tb [Wed, 5 Jan 2022 17:27:40 +0000 (17:27 +0000)]
Move variable declarations in X509v3_addr_canonize() to the top of
the function and unindent some code.
ok inoguchi jsing
jsing [Wed, 5 Jan 2022 17:10:59 +0000 (17:10 +0000)]
Revise for tls13_key_share rename.
jsing [Wed, 5 Jan 2022 17:10:02 +0000 (17:10 +0000)]
Rename tls13_key_share to tls_key_share.
In preparation to use the key share code in both the TLSv1.3 and legacy
stacks, rename tls13_key_share to tls_key_share, moving it into the shared
handshake struct. Further changes will then allow the legacy stack to make
use of the same code for ephemeral key exchange.
ok inoguchi@ tb@
stsp [Wed, 5 Jan 2022 17:06:20 +0000 (17:06 +0000)]
Remove unused function arguments in iwm/iwx interrupt handlers.
pointed out by + ok millert@
tb [Wed, 5 Jan 2022 17:01:06 +0000 (17:01 +0000)]
Switch snmpd(8) to using EVP_Digest{Init,Final}_ex() and drop a no
longer needed EVP_MD_CTX_reset().
ok martijn
deraadt [Wed, 5 Jan 2022 16:46:55 +0000 (16:46 +0000)]
sync
deraadt [Wed, 5 Jan 2022 16:46:11 +0000 (16:46 +0000)]
Compensate for i386 pcitag_t union
ok jsg kettenis
tb [Wed, 5 Jan 2022 16:41:42 +0000 (16:41 +0000)]
Add error checking for EVP_Digest* to snmpd(8).
ok martijn
tb [Wed, 5 Jan 2022 16:41:07 +0000 (16:41 +0000)]
Add error checking for EVP_Digest*() to snmp(1).
ok martijn
deraadt [Wed, 5 Jan 2022 16:35:33 +0000 (16:35 +0000)]
Use new shell-based fw_update(8)
with afresh1
stsp [Wed, 5 Jan 2022 16:33:42 +0000 (16:33 +0000)]
In iwx(4), fix wrong pointer assignment in iwx_bar_frame_release().
This bug caused the driver to read block ack request information sent
by firmware from the wrong offset. The driver flushes buffered frames
and moves its Rx block ack window based on this information. Possible
consequences of this bug are packet loss or even stalled traffic if
the Rx BA window gets out of sync between driver and firmware. Though
this effect might get cancelled out when the driver re-syncs the BA
window in its regular Rx code path.
Spotted by Christian Ehrhardt.
afresh1 [Wed, 5 Jan 2022 16:32:46 +0000 (16:32 +0000)]
Commit the correct version of fw_update
Sigh.
afresh1 [Wed, 5 Jan 2022 16:28:19 +0000 (16:28 +0000)]
Add the shell based fw_update and updated man page
This allows installing firmware from the installer without having
to wait to boot into a live system.
commit deraadt@
deraadt [Wed, 5 Jan 2022 16:24:29 +0000 (16:24 +0000)]
We are moving back to a shell-script based fw_update, written in such a
way that the install script can also run it. This allows earlier retrieval
of downloaded firmwares, based upon patterns found in dmesg.
many iterations of this in snaps for about a month.
espie [Wed, 5 Jan 2022 14:50:03 +0000 (14:50 +0000)]
formatting nit, noticed by jmc@, thx!
inoguchi [Wed, 5 Jan 2022 13:41:12 +0000 (13:41 +0000)]
Wrap long lines and add some braces
martijn [Wed, 5 Jan 2022 13:27:04 +0000 (13:27 +0000)]
Use LC_CTYPE instead of LC_ALL.
Makes regress pass when LC_CTYPE is set.
Found by and OK tb@
inoguchi [Wed, 5 Jan 2022 12:51:49 +0000 (12:51 +0000)]
Check function return value
inoguchi [Wed, 5 Jan 2022 11:38:19 +0000 (11:38 +0000)]
Checking pointer variable with NULL
claudio [Wed, 5 Jan 2022 11:07:35 +0000 (11:07 +0000)]
Switch proc_parser_root_cert() to not pass the entity but instead the
file, pkey and tal id. This is the last proc_parser function that needed
to be converted.
