otto [Sun, 4 Jun 2023 06:58:33 +0000 (06:58 +0000)]
More thorough write-afetr-free checks.
On free, chunks (the pieces of a pages used for smaller allocations)
are junked and then validated after they leave the delayed free
list. So after free, a chunk always contains junk bytes. This means
that if we start with the right contents for a new page of chunks,
we can *validate* instead of *write* junk bytes when (re)-using a
chunk.
With this, we can detect write-after-free when a chunk is recycled,
not justy when a chunk is in the delayed free list. We do a little
bit more work on initial allocation of a page of chunks and when
re-using (as we validate now even on junk level 1).
Also: some extra consistency checks for recallocaray(3) and fixes
in error messages to make them more consistent, with man page bits.
Plus regress additions.
krw [Sat, 3 Jun 2023 21:37:53 +0000 (21:37 +0000)]
Remove declarations of unused local variables, an unused function
(get_long) and add missing {} in devsw[1] initialization.
Most from 2011 NetBSD commit by tsutui.
No functional change.
Build tested and ok kn@
tb [Sat, 3 Jun 2023 21:20:29 +0000 (21:20 +0000)]
Add regress coverage for BN_mod_inverse()
This would detect the aliasing issue reported by Guido Vranken fixed
in bn_gcd.c r1.28. Most testcases are from BoringSSL's regress test.
op [Sat, 3 Jun 2023 15:19:38 +0000 (15:19 +0000)]
drop `uptodate()' check from hack(6)
hack(6) scrapes $PATH to find its executable and compare the mtime to
the save file and bone file. If the game is newer than those, they're
not loaded.
Drop this feature. /usr/games is not in the default $PATH anymore, and
the format for those file didn't change since the import.
Diff from Anton Konyahin (me [at] konyahin [dot] xyz)
cheloha [Fri, 2 Jun 2023 17:44:29 +0000 (17:44 +0000)]
pledge(2): stdio: permit restricted profil(2) for moncontrol(3)
Currently, pledged '-pg' binaries get killed in _mcleanup() when they
try to disable profil(2) via moncontrol(3).
Disabling profil(2) is harmless. Add profil(2) to the "stdio"
pledge(2) promise and permit profil(2) calls when the scale argument
is zero. Enabling profil(2) remains forbidden in pledged processes.
This gets us one step closer to making '-pg' binaries compatible with
pledge(2). The next step is to decide how to exfiltrate the profiling
data from the process during _mcleanup().
Prompted by semarie@. Cleaned up by deraadt@. With input from
deraadt@, espie@, and semarie@.
"Looks good" deraadt@
pledge(2) pieces ok semarie@
tb [Fri, 2 Jun 2023 17:15:30 +0000 (17:15 +0000)]
Fix variable reuse in BN_mod_inverse()
The somewhat strange calculation m = a^{-1} (mod m) can return 0. This
breaks because of BN_nnmod() having delicate semantics of which variable
can be reused. BN_nnmod(a, a, m, ctx) works and the library relies on that.
Here, the code ends up doing BN_nnmod(m, a, m, ctx) and this doesn't work.
If the result of the initial BN_mod() is negative, then BN_nnmod() will
return 0.
Problem reported by Guido Vranken in
https://github.com/openssl/openssl/issues/21110
This code is well covered by regress, but it does not currently have
explicit test coverage. Such will be added soon.
ok beck jsing
tb [Fri, 2 Jun 2023 08:35:10 +0000 (08:35 +0000)]
fix typo
krw [Fri, 2 Jun 2023 00:10:25 +0000 (00:10 +0000)]
Missed a trailing space.
kn [Thu, 1 Jun 2023 18:57:53 +0000 (18:57 +0000)]
Add support for wireguard peer descriptions
"wgdescr[iption] foo" to label one peer (amongst many) on a wg(4) interface,
"-wgdescr[iption]" or "wgdescr ''" to remove the label, completely analogous
to existing interface discriptions.
Idea/initial diff from Mikolaj Kucharski (OK sthen)
Tests/prodded by Hrvoje Popovski
Tweaks/manual bits from me
Feedback deraadt sthen mvs claudio
OK claudio
krw [Thu, 1 Jun 2023 17:24:56 +0000 (17:24 +0000)]
Expunge a bunch of eye searing trailing whitespace.
claudio [Thu, 1 Jun 2023 10:21:26 +0000 (10:21 +0000)]
Change wakeup_proc() to no longer grab the SCHED_LOCK() instead it must
be called with SCHED_LOCK() held. Also add an extra argument to update
the process flags p_flag so that the timeout handler can set the
P_TIMEOUT flag before making the process runnable.
OK mpi@
claudio [Thu, 1 Jun 2023 09:47:34 +0000 (09:47 +0000)]
Check the F_NEXTHOP flag on the right kroute6 object.
On multipath routes the check ended up checking the wrong route for the
nexthop update. This resulted in a use-after-free in kroute_detach_nexthop().
This only affects IPv6 in the IPv4 code path the right object was already used.
Thanks to sthen@ for providing the debug information to track this down.
OK sthen@ tb@
tb [Thu, 1 Jun 2023 09:46:00 +0000 (09:46 +0000)]
Remove the speed test again
It takes too much time and we now know that all covered ciphers can cope
with unaligned input and output on all tested architectures.
jan [Thu, 1 Jun 2023 09:05:33 +0000 (09:05 +0000)]
Enable forwarding of ix(4) LRO Pakets via TSO
Also fix ip6_forwarding of TSO packets with tcp_if_output_tso().
With a lot of testing from Hrvoje Popovski
and a lot of tweaks from bluhm@
ok bluhm@
tb [Thu, 1 Jun 2023 07:32:25 +0000 (07:32 +0000)]
Rework tls_check_subject_altname() error handling
Default to having rv = -1 and explicitly goto done to set rv = 0.
This matches other code better.
ok jsing
tb [Thu, 1 Jun 2023 07:29:15 +0000 (07:29 +0000)]
Check for X509_get_ext_d2i() failure
X509_get_ext_d2i() (or rather X509V3_get_d2i()) can return NULL for
various reasons. If it fails because the extension wasn't found, it
sets *crit = -1. In any other case, e.g., the cert is bad or we ran
out of memory in X509V3_EXT_d2i(), crit is set to something else, so
we should actually error.
ok jsing
tb [Thu, 1 Jun 2023 02:34:23 +0000 (02:34 +0000)]
Avoid a potentially overflowing check
This doesn't actually overflow, but still is poor style.
Speaking of which: this is now the second time I get to fix something
reported by Nicky Mouha by way of a blog post. The first time was the
actual SHA-3 buffer overflow in Python where it is not entirely clear
who screwed up and how. Hopefully next time proper communication will
happen and work.
ok jsing
op [Wed, 31 May 2023 16:51:46 +0000 (16:51 +0000)]
add missing include of time.h
spotted after a report on OpenSMTPD-portable. While here include
sys/time.h in smtpd.h, as noted in event_init(3), since it includes
event.h.
ok millert@
jeremy [Wed, 31 May 2023 15:27:45 +0000 (15:27 +0000)]
Update to reflect that default Ruby version in ports is now 3.2
krw [Wed, 31 May 2023 13:49:56 +0000 (13:49 +0000)]
Remove declarations for unused search_label() local variables
error, i, p and poff