openbsd
6 years agoForgot to rename pf_frent_holes() prototype in previous commit.
bluhm [Tue, 4 Sep 2018 20:34:10 +0000 (20:34 +0000)]
Forgot to rename pf_frent_holes() prototype in previous commit.

6 years agoAvoid traversing the list of fragment entris to check whether the
bluhm [Tue, 4 Sep 2018 19:09:39 +0000 (19:09 +0000)]
Avoid traversing the list of fragment entris to check whether the
pf(4) reassembly is complete.  Instead count the holes that are
created when inserting a fragment.  If there are no holes left, the
fragments are continuous.
idea from claudio@; OK claudio@ sashan@

6 years agoremove useless session flag
eric [Tue, 4 Sep 2018 17:19:00 +0000 (17:19 +0000)]
remove useless session flag

ok gilles@

6 years agomove readmes detection to find_extractible, so that only changed/new readmes
espie [Tue, 4 Sep 2018 14:46:12 +0000 (14:46 +0000)]
move readmes detection to find_extractible, so that only changed/new readmes
are recorded.

Be more specific and list those readmes in alphabetic order at end.

6 years agodocument PKGSTEM changes
espie [Tue, 4 Sep 2018 13:57:46 +0000 (13:57 +0000)]
document PKGSTEM changes

6 years agoupon mda failure, smtpd would assume tempfail and retry. this is at odds
gilles [Tue, 4 Sep 2018 13:04:42 +0000 (13:04 +0000)]
upon mda failure, smtpd would assume tempfail and retry. this is at odds
with the other MTA which assume a permfail unless the exit status is one
of a specific set. make smtpd honour the same exit statuses as postfix.

note that all errors that occur before the user mda is executed (fork, pipe
and related) are still considered tempfail, only errors coming from the mda
itself are handled as permfail.

this commit is a temporary solution as i believe the SIGCHLD handler is way
more complex than it should be and we'll simplify it after 6.4 is out.

ok eric@

6 years agoCorrect usbd_abort_pipe() signature.
mpi [Tue, 4 Sep 2018 12:46:32 +0000 (12:46 +0000)]
Correct usbd_abort_pipe() signature.

From Artturi Alm.

6 years agoIntroduce inet4applymask() which does the same as inet6applymask() and
claudio [Tue, 4 Sep 2018 12:00:29 +0000 (12:00 +0000)]
Introduce inet4applymask() which does the same as inet6applymask() and
can be used instead of doing direct fiddling around with struct in_addr.
Use it in a few cases where it makes the code more similar between INET
and INET6 case.
OK denis@

6 years agoInstead of bzero() use an assinment to zero struct in_addr.
claudio [Tue, 4 Sep 2018 10:48:39 +0000 (10:48 +0000)]
Instead of bzero() use an assinment to zero struct in_addr.

6 years agoremove unused flags
eric [Tue, 4 Sep 2018 10:08:22 +0000 (10:08 +0000)]
remove unused flags

6 years agokevent: Don't poll for nonzero timeouts < 1us.
cheloha [Tue, 4 Sep 2018 02:38:25 +0000 (02:38 +0000)]
kevent: Don't poll for nonzero timeouts < 1us.

Instead of truncating nanosecond timeouts to zero here and polling, we
should round up to a delay of at least a tick, just like all the other
timespec syscalls.

Fixed in NetBSD kern_event.c v1.62 and FreeBSD r247804.

ok millert@ visa@

6 years agoRemove bits about MODMONO_NANT in the lang/mono section; devel/nant does
bcallah [Mon, 3 Sep 2018 22:12:37 +0000 (22:12 +0000)]
Remove bits about MODMONO_NANT in the lang/mono section; devel/nant does
not exist anymore.
ok jca@ daniel@

6 years agoregen
jcs [Mon, 3 Sep 2018 19:52:51 +0000 (19:52 +0000)]
regen

6 years agoadd another toshiba nvme
jcs [Mon, 3 Sep 2018 19:51:48 +0000 (19:51 +0000)]
add another toshiba nvme

6 years agosimplify the smtp_command() function by splitting cmd handlers into
gilles [Mon, 3 Sep 2018 19:01:29 +0000 (19:01 +0000)]
simplify the smtp_command() function by splitting cmd handlers into
two sets of functions smtp_check_<verb> and smtp_proceed_<verb>. we
use the check functions to validate that verb is acceptable at this
point of a session and proceed to actually move the session forward

ok eric@

6 years agoStop using composite EVP_CIPHER AEADs.
jsing [Mon, 3 Sep 2018 18:00:50 +0000 (18:00 +0000)]
Stop using composite EVP_CIPHER AEADs.

