eric [Tue, 10 Sep 2019 12:08:26 +0000 (12:08 +0000)]
immediatly close the connection if the cert is not valid, rather than
ending the smtp session.
ok gilles@
nicm [Tue, 10 Sep 2019 07:50:33 +0000 (07:50 +0000)]
Set up format tree for %if, GitHub issue 1896.
ratchov [Tue, 10 Sep 2019 07:49:15 +0000 (07:49 +0000)]
Prevents the block size from exceeding the requested one when a huge
block size is requested. This avoids exceeding half the buffer size.
ratchov [Tue, 10 Sep 2019 07:39:39 +0000 (07:39 +0000)]
If the requensted block size is too large, then use half the
buffer size instead of failing. Problem found and fix tested
by Joe Davis <me at jo.ie>, thanks!
gilles [Tue, 10 Sep 2019 07:03:16 +0000 (07:03 +0000)]
simplify lka reporting functions by moving the reqid inside the broadcast
function: reqid is a mandatory part of the header. lka reporting functions
now only append their own specific parameters.
ok martijn@
krw [Mon, 9 Sep 2019 23:28:41 +0000 (23:28 +0000)]
SET/CLR/ISSET all at once for the far less numerous link->flags.
tb [Mon, 9 Sep 2019 20:26:16 +0000 (20:26 +0000)]
Plug memory leak in error paths. Found while comparing this file
with OpenSSL 1.1.1's version which contains a similar fix.
ok jsing
bluhm [Mon, 9 Sep 2019 20:02:26 +0000 (20:02 +0000)]
Inform about system call memory write protection and stack mapping
violations in system accounting. This will help to find missbehaving
programs and possible attacks. The flags bit field is full, so
recycle the PDP-11 compatibility on VAX. lastcomm(1) prints the
AMAP flag as 'M'. daily(8) prints a list of affected processes.
OK deraadt@
patrick [Mon, 9 Sep 2019 20:00:51 +0000 (20:00 +0000)]
Update the bindings for imxsrc(4), since they changed when they were
upstreamed.
ok kettenis@
patrick [Mon, 9 Sep 2019 20:00:27 +0000 (20:00 +0000)]
Attach imxsrc(4) early and raise its priority to make sure it beats
syscon(4).
ok kettenis@
krw [Mon, 9 Sep 2019 18:59:20 +0000 (18:59 +0000)]
Continue bit twiddling tweaks. Use SET() with st->flags.
krw [Mon, 9 Sep 2019 18:25:47 +0000 (18:25 +0000)]
Lots of st->flags bit clearing in a mix of old school and CLR() makes
code harder to read. Switch to CLR().
No change to executable.
jsing [Mon, 9 Sep 2019 18:06:25 +0000 (18:06 +0000)]
Provide EVP_PKEY_CTX_get_signature_md() macro and implement the
EVP_PKEY_CTRL_GET_MD control for DSA, EC and RSA.
This is used by the upcoming RSA CMS code.
ok inoguchi@ tb@
krw [Mon, 9 Sep 2019 18:02:12 +0000 (18:02 +0000)]
Lots of st->flags checking in a mix of old school and ISSET() makes
code harder to read. Switch to ISSET() where obvious.
No change to executable.
jsing [Mon, 9 Sep 2019 17:56:21 +0000 (17:56 +0000)]
Load CMS error strings.
jsing [Mon, 9 Sep 2019 17:56:00 +0000 (17:56 +0000)]
Move #include <openssl/cms.h> to more appropriate location (since it is
now being installed).
jsing [Mon, 9 Sep 2019 16:49:34 +0000 (16:49 +0000)]
Sync for openssl/cms.h.
jsing [Mon, 9 Sep 2019 16:49:03 +0000 (16:49 +0000)]
Install the openssl/cms.h header.