OK job@
tb [Wed, 5 Jan 2022 11:01:59 +0000 (11:01 +0000)]
dhcpd: straightforward conversion to HMAC_CTX on the heap, similar
to what was done in spamd a while back.
ok florian
tb [Wed, 5 Jan 2022 11:00:49 +0000 (11:00 +0000)]
snmpd: Straightforward conversion to EVP_* on the heap.
It would be nice if someone added error checking for the EVP_Digest*
calls.
tested by & ok martijn
tb [Wed, 5 Jan 2022 10:59:21 +0000 (10:59 +0000)]
Straightforward conversion to EVP_* on the heap.
ok martijn
inoguchi [Wed, 5 Jan 2022 10:33:36 +0000 (10:33 +0000)]
Use calloc instead of malloc
suggested by tb@
inoguchi [Wed, 5 Jan 2022 10:29:08 +0000 (10:29 +0000)]
Check NULL first and unindent the rest of the code
suggested by tb@
inoguchi [Wed, 5 Jan 2022 10:01:39 +0000 (10:01 +0000)]
Convert openssl(1) cms option handling
Just applying new option handling and no functional changes.
Referred to verify.c and using 'verify_shared_options'.
ok and comments from jsing@ and tb@
jsing [Wed, 5 Jan 2022 09:59:39 +0000 (09:59 +0000)]
Provide regress for SSL public APIs.
This will largely test curly and inconsistent APIs that are not covered by
other regress tests. Currently, this tests the wonder that is
SSL_get_peer_cert_chain().
espie [Wed, 5 Jan 2022 09:19:15 +0000 (09:19 +0000)]
document -m
djm [Wed, 5 Jan 2022 08:25:05 +0000 (08:25 +0000)]
select all RSA hostkey algorithms for UpdateHostkeys tests, not just
RSA-SHA1
tb [Wed, 5 Jan 2022 07:50:40 +0000 (07:50 +0000)]
Remove bandaid to work around expected range_should_be_prefix() problem.
tb [Wed, 5 Jan 2022 07:47:15 +0000 (07:47 +0000)]
Remove a bogus memcmp in range_should_be_prefix()
range_should_be_prefix() currently always fails. The reason for this
is that OpenSSL commit
42d7d7dd incorrectly moved a memcmp() out of
an assertion. As a consequence, the library emits and accepts
incorrectly encoded ipAddrBlock extensions since it will never detect
ranges that MUST be encoded as a prefix according to RFC 3779, 2.2.3.7.
The return -1 from this memcmp() indicates to the callers that the
range should be expressed as a range, so callers must check beforehand
that min <= max to be able to fail. Thus, remove this memcmp() and
add a check to make_addressRange(), the only caller that didn't already
ensure that min <= max.
This fixes the noisy output in regress/lib/libcrypto/x509/rfc3779.
ok inoguchi jsing
tb [Wed, 5 Jan 2022 07:37:01 +0000 (07:37 +0000)]
Polish X509v3_addr_subset() a bit
Use child and parent instead of a and b. Split unrelated checks. Use
accessors and assign to local variables to avoid ugly line wrapping.
Declare vriables up front instead of mixing declarations with
assignments from function returns.
ok inoguchi jsing
tb [Wed, 5 Jan 2022 07:29:47 +0000 (07:29 +0000)]
Readability tweaks in addr_contains()
Assign to local variables to avoid ugly line wrapping.
ok inoguchi jsing
tb [Wed, 5 Jan 2022 07:28:41 +0000 (07:28 +0000)]
Fix a bug in addr_contains() introduced in OpenSSL commit
be71c372
by returning 0 instead of -1 on extract_min_max() failure. Callers
would interpret -1 as success of addr_contains().
ok inoguchi jsing
dlg [Wed, 5 Jan 2022 05:53:03 +0000 (05:53 +0000)]
add a basic printer for EAPOL packets.
