tb [Sun, 16 Sep 2018 19:22:50 +0000 (19:22 +0000)]
add missing default case to switch
bluhm [Sun, 16 Sep 2018 18:58:36 +0000 (18:58 +0000)]
If a prefix without length was specifed in the config file, then
128 was used. This does not make much sense for rad(8). Change
the default prefixlen to 64.
OK florian@
tb [Sun, 16 Sep 2018 18:44:33 +0000 (18:44 +0000)]
EVP_aead_chacha20_poly1305() can't actually fail.
krw [Sun, 16 Sep 2018 17:37:11 +0000 (17:37 +0000)]
Unbreak tree with 'char *' -> 'const char *' for
userid().
kettenis [Sun, 16 Sep 2018 14:27:32 +0000 (14:27 +0000)]
Add support for setting NVRAM variables.
millert [Sun, 16 Sep 2018 12:43:40 +0000 (12:43 +0000)]
Add missing includes for setpassent() and setgroupent().
kettenis [Sun, 16 Sep 2018 12:26:06 +0000 (12:26 +0000)]
Adjust sparc64 bits after libunwind 6.0.0 update.
ok visa@
kettenis [Sun, 16 Sep 2018 12:17:05 +0000 (12:17 +0000)]
Various improvements to generate logical domain configurations that are
accepted by more modern firmwares. In particular on SPARC T3 machines.
Tested on a t1k and t5120.
tb [Sun, 16 Sep 2018 11:45:08 +0000 (11:45 +0000)]
Rename *AesCcmOrGcm* into the slightly less ugly *AesAead*.
visa [Sun, 16 Sep 2018 11:41:44 +0000 (11:41 +0000)]
Move vfsconf lookup code into dedicated functions.
OK bluhm@
bluhm [Sun, 16 Sep 2018 08:53:02 +0000 (08:53 +0000)]
With prefixlen 128, mask_prefix() in rad(8) caused a stack overflow
in the config parser. Add an out of bounds check for the in6_addr.
OK florian@
millert [Sun, 16 Sep 2018 02:44:06 +0000 (02:44 +0000)]
Use uid_from_user(3) and gid_from_group(3) in utilities that
do repeated lookups. OK tb@
millert [Sun, 16 Sep 2018 02:43:11 +0000 (02:43 +0000)]
Use user_from_uid(3) and group_from_gid(3) in a few more places
that do repeated lookups. OK tb@
millert [Sun, 16 Sep 2018 02:42:19 +0000 (02:42 +0000)]
Use user_from_uid(3) and group_from_gid(3) to avoid extra passwd
and group file lookups. This required a bit of reordering of the
file mode handling bits to deal with the const char *. OK tb@
millert [Sun, 16 Sep 2018 02:41:16 +0000 (02:41 +0000)]
Use user_from_uid(3), group_from_gid(3), uid_from_user(3) and
gid_from_group(3) to avoid repeatedly looking up the same user/group.
Also keep the passwd and group files open to avoid opening and
closing them all the time. OK tb@
millert [Sun, 16 Sep 2018 02:38:57 +0000 (02:38 +0000)]
Replace the local getname() and getuserid() functions with calls
to user_from_uid(3) and uid_from_user(3). This requires sprinkling
const in a few places to match the return value of user_from_uid(3).
OK tb@
tb [Sat, 15 Sep 2018 22:09:08 +0000 (22:09 +0000)]
another typo. time to sleep
tb [Sat, 15 Sep 2018 22:07:52 +0000 (22:07 +0000)]
typos
tb [Sat, 15 Sep 2018 22:03:28 +0000 (22:03 +0000)]
add a brief comment on the acceptable AES CCM and AES GCM cases
tb [Sat, 15 Sep 2018 19:12:31 +0000 (19:12 +0000)]
Also exercise EVP_aead_aes_128_gcm() and EVP_aead_aes_256_gcm().
tb [Sat, 15 Sep 2018 19:09:07 +0000 (19:09 +0000)]
rename checkChaCha20Poly1305{Open,Seal}() into checkAead{Open,Seal}().
tb [Sat, 15 Sep 2018 19:06:47 +0000 (19:06 +0000)]
Merge AES CCM, AES GCM and ChaCha20 data structures into
single AEAD types. Will be used in upcoming commits.
