLS the meaning of that protocol type
is supposed to have changed. It's no longer meant to indicate
multicast mpls, but mpls with a label selected by an upstream. So
what's the right thing to do?
noticed by Mitchell Krome, who used this breakage to identify another
problem, which won't happen anymore.
deraadt [Tue, 2 Apr 2019 11:05:55 +0000 (11:05 +0000)]
annoying white space
deraadt [Tue, 2 Apr 2019 11:02:01 +0000 (11:02 +0000)]
BOGO_PC is an invalid userland address, which indicates kbind() is now
disabled in the process. Rather than tying it to KERNBASE, make it simply
-1, which means it even more invalid..
ok tedu
deraadt [Tue, 2 Apr 2019 11:00:22 +0000 (11:00 +0000)]
retguard has now replaced the stack protector on clang architectures,
the kernel does not need a __stack_smash_handler function.
WARNING: You need a fairly new clang, approximately > March 31.
with mortimer
dlg [Tue, 2 Apr 2019 10:52:33 +0000 (10:52 +0000)]
call pf_pkt_addr_changed on input
makes this consistent with other tunnel drivers, but mostly to avoid
having state info leak between layers of encapsulation.
yasuoka [Tue, 2 Apr 2019 10:50:20 +0000 (10:50 +0000)]
Fix ddb not to write its history to out of the region. When the
inputted line just ends at sizeof(db_history), ddb started writing the
histories to out of the region. diff from IIJ.
ok deraadt anton
dlg [Tue, 2 Apr 2019 10:50:16 +0000 (10:50 +0000)]
clear the BCAST and MCAST mbuf flags for "outgoing" packets.
if these remain set then output on the underlying interface may
mistakenly be done with the wrong protocol type (eg, MPLS_MCAST
instead of MPLS), and to the wrong link layer address.
reported by Lee Nelson
the specific problem was identified by Mitchell Krome
dlg [Tue, 2 Apr 2019 10:46:02 +0000 (10:46 +0000)]
some mbuf ph_rtableid fixes
consistently set the rtabled for "outgoing" packets to the encap
rdomain. use this for rtallocs in mpip too instead of assuming 0.
sthen [Tue, 2 Apr 2019 09:42:55 +0000 (09:42 +0000)]
When curve25519 was added to iked, it was based on the internet-draft and
used a private-use group number. Switch to the group number assigned in
RFC8031 as used in other implementations.
"this is the right time" deraadt@ "I like the idea" reyk@
If you use iked<>iked and have configured curve25519 in iked.conf (this
is not the default), you can switch to another PFS group before updating
then switch back. OpenBSD 6.3+ allows multiple "ikesa" lines so the
initiator can choose which to use.
jmc [Tue, 2 Apr 2019 09:20:52 +0000 (09:20 +0000)]
move the captive portal text into the actual block;
add an entry for /etc/example/unwind.conf;
sort;
ok florian
nicm [Tue, 2 Apr 2019 09:03:39 +0000 (09:03 +0000)]
Add an argument to copy commands to set the prefix for the buffer name,
allows buffers for different sessions to be named separately.
jmc [Tue, 2 Apr 2019 08:52:26 +0000 (08:52 +0000)]
use a compact list for previous, and tweak a little; ok florian
deraadt [Tue, 2 Apr 2019 08:51:49 +0000 (08:51 +0000)]
After upgrade time, delete contents of /var/syspatch. Either
we are on a release (and the rollbacks are meaningless) or it's a snapshot
(and they are even more meaningless)
It appears the first syspatch of a release does that cleaning, but why not
get ahead..
ok phessler
nicm [Tue, 2 Apr 2019 08:45:32 +0000 (08:45 +0000)]
Store and restore cursor across reflow by working out a position based
on unwrapped lines, rather than a grid offset. Fixes problems reported
by Thomas Sattler and Paul de Weerd.
deraadt [Tue, 2 Apr 2019 08:30:38 +0000 (08:30 +0000)]
Move to 6.5 release rathe than -beta. That means "pkg_add -u -Dsnap"
becomes the norm until release is out.
florian [Tue, 2 Apr 2019 08:28:20 +0000 (08:28 +0000)]
the fiendishly that needed some extra space here didn't pan out
florian [Tue, 2 Apr 2019 08:04:13 +0000 (08:04 +0000)]
Simplify code; we can loop over our resolvers now.
florian [Tue, 2 Apr 2019 07:47:22 +0000 (07:47 +0000)]
Add a config option to specify the preference of name servers.