EAPOL turns out to be a little container for a bunch of other types
of packets, including EAP for use with vanilla 802.1X, a kind of
capability announcement thing, and MACsec Key Agreement.
it's not worth adding a separately file for such a small chunk of
functionality, and it seems specific to ethernet. parsers/printers
for the sub protocols can come later if needed.
ok deraadt@ visa@
dlg [Wed, 5 Jan 2022 05:47:53 +0000 (05:47 +0000)]
ethertypes.h has now added the missing ethertypes we're looking at here.
dlg [Wed, 5 Jan 2022 05:46:18 +0000 (05:46 +0000)]
fix some more -Wunused-but-set-variable
dlg [Wed, 5 Jan 2022 05:41:25 +0000 (05:41 +0000)]
fix another -Wunused-but-set-variable
dlg [Wed, 5 Jan 2022 05:37:37 +0000 (05:37 +0000)]
more -Wunused-but-set-variable fixes
dlg [Wed, 5 Jan 2022 05:36:37 +0000 (05:36 +0000)]
clean up another -Wunused-but-set-variable thing
dlg [Wed, 5 Jan 2022 05:35:19 +0000 (05:35 +0000)]
clean up some -Wunused-but-set-variable.
this file had it's own verison of TCHECK, but also not. not the best.
dlg [Wed, 5 Jan 2022 05:33:14 +0000 (05:33 +0000)]
fix some -Wunused-but-set-variable stuff.
dlg [Wed, 5 Jan 2022 05:29:54 +0000 (05:29 +0000)]
silence a -Wunused-but-set-variable
dlg [Wed, 5 Jan 2022 05:19:22 +0000 (05:19 +0000)]
add NSH and NHRP ethertypes, mostly for tcpdump stuff.
ok deraadt@
dlg [Wed, 5 Jan 2022 05:18:24 +0000 (05:18 +0000)]
rename ETHERTYPE_PAE to ETHERTYPE_EAPOL.
everyone else seems to use ETHERTYPE_EAPOL, and as a bonus it also
appears to be more correct.
ok deraadt@ stsp@
djm [Wed, 5 Jan 2022 04:56:15 +0000 (04:56 +0000)]
regress test both sshsig message hash algorithms, possible now because
the algorithm is controllable via the CLI
djm [Wed, 5 Jan 2022 04:50:11 +0000 (04:50 +0000)]
allow selection of hash at sshsig signing time; code already supported
either sha512 (default) or sha256, but plumbing wasn't there
mostly by Linus Nordberg
djm [Wed, 5 Jan 2022 04:27:54 +0000 (04:27 +0000)]
add missing -O option to usage() for ssh-keygen -Y sign;
from Linus Nordberg
djm [Wed, 5 Jan 2022 04:27:01 +0000 (04:27 +0000)]
move sig_process_opts() to before sig_sign(); no functional code change
djm [Wed, 5 Jan 2022 04:10:39 +0000 (04:10 +0000)]
regression test for find-principals NULL deref; from Fabian Stelzer
guenther [Wed, 5 Jan 2022 04:10:36 +0000 (04:10 +0000)]
Delete 'emul' keyword: it's been just returned 'native' for a long time
ok jsg@ deraadt@
djm [Wed, 5 Jan 2022 04:02:42 +0000 (04:02 +0000)]
NULL deref when using find-principals when matching an allowed_signers
line that contains a namespace restriction, but no restriction
specified on the command-line; report and fix from Fabian Stelzer
dlg [Wed, 5 Jan 2022 03:53:26 +0000 (03:53 +0000)]
use the index provided by the txcompletion descriptor in txeof.
this replaces the use of a register in txeof to figure out how much
of the ring has been completed by the hardware. that register isn't
reset when an interface is taken down and brought up again, which
messes with the calculations of free slots on the ring, which in
turn messes up the ability to transmit packets.
this means the watchdog can't know where the hardware is up to
anymore, so just restartthe chip if the watchdog fires.
it seems to be important to start the tx ring in msk_init on the
0th ring entry that we use to reset the high address value to 0.
we still fill the 0th descriptor, but we let the first msk_start
call post it for us when a packet goes on the ring.
a slight tweak and ok jmatthew@
visa [Wed, 5 Jan 2022 03:32:43 +0000 (03:32 +0000)]
Add mpfclock(4), a driver for the PolarFire SoC MSS clock controller.
