nicm [Fri, 20 Aug 2021 19:08:36 +0000 (19:08 +0000)]
Remove some unnecessary blank lines.
nicm [Fri, 20 Aug 2021 18:59:53 +0000 (18:59 +0000)]
Expose args_value struct (will be needed soon) and add some missing frees.
nicm [Fri, 20 Aug 2021 17:53:54 +0000 (17:53 +0000)]
Add a way to create an empty arguments set.
nicm [Fri, 20 Aug 2021 17:50:42 +0000 (17:50 +0000)]
Remove stray spaces after function names.
nicm [Fri, 20 Aug 2021 17:36:03 +0000 (17:36 +0000)]
Fill colour palette correctly from option for new panes, GitHub issue
2831.
nicm [Fri, 20 Aug 2021 09:06:26 +0000 (09:06 +0000)]
Actually parse contents of {} as a command and then convert to a string
instead of just copying arguments around as strings.
nicm [Fri, 20 Aug 2021 06:30:57 +0000 (06:30 +0000)]
Add a helper function for actually parsing the command.
anton [Fri, 20 Aug 2021 05:26:25 +0000 (05:26 +0000)]
sync
anton [Fri, 20 Aug 2021 05:25:23 +0000 (05:25 +0000)]
add ucc(4) manual, looks good to jmc@
anton [Fri, 20 Aug 2021 05:23:18 +0000 (05:23 +0000)]
Add ucc(4), a driver for USB HID Consumer Control keyboards. Such
keyboard is a pseudo device which is used to expose audio and
application launch keys. My prime motivation is to get the volume mute,
increment and decrement keys to just work on my keyboard without the
need to use usbhidaction(1).
Looks reasonable to kettenis@ mpi@ and ok jcs@
anton [Fri, 20 Aug 2021 05:19:08 +0000 (05:19 +0000)]
add consumer control hid usages, will be used by ucc(4)
kevlo [Fri, 20 Aug 2021 04:54:10 +0000 (04:54 +0000)]
Back out last commit.
deraadt@ pointed out that the operation is correct.
It attaches, but by then it is being pulled. The dying flag is set to
avoid half the kinds of failure conditions. The other half of the failure
conditions are detect on-the-spot.
And it seems when compiled for debug, this on-the-spot check which notices
the device isn't there, has a printf.
djm [Fri, 20 Aug 2021 03:22:55 +0000 (03:22 +0000)]
openssh-8.7
kevlo [Fri, 20 Aug 2021 01:40:51 +0000 (01:40 +0000)]
When plugging and unplugging an adapter continuously, it waits for a timeout
over and over again. The problem is that the device has been unplugged,
but the attach handler still issues USB requests via ure_ctl().
The easiest solution is to deactivate the device proactively in
ure_ctl() when we see that error there.
From Christian Ludwig
kevlo [Fri, 20 Aug 2021 01:33:44 +0000 (01:33 +0000)]
In the attach routine, the hardware bug workaround is only for legacy PCI
interrupts. Move it to the right place.
stsp [Fri, 20 Aug 2021 00:22:54 +0000 (00:22 +0000)]
Revert previous; Turns out the start_hw() functions don't always
work right during the resume path. To be revisited later.
stsp [Thu, 19 Aug 2021 18:45:11 +0000 (18:45 +0000)]
Run more device init code when resuming iwm(4) and iwx(4) devices.
deraadt@ pointed out that the resume code path only ran a small part
of the entire hardware init sequence which runs when the device is
first attached at boot time.
In particular, we didn't wait for device stabilization, and MSI-X
initialization was attempted too early, and the "persistence bit"
workaround was skipped (which, ironically, is supposed to address
a known hardware bug during resume).