The composite AEADs are "stitched" mode ciphers, that are only supported on
some architectures/CPUs and are designed to be faster than a separate
EVP_CIPHER and EVP_MD implementation. The three AEADs are used for less
than ideal cipher suites (if you have hardware support that these use
there are better cipher suite options), plus continuing to support AEADs
via EVP_CIPHER is creating additional code complexity.

ok inoguchi@ tb@

6 years agoStop handling AES-GCM via ssl_cipher_get_evp().
jsing [Mon, 3 Sep 2018 17:45:24 +0000 (17:45 +0000)]
Stop handling AES-GCM via ssl_cipher_get_evp().

All of the AES-GCM ciphersuites use the EVP_AEAD interface, so there is no
need to support them via EVP_CIPHER.

ok inoguchi@ tb@

6 years agobump max line size
gilles [Mon, 3 Sep 2018 17:41:55 +0000 (17:41 +0000)]
bump max line size

ok eric@

6 years agoClean up SSL_DES and SSL_IDEA remnants.
jsing [Mon, 3 Sep 2018 17:41:13 +0000 (17:41 +0000)]
Clean up SSL_DES and SSL_IDEA remnants.

All ciphersuites that used these encryption algorithms were removed some
time ago.

ok bcook@ inoguchi@ tb@

6 years agodocument "wrapper" in the local section; from matt schwartz
jmc [Mon, 3 Sep 2018 13:27:43 +0000 (13:27 +0000)]
document "wrapper" in the local section; from matt schwartz
ok gilles

while here, fix some spelling at eol whitespace introduced in previous;

6 years agouse new smarthost syntax in log message
eric [Mon, 3 Sep 2018 11:48:27 +0000 (11:48 +0000)]
use new smarthost syntax in log message

6 years agoUpdate and document syntax for smarthost string in smtpd.conf:
eric [Mon, 3 Sep 2018 11:30:14 +0000 (11:30 +0000)]
Update and document syntax for smarthost string in smtpd.conf:

- the +auth specifier is removed: it is implied by the presence of an
  auth label in the rest of the string
- secure:// is removed: use smtp+tls:// or smtps://
- tls:// is replaced by smtp+tls://
- smtp:// becomes SMTP with opportunistic STARTTLS
- smtp+tls:// becomes SMTP with mandatory STARTTLS

Adjust your config file accordingly.

ok gilles@

6 years agoAllow a large line number to go to the end with goto-line, from Mark
nicm [Mon, 3 Sep 2018 08:51:43 +0000 (08:51 +0000)]
Allow a large line number to go to the end with goto-line, from Mark
Kelly in GitHub issue 1460.

6 years agoFix selection test, from Takeshi Banse.
nicm [Mon, 3 Sep 2018 08:47:27 +0000 (08:47 +0000)]
Fix selection test, from Takeshi Banse.

6 years agoRemove round_buffersize function from drivers using the default buffersize
miko [Mon, 3 Sep 2018 05:37:32 +0000 (05:37 +0000)]
Remove round_buffersize function from drivers using the default buffersize
provided by the upper layer audio driver.

ok ratchov@

6 years agomailaddr matching was working for the exact and domain match, but not when
gilles [Sun, 2 Sep 2018 21:06:44 +0000 (21:06 +0000)]
mailaddr matching was working for the exact and domain match, but not when
only a user-part was supplied.

ok millert@ and eric@

6 years agoRemove a few unnecessary casts
tb [Sun, 2 Sep 2018 20:29:01 +0000 (20:29 +0000)]
Remove a few unnecessary casts

6 years agoPrint SKIPPED if package wycheproof-testvectors is missing. This
bluhm [Sun, 2 Sep 2018 20:09:29 +0000 (20:09 +0000)]
Print SKIPPED if package wycheproof-testvectors is missing.  This
is the magic string that is recognized by my test framework.
OK tb@

6 years agoRemove ECDH from TODO list. Done!
tb [Sun, 2 Sep 2018 17:29:17 +0000 (17:29 +0000)]
Remove ECDH from TODO list. Done!