OK kettenis@
jsg [Wed, 5 Jan 2022 02:00:55 +0000 (02:00 +0000)]
remove unused defines
ok gnezdo@
kettenis [Tue, 4 Jan 2022 20:55:48 +0000 (20:55 +0000)]
Future-proof by adding some proposed compatible strings to match on.
tb [Tue, 4 Jan 2022 20:52:34 +0000 (20:52 +0000)]
Readability tweaks in the print helper i2r_IPAddressOrRanges.
Assign repeated nested expressions to local variables and avoid some
awkward line wrapping.
deraadt [Tue, 4 Jan 2022 20:43:44 +0000 (20:43 +0000)]
struct pci_matchid is useful an upcoming userland program that wants
to includes pcivar.h because it inspects vid/pid tables
deraadt [Tue, 4 Jan 2022 20:41:42 +0000 (20:41 +0000)]
hide more things behind _KERNEL, in case userland manages to include
this file
tb [Tue, 4 Jan 2022 20:40:43 +0000 (20:40 +0000)]
Consistently name variables with a _len suffix instead of mixing
things like prefixlen, afi_length, etc.
suggested by jsing
tb [Tue, 4 Jan 2022 20:33:02 +0000 (20:33 +0000)]
Only check the parent to be canonical once we know it is non-NULL.
suggested by jsing during review
tb [Tue, 4 Jan 2022 20:30:30 +0000 (20:30 +0000)]
Refactor extract_min_max()
extract_min_max() crammed all the work in two return statements
inside a switch. Make this more readable by splitting out the
extraction of the min and max as BIT STRINGs from an addressPrefix
or an addressRange and once that's done expanding them to raw
addresses.
ok inoguchi jsing
tb [Tue, 4 Jan 2022 20:23:05 +0000 (20:23 +0000)]
Remove checks that are duplicated in extract_min_max()
The NULL checks and the checks that aor->type is reasonable are already
performed in extract_min_max(), so it is unnecessary to repeat them
in X509v3_addr_get_range()
ok inoguchi jsing
deraadt [Tue, 4 Jan 2022 20:21:14 +0000 (20:21 +0000)]
sync
tb [Tue, 4 Jan 2022 20:21:04 +0000 (20:21 +0000)]
Make X509v3_addr_get_range() readable.
Instead of checking everything in a single if statement, group the
checks according to their purposes.
ok inoguchi jsing
tb [Tue, 4 Jan 2022 20:17:07 +0000 (20:17 +0000)]
Add a length check to make_addressPrefix()
Make the callers pass in the afi so that make_addressPrefix() can check
prefixlen to be reasonable. If the afi is anything else than IPv4 or
IPv6, cap its length at the length needed for IPv6. This way we avoid
arbitrary out-of-bounds reads if the caller decides to pass in something
stupid.
ok inoguchi jsing
tb [Tue, 4 Jan 2022 20:04:38 +0000 (20:04 +0000)]
Remove some dead code
IPAddressRange_new() populates both its min and max members, so
they won't ever be NULL and will never need to be allocated.
ok inoguchi jsing
tb [Tue, 4 Jan 2022 20:02:22 +0000 (20:02 +0000)]
Drop a pointless NULL check
IPAddressOrRange_new() instantiates a choice type, so we need to
allocate one member of the union ourselves, so aor->u.addressPrefix
will always be NULL.
ok inoguchi jsing
tb [Tue, 4 Jan 2022 19:56:53 +0000 (19:56 +0000)]
First pass over x509_addr_validate_path()
Replace reaching into the structs with IPAddressFamily accessors
and add a few comments that explain what the code is actually doing.
ok inoguchi jsing
jeremy [Tue, 4 Jan 2022 19:51:24 +0000 (19:51 +0000)]
Document ruby31 FLAVOR
tb [Tue, 4 Jan 2022 19:49:14 +0000 (19:49 +0000)]
Refactor IPAddressFamily accessors
Introduce a helper function that allows fetching the AFI and the
optional SAFI out of an IPAddressFamily. Also add two wrappers that
only fetch and validate the AFI, where validation currently only
means that the length is between 2 and 3.