tb [Sat, 15 Sep 2018 16:05:17 +0000 (16:05 +0000)]
Add missing conversion specifier
inoguchi [Sat, 15 Sep 2018 13:26:13 +0000 (13:26 +0000)]
Modify regress appstest.sh for interoperability testing with OpenSSL 1.1.x
- add sleep before s_client instead of removing -pause option
- change check words for s_client output messages
- replace CAfile to CApath for s_time
- remove -prexit from s_client
- confirm openssl command path is executable
kettenis [Sat, 15 Sep 2018 13:20:16 +0000 (13:20 +0000)]
Add support for version 2.0 of the mdstore protocol.
miko [Sat, 15 Sep 2018 12:15:32 +0000 (12:15 +0000)]
strsave() is hard-fail strdup() so simplify and rename to xstrdup().
with help from martijn@.
ok millert@ martijn@
bentley [Sat, 15 Sep 2018 09:44:19 +0000 (09:44 +0000)]
Fix fortune underlines and use the right number of underscores in Notes.
ok sthen@
mestre [Fri, 14 Sep 2018 23:40:10 +0000 (23:40 +0000)]
Initialize the TDB to NULL in ipsec_common_input() and
ipsec_{input,output}_cb() so that in the case of sending or receiving a bogus
mbuf (NULL) we don't end up trying to dereference the TDB, while being an
uninitialized pointer, to increase the drops.
Coverity IDs
1473312,
1473313 and
1473317.
OK mpi@ visa@
bluhm [Fri, 14 Sep 2018 18:17:46 +0000 (18:17 +0000)]
Do not leak a file descriptor when opening nohup.out. Make sure
that stdout and stderr are not closed.
from Nan Xiao
claudio [Fri, 14 Sep 2018 13:58:20 +0000 (13:58 +0000)]
Unify and bump some of the NMBCLUSTERS defines. Some archs had it set to
4MB which is far too low especially when the platform is able to run MP.
New limits are, amd64 = 256M; arm64, mips64, sparc64 = 64M; alpha, arm,
hppa, i386, powerpc = 32M; m88k, sh = 8M
Still rather conservative numbers but much better than before. At least
some hangs of arm64 build boxes was caused by this.
OK kettenis@, visa@
inoguchi [Fri, 14 Sep 2018 13:54:57 +0000 (13:54 +0000)]
Add interoperability test mode for regress appstest.sh
- test s_server and s_client between different version by option -i
- indicate other version by defining OTHER_OPENSSL environment variable
- fix "SSL/TLS" to "TLS/SSL", since TLS is correct as technical term
- s/SKIPPNG/SKIPPING/
naddy [Fri, 14 Sep 2018 13:49:01 +0000 (13:49 +0000)]
add gapdummy.c to the "clean" target like other generated files
ok visa@ jsg@ phessler@
naddy [Fri, 14 Sep 2018 13:44:18 +0000 (13:44 +0000)]
Pass -L/usr/lib to the linker in preparation for switching to lld, which
does not have a default search path. ok kettenis@ jsg@
naddy [Fri, 14 Sep 2018 13:37:52 +0000 (13:37 +0000)]
Pass CC/CFLAGS/LDFLAGS to the configure script. ok millert@
bluhm [Fri, 14 Sep 2018 12:55:17 +0000 (12:55 +0000)]
In general it is a bad idea to use one random secret for two things.
The inet PCB uses one hash with local and foreign addresses, and
one with local port numbers. Give both hashes separate keys. Also
document the struct fields.
OK visa@
florian [Fri, 14 Sep 2018 10:28:55 +0000 (10:28 +0000)]
No longer needed since the ping/ping6 unification.
Pointed out by Clemens Goessnitzer (clemens AT goessnitzer.info), thanks!
claudio [Fri, 14 Sep 2018 10:23:42 +0000 (10:23 +0000)]
Adjust and extend unit test for as_set after last bgpd commit
OK benno@
claudio [Fri, 14 Sep 2018 10:22:55 +0000 (10:22 +0000)]
Adjust the dummy as_set_match function to new prototype after last bgpd commit
OK benno@
claudio [Fri, 14 Sep 2018 10:22:11 +0000 (10:22 +0000)]
Extend as_set to allow for different sized objects to be added. The only
requirement is that the first value of the struct is a 32bit ID which is
used in the bsearch. This allows to add more than just as numbers to a
set. as_set_match now returns a pointer to this data or NULL if not found.