6 years agoUnify FAIL printfs.
tb [Sun, 2 Sep 2018 17:28:01 +0000 (17:28 +0000)]
Unify FAIL printfs.

6 years agoAfter libcrypto/ecdh/ech_key.c -r1.8 fixed the failing test cases, remove
tb [Sun, 2 Sep 2018 17:24:02 +0000 (17:24 +0000)]
After libcrypto/ecdh/ech_key.c -r1.8 fixed the failing test cases, remove
two noisy INFO and reorder things a bit.

6 years agoElliptic curve arithmetic only makes sense between points that belong to
tb [Sun, 2 Sep 2018 17:20:31 +0000 (17:20 +0000)]
Elliptic curve arithmetic only makes sense between points that belong to
the same curve. Some Wycheproof tests violate this assumption, making
ECDH_compute_key() compute and return garbage. Check that pub_key lies
on the curve of the private key so that the calculations make sense.
Most paths that get here have this checked (in particular those from
OpenSSH and libssl), but one might get here after using d2i_* or manual
computation.

discussed with & ok jsing;
"good catch!" markus

6 years agoRun Wycheproof ECDH tests against libcrypto. Some tests currently fail,
tb [Sun, 2 Sep 2018 17:12:01 +0000 (17:12 +0000)]
Run Wycheproof ECDH tests against libcrypto. Some tests currently fail,
will be fixed with the next commit to libcrypto.

6 years agoUse a Boolean rather than repeated string comparison.
tb [Sun, 2 Sep 2018 17:05:51 +0000 (17:05 +0000)]
Use a Boolean rather than repeated string comparison.

6 years agoDocument how SIGHUP and log rotation works with syslogd(8). Do not
bluhm [Sun, 2 Sep 2018 14:32:12 +0000 (14:32 +0000)]
Document how SIGHUP and log rotation works with syslogd(8).  Do not
mention the PID file.  While there, use .Nm consistently.
wording from schwarze@; deraadt@ cheloha@ millert@ agree

6 years agoupdate tradcpp to 0.5.2
jsg [Sun, 2 Sep 2018 08:28:05 +0000 (08:28 +0000)]
update tradcpp to 0.5.2
ok miko@ bcallah@ deraadt@

6 years agoUpdate nmea(4) wrt talker ids.
landry [Sun, 2 Sep 2018 08:14:25 +0000 (08:14 +0000)]
Update nmea(4) wrt talker ids.

Prodded by semarie@, reworded by jmc@

6 years agoproperly deal with MAILER-DAEMON sender in LMTP
gilles [Sat, 1 Sep 2018 21:20:32 +0000 (21:20 +0000)]
properly deal with MAILER-DAEMON sender in LMTP

reported and fix tested by Mark Kane

6 years agoMatch on interfaces with Digitizers/Touchscreen collections, as long
jcs [Sat, 1 Sep 2018 20:50:16 +0000 (20:50 +0000)]
Match on interfaces with Digitizers/Touchscreen collections, as long
as they have an X usage.

Should get basic touchscreen functionality on some common
HID-over-i2c devices like laptops.

6 years agoFor touchscreen devices, use the first HUG_X/Y usage pages found
jcs [Sat, 1 Sep 2018 20:48:00 +0000 (20:48 +0000)]
For touchscreen devices, use the first HUG_X/Y usage pages found
when looking for logical min/max of screen.

ok kettenis

6 years agoclarify the forward-only delivery action
gilles [Sat, 1 Sep 2018 19:56:28 +0000 (19:56 +0000)]
clarify the forward-only delivery action

6 years agoPrevent ospf6d from starting when another process is listening on the
remi [Sat, 1 Sep 2018 19:21:10 +0000 (19:21 +0000)]
Prevent ospf6d from starting when another process is listening on the
control socket.

ok florian@

6 years agoAccidentally relayd(8) closed file descriptor 0 in the pfe child
bluhm [Sat, 1 Sep 2018 18:09:14 +0000 (18:09 +0000)]
Accidentally relayd(8) closed file descriptor 0 in the pfe child
process.  If env->sc_snmp is initialized with 0, snmp_init() closes
it.  Set it to -1 to prevent the close(2).
OK reyk@ benno@ millert@

6 years agoFix errno for post-lock unveil calls
deraadt [Sat, 1 Sep 2018 17:02:12 +0000 (17:02 +0000)]
Fix errno for post-lock unveil calls
from Jan Klemkow

6 years agoRemove 'overlap_allowance' parameter from bootstrap().
krw [Sat, 1 Sep 2018 16:55:29 +0000 (16:55 +0000)]
Remove 'overlap_allowance' parameter from bootstrap().