Unfortunately the nameserver types enums needed to be renamed
to not collide with yacc tokens.
florian [Tue, 2 Apr 2019 07:46:03 +0000 (07:46 +0000)]
Collapse case statements to a common fall through and use the passed
in type to identify the resolver now that they can be addressed by
index.
florian [Tue, 2 Apr 2019 07:45:11 +0000 (07:45 +0000)]
Introduce array of resolvers in preparation of custom ordering
stsp [Tue, 2 Apr 2019 07:08:39 +0000 (07:08 +0000)]
Don't detach non-removable devices during resume on "sdhc* at acpi?".
Makes hibernate work with rootfs on built-in emmc storage.
Tested on King Jim Portabook.
ok deraadt@ kettenis@
deraadt [Tue, 2 Apr 2019 07:03:17 +0000 (07:03 +0000)]
sync
jmatthew [Tue, 2 Apr 2019 07:01:29 +0000 (07:01 +0000)]
add some msi and msi-x details
ok dlg@ kettenis@
deraadt [Tue, 2 Apr 2019 05:32:08 +0000 (05:32 +0000)]
Pull in addargs() API from ssh to replace complicated hand-rolled argument
composition code for the remote process.
ok florian naddy
mlarkin [Tue, 2 Apr 2019 05:06:39 +0000 (05:06 +0000)]
vmm(4): Fix some broken event injection code for SVM
Rework some event injection code which was using an incorrect v_eventinj
field format.
ok deraadt
mlarkin [Tue, 2 Apr 2019 05:03:00 +0000 (05:03 +0000)]
vmm(4): Inject #UD on read of MSR_LS_CFG on AMD SVM
The LS_CFG MSR is used as a different way to determine if SSBD controls
are present. Since we don't implement this, inject a #UD on read of this
MSR, which Linux interprets as "not having SSBD controls".
ok deraadt
kettenis [Tue, 2 Apr 2019 04:03:48 +0000 (04:03 +0000)]
mvgicp(4)
kn [Tue, 2 Apr 2019 03:58:57 +0000 (03:58 +0000)]
Report reliable VM state on status command
`vmctl status id' reports the VM's VCPU states, that is whether they are
running or halted.
Given that only one VCPU is currently supported per VM and the undocumented
states change across repetitive status commands, change it to indicate the
VM's general state based on whether the respective host procces is running
or not.
This makes vmctl reliably report "STATE: RUNNING" and "STATE: STOPPED" for
started and stopped VMs respectively.
OK mlarkin
kettenis [Tue, 2 Apr 2019 03:53:06 +0000 (03:53 +0000)]
acpipci(4)
mortimer [Tue, 2 Apr 2019 03:35:08 +0000 (03:35 +0000)]
Add variable length trap padding between the retguard epilogue and the
following return.
This change adds a constraint that the name passed to the RETGUARD_* macros
must correspond to the name in the corresponding ENTRY which starts the
function (or a function which appears beforehand in the same file). Since
we use the distance from the ENTRY definition to calculate how much padding
to insert, the ENTRY symbol must be in scope at assembly time. This is
almost always the case already, since it is the natural way to name the
retguard symbols so they remain unique.
ok deraadt@
krw [Tue, 2 Apr 2019 03:21:21 +0000 (03:21 +0000)]
Document new domain-search syntax available to
dhclient(8).
mortimer [Tue, 2 Apr 2019 03:02:47 +0000 (03:02 +0000)]
Emit variable length trap padding in retguard epilogue.
This adds more trap padding before the return while ensuring that the
return is still in the same cache line.
ok deraadt@
krw [Tue, 2 Apr 2019 02:59:43 +0000 (02:59 +0000)]
Add human readable parsing/display of RFC1035 data in domain-search
option. Replace handrolled dn_expand() with the system
version. Existing hex octet versions still accepted. New format is
option domain-search "my.domain.org", "fw.my.domain.org";
It is now possible to append and prepend domains to the list provided
by the server.
Documention update to dhcp-options(5) in the pipeline!
Inspired by dhcpd(8) domain-search diff from William Ahern.
Code peered at by florian@ and kn@.
krw [Tue, 2 Apr 2019 01:47:49 +0000 (01:47 +0000)]
Use consistant idiom for checking return value of
ioctl() (i.e. == -1). Don't prepend 'ioctl ' to ioctl
name in error or warning messages. Always put the
ioctl name in emitted messages.
No intentional functional change.
deraadt [Tue, 2 Apr 2019 01:40:40 +0000 (01:40 +0000)]
After upgrade, cleanup old clang and gcc compoents.
Test by phessler
krw [Tue, 2 Apr 2019 01:10:29 +0000 (01:10 +0000)]
No need for RESTORE (-R) to call DIOCGINFO and mpsave() twice in
normal mode.
And not overwriting any -F/-f file when in -n (donothing) mode is
likely less surprising.