OK benno@
miko [Fri, 14 Sep 2018 08:45:46 +0000 (08:45 +0000)]
ansify auich_trigger_input() and remove #ifdef around DPRINTF(); ok ratchov@
miko [Fri, 14 Sep 2018 08:37:34 +0000 (08:37 +0000)]
mark some suspend/resume functions always returning zero as void; ok ratchov@
jsg [Fri, 14 Sep 2018 07:25:02 +0000 (07:25 +0000)]
unbreak userland uses of in_pcb.h by including sys/refcnt.h
ok visa@
djm [Fri, 14 Sep 2018 05:26:27 +0000 (05:26 +0000)]
second try, deals properly with missing and private-only keys:
Use consistent format in debug log for keys readied, offered and
received during public key authentication.
This makes it a little easier to see what is going on, as each message
now contains (where available) the key filename, its type and fingerprint,
and whether the key is hosted in an agent or a token.
djm [Fri, 14 Sep 2018 04:44:04 +0000 (04:44 +0000)]
revert following; deals badly with agent keys
revision 1.285
date: 2018/09/14 04:17:12; author: djm; state: Exp; lines: +47 -26; commitid: lflGFcNb2X2HebaK;
Use consistent format in debug log for keys readied, offered and
received during public key authentication.
This makes it a little easier to see what is going on, as each message
now contains the key filename, its type and fingerprint, and whether
the key is hosted in an agent or a token.
djm [Fri, 14 Sep 2018 04:17:44 +0000 (04:17 +0000)]
garbage-collect moribund ssh_new_private() API.
djm [Fri, 14 Sep 2018 04:17:12 +0000 (04:17 +0000)]
Use consistent format in debug log for keys readied, offered and
received during public key authentication.
This makes it a little easier to see what is going on, as each message
now contains the key filename, its type and fingerprint, and whether
the key is hosted in an agent or a token.
bluhm [Thu, 13 Sep 2018 19:53:58 +0000 (19:53 +0000)]
Add reference counting for inet pcb, this will be needed when we
start locking the socket. An inp can be referenced by the PCB queue
and hashes, by a pf mbuf header, or by a pf state key.
OK visa@
jmc [Thu, 13 Sep 2018 16:50:54 +0000 (16:50 +0000)]
missing comma in previous;
sthen [Thu, 13 Sep 2018 16:34:33 +0000 (16:34 +0000)]
Unbreak ramdisks by catching up with the change to use the new libc
uid_from_user() and gid_from_group() instead of the pax-specific
functions in cache.c.
sthen [Thu, 13 Sep 2018 16:19:26 +0000 (16:19 +0000)]
oops, missed the libc.so.XX.a relink file
millert [Thu, 13 Sep 2018 15:23:32 +0000 (15:23 +0000)]
Fix warnings caused by user_from_uid() and group_from_gid() now
returning const char *.
sthen [Thu, 13 Sep 2018 13:59:49 +0000 (13:59 +0000)]
sync
millert [Thu, 13 Sep 2018 12:33:43 +0000 (12:33 +0000)]
Use the new libc uid_from_user() and gid_from_group() instead of
the pax-specific functions in cache.c. OK guenther@
millert [Thu, 13 Sep 2018 12:31:15 +0000 (12:31 +0000)]
Add uid_from_user() and gid_from_group(), derived from pax's cache.c.
It replaces the existing pwcache.c functions user_from_uid(3) and
group_from_gid(3) with the pax equivalents. Adapted from NetBSD
(mycroft) changes from our own pax's cache.c. OK guenther@
mpi [Thu, 13 Sep 2018 12:29:43 +0000 (12:29 +0000)]
Include the size of IPCOMP header when checking for compression.
Problem found and anaylyzed by Romain Gabet, ok markus@
claudio [Thu, 13 Sep 2018 11:25:41 +0000 (11:25 +0000)]
Noticed that the equal case is a bit wrong while adjusting the regress tests.
This is better. There is no need to check for the prefix length p->len.
claudio [Thu, 13 Sep 2018 11:24:30 +0000 (11:24 +0000)]
Adjust after introduction of maxlen
claudio [Thu, 13 Sep 2018 11:18:18 +0000 (11:18 +0000)]
Similar to as-set factor out the code to create a prefix-set into a function.
Makes all a bit nicer and as an added bonus fixes a memory leak.
OK phessler@
claudio [Thu, 13 Sep 2018 11:16:21 +0000 (11:16 +0000)]
ROA entires are allowing to define a prefix with a maxlen.
In the end this is just another way to specify a prefixlen range
and kind of an or-longer case with an upper limit.
So these two prefix statements are equivalent:
prefix 10.0.0.0/8 prefixlen 8 - 24
prefix 10.0.0.0/8 maxlen 24
While there also make 'prefixlen = 17' a OP_RANGE and because of that also
usable in prefix-set tables. Finally adjust printconf.c for those to
changes to print them nicely.