Use these accessors throughout to simplify and streamline the code.
ok inoguchi jsing
claudio [Tue, 4 Jan 2022 18:41:32 +0000 (18:41 +0000)]
Pass the filename back from the parser to the parent.
The parent will then add the filename to the filepath tree instead
of doing that in entity_write_req(). In the parser pass the filename
instead of the full entity object to various proc_parser functions.
With this it will be possible to check more then one file in the parser.
OK tb@, earlyer version OK benno@
claudio [Tue, 4 Jan 2022 18:16:09 +0000 (18:16 +0000)]
Fix -n mode by setting repository states to REPO_DONE instead of the
more correct REPO_FAILED. The problem is that with REPO_FAILED the repo
no longer matches a subsequent rrdp_get lookup and as a result the repo
uses rsync which wrong in this case.
OK benno@ tb@
guenther [Tue, 4 Jan 2022 18:13:31 +0000 (18:13 +0000)]
Use the device we read the hibernate signature from for the entire
resume. This fixes setups where a umass device no longer attaching
at resume results in a softraid device being renumbered so the
hibernate-time device is no longer correct
ok mlarkin@ jsing@
kettenis [Tue, 4 Jan 2022 16:15:28 +0000 (16:15 +0000)]
Restrict the pci(4) ioctl interface to devices detected by the kernel.
This fixes issues on the M1 Macs where the PCI probe done by Xorg
breaks the WiFi chip.
ok patrick@
stsp [Tue, 4 Jan 2022 15:55:28 +0000 (15:55 +0000)]
fix length boundary checks for incoming packets in iwm/iwx
The minimum length and the maximum length required were both too low,
due to an error in accounting for the 4-byte packet length+flags header.
Patch by Christian Ehrhardt
stsp [Tue, 4 Jan 2022 15:53:57 +0000 (15:53 +0000)]
fix Rx Block Ack session validity checks in iwm(4) and iwx(4)
I ported a NULL pointer check from iwlwifi rxmq.c which cannot
happen in our version of this code. Instead we need to check
the BA session ID to determine whether a BA session is valid.
Patch by Christian Ehrhardt.
tb [Tue, 4 Jan 2022 15:37:23 +0000 (15:37 +0000)]
Simplify the verify callback
The final warnx() is very noisy and essentially a remnant of earlier
debugging code. By ditching it and erroring directly on encountering
an unknown critical extension, the code becomes a bit simpler.
ok claudio
claudio [Tue, 4 Jan 2022 15:25:05 +0000 (15:25 +0000)]
Try to handle possible vm_register() failures and return an error back
to the caller instead of most probably crashing because of a NULL pointer
access. This fixes also another -Wunused-but-set-variable warning.
OK benno@ dv@
claudio [Tue, 4 Jan 2022 15:22:53 +0000 (15:22 +0000)]
Remove unused imsg type IMSG_VMDOP_RECEIVE_VM_RESPONSE.
OK benno@ dv@
claudio [Tue, 4 Jan 2022 15:21:40 +0000 (15:21 +0000)]
Another -Wunused-but-set-variable fix.
Based on input from dv@
claudio [Tue, 4 Jan 2022 15:18:44 +0000 (15:18 +0000)]
Fix some simple -Wunused-but-set-variable warnings.
OK benno@ dv@
jmc [Tue, 4 Jan 2022 13:43:14 +0000 (13:43 +0000)]
- add LDAP
- capitalise RADIUS when referring to the protocol
- remove tis
from raf czlonka
ok sthen ajacoutot
patrick [Tue, 4 Jan 2022 13:40:58 +0000 (13:40 +0000)]
acpi_getprop() needs to actually make sure that we're looking at the
correct property. While there adjust acpi_getpropint() as well to
increase similarity with acpi_getprop().
ok kettenis@
tb [Tue, 4 Jan 2022 13:39:01 +0000 (13:39 +0000)]
Stop setting X509_V_FLAG_IGNORE_CRITICAL
Since the last bump, libcrypto knows about the RFC 3779 extensions.
Therefore, setting X509_V_FLAG_IGNORE_CRITICAL is no longer needed.