This header includes OPENSSL_NO_CMS guards, so even if things find the
header it provides no useful content (and other code should technically
also be using OPENSSL_NO_CMS...).
ok deraadt@ inoguchi@
deraadt [Mon, 9 Sep 2019 13:50:06 +0000 (13:50 +0000)]
use PATH_MAX instead of 1024; ok tedu
nicm [Mon, 9 Sep 2019 11:47:25 +0000 (11:47 +0000)]
Clarify server options slightly.
dlg [Mon, 9 Sep 2019 09:38:04 +0000 (09:38 +0000)]
revert back to using 0x6c for the reply post index register.
this was mistakenly committed as part of the powerdown hook stuff,
and caused "older" chips like SAS2208s to get stuck in biowait.
reported by Mischa <obsdml at high5 dot nl> and hrvoje popovski
discussed with jmatthew@ and alex wilson.
fix tested by me on a SAS3508 and SAS2208
nicm [Mon, 9 Sep 2019 08:01:21 +0000 (08:01 +0000)]
Add cursor-down-and-cancel, from Mark Kelly.
nicm [Mon, 9 Sep 2019 07:57:30 +0000 (07:57 +0000)]
Put default keys in alphabetical order, from Barry Rountree.
nicm [Mon, 9 Sep 2019 07:53:19 +0000 (07:53 +0000)]
Use a zero width space to prevent mandoc warning about trailing
delimeters in Ql, from orbea at riseup dot net in GitHub issue 1892.
jmatthew [Mon, 9 Sep 2019 02:57:20 +0000 (02:57 +0000)]
Specify the right opcode when trying to set the Interface Step Sequence ID.
Newer firmware (version 14.25) doesn't allow the driver to create TIS
contexts without setting the ISSI, but earlier firmware doesn't seem to care.
Big thanks to Xiaofeng Liu from Mellanox support who identified the problem.
fixes problems first reported by patrick@
dtucker [Mon, 9 Sep 2019 02:31:19 +0000 (02:31 +0000)]
Fix potential truncation warning. ok deraadt.
krw [Mon, 9 Sep 2019 01:37:27 +0000 (01:37 +0000)]
Tweak some comments to be useful. Sort switch cases into
numerical order. Move bit of logic before calling
st_unmount() into st_unmount().
jsg [Mon, 9 Sep 2019 01:35:43 +0000 (01:35 +0000)]
When no display outputs are connected on boot linux 4.19 drm relies on
deferred setup to handle the console framebuffer where as linux 4.4 drm
created a 1024x768 console framebuffer in this situation.
As we only handle setting up rasops and wsdisplay on attach go back to
the old behaviour for now so a display can be connected after booting
with none attached to interact with the console.
This partly reverts linux commit
drm/fb-helper: Support deferred setup
ca91a2758fcef6635626993557dd51cfbb6dd134
Reported and tested by Marcus MERIGHI.
Tested by and ok kettenis@
schwarze [Sun, 8 Sep 2019 22:50:59 +0000 (22:50 +0000)]
sbrk(2) already existed in Version 4 AT&T UNIX;
source: https://minnie.tuhs.org/cgi-bin/utree.pl?file=V4/man/man2/break.2
pointed out by Sevan Janiyan <venture37 at geeklan dot co dot uk>
schwarze [Sun, 8 Sep 2019 22:43:46 +0000 (22:43 +0000)]
Be more specific regarding which encoding this simulator generates:
the existing text encouraged the misconception that punched card
encodings were somehow standardized, which they were not. Lots of
different encodings existed, and the one simulated here was a
proprietary one of one specific vendor (GE). Also correct the
history: the folks at Bell Labs already played with this.
HISTORY mistake noticed by Sevan Janiyan <venture37 at geeklan
dot co dot uk>, who also agrees with the final patch.
tweak and OK jmc@
beck [Sun, 8 Sep 2019 22:43:14 +0000 (22:43 +0000)]
Pull out and expand openat tests into their own routine.
daniel [Sun, 8 Sep 2019 17:20:51 +0000 (17:20 +0000)]
reverting for now; requested by deraadt
daniel [Sun, 8 Sep 2019 17:13:44 +0000 (17:13 +0000)]
Only install the gcc-local(1) and clang-local(1) man pages if we also have
the corresponding compiler on a given platform.
ok deraadt@
jsing [Sun, 8 Sep 2019 17:00:05 +0000 (17:00 +0000)]
Add CMS ECC support.
