anton [Sun, 23 Sep 2018 08:56:19 +0000 (08:56 +0000)]
Document bufferevent_setwatermark(). Initial diff from Geoff Hill on tech@ with
some tweaks.
With feedback and ok jmc@
robert [Sun, 23 Sep 2018 03:54:47 +0000 (03:54 +0000)]
add 6.5 syspatch public key
kettenis [Sat, 22 Sep 2018 17:41:52 +0000 (17:41 +0000)]
Remap the UEFI buffer early such that we can use a write combining mapping
which speeds things up considerably compared to an uncached mapping.
ok deraadt@
kettenis [Sat, 22 Sep 2018 17:40:57 +0000 (17:40 +0000)]
Only clear the character backing store when the RI_CLEAR flag is set.
ok deraadt@
millert [Sat, 22 Sep 2018 17:10:28 +0000 (17:10 +0000)]
Remove unused Table_size define and digits() function.
millert [Sat, 22 Sep 2018 16:50:35 +0000 (16:50 +0000)]
Use user_from_uid() and uid_from_user() directly. The wrappers
in username.c are now so simple there is no longer a good reason
to use them. OK deraadt@
tb [Sat, 22 Sep 2018 15:53:38 +0000 (15:53 +0000)]
Add a comment on the acceptable RSASSA cases.
tb [Sat, 22 Sep 2018 14:12:47 +0000 (14:12 +0000)]
gather statistics in checkAead{Open,Seal}() as well.
stsp [Sat, 22 Sep 2018 13:55:55 +0000 (13:55 +0000)]
Back out the following if_iwm.c revisions:
r1.232 Fix length checks in the receive path of iwm(4)
r1.230 Add monitor mode support to iwm(4)
r1.229 Implement Rx of multiple frames per interrupt in the iwm(4) driver
There is an apparent block-ack problem (base.tgz takes 8 hours to download)
which goes away when these changes are reverted. To be revisited after release.
ok deraadt@
tb [Sat, 22 Sep 2018 13:42:46 +0000 (13:42 +0000)]
remove some unneeded checks
tb [Sat, 22 Sep 2018 11:00:25 +0000 (11:00 +0000)]
gather and print some statistics on the acceptable cases we need to
look into
stsp [Sat, 22 Sep 2018 10:12:42 +0000 (10:12 +0000)]
Enable USB bwfm(4) on macppc RAMDISK as well, matching GENERIC.
Passes 'make release' build.
fcambus [Sat, 22 Sep 2018 09:12:36 +0000 (09:12 +0000)]
Harmonize spacing after ellipses in displayed messages.
We were using spacing after ellipses in an inconsistent way in the
installer. Standardize on using "... " everywhere and take into account
the cursor position while we are waiting for the task to complete: the
cursor is now always positioned after the last dot, and the space is
added when displaying completion confirmation.
While there, also take cursor position into account in vfs_shutdown(),
and remove the extra leading space before ticks in dhclient.
OK deraadt@
tb [Sat, 22 Sep 2018 06:06:36 +0000 (06:06 +0000)]
more flags printing
millert [Sat, 22 Sep 2018 02:47:23 +0000 (02:47 +0000)]
Correct the uid_from_user() and gid_from_group() comments.
visa [Sat, 22 Sep 2018 02:20:44 +0000 (02:20 +0000)]
Revert previous. It broke /etc/rc.
Prompted by kn@
procter [Sat, 22 Sep 2018 02:18:19 +0000 (02:18 +0000)]
fix uid -> username lookup
ok deraadt@
tb [Sat, 22 Sep 2018 00:29:13 +0000 (00:29 +0000)]
improve logic involving acceptableAudit
tb [Sat, 22 Sep 2018 00:14:37 +0000 (00:14 +0000)]
Swap order of "action" and "wt.Flags" in a few fmt.Printfs.
tb [Sat, 22 Sep 2018 00:10:18 +0000 (00:10 +0000)]
Introduce a couple of convenience targets to help with auditing the
acceptable cases.
tb [Fri, 21 Sep 2018 23:16:16 +0000 (23:16 +0000)]
Print the flags field in INFO: and FAIL: messages. It's helpful in
identifying the important failures while auditing.
kn [Fri, 21 Sep 2018 20:45:50 +0000 (20:45 +0000)]
sync host*() changes from pfctl
This simplifies host() and merges host_v{4,6}() into host_ip() as recently
done for pfctl and ntpd.