In fact, we want to error on critical extensions neither rpki-client
nor libcrypto knows about.
On older LibreSSL versions with the default verify callback, this
causes verification failures. Implement a verify callback that
intercepts X509_V_ERR_UNHANDLED_CRITICAL_EXTENSIONS and checks that
the cert doesn't contain critical extensions not supported by libcrypto
other than the expected RFC 3779 extensions.
Tested with LibreSSL 3.3 and 3.4 on OpenBSD 6.9 and 7.0-stable by me
and with LibreSSL 3.2 on Linux by claudio.
input/ok claudio
jsing [Tue, 4 Jan 2022 12:53:31 +0000 (12:53 +0000)]
Return 0 on failure from send/get kex functions in the legacy stack.
In the legacy stack, a message handling function returns -1 for failure,
0 for need more data and 1 for success (although in extra special cases
2 may also be used). However, the various send/get kex functions only
need to indicate success or failure - switch these to return 0 on failure
(rather than -1) and use normal result testing.
This leaves GOST unchanged for now, as that code is special and needs
extra work.
ok inoguchi@ tb@
jsg [Tue, 4 Jan 2022 11:31:42 +0000 (11:31 +0000)]
remove the last part of driver for aic-6250 scsi on aviion
jsing [Tue, 4 Jan 2022 11:17:11 +0000 (11:17 +0000)]
Use normal result testing for tls1_check_curve().
jsing [Tue, 4 Jan 2022 11:14:54 +0000 (11:14 +0000)]
Refactor ssl3_get_server_kex_ecdhe() to separate parsing and validation.
If we receive something other than a "named curve", send a handshake
failure alert as we're unable to complete the handshake with the given
parameters. If the server responded with a curve that we did not advertise
send an illegal parameter alert.
ok inoguchi@ tb@
jsing [Tue, 4 Jan 2022 11:01:58 +0000 (11:01 +0000)]
Pull key share group/length CBB code up from tls13_key_share_public()
This provides better symmetry with the parsing code and will allow for
better reuse with the legacy stack, which has different message structures.
ok inoguchi@ tb@
jsg [Tue, 4 Jan 2022 11:01:24 +0000 (11:01 +0000)]
remove files missed when sgi was removed
ok visa@
jsing [Tue, 4 Jan 2022 10:34:16 +0000 (10:34 +0000)]
Only allow zero length key shares when we know we're doing HRR.
ok inoguchi@ tb@
dtucker [Tue, 4 Jan 2022 08:38:53 +0000 (08:38 +0000)]
Log command invocation in debugging log to aid in manually reproducing
failing commands.
djm [Tue, 4 Jan 2022 07:20:33 +0000 (07:20 +0000)]
unbreak test: was picking up system ssh-add instead of the one supposedly
being tested. Spotted by dtucker and using his VM zoo (which includes
some systems old enough to lack
ed25519 key support)
jsg [Tue, 4 Jan 2022 06:48:46 +0000 (06:48 +0000)]
remove unused file
yasuoka [Tue, 4 Jan 2022 06:32:39 +0000 (06:32 +0000)]
Add `ipsec_flows_mtx' mutex(9) to protect `ipsp_ids_*' list and
trees. ipsp_ids_lookup() returns `ids' with bumped reference
counter. original diff from mvs
ok mvs
florian [Tue, 4 Jan 2022 06:20:37 +0000 (06:20 +0000)]
Make host name DHCP option configurable.
Diff from hagen@sdf.org, tweaks by me.
OK phessler
testing & OK bket
florian [Tue, 4 Jan 2022 06:17:46 +0000 (06:17 +0000)]
Wait for the link-local address to become non-tentative.
Otherwise we just end up with an ugly "Can't assign requested address"
error from sendmsg(2).
OK phessler
gnezdo [Tue, 4 Jan 2022 05:50:43 +0000 (05:50 +0000)]
Remove commented out gtp from amd64 GENERIC
It has long been irrelevant as pointed out by Crystal Kolipe
kolipe.c AT exoticsilicon.com
ok mlarkin
deraadt [Mon, 3 Jan 2022 23:48:03 +0000 (23:48 +0000)]
sync