This brings in EC code from OpenSSL 1.1.1b, with style(9) and whitespace
cleanups. All of this code is currently under OPENSSL_NO_CMS hence is a
no-op.
ok inoguchi@
landry [Sun, 8 Sep 2019 16:57:17 +0000 (16:57 +0000)]
Bump datasize-cur to 4Gb for pbuild class on sparc64, rust is a pig.
ok semarie@
kettenis [Sun, 8 Sep 2019 16:45:21 +0000 (16:45 +0000)]
Add Allwinner H6 support.
ajacoutot [Sun, 8 Sep 2019 13:30:46 +0000 (13:30 +0000)]
Have sm_install() use install -F (fsync).
kettenis [Sat, 7 Sep 2019 19:05:44 +0000 (19:05 +0000)]
We can't call cpu_ucode_apply() before we call cpu_ucode_setup().
Fallout spotted by naddy@
ok deraadt@
florian [Sat, 7 Sep 2019 18:57:47 +0000 (18:57 +0000)]
Do not try to send router advertisements on interfaces that have no link.
Keeps log noise down on carp backup interfaces.
Reported by matthieu
Previous version OK deraadt
Input & OK benno
krw [Sat, 7 Sep 2019 15:26:07 +0000 (15:26 +0000)]
Fold st_identify_drive() into stattach().
visa [Sat, 7 Sep 2019 14:03:52 +0000 (14:03 +0000)]
Enable nvme(4) on octeon.
visa [Sat, 7 Sep 2019 14:02:31 +0000 (14:02 +0000)]
Mention octpcie(4).
visa [Sat, 7 Sep 2019 14:01:27 +0000 (14:01 +0000)]
sync
visa [Sat, 7 Sep 2019 14:00:32 +0000 (14:00 +0000)]
Add a manual page for octpcie(4).
visa [Sat, 7 Sep 2019 13:58:58 +0000 (13:58 +0000)]
Add a driver for the PCIe controller found on OCTEON II and OCTEON III.
kettenis [Sat, 7 Sep 2019 13:46:19 +0000 (13:46 +0000)]
Revert acpipci(4) changes. Too many systems are broken. We'll try again
after 6.6 as been released. The acpireg.h change stays behind.
patrick [Sat, 7 Sep 2019 13:33:00 +0000 (13:33 +0000)]
Retrieve SFP information using the SFP framework.
Feedback from dlg@
ok kettenis@
patrick [Sat, 7 Sep 2019 13:32:36 +0000 (13:32 +0000)]
Add sfp(4), a driver that allows talking to SFPs connected over
an I2C bus and provides a method to read its pages over the SFP
framework.
Feedback from dlg@
ok kettenis@
patrick [Sat, 7 Sep 2019 13:29:08 +0000 (13:29 +0000)]
Add an SFP framework which allows SFP providers to provide a
method to access its pages.
ok kettenis@
patrick [Sat, 7 Sep 2019 13:27:40 +0000 (13:27 +0000)]
Register mviic(4) in the I2C framework.
ok kettenis@
patrick [Sat, 7 Sep 2019 13:27:23 +0000 (13:27 +0000)]
Add an I2C framework, so that nodes that need to use I2C but
are not children of the I2C controller are able to use it.
ok kettenis@
schwarze [Sat, 7 Sep 2019 10:28:27 +0000 (10:28 +0000)]
more Version 1 AT&T UNIX history:
a few cases that weren't altogether straightforward;
tweak and OK jmc@, OK sobrado@
ajacoutot [Sat, 7 Sep 2019 10:02:19 +0000 (10:02 +0000)]
Remove dependency on basename(1).