It was added to support Vax. Vax is dead Jim.

ok deraadt@

6 years agoTweak comment.
tb [Sat, 1 Sep 2018 16:23:15 +0000 (16:23 +0000)]
Tweak comment.

6 years agosync fatalx() message between smtp_client_{state,response}; ok stsp@ gilles@
miko [Sat, 1 Sep 2018 12:03:31 +0000 (12:03 +0000)]
sync fatalx() message between smtp_client_{state,response}; ok stsp@ gilles@

6 years agoMake 'ifconfig nwid' override 'ifconfig join'.
stsp [Sat, 1 Sep 2018 08:20:56 +0000 (08:20 +0000)]
Make 'ifconfig nwid' override 'ifconfig join'.

There was no way to override a decision made by join's network
selection algorithm (apart from disabling the feature by clearing
the join list). Automatic selection is based on heuristics which
cannot always guess correctly so we need to provide an override.

One specific problem was that if 'nwid foo' and 'nwid bar' were
both visible in the scan and only 'nwid foo' was a member of the
join list, then there was no way at all to connect to 'nwid bar'.
The wireless stack would keep selecting 'nwid foo' instead.

'ifconfig iwm0 nwid bar' command will now disable automatic
network selection and force the use of ESSID 'bar'.
Any of these commands will re-enable automatic network selection:
   ifconfig iwm0 -nwid
   ifconfig iwm0 nwid ''
   ifconfig iwm0 join some-network-id

ok phessler@ deraadt@

6 years agoupdate currency exchange rates;
jmc [Sat, 1 Sep 2018 06:27:32 +0000 (06:27 +0000)]
update currency exchange rates;

6 years agoRecognize more talker IDs when parsing NMEA RMC messages
landry [Sat, 1 Sep 2018 06:09:26 +0000 (06:09 +0000)]
Recognize more talker IDs when parsing NMEA RMC messages

The NMEA 0183 standard says that the first two chars correspond to the
'source' of the message, right now we were only looking for 'GP' prefix
for 'GPS', but this can also be 'GL' for Glonass, 'BD' for BeiDou, 'GA'
for Galileo, or 'GN' for a generic GNSS source.

Match the RMC messages from all those variants, with this i'm able to
use my navilock nl-8002u (which uses GNRMC) as a timedelta sensor for
ntpd, and i have my GPS position in the nmea(4) sensors.

ok deraadt@

6 years agoRemove RSA-PSS from todo-list
tb [Sat, 1 Sep 2018 05:57:23 +0000 (05:57 +0000)]
Remove RSA-PSS from todo-list

6 years agoRun Wycheproof RSASSA-PSS testvectors against libcrypto.
tb [Sat, 1 Sep 2018 05:56:24 +0000 (05:56 +0000)]
Run Wycheproof RSASSA-PSS testvectors against libcrypto.

6 years agojmc points out that the Nd should match the description in pci(4)
jmatthew [Fri, 31 Aug 2018 22:35:01 +0000 (22:35 +0000)]
jmc points out that the Nd should match the description in pci(4)

6 years agoUnsetting Initialized during syslogd die() is a relic from the time
bluhm [Fri, 31 Aug 2018 19:06:08 +0000 (19:06 +0000)]
Unsetting Initialized during syslogd die() is a relic from the time
when we had real signal handlers.  But now we use libevent, so
remove the old logic.
OK deraadt@ millert@

6 years agothe main process must chdir to /, since it cannot have daemon() do the
deraadt [Fri, 31 Aug 2018 18:45:02 +0000 (18:45 +0000)]
the main process must chdir to /, since it cannot have daemon() do the
job at startup.  After much anguish I accept dlg's solution of chdir
for the problem ("starting ntpd on a filesystem I want to unmount"),
but we cannot change the main-process daemon() call.  Why?  Because
the ntpd privsep design predates more modern designs where the config
file is parsed once, and configuration marshalled to the fork+exec
children.  Instead each ntpd process re-parses the config, and if
we chdir before fork+exec startup, it will move the basedir causing
-f "relativepath" to fail.

discussed with florian

6 years agoRemove unused argument to tls1_change_cipher_state_cipher().
jsing [Fri, 31 Aug 2018 18:31:34 +0000 (18:31 +0000)]
Remove unused argument to tls1_change_cipher_state_cipher().