Tested by jcs on AX201 and by myself on 8265 with no regressions seen.
schwarze [Thu, 19 Aug 2021 16:55:24 +0000 (16:55 +0000)]
do not crash when a manpath directory contains a symbolic link
that points to a directory rather than to a regular file;
bug reported by Lukas Epple <sternenseemann at systemli dot org>,
and my patch also tested by him on NixOS
schwarze [Thu, 19 Aug 2021 15:21:32 +0000 (15:21 +0000)]
fix the section number in the <title> element for preformatted pages;
minibug reported by Ian <Ropers at gmail dot com> on misc@
patrick [Thu, 19 Aug 2021 14:13:39 +0000 (14:13 +0000)]
Purge the cached nodes when we switch back to SCAN. In case we lost
the connection because an AP is gone away, this makes sure we don't
end up picking a now non-existant AP over and over.
Another problem this works around is that when trying to re-join that
AP, the attempt to connect would not yield any event message from the
firmware, and then we would get stuck. Since we now definitely choose
a different AP, this issue does not show up.
ok stsp@
patrick [Thu, 19 Aug 2021 13:58:40 +0000 (13:58 +0000)]
Consistently use ieee80211_begin_scan() to switch to SCAN.
React to deauth/disassoc and link state events if we're
already past the SCAN stage.
ok stsp@
dlg [Thu, 19 Aug 2021 10:22:00 +0000 (10:22 +0000)]
implement reception of "VLAN 0 priority tagged" packets.
according to 802.1Q, vlan 0 on the wire is special and should be
interpreted as if it was a packet received on the parent interface,
but you get the packet priority information encoded in the vlan
header.
historically we drop vlan tagged packets that don't have a vlan
interface configured for the received tag number. historically we
have deviated from 802.1Q by allowing for the configuration of a
vlan subinterface with the vnetid "unset". this works just like any
other vlan interface, but it uses tag 0 on the wire. however, if
you're in a situation where you're receiving vlan tagged 0 packets
that really are part of the same layer 2 ethernet domain as the
parent inteface, this doesnt work well.
landry@ is in such a situation at work where the network is sending
his OpenBSD boxes packets with VLAN tag 0. sometimes. most of the
time the packets are untagged, which is expected, but sometimes
they have a VLAN header set. this causes problems, particularly
with arp.
this diff does the smallest possible change to enable reception of
these vlan 0 priority tagged packets. if an "unset" vlan interface
is not configured on the parent, then vlan 0 tagged packets get
their header stripped and continue stack processing as if they didnt
have the tag at all.
landry has been running this for months.
ok sthen@ claudio@
stsp [Thu, 19 Aug 2021 06:02:39 +0000 (06:02 +0000)]
Fix an off-by-one in bwfm(4) SSID length checks.
SSIDs are binary data, not NUL-terminated strings.
ok patrick@
stsp [Thu, 19 Aug 2021 06:02:04 +0000 (06:02 +0000)]
Hide detailed firmware error reports from iwn/iwm/iwx by default.
These drivers will now report "fatal firmware error" on a single line
by default and provide the full trace when debugging has been enabled
with e.g. 'ifconfig iwm0 debug'.
ok sthen@ kevlo@
beck [Thu, 19 Aug 2021 03:44:00 +0000 (03:44 +0000)]
Pull roots out of the trust store in the legacy xsc when building chains
to handly by_dir and fun things correctly. - fixes dlg@'s case and
by_dir regress in openssl-ruby
ok jsing@
solene [Wed, 18 Aug 2021 18:10:33 +0000 (18:10 +0000)]
Add veb(4) to the list of pseudo devices that can be created.
ok deraadt@
tb [Wed, 18 Aug 2021 16:06:56 +0000 (16:06 +0000)]
Import regress tests for SM2. Not yet linked to the build.
Part of Github PR #105
tb [Wed, 18 Aug 2021 16:04:32 +0000 (16:04 +0000)]
Import initial code for the SM2 cipher
This adds the SM2 algorithm defined in the Chinese standards
GB/T 32918.1-2016, GB/T 32918.2-2016, GB/T 32918.3-2016,
GB/T 32918.4-2016 and GB/T 32918.5-2017.