prodded by deraadt@
ok kn@ deraadt@ tb@
tobhe [Sat, 7 Sep 2019 09:11:14 +0000 (09:11 +0000)]
Remove unused VMD_DISK_INVALID message type and mark it obsolete.
ok mlarkin@
krw [Sat, 7 Sep 2019 02:30:40 +0000 (02:30 +0000)]
Tweak some comments and code layout to make MTSETBSIZE and MTSETDENSITY
cases look the same.
krw [Sat, 7 Sep 2019 02:07:08 +0000 (02:07 +0000)]
ST_USER_BLKSIZE and ST_QUIRK_BLKSIZE are only tested as an or'd pair.
Ditto for ST_USER_DENSITY and ST_QUIRK_DENSITY. Thus the origin of the
value stored in the mode is irrelevant. Just use two flags
(ST_MODE_BLKSIZE and ST_MODE_DENSITY) instead of the existing four.
krw [Sat, 7 Sep 2019 01:47:48 +0000 (01:47 +0000)]
ST_Q_FORCE_BLKSIZE is superfluous since it is only used in combination with
a non-zero value of blksize in the quirkdata. So just check for non-zero
value of blksize, as is already done for density.
krw [Sat, 7 Sep 2019 01:27:02 +0000 (01:27 +0000)]
Nuke unused flag ST_BLOCK_SET.
cheloha [Sat, 7 Sep 2019 01:23:23 +0000 (01:23 +0000)]
Remove tz, the kernel timezone.
After removing timezone support from date(1), msdosfs, gettimeofday(2),
settimeofday(2), boot_config(8), config(8), and introducing kern.utc_offset
sysctl(2) to replace the DST/TIMEZONE options(4) to support running the RTC
at an offset from UTC, the kernel timezone is unused and we can delete it.
With this commit the kernel timezone now appears to be extinct in
upstream BSD, though it does survive in XNU:
https://github.com/opensource-apple/xnu/blob/
0a798f6738bc1db01281fc08ae024145e84df927/bsd/conf/param.c#L83
Prompted by tedu@ some time back, and inspired by NetBSD and FreeBSD's
work in this area.
ok deraadt@
krw [Sat, 7 Sep 2019 01:15:41 +0000 (01:15 +0000)]
Various struct/variables "modes" -> "mode" since there is only one.
jsg [Sat, 7 Sep 2019 00:51:40 +0000 (00:51 +0000)]
drm/i915: Call dma_set_max_seg_size() in i915_driver_hw_probe()
From Lyude Paul
68b58d3924645356415ae4f6922da8fa14ea1642 in linux 4.19.y/4.19.70
32f0a982650b123bdab36865617d3e03ebcacf3b in mainline linux
jsg [Sat, 7 Sep 2019 00:48:55 +0000 (00:48 +0000)]
define away dma_set_max_seg_size()
needed for a future linux 4.19 change
jsg [Sat, 7 Sep 2019 00:43:05 +0000 (00:43 +0000)]
drm/i915: Don't deballoon unused ggtt drm_mm_node in linux guest
From Xiong Zhang
c7615333645de572cae573da0a92dd3fc8e099e0 in linux 4.19.y/4.19.70
0a3dfbb5cd9033752639ef33e319c2f2863c713a in mainline linux
jsg [Sat, 7 Sep 2019 00:39:29 +0000 (00:39 +0000)]
drm/i915: fix broadwell EU computation
From Lionel Landwerlin
a3eb2eba2f79950b788342562604d36c5a260da7 in linux 4.19.y/4.19.70
63ac3328f0d1d37f286e397b14d9596ed09d7ca5 in mainline linux
cheloha [Fri, 6 Sep 2019 21:30:31 +0000 (21:30 +0000)]
options(4), config(8): drop dst/timezone knobs; ok kettenis@ deraadt@ jca@
schwarze [Fri, 6 Sep 2019 19:25:08 +0000 (19:25 +0000)]
More Version 1 AT&T UNIX history.