OK phessler@
kettenis [Thu, 13 Sep 2018 09:32:27 +0000 (09:32 +0000)]
In drm_wait_one_vblank() add a delay when we're "cold". Interrupts aren't
enabled at that point, so we cannot wait for one to happen. But having no
delay at all breaks detection of some output connectors.
Thanks to Philippe Meunier for tracking down the issue.
ok millert@, jsg@
kevlo [Thu, 13 Sep 2018 09:28:07 +0000 (09:28 +0000)]
- There's no need to set R92C_HSSI_PARAM2_READ_EDGE for R92C_HSSI_PARAM2(0)
for rtl8188eu
- Fix typo in structure r92c_rom in comment: s/0x8192/0x8129/
- Add id member to struct r88e_rom which identifies eeprom
- Replace magic numbers with something more readable
- Cosmetic tweaking
ok stsp@
djm [Thu, 13 Sep 2018 09:03:20 +0000 (09:03 +0000)]
missed a bit of openssl-1.0.x API in this unittest
mestre [Thu, 13 Sep 2018 07:49:33 +0000 (07:49 +0000)]
When unveil(2) was introduced one break from SYS_access case was removed
here, this adds it back. Noticed by Coverity
1471854.
feedback from semarie@ OK deraadt@
kn [Thu, 13 Sep 2018 06:03:27 +0000 (06:03 +0000)]
Avoid unneeded variable in gen_dynnode()
OK bluhm
djm [Thu, 13 Sep 2018 05:06:51 +0000 (05:06 +0000)]
use only openssl-1.1.x API here too
pd [Thu, 13 Sep 2018 04:23:36 +0000 (04:23 +0000)]
vmd: set irq and vm_id in virtio dev structs on restore
This unbreaks vmctl receive.
ok ccardenas@
miko [Thu, 13 Sep 2018 04:07:20 +0000 (04:07 +0000)]
clarify that config_activate_children() is called unconditionally in *activate().
ok ratchov@
jsg [Thu, 13 Sep 2018 03:55:17 +0000 (03:55 +0000)]
drm/drivers: add support for using the arch wc mapping API.
From Dave Airlie
c59fdc4cfbda52ce081c59540762185d765c3369 in linux 4.4.y/4.4.155
7cf321d118a825c1541b43ca45294126fd474efa in mainline linux
ccardenas [Thu, 13 Sep 2018 03:53:33 +0000 (03:53 +0000)]
Add initial set of unveil's to vmctl.
Was in snaps for a while.
Ok mlarkin@ and reyk@
jsg [Thu, 13 Sep 2018 03:45:08 +0000 (03:45 +0000)]
drm/i915/userptr: reject zero user_size
From Matthew Auld
182e963432d867384f2e55487ec60ca7a9f99cd1 in linux 4.4.y/4.4.155
c11c7bfd213495784b22ef82a69b6489f8d0092f in mainline linux
jsg [Thu, 13 Sep 2018 03:38:15 +0000 (03:38 +0000)]
avoid sequence-point warning with gcc 4.9
ok kettenis@
djm [Thu, 13 Sep 2018 02:08:33 +0000 (02:08 +0000)]
hold our collective noses and use the openssl-1.1.x API in OpenSSH;
feedback and ok tb@ jsing@ markus@
sthen [Wed, 12 Sep 2018 22:17:08 +0000 (22:17 +0000)]
sync with mozilla-release (one removal, TURKTRUST, more details at
https://bugzilla.mozilla.org/show_bug.cgi?id=
1439127)
ok danj guenther millert
jmc [Wed, 12 Sep 2018 15:09:22 +0000 (15:09 +0000)]
tweak previous;
kettenis [Wed, 12 Sep 2018 11:59:40 +0000 (11:59 +0000)]
Make this work on arm64.
kettenis [Wed, 12 Sep 2018 11:58:28 +0000 (11:58 +0000)]
Make pmap_protect(9) actually remove exec permission if the new permissions
include PROT_READ but not PROT_EXEC.
ok patrick@
mpi [Wed, 12 Sep 2018 11:24:38 +0000 (11:24 +0000)]
Stop exporting TDB counters to userland, this change introduced a
regression with iked(8).