Tested and OK denis, OK deraadt, "go ahead" benno
millert [Fri, 21 Sep 2018 19:13:49 +0000 (19:13 +0000)]
Use password/group cache functions and avoid stashing a pointer to
the return value of getgrgid(3) or getgrnam(3) which relies on
undefined behavior. The rdist server will now use getgroups(2) to
determine group membership of the invoking user. In addition, there
is now one implementation of tilde expansion instead of two.
OK tb@ tim@
millert [Fri, 21 Sep 2018 19:00:45 +0000 (19:00 +0000)]
Use password/group cache functions and avoid stashing a pointer to
the return value of getgrgid(3) or getgrnam(3) which relies on
undefined behavior. The rdist server will now use getgroups(2) to
determine group membership of the invoking user. In addition, there
is now one implementation of tilde expansion instead of two.
OK tb@ tim@
visa [Fri, 21 Sep 2018 14:31:29 +0000 (14:31 +0000)]
Stop displaying vfsconf reference counts so that the vfc_refcount field
can be removed from struct mount.
As a result of this diff, arrays vfsname[] and vfsvars[] are indexed
by filesystem typenum. This makes the vfs_typenums[] array redundant.
OK bluhm@ mpi@
solene [Fri, 21 Sep 2018 14:07:34 +0000 (14:07 +0000)]
Increase /usr/local max size to 20 GB in default template
ok krw@ deraadt@ jca@
solene [Fri, 21 Sep 2018 14:04:37 +0000 (14:04 +0000)]
Add explanations about vmctl send command
ok jmc@ jca@ mlarkin@
mdoc tip from bentley@
djm [Fri, 21 Sep 2018 12:46:22 +0000 (12:46 +0000)]
Allow ssh_config ForwardX11Timeout=0 to disable the timeout and allow
X11 connections in untrusted mode indefinitely. ok dtucker@
stsp [Fri, 21 Sep 2018 12:42:34 +0000 (12:42 +0000)]
Enable bwfm(4) in GENERIC on macppc.
Tested by Christian Hammerschmidt.
djm [Fri, 21 Sep 2018 12:23:17 +0000 (12:23 +0000)]
when compiled with GSSAPI support, cache supported method OIDs by
calling ssh_gssapi_prepare_supported_oids() regardless of whether
GSSAPI authentication is enabled in the main config.
This avoids sandbox violations for configurations that enable GSSAPI
auth later, e.g.
Match user djm
GSSAPIAuthentication yes
bz#2107; ok dtucker@
djm [Fri, 21 Sep 2018 12:20:12 +0000 (12:20 +0000)]
In sshkey_in_file(), ignore keys that are considered for being too
short (i.e. SSH_ERR_KEY_LENGTH). These keys will not be considered to
be "in the file". This allows key revocation lists to contain short
keys without the entire revocation list being considered invalid.
bz#2897; ok dtucker
claudio [Fri, 21 Sep 2018 08:17:15 +0000 (08:17 +0000)]
Both AS 23456 and AS 0 are reserved and can nor be used. Extend check for
AS 0 and adjust yyerror message to print the right number.
With input and OK denis@
sthen [Fri, 21 Sep 2018 08:17:04 +0000 (08:17 +0000)]
6.5 firmware key
claudio [Fri, 21 Sep 2018 08:15:33 +0000 (08:15 +0000)]
better yyerror messages. "syntax error" is generally not very helpful.