6 years agoInstead of enumerating the files to clean by hand, set PROGS=${TESTS}.
tb [Fri, 31 Aug 2018 17:35:21 +0000 (17:35 +0000)]
Instead of enumerating the files to clean by hand, set PROGS=${TESTS}.

Suggested by jsing

6 years agoremove shadow variables for 'error' in usbioctl(); ok mpi@ ratchov@
miko [Fri, 31 Aug 2018 16:32:31 +0000 (16:32 +0000)]
remove shadow variables for 'error' in usbioctl(); ok mpi@ ratchov@

6 years agoclarify that patterns are extended regular expressions; ok millert@ phessler@
naddy [Fri, 31 Aug 2018 15:56:42 +0000 (15:56 +0000)]
clarify that patterns are extended regular expressions; ok millert@ phessler@

6 years agoDont print "default" for "0.0.0.0" if it is a host route.
yasuoka [Fri, 31 Aug 2018 15:18:02 +0000 (15:18 +0000)]
Dont print "default" for "0.0.0.0" if it is a host route.
diff from asou at soum.co.jp

ok claudio kn

6 years agoCadd a TDB `tdb' instead of `sa'.
mpi [Fri, 31 Aug 2018 12:55:46 +0000 (12:55 +0000)]
Cadd a TDB `tdb' instead of `sa'.

No functionnal change.

ok deraadt@, bluhm@, visa@

6 years agoPass the correct size to free(9) in the error path of db_ctf_decompress().
bluhm [Fri, 31 Aug 2018 11:57:04 +0000 (11:57 +0000)]
Pass the correct size to free(9) in the error path of db_ctf_decompress().
OK jasper@

6 years agoswitch to improved incoming message parser:
eric [Fri, 31 Aug 2018 07:28:27 +0000 (07:28 +0000)]
switch to improved incoming message parser:
- simpler interface not using callbacks
- no hard-coded line length
- avoid unnecessary string copy

ok gilles@

6 years agoich variable is only used in DPRINTF() and uaudio_get_cluster_nchan()
miko [Fri, 31 Aug 2018 07:18:18 +0000 (07:18 +0000)]
ich variable is only used in DPRINTF() and uaudio_get_cluster_nchan()
has no side effect, so setting value for ich can be moved under
UAUDIO_DEBUG.

ok ratchov

6 years agofix fd leak in an error path
jsg [Fri, 31 Aug 2018 05:20:36 +0000 (05:20 +0000)]
fix fd leak in an error path
ok mpi@

6 years agoMove kcov device definitions into <sys/conf.h>.
visa [Fri, 31 Aug 2018 04:20:37 +0000 (04:20 +0000)]
Move kcov device definitions into <sys/conf.h>.

OK anton@ deraadt@ kettenis@ mpi@

6 years agosync
jcs [Fri, 31 Aug 2018 02:32:57 +0000 (02:32 +0000)]
sync

6 years agoAdd more uhid devices, 8 total
jcs [Fri, 31 Aug 2018 02:32:29 +0000 (02:32 +0000)]
Add more uhid devices, 8 total

ok deraadt

6 years agomention bnxt(4)
jmatthew [Fri, 31 Aug 2018 00:10:20 +0000 (00:10 +0000)]
mention bnxt(4)

6 years agoenable bnxt(4)
jmatthew [Fri, 31 Aug 2018 00:09:13 +0000 (00:09 +0000)]
enable bnxt(4)

6 years agoUse correct environment variable to access remote machine.
bluhm [Thu, 30 Aug 2018 19:34:37 +0000 (19:34 +0000)]
Use correct environment variable to access remote machine.