This is an ISC licensed implementation contributed by Ribose.inc, based
on the same code that was contributed to OpenSSL by Jack Lloyd. The port
to LibreSSL was done by Ronald Tse and Nickolay Olshevsky.
Github PR #105
I made quite a few cleanup passes on this, but more is needed, some
of which will happen in-tree before this is linked to the build.
ok deraadt inoguchi (a long time ago), jsing
beck [Wed, 18 Aug 2021 15:32:38 +0000 (15:32 +0000)]
Add a check_trust call to the legacy chain validation on chain add, remembering
the result in order to return the same errors as OpenSSL users expect to override
the generic "Untrusted cert" error.
This fixes the openssl-ruby timestamp test.
ok tb@
nicm [Wed, 18 Aug 2021 15:16:33 +0000 (15:16 +0000)]
Need to flatten arguments for aliases.
beck [Wed, 18 Aug 2021 15:10:46 +0000 (15:10 +0000)]
Refactor the legacy chain validation from the chain adding code into its
own function, in preparation for subesquent change. No functional change.
ok tb@
nicm [Wed, 18 Aug 2021 10:15:08 +0000 (10:15 +0000)]
Push the conversion of {} to string up out of the parser and into the
command builder.
jsg [Wed, 18 Aug 2021 10:02:49 +0000 (10:02 +0000)]
drm/i915: Only access SFC_DONE when media domain is not fused off
From Matt Roper
65395b053d03cb662e63cbf2c7e0faef8c15cb8c in linux 5.10.y/5.10.60
24d032e2359e3abc926b3d423f49a7c33e0b7836 in mainline linux
jsg [Wed, 18 Aug 2021 09:59:05 +0000 (09:59 +0000)]
drm/amdgpu: don't enable baco on boco platforms in runpm
From Alex Deucher
95de3592f87e46df63119dd52b4a0e544e519c6b in linux 5.10.y/5.10.60
202ead5a3c589b0594a75cb99f080174f6851fed in mainline linux
jsg [Wed, 18 Aug 2021 09:56:59 +0000 (09:56 +0000)]
drm/amd/display: use GFP_ATOMIC in amdgpu_dm_irq_schedule_work
From Anson Jacob
bd80d11a516c78fb74d11e69c67082f36f8ef8e3 in linux 5.10.y/5.10.60
0cde63a8fc4d9f9f580c297211fd05f91c0fd66d in mainline linux
jsg [Wed, 18 Aug 2021 09:54:44 +0000 (09:54 +0000)]
drm/amd/display: Remove invalid assert for ODM + MPC case
From Eric Bernstein
ae311a7418f13be375c29ec4178baa51cd9101ba in linux 5.10.y/5.10.60
c90f6263f58a28c3d97b83679d6fd693b33dfd4e in mainline linux
deraadt [Tue, 17 Aug 2021 20:32:26 +0000 (20:32 +0000)]
sync
nicm [Tue, 17 Aug 2021 20:17:21 +0000 (20:17 +0000)]
Treat a pane that has died the same as no pane when sending data to
control mode clients, GitHub issue 2828.
tb [Tue, 17 Aug 2021 19:44:01 +0000 (19:44 +0000)]
Sync amd64's est.c r1.40 to i386
Tested by Sven F., James Jerkins and myself.
nicm [Tue, 17 Aug 2021 19:37:55 +0000 (19:37 +0000)]
Fix pipe-pane usage.
nicm [Tue, 17 Aug 2021 19:26:42 +0000 (19:26 +0000)]
Do not block with incremental command prompt.
sthen [Tue, 17 Aug 2021 19:19:59 +0000 (19:19 +0000)]
sync
sthen [Tue, 17 Aug 2021 19:19:30 +0000 (19:19 +0000)]
7.1 fw key
anton [Tue, 17 Aug 2021 18:08:51 +0000 (18:08 +0000)]
oops, fix spelling of my own name
nicm [Tue, 17 Aug 2021 16:19:00 +0000 (16:19 +0000)]
Set the right session if detach-on-destroy is off.