OK denis@
jmc [Fri, 21 Sep 2018 06:06:56 +0000 (06:06 +0000)]
update rtwn;
claudio [Fri, 21 Sep 2018 05:14:07 +0000 (05:14 +0000)]
Basic testing of roa-sets.
claudio [Fri, 21 Sep 2018 05:13:35 +0000 (05:13 +0000)]
Move setting of the PREFIXSET_FLAG_OPS higher up since prefixset_item rule
is now also used by roa-set. Also set the prefix operation for roa-set
items to OP_NONE since that what it actually needs to be.
claudio [Fri, 21 Sep 2018 05:06:30 +0000 (05:06 +0000)]
Add some more prefix-set test cases. Mainly to test edge cases in the RB
tree implementation now used.
claudio [Fri, 21 Sep 2018 04:55:27 +0000 (04:55 +0000)]
Implement code to parse, print and reload roa-set tables.
This is sharing a lot of code with prefixset which makes all a bit easier.
A roa-set is defined like this:
roa-set "test2" {
1.2.3.0/24 source-as 1,
1.2.8.0/22 maxlen 24 source-as 3
}
No support for acting on this data yet.
Put it in deraadt@, OK benno@, input and OK denis@
djm [Fri, 21 Sep 2018 03:11:36 +0000 (03:11 +0000)]
Treat connections with ProxyJump specified the same as ones with a
ProxyCommand set with regards to hostname canonicalisation (i.e.
don't try to canonicalise the hostname unless CanonicalizeHostname
is set to 'always').
Patch from Sven Wegener via bz#2896
jsg [Fri, 21 Sep 2018 02:21:53 +0000 (02:21 +0000)]
U-Boot 2018.05 and later will attempt to load a dtb for PocketBeagle if
the hardware is detected. Add this to the miniroot/ramdisk.
requires dtb 4.18
U-Boot 2018.09 and later will load a dtb for 'SanCloud BeagleBone
Enhanced' if required which will be in dtb 4.19 after linux 4.19 is
released and can be added then.
jmatthew [Fri, 21 Sep 2018 02:14:37 +0000 (02:14 +0000)]
mention RTL8188EE support
jmatthew [Fri, 21 Sep 2018 01:45:53 +0000 (01:45 +0000)]
Add support for RTL8188EE.
This needs a new firmware image, which should be added to the rtwn
firmware package shortly.
testing and lots of help from kevlo@
ok kevlo@ stsp@
jsg [Fri, 21 Sep 2018 01:33:55 +0000 (01:33 +0000)]
add missing braces implied by indentation
ok millert@ claudio@
djm [Thu, 20 Sep 2018 23:40:16 +0000 (23:40 +0000)]
actually make CASignatureAlgorithms available as a config option
sthen [Thu, 20 Sep 2018 23:15:39 +0000 (23:15 +0000)]
merge unbound 1.8.0
sthen [Thu, 20 Sep 2018 23:14:36 +0000 (23:14 +0000)]
import unbound 1.8.0, tested by myself and benno@
bluhm [Thu, 20 Sep 2018 18:59:10 +0000 (18:59 +0000)]
As a step towards per inpcb or socket locks, remove the net lock
for netstat -a. Introduce a global mutex that protects the tables
and hashes for the internet PCBs. To detect detached PCB, set its
inp_socket field to NULL. This has to be protected by a per PCB
mutex. The protocol pointer has to be protected by the mutex as
netstat uses it.
Always take the kernel lock in in_pcbnotifyall() and in6_pcbnotify()
before the table mutex to avoid lock ordering problems in the notify
functions.