6 years agoMake sure to clean up the .d files with 'make clean'
tb [Thu, 30 Aug 2018 18:36:59 +0000 (18:36 +0000)]
Make sure to clean up the .d files with 'make clean'

6 years agoNuke ssl_pending/ssl_shutdown function pointers.
jsing [Thu, 30 Aug 2018 16:56:16 +0000 (16:56 +0000)]
Nuke ssl_pending/ssl_shutdown function pointers.

ssl3_pending() is used for all protocols and dtls1_shutdown() just calls
ssl3_shutdown(), so just call the appropriate function directly instead.

ok beck@ inoguchi@ tb@

6 years agoMove .dynstr before _edata to fix image size calculation. The PE
patrick [Thu, 30 Aug 2018 14:39:46 +0000 (14:39 +0000)]
Move .dynstr before _edata to fix image size calculation.  The PE
header contains an image size field that is calculated using the
difference between the start of the header and edata.  Since we
copy out .dynstr into the EFI binary, make sure that .dynstr is
before edata so that it's included in the image size.  This makes
it consistent with efiboot on armv7.

ok kettenis@

6 years agoPractice safe free()'ing by setting alloc to NULL to
krw [Thu, 30 Aug 2018 13:07:19 +0000 (13:07 +0000)]
Practice safe free()'ing by setting alloc to NULL to
prevent double free.

Spotted by jsg@, simpler fix by myself and miko@.

ok jsg@

6 years agoNever let FS_RAID partition be named, a.k.a. given a mount point.
krw [Thu, 30 Aug 2018 12:30:08 +0000 (12:30 +0000)]
Never let FS_RAID partition be named, a.k.a. given a mount point.

Nuke pointless and inconsistant error message before one get_mp()
invocation.

ok otto@

6 years agofix double free in mib_carpgroupget() error path
jsg [Thu, 30 Aug 2018 12:25:40 +0000 (12:25 +0000)]
fix double free in mib_carpgroupget() error path
ok claudio@ miko@

6 years agofix memory leak in an error path
jsg [Thu, 30 Aug 2018 12:14:30 +0000 (12:14 +0000)]
fix memory leak in an error path
ok patrick@

6 years agofix use-after-free
eric [Thu, 30 Aug 2018 11:58:01 +0000 (11:58 +0000)]
fix use-after-free

spotted by jsg@
input from jsg@ miko@

6 years agoSet up an aggregation buffer ring and configure placement mode so data
jmatthew [Thu, 30 Aug 2018 11:18:21 +0000 (11:18 +0000)]
Set up an aggregation buffer ring and configure placement mode so data
that doesn't fit in the rx buffer goes into an aggregation buffer,
allowing jumbo frames to be received.  Using 8k aggregation buffers
means we'll only ever need one per packet.

When receiving jumbos, sometimes we get an interrupt before all three
of the completion events are ready, in which case we should not consume
the events that are ready.  Expanding the completion ring makes this
happen less frequently, so allocate four cp ring pages instead of one.

6 years agoMove softbutton check before the check that skips attaching certain acpi
kettenis [Thu, 30 Aug 2018 10:11:34 +0000 (10:11 +0000)]
Move softbutton check before the check that skips attaching certain acpi
devices.  Fixes regression caused by matching on _CID in addition to
matching on _HID.

ok matthieu@, mlarkin@

6 years agoUse past tense for treated as withdrawn since the withdraw happened in the past.
claudio [Thu, 30 Aug 2018 08:30:05 +0000 (08:30 +0000)]
Use past tense for treated as withdrawn since the withdraw happened in the past.
From and with jmc@

6 years agoRemove dead code related to tty allocation for ptys. Back in revision 1.17 of
anton [Thu, 30 Aug 2018 06:16:30 +0000 (06:16 +0000)]
Remove dead code related to tty allocation for ptys. Back in revision 1.17 of
kern/tty_pty.c, allocation of ptys was made dynamic with the introduction of
check_pty(). Every time a new struct pty is allocated its corresponding struct
tty is also allocated. It's therefore no longer necessary to ensure that a pty
has a tty allocated after calling check_pty().

ok deraadt@ millert@ mpi@ visa@

6 years agoSplit the system-wide list of all futexes into process-specific lists
visa [Thu, 30 Aug 2018 03:30:25 +0000 (03:30 +0000)]
Split the system-wide list of all futexes into process-specific lists
of private futexes and a shared list of shared futexes. This speeds up
futex lookups.

Tested by and OK krw@
OK mpi@

6 years ago/dev/usbN devices no longer support poll as of 2011 when "USB
jcs [Thu, 30 Aug 2018 03:26:08 +0000 (03:26 +0000)]
/dev/usbN devices no longer support poll as of 2011 when "USB
events" were ripped out, so remove the sentence about monitoring for
USB device attachment to avoid filling developers with hope only to
crush their spirit when they implement the poll code only to find
out it doesn't do anything at all.