deraadt [Tue, 17 Aug 2021 15:29:09 +0000 (15:29 +0000)]
sync
naddy [Tue, 17 Aug 2021 15:25:27 +0000 (15:25 +0000)]
7.1 packages key
deraadt [Tue, 17 Aug 2021 15:13:08 +0000 (15:13 +0000)]
sync
deraadt [Tue, 17 Aug 2021 15:11:48 +0000 (15:11 +0000)]
7.1 base key
deraadt [Tue, 17 Aug 2021 15:03:55 +0000 (15:03 +0000)]
7.0-beta
anton [Tue, 17 Aug 2021 11:30:45 +0000 (11:30 +0000)]
For devices with rechargeable batteries, expose a new boolean sensor
indicating the charging state.
Thanks to Laurence Tratt <laurie at tratt dot net> for testing.
nicm [Tue, 17 Aug 2021 11:20:13 +0000 (11:20 +0000)]
calloc for confirm-before data since the item needs to start NULL.
nicm [Tue, 17 Aug 2021 08:44:52 +0000 (08:44 +0000)]
Be more sophisticated about enabling synchronized updates when there is
an overlay and treat it like the active pane (use for commands which
move the cursor only). When there is an overlay also use it for all
panes and not just the active pane. GitHub issue 2826.
nicm [Tue, 17 Aug 2021 08:22:44 +0000 (08:22 +0000)]
Revert previous; this is not how it should work.
jmc [Tue, 17 Aug 2021 08:19:55 +0000 (08:19 +0000)]
document 11ac as a valid wireless mode; ok stsp
nicm [Tue, 17 Aug 2021 07:14:33 +0000 (07:14 +0000)]
Start sync before drawing popup.
anton [Tue, 17 Aug 2021 05:56:24 +0000 (05:56 +0000)]
Set the battery level sensor status to unknown while the battery is
charging as some devices reports a level of zero during charging.
Prevents certain sensorsd.conf actions from triggering as a battery
level of zero would otherwise set the sensor status to critical.
Thanks to Laurence Tratt <laurie at tratt dot net> for the report.
kevlo [Mon, 16 Aug 2021 14:54:50 +0000 (14:54 +0000)]
Remove obsolete pre-release support code about two early versions of 8260.
No need to access iwx_read_prph and could get the rid of the obsolete comment.
See Linux commit
4adfaf9b2de3a04a9ee9adff6e000e8dbb37bed5
ok stsp@
tb [Mon, 16 Aug 2021 13:54:38 +0000 (13:54 +0000)]
typo in comment
jsg [Mon, 16 Aug 2021 10:38:57 +0000 (10:38 +0000)]
add missing break
ok claudio@
kevlo [Mon, 16 Aug 2021 01:30:27 +0000 (01:30 +0000)]
Remove redundant SIOCSIFMTU ioctl handler already handlded by ether_ioctl().
From Brad
krw [Sun, 15 Aug 2021 13:45:42 +0000 (13:45 +0000)]
Don't check & errx() after each and every invocation of
string_from_line(). Just errx() inside string_from_line() if
getline() fails.
Use strcspn() idiom to nuke '\n' returned by getline().
No functional change.
nicm [Sat, 14 Aug 2021 16:26:29 +0000 (16:26 +0000)]
Some other missing palette NULL checks, from oss-fuzz.
millert [Sat, 14 Aug 2021 14:22:26 +0000 (14:22 +0000)]
Fix iostat per-device values when in boot time mode ('b').