OK visa@
tb [Thu, 20 Sep 2018 16:16:52 +0000 (16:16 +0000)]
missing space after comma
deraadt [Thu, 20 Sep 2018 15:19:36 +0000 (15:19 +0000)]
grow alpha and hppa media to accomodate some recent growth
brynet [Thu, 20 Sep 2018 14:32:59 +0000 (14:32 +0000)]
vmm(4): Clear the guest MWAITX/MONITORX extended CPUID feature bit,
like we already do for MWAIT/MONITOR. Also match Intel here by not
exposing the SVM capability to AMD guests.
Allows Linux guests to boot in vmd(8) on Ryzen CPUs.
ok mlarkin@
solene [Thu, 20 Sep 2018 12:24:14 +0000 (12:24 +0000)]
add missing explanation about daemon_variables removed when disabling a pkg
script
ok aja@ jca@
jsg [Thu, 20 Sep 2018 12:23:13 +0000 (12:23 +0000)]
add missing braces implied by indentation
ok millert@
jsg [Thu, 20 Sep 2018 11:49:55 +0000 (11:49 +0000)]
add missing braces implied by indentation
ok millert@ mpi@
claudio [Thu, 20 Sep 2018 11:47:50 +0000 (11:47 +0000)]
Adjust unittests to the adjustments done to the as_set code.
OK benno@
claudio [Thu, 20 Sep 2018 11:46:40 +0000 (11:46 +0000)]
as_set_match() changed again, so adjust it here too.
OK benno@
claudio [Thu, 20 Sep 2018 11:45:59 +0000 (11:45 +0000)]
Split up as_set into a set_table and an as_set. The first is what does
the lookup and will now also be used in roa-set tries. The as_set is glue
to add the name and dirty flag. Add an accessor to get the set data so
that the imsg sending and printing can be moved into the right places.
This is done mainly because roa-sets need similar but slightly different
versions and making the code more generic is the best way fixing this.
OK benno@
jsg [Thu, 20 Sep 2018 11:42:42 +0000 (11:42 +0000)]
fix indentation
ok krw@ millert@
eric [Thu, 20 Sep 2018 11:42:28 +0000 (11:42 +0000)]
properly handle credentials and fix auth in smtp(1)
ok gilles@
jsg [Thu, 20 Sep 2018 11:41:28 +0000 (11:41 +0000)]
fix indentation
ok krw@ millert@
benno [Thu, 20 Sep 2018 11:06:04 +0000 (11:06 +0000)]
whitespace cleanup, ok claudio@
eric [Thu, 20 Sep 2018 10:22:14 +0000 (10:22 +0000)]
fix indentation
dtucker [Thu, 20 Sep 2018 08:07:03 +0000 (08:07 +0000)]
Import updated moduli.
claudio [Thu, 20 Sep 2018 07:58:22 +0000 (07:58 +0000)]
Sort order changed because an RB tree is now used for prefixsets.
claudio [Thu, 20 Sep 2018 07:46:39 +0000 (07:46 +0000)]
Switch prefixset to an RB_TREE instead of a SIMPLEQ. This allows to trigger
on duplicates (which are only reported) but is needed as a preparation step
for roa-sets.
OK benno@ denis@
claudio [Thu, 20 Sep 2018 07:41:25 +0000 (07:41 +0000)]
Fix the empty aspath segments check. seg_size is never 0, this needs to use
seg_len instead. Since seg_len is known early move the check up.
Found while hunting for the other bug in aspath_verify.
claudio [Thu, 20 Sep 2018 07:37:06 +0000 (07:37 +0000)]
Fix an out of bound read that could crash the RDE because it touched
unallocated memory while looking for AS 0.
Found by and debugged with Aaron A. Glenn. Thanks a lot.
jmc [Thu, 20 Sep 2018 06:58:48 +0000 (06:58 +0000)]
reorder CASignatureAlgorithms, and add them to the various -o lists;
ok djm
djm [Thu, 20 Sep 2018 03:31:49 +0000 (03:31 +0000)]
fix "ssh -Q sig" to show correct signature algorithm list (it was
erroneously showing certificate algorithms); prompted by markus@
djm [Thu, 20 Sep 2018 03:30:44 +0000 (03:30 +0000)]
add CASignatureAlgorithms option for the client, allowing it to specify
which signature algorithms may be used by CAs when signing certificates.