6 years agoFix "route add 0.0.0.0/0 <gateway>" to work properly. Shift count
yasuoka [Thu, 30 Aug 2018 01:32:53 +0000 (01:32 +0000)]
Fix "route add 0.0.0.0/0 <gateway>" to work properly.  Shift count
beyonds the integer width results undefined behavior in C spec.
Actually the count is masked by 0x1f on amd64.  Found by asou at
soum.co.jp

ok claudio kn schwarze

6 years agoRemove unused struct member.
kettenis [Wed, 29 Aug 2018 20:18:14 +0000 (20:18 +0000)]
Remove unused struct member.

ok deraadt@, mlarkin@

6 years agoIntroduce 'bgpctl show rib error' to show all prefixes that got flagged
claudio [Wed, 29 Aug 2018 19:52:23 +0000 (19:52 +0000)]
Introduce 'bgpctl show rib error' to show all prefixes that got flagged
invalid because of a soft parsing error. These prefixes are never eligible
or valid but the listing may help to understand what is going on.
'bgpctl show rib error' run automatically on Adj-RIB-In since that is the only
RIB that has such prefixes.
OK sthen@ denis@

6 years agoMake it possible to filter the RIB for invalid path which resulted in withdrawn
claudio [Wed, 29 Aug 2018 19:47:47 +0000 (19:47 +0000)]
Make it possible to filter the RIB for invalid path which resulted in withdrawn
prefixes. This way it should be possible to inspect which prefixes have been
removed. While there also fix dumping of the Adj-RIB-In when requested to do so.
OK sthen@ denis@

6 years agoAES is now done also.
tb [Wed, 29 Aug 2018 19:22:32 +0000 (19:22 +0000)]
AES is now done also.

6 years agoPass algorithm as a string to all *TestGroup functions for consistency.
tb [Wed, 29 Aug 2018 19:20:22 +0000 (19:20 +0000)]
Pass algorithm as a string to all *TestGroup functions for consistency.

6 years agoRun Wycheproof AES-GCM testvectors against libcrypto.
tb [Wed, 29 Aug 2018 19:18:20 +0000 (19:18 +0000)]
Run Wycheproof AES-GCM testvectors against libcrypto.

6 years agoCalculate and check tag during AES-CCM encryption test.
tb [Wed, 29 Aug 2018 19:00:41 +0000 (19:00 +0000)]
Calculate and check tag during AES-CCM encryption test.

6 years agotypo
tb [Wed, 29 Aug 2018 18:59:22 +0000 (18:59 +0000)]
typo

6 years agoDon't fatal on keys of invalid sice, just print an INFO.
tb [Wed, 29 Aug 2018 18:58:46 +0000 (18:58 +0000)]
Don't fatal on keys of invalid sice, just print an INFO.

6 years agoReset line flag when clearing selection, GitHub issue 1454.
nicm [Wed, 29 Aug 2018 18:54:23 +0000 (18:54 +0000)]
Reset line flag when clearing selection, GitHub issue 1454.

6 years agoclean up the imsg structure for getnameinfo lookup results.
eric [Wed, 29 Aug 2018 17:56:17 +0000 (17:56 +0000)]
clean up the imsg structure for getnameinfo lookup results.
fix a crash with ill-configured reverse address.

found by jturner@

ok jturner@ gilles@

6 years agorecognize kcov ioctl commands; ok deraadt@ mpi@ visa@
anton [Wed, 29 Aug 2018 17:47:07 +0000 (17:47 +0000)]
recognize kcov ioctl commands; ok deraadt@ mpi@ visa@

6 years agomake ld.so build reproducible by including a FILE symbol and by removing
robert [Wed, 29 Aug 2018 15:56:46 +0000 (15:56 +0000)]
make ld.so build reproducible by including a FILE symbol and by removing
-x from ld; this is only used for syspatch creation so builds are still
randomized just as before

ok guenther@

6 years agoBetter description for inet6 eui64, plus consistency tweaks
jca [Wed, 29 Aug 2018 13:47:19 +0000 (13:47 +0000)]
Better description for inet6 eui64, plus consistency tweaks

from Callum R. Davies, ok jmc@