This became possible because copies of the original v1 manuals
have shown up on the Internet some time ago.
Reminded by Sevan Janiyan <venture37 at geeklan dot co dot uk>.
patrick [Fri, 6 Sep 2019 19:18:24 +0000 (19:18 +0000)]
Mention imxiic(4), mviic(4), rkiic(4), sxitwi(4) and mcprtc(4).
Requested by and ok jmc@
jsing [Fri, 6 Sep 2019 17:59:24 +0000 (17:59 +0000)]
Add various macros and controls for EC_PKEY_CTX.
These are needed for the upcoming EC CMS support (nothing else appears
to use them). This largely syncs our ec_pmeth.c with OpenSSL 1.1.1b.
With input from inoguchi@ and tb@.
ok inoguchi@ tb@
jsing [Fri, 6 Sep 2019 17:41:05 +0000 (17:41 +0000)]
Handle CMS PEM headers.
ok inoguchi@ tb@
visa [Fri, 6 Sep 2019 16:22:40 +0000 (16:22 +0000)]
Do not stop prologue scan at branch instructions. This lets stack trace
saving work with the twisted code that clang occasionally emits; stack
frame setup can happen very late after blocks of conditional code.
visa [Fri, 6 Sep 2019 16:06:30 +0000 (16:06 +0000)]
Stop stack trace saving if the current subroutine is u_general or u_intr
because the next frame is in userspace.
deraadt [Fri, 6 Sep 2019 15:22:24 +0000 (15:22 +0000)]
missing ;
found by wilfried meindl
ok kettenis
naddy [Fri, 6 Sep 2019 15:07:19 +0000 (15:07 +0000)]
disable base-gcc on armv7, sets from deraadt@; ok deraadt@ kettenis@ jca@
deraadt [Fri, 6 Sep 2019 14:50:23 +0000 (14:50 +0000)]
sync
naddy [Fri, 6 Sep 2019 14:45:34 +0000 (14:45 +0000)]
Allow prepending a list of algorithms to the default set by starting
the list with the '^' character, e.g.
HostKeyAlgorithms ^ssh-
ed25519
Ciphers ^aes128-gcm@openssh.com,aes256-gcm@openssh.com
ok djm@ dtucker@
deraadt [Fri, 6 Sep 2019 13:45:04 +0000 (13:45 +0000)]
delete two decades of debugging code and further simplify the main
trap() switch statement
ok kettenis
kettenis [Fri, 6 Sep 2019 13:10:44 +0000 (13:10 +0000)]
Only do the timecounter initialization on the primary CPU. Avoids using a
bogus CPU frequency from a secondary CPU to set the TSC frequency.
ok deraadt@, mikeb@
deraadt [Fri, 6 Sep 2019 12:55:26 +0000 (12:55 +0000)]
oops the label is actually out:
deraadt [Fri, 6 Sep 2019 12:23:49 +0000 (12:23 +0000)]
oops incorrect goto label
deraadt [Fri, 6 Sep 2019 12:22:01 +0000 (12:22 +0000)]
If uvm_map_inentry returns false then a signal has been delivered, and
userret() must be called on trap() exit to deliver it, rather than
repeating the same cause infinitely. discovered by George Koehler
ok kettenis bluhm visa
deraadt [Fri, 6 Sep 2019 12:13:41 +0000 (12:13 +0000)]
fpuinit() comments should precede cpu_init() call
benno [Fri, 6 Sep 2019 09:40:59 +0000 (09:40 +0000)]
this bit about soii and link-local addresses is no longer needed since
florian@ removed soii for link-local adresses in
sys/netinet6/in6_ifattach.c rev 1.114
ok florian@ sthen@ kn@
patrick [Fri, 6 Sep 2019 09:38:19 +0000 (09:38 +0000)]
mcprtc(4), a driver for the Microchip MCP79400 RTC and similar variants.
ok kettenis@
patrick [Fri, 6 Sep 2019 08:45:37 +0000 (08:45 +0000)]
Add mviic(4), a driver to support the I2C controller on the
Armada 3700 SoC.