The values should be displayed as-is and not normalized based on
the sleep interval. From Anindya Mukherjee
nicm [Sat, 14 Aug 2021 14:00:07 +0000 (14:00 +0000)]
Do not use NULL palette when clearing.
schwarze [Sat, 14 Aug 2021 13:51:46 +0000 (13:51 +0000)]
print a BAGARG message if -T markdown is requested on man(7) input;
suggested by Michael Stapelberg at debian dot org
schwarze [Sat, 14 Aug 2021 11:30:12 +0000 (11:30 +0000)]
when talking about non-blocking I/O, .Xr both FIONBIO and O_NONBLOCK
in the most concise way i managed to come up with;
suggested by martijn@ and OK millert@ martijn@
nicm [Sat, 14 Aug 2021 08:06:37 +0000 (08:06 +0000)]
Missing argument specifier for -c.
florian [Sat, 14 Aug 2021 07:40:02 +0000 (07:40 +0000)]
Use new val-max-restart: 0 option to stop unwind from trying different
authorities when encountering a validation error.
This only helps one particular case of validation errors: When
authorities are out of sync and some carry old zones. In all other
cases this causes a huge amount of work that will just end in a
SERVFAIL because the result will still be bogus.
OK sthen
florian [Sat, 14 Aug 2021 07:33:35 +0000 (07:33 +0000)]
Fix "no previous prototype" warning
OK sthen as part of a larger diff
florian [Sat, 14 Aug 2021 07:32:46 +0000 (07:32 +0000)]
sync to unbound 1.13.2
OK sthen
jsg [Sat, 14 Aug 2021 03:12:51 +0000 (03:12 +0000)]
call task_del() from destroy_work_on_stack()
prompted by jcs@ reporting a protection fault trap in
drm_mode_rmfb_work_fn() while playing a youtube video in firefox on a
kaby lake machine. He later saw the same trace on tiger lake.
The previous attempt to avoid this situation by changing work flush
functions from taskq_barrier() to taskq_del_barrier() resulted in
suspend sometimes not working on various intel based thinkpads.
The only code we build which calls destroy_work_on_stack() is in
drm_framebuffer.c so the scope of this change is more limited.
Linux only uses destroy_work_on_stack() for debugging so the workqueue
behaviour still doesn't match.
This version is confirmed to not break suspend on x260 by sthen@ and
x280 by tb@ and still avoids the original problem according to jcs@
nicm [Fri, 13 Aug 2021 23:05:40 +0000 (23:05 +0000)]
Tweak how mouse works on popup: only Meta alone resizes or moves, not
Meta with other modifiers; button 2 on the left or top border opens
menu, right or bottom resizes; button 1 on any border moves.
nicm [Fri, 13 Aug 2021 20:04:45 +0000 (20:04 +0000)]
Check callback needs to only return 0 (text should be suppressed) if
menu returns 0, otherwise it should check the popup also.
sthen [Fri, 13 Aug 2021 19:58:45 +0000 (19:58 +0000)]
merge unbound-1.13.2 / regen configure, ok florian
sthen [Fri, 13 Aug 2021 19:55:24 +0000 (19:55 +0000)]
import unbound 1.13.2, ok florian@
nicm [Fri, 13 Aug 2021 19:55:11 +0000 (19:55 +0000)]
Fill in some other bits on new panes.
nicm [Fri, 13 Aug 2021 19:27:25 +0000 (19:27 +0000)]
Add menu options to convert a popup into a pane.
nicm [Fri, 13 Aug 2021 19:25:24 +0000 (19:25 +0000)]
Adjust overlay check callback before drawing data from pty.
nicm [Fri, 13 Aug 2021 18:54:54 +0000 (18:54 +0000)]
Add a menu when a popup is present (mouse only for now).
nicm [Fri, 13 Aug 2021 17:03:29 +0000 (17:03 +0000)]
Add -B flag to remove border from popup.
stsp [Fri, 13 Aug 2021 13:13:11 +0000 (13:13 +0000)]
Plug memory leak in an unlikely error path of iwx_umac_scan_v14().