Useful if you want to ban RSA/SHA1; ok markus@
djm [Thu, 20 Sep 2018 03:28:06 +0000 (03:28 +0000)]
Add sshd_config CASignatureAlgorithms option to allow control over
which signature algorithms a CA may use when signing certificates.
In particular, this allows a sshd to ban certificates signed with
RSA/SHA1.
ok markus@
jsg [Thu, 20 Sep 2018 01:19:56 +0000 (01:19 +0000)]
fix a memory leak in ihidev_hid_command()
ok claudio@
millert [Wed, 19 Sep 2018 18:55:33 +0000 (18:55 +0000)]
If getcwd() fails in dinit(), the stat buffer 'swd' is used
uninitialized by the else clause. Since it is used in both clauses
we should perform the stat before the if(). However, fixing this
causes 'cp' to be unitialized in some case so initialize cp to NULL
and move the "cp == NULL" check out of the first if() clause now
that it can be true in either case. OK miko@ deraadt@
millert [Wed, 19 Sep 2018 18:48:55 +0000 (18:48 +0000)]
Fix last commit, I made one of the changes to the wrong line.
Noticed by martijn@
bluhm [Wed, 19 Sep 2018 18:35:21 +0000 (18:35 +0000)]
Update disklabel(8) man page with the new 5G minimum for /usr/obj.
Remove Tn macro to make mandoc lint happy.
requested by jmc@
miko [Wed, 19 Sep 2018 16:21:00 +0000 (16:21 +0000)]
fix message to reflect "rmidi" is the expected string; ok ratchov@
millert [Wed, 19 Sep 2018 15:14:35 +0000 (15:14 +0000)]
Compare against NULL, not '\0' for pointers. Quiets a warning on
newer gcc.
miko [Wed, 19 Sep 2018 14:01:52 +0000 (14:01 +0000)]
sys/stat.h not needed here; ok ratchov@
mpi [Wed, 19 Sep 2018 13:17:21 +0000 (13:17 +0000)]
Always call bridge_iflist `bif'.
ok bluhm@, visa@
claudio [Wed, 19 Sep 2018 13:09:30 +0000 (13:09 +0000)]
Wrap sending imsg to the RDE in a function and make sure that the ibuf
to the RDE is valid. The SE is stopping all sessions on exit and so
session_stop() is called which will send an imsg to the RDE which is no
longer there. Instead of fixing just one call fix all. Now the SE should
no longer crash when the RDE crashes.
OK sthen@
krw [Wed, 19 Sep 2018 12:20:47 +0000 (12:20 +0000)]
Try to turn this into an actual regress test.
with & ok bluhm@
reyk [Wed, 19 Sep 2018 11:28:02 +0000 (11:28 +0000)]
Do not abort when the ca privenc runs into a timeout.
OK claudio@
jmatthew [Wed, 19 Sep 2018 10:26:17 +0000 (10:26 +0000)]
Report duplex state correctly for adapters with firmware interface versions
older than 1.08.
tested by and ok ccardenas@
claudio [Wed, 19 Sep 2018 08:12:39 +0000 (08:12 +0000)]
Free edid_buf after use so to fix possible memory leak.
With and ok jsg@
mestre [Wed, 19 Sep 2018 07:47:54 +0000 (07:47 +0000)]
In mue_iff() move the initialization of the hash table up so that is done
unconditionally, like it's done on other devices, so that we don't end up in
the situation of the card being put in promiscuous mode or if
ac->ac_multirangecnt > 0 and then we would write into the table while being
uninitialized a little bit later.