Reported by Mark Patruck.
krw [Wed, 12 Sep 2018 09:20:34 +0000 (09:20 +0000)]
Fix obvious cut&pasto in comment (ifa_msghdr -> if_announcemsghdr).
ok claudio@
guenther [Wed, 12 Sep 2018 07:00:51 +0000 (07:00 +0000)]
Whitespace fixes
djm [Wed, 12 Sep 2018 06:37:23 +0000 (06:37 +0000)]
sync for libcrypto/libssl/libtls minor bumps; from tb@
djm [Wed, 12 Sep 2018 06:36:15 +0000 (06:36 +0000)]
crank to follow minor crank in libcrypto; ok tb@ jsing@
djm [Wed, 12 Sep 2018 06:35:38 +0000 (06:35 +0000)]
Add some accessor functions:
RSA_meth_get_finish() RSA_meth_set1_name() EVP_CIPHER_CTX_(get|set)_iv()
feedback and ok jsing@ tb@
djm [Wed, 12 Sep 2018 06:18:59 +0000 (06:18 +0000)]
fix edit mistake; spotted by jmc@
guenther [Wed, 12 Sep 2018 06:12:59 +0000 (06:12 +0000)]
Now that the pmap is more paranoid about some shootdowns (pmap.c rev 1.119),
avoid some TLB flushes by not reloading %cr3 when the value isn't changing.
original diff by and ok mlarkin@
guenther [Wed, 12 Sep 2018 06:09:39 +0000 (06:09 +0000)]
When shooting pages in the KVA range, all pmaps have the page mapped,
not just pmap_kernel() and this CPUs pmap. Meanwhile, when mapping
another pmap's tables, order the locking so that we don't need IPIs
specific to the temp pmap.
tested in snaps for a bit
ok mlarkin@
jsg [Wed, 12 Sep 2018 04:34:59 +0000 (04:34 +0000)]
Move -Wno-address-of-packed-member to the clang block as well and sync
armv7 and i386 with amd64.
ok guenther@
djm [Wed, 12 Sep 2018 01:36:45 +0000 (01:36 +0000)]
Include certs with multiple RSA signature variants in test data
Ensure that cert->signature_key is populated correctly
djm [Wed, 12 Sep 2018 01:34:02 +0000 (01:34 +0000)]
add SSH_ALLOWED_CA_SIGALGS - the default list of signature algorithms
that are allowed for CA signatures. Notably excludes ssh-dsa.
ok markus@
djm [Wed, 12 Sep 2018 01:32:54 +0000 (01:32 +0000)]
add sshkey_check_cert_sigtype() that checks a cert->signature_type
against a supplied whitelist; ok markus
djm [Wed, 12 Sep 2018 01:31:30 +0000 (01:31 +0000)]
add cert->signature_type field and keep it in sync with certificate
signature wrt loading and certification operations; ok markus@
djm [Wed, 12 Sep 2018 01:30:10 +0000 (01:30 +0000)]
Add "ssh -Q sig" to allow listing supported signature algorithms
ok markus@
djm [Wed, 12 Sep 2018 01:23:48 +0000 (01:23 +0000)]
test revocation by explicit hash and by fingerprint
djm [Wed, 12 Sep 2018 01:22:43 +0000 (01:22 +0000)]
s/sshkey_demote/sshkey_from_private/g
djm [Wed, 12 Sep 2018 01:21:34 +0000 (01:21 +0000)]
allow key revocation by SHA256 hash and allow ssh-keygen to create KRLs
using SHA256/base64 key fingerprints; ok markus@
djm [Wed, 12 Sep 2018 01:19:12 +0000 (01:19 +0000)]
log certificate fingerprint in authentication success/failure message
(previously we logged only key ID and CA key fingerprint).
ok markus@
bluhm [Tue, 11 Sep 2018 21:04:03 +0000 (21:04 +0000)]
Convert inetctlerrmap to u_char like inet6ctlerrmap. That is also
what FreeBSD does. Remove old #if 0 version of inet6ctlerrmap.
OK mpi@
kettenis [Tue, 11 Sep 2018 20:25:58 +0000 (20:25 +0000)]
Only look for acpi tables if acpi(4) attached. Fixes radeondrm(4) crash
on arm64 when using a device tree.
sthen [Tue, 11 Sep 2018 19:43:15 +0000 (19:43 +0000)]
sync (libc++/libc++abi update)
phessler [Tue, 11 Sep 2018 19:25:54 +0000 (19:25 +0000)]
ues the format string for signed ints, for signed ints
reminded by stsp@
robert [Tue, 11 Sep 2018 18:36:58 +0000 (18:36 +0000)]
define _LIBCXXABI_BUILDING_LIBRARY to avoid the same issues as in libcxx