CID
1506261
jmc [Fri, 13 Aug 2021 12:37:28 +0000 (12:37 +0000)]
update usage() to reflect s->step change;
martijn [Fri, 13 Aug 2021 11:27:33 +0000 (11:27 +0000)]
Make -b, -c, and -b mutually exclusive.
Help/Feedback/OK schwarze@, tb@
schwarze [Fri, 13 Aug 2021 10:56:54 +0000 (10:56 +0000)]
During line editing, let Ctrl-C discard the current input line and
provide a fresh prompt instead of exiting the program. This aligns
behaviour with bc(1), ftp(1), sftp(1), and all the shells.
OK naddy@
Both martijn@ and naddy@ point out that this program might profit
from signal handling during more of its code, but that would require
more code inspection, design work, and testing which i'm not planning
to do right now. Consider standard behaviour during line editing
as a first step.
tb [Fri, 13 Aug 2021 10:45:26 +0000 (10:45 +0000)]
Simplify an if clause using De Morgan's law.
Using the fact that since r1.53 prec is always non-negative and that
r1.50 added an initialization of use_unif, we can eliminate a couple
of lines and make the intention clearer.
ok martijn
tb [Fri, 13 Aug 2021 10:40:23 +0000 (10:40 +0000)]
Make sure prec is non-negative before calling getformat()
If jot is called without arguments, getformat() produces the illegal
"%.-1f" format string. __vfprintf() misinterprets the - as a left
adjust flag and uses a precision of 0. Set prec = 0 so that we use
"%.0f" in this situation.
ok martijn
schwarze [Fri, 13 Aug 2021 10:21:25 +0000 (10:21 +0000)]
Stop using a while loop for code that runs at most once,
and garbage collect an automatic variable that is no longer needed
and a comment that was probably contributed by Capt. Obvious.
No functional change.
OK millert@ martijn@
martijn [Fri, 13 Aug 2021 07:59:04 +0000 (07:59 +0000)]
The s argument isn't used as seed anymore, so increase readability by
renaming it to step, which matches the name in the code.
OK tb@
martijn [Fri, 13 Aug 2021 07:56:34 +0000 (07:56 +0000)]
Apply de Morgan's rule to shave of 4 LoC.
OK tb@
nicm [Fri, 13 Aug 2021 07:37:58 +0000 (07:37 +0000)]
Break message type stuff out into its own header.
nicm [Fri, 13 Aug 2021 06:52:51 +0000 (06:52 +0000)]
Change focus to be driven by events rather than walking all panes at end
of event loop, this way the ordering of in and out can be enforced.
GitHub issue 2808.
nicm [Fri, 13 Aug 2021 06:50:42 +0000 (06:50 +0000)]
Set return code for confirm-before and make command-prompt also block,
GitHub issue 2822.
gnezdo [Fri, 13 Aug 2021 01:24:22 +0000 (01:24 +0000)]
Remove unused ure_stop_task (from Christian Ludwig of genua.de)
The ure_stop_task is not scheduled anywhere.
Compile tested here.
djm [Thu, 12 Aug 2021 23:59:25 +0000 (23:59 +0000)]
mention that CASignatureAlgorithms accepts +/- similarly to the
other algorithm list directives; ok jmc bz#3335
nicm [Thu, 12 Aug 2021 20:46:30 +0000 (20:46 +0000)]
Make newline a style delimiter as well so they can cross multiple lines
for readability.
nicm [Thu, 12 Aug 2021 20:44:49 +0000 (20:44 +0000)]
Now that styles can contain formats, they need to be expanded when
inserted into the status line.
nicm [Thu, 12 Aug 2021 20:09:34 +0000 (20:09 +0000)]
Use COLOUR_DEFAULT not hardcoded 8.
nicm [Thu, 12 Aug 2021 19:47:05 +0000 (19:47 +0000)]
Evaluate styles with the pane variables.
krw [Thu, 12 Aug 2021 17:30:52 +0000 (17:30 +0000)]
Oops. Lost a 'nitems()' chunk during last commit. Don't
segfault on invalid edit commands.