Noticed on Coverity ID
1473316.
hint from claudio@ and kettenis@
OK kevlo@ claudio@
eric [Wed, 19 Sep 2018 05:31:12 +0000 (05:31 +0000)]
rename the ill-named "flags" member to "as_host" in domain structure.
remove yet another useless relay flag while there.
ok gilles@
guenther [Wed, 19 Sep 2018 05:23:16 +0000 (05:23 +0000)]
Delete checks of the CST bit_width ('vendor') field, as new Ryzen machines
are showing up with vendor 2 and it doesn't seem to mean anything: Linux
doesn't check it at all, so this exhibits all the normal ACPI spec lossage
tested in snaps for a few weeks
shrug kettenis@
ccardenas [Wed, 19 Sep 2018 04:29:21 +0000 (04:29 +0000)]
Various clean up items for disks.
- qcow2: general cleanup
- vioraw: check malloc
- virtio: add function to sync disks
- vm: call virtio_shutdown to sync disks when vm is finished executing
Thanks to Ori Bernstein.
Ok miko@
djm [Wed, 19 Sep 2018 02:03:02 +0000 (02:03 +0000)]
Make "ssh-add -q" do what it says on the tin: silence output from
successful operations.
Based on patch from Thijs van Dijk; ok dtucker@ deraadt@
bluhm [Tue, 18 Sep 2018 20:47:11 +0000 (20:47 +0000)]
Updating time counters without memory barriers is wrong. Put
membar_producer() into tc_windup() and membar_consumer() into the
uptime functions. They order the visibility of the time and
generation number updates.
This is a combination of what NetBSD and FreeBSD do.
OK kettenis@
patrick [Tue, 18 Sep 2018 20:21:40 +0000 (20:21 +0000)]
Update to compiler-rt 6.0.0.
tested by naddy@
ok kettenis@
deraadt [Tue, 18 Sep 2018 19:40:15 +0000 (19:40 +0000)]
sync
naddy [Tue, 18 Sep 2018 19:15:40 +0000 (19:15 +0000)]
6.5 packages key
deraadt [Tue, 18 Sep 2018 18:48:16 +0000 (18:48 +0000)]
6.5 base key
anton [Tue, 18 Sep 2018 18:36:27 +0000 (18:36 +0000)]
whitespace fix; no binary change
anton [Tue, 18 Sep 2018 18:25:31 +0000 (18:25 +0000)]
Backport support for option -a to addr2line, which will display each address
given as input before the resolved source location. This change was introduced
to bintutils after the switch from GPLv2 but the author Tristan Gingold granted
me permission relicense the diff under GPLv2; thanks!
Taken as is from binutils commit
be6f64938f985dfb0eaa2107b99f193bb865ce04
This option is used by the syzkaller kernel fuzzer to produce a human readable
representation of the coverage collected by kcov.
ok deraadt@ jca@ visa@
millert [Tue, 18 Sep 2018 17:48:22 +0000 (17:48 +0000)]
Restore the xmalloc(), xcalloc(), xreallocarray() and xstrdup() changes.
OK deraadt@
naddy [Tue, 18 Sep 2018 17:43:40 +0000 (17:43 +0000)]
The dmesg buffer is sufficiently large now that using the accumulated
output from several boots in the dmesg listener can overflow the
ramdisk, so use only one boot's worth of dmesg in the listener.
sed(1) expression from kn@. ok deraadt@
sthen [Tue, 18 Sep 2018 16:54:01 +0000 (16:54 +0000)]
s/mulitprotocol/multiprotocol/ in error text
mlarkin [Tue, 18 Sep 2018 16:02:08 +0000 (16:02 +0000)]
vmm(4): Reset host LDTR on exit for SVM
For SVM machines, the LDT content remains set to that of the guest VM on
exit (as compared to Intel/VMX which resets the LDTR to 0). This fix
ensures the LDT is reset to 0 on SVM exits.
Leaving the LDT set to the guest's choice could allow a malicious process
to escalate its privileges with the help of a malicious VM that they
also are able to run on the machine.
This was reported by Maxime Villard; thanks!