With feedback from and ok kettenis@
patrick [Fri, 6 Sep 2019 08:44:21 +0000 (08:44 +0000)]
Add I2C clock gates.
ok kettenis@
asou [Fri, 6 Sep 2019 08:39:27 +0000 (08:39 +0000)]
Correct the description of EINTR and EINVAL. This looks like a mis-merge
in revision 1.30.
ok deraadt@ tb@
martijn [Fri, 6 Sep 2019 08:23:56 +0000 (08:23 +0000)]
Add support for filter-reports. These allow filters to send freetext
reports to other filters/report handlers.
Builtin filters work via the new "report" keyword.
Proc filters can send reports via:
"report|<seconds.<microseconds>|smtp-in|<reqid>|<message>"
Subscribing to these reports can be done via the the "filter-report"
keyword.
The reports themselves contain the usual elements followed by:
<type>|<name>|<message>
Type can be builtin or proc.
Name is the process name for type proc and the filter name for type builtin.
OK gilles@
djm [Fri, 6 Sep 2019 07:53:40 +0000 (07:53 +0000)]
key conversion should fail for !openssl builds, not fall through to
the key generation code
jsg [Fri, 6 Sep 2019 06:40:54 +0000 (06:40 +0000)]
zero hash bits early in dwxe_iff()
ok kettenis@
djm [Fri, 6 Sep 2019 06:08:11 +0000 (06:08 +0000)]
typo in previous
djm [Fri, 6 Sep 2019 05:59:41 +0000 (05:59 +0000)]
avoid compiling certain files that deeply depend on libcrypto when
WITH_OPENSSL isn't set
djm [Fri, 6 Sep 2019 05:23:55 +0000 (05:23 +0000)]
fixes for !WITH_OPENSSL compilation; ok dtucker@
djm [Fri, 6 Sep 2019 04:53:27 +0000 (04:53 +0000)]
lots of things were relying on libcrypto headers to transitively
include various system headers (mostly stdlib.h); include them
explicitly
dtucker [Fri, 6 Sep 2019 04:24:06 +0000 (04:24 +0000)]
Check for RSA support before using it for the user key, otherwise use
ed25519 which is supported when built without OpenSSL.
djm [Fri, 6 Sep 2019 03:30:42 +0000 (03:30 +0000)]
remove leakmalloc reference; we used this early when refactoring but
not since
djm [Fri, 6 Sep 2019 01:58:50 +0000 (01:58 +0000)]
better error code for bad arguments; inspired by
https://github.com/openssh/openssh-portable/pull/144 by Ed Maste
deraadt [Thu, 5 Sep 2019 23:09:31 +0000 (23:09 +0000)]
sync
kn [Thu, 5 Sep 2019 22:01:14 +0000 (22:01 +0000)]
Document pledge keyword
OK millert
kettenis [Thu, 5 Sep 2019 20:47:49 +0000 (20:47 +0000)]
Fix previous commit.
ok millert@, sf@, deraadt@
fcambus [Thu, 5 Sep 2019 19:18:32 +0000 (19:18 +0000)]
Update Spleen kernel fonts to the latest released version, bringing
the following improvements:
- Character at position CB is 'E' with diaeresis, not with tilde (12x24 version)
- Character at position D6 is 'O' with diaeresis, not with tilde (12x24 version)
- Character at position DC is 'U' with diaeresis, not with tilde (12x24 version)
- Fix middle bar thickness in the upper case 'eth' character (12x24 version)
- Fix ring alignment for the 'A' and 'a' characters (16x32 and 32x64 versions)
- Fix tilde alignment in the 'o' with tilde character (32x64 version)
- Remove strain pixel on the '3' and 'k' characters (32x64 version)
OK deraadt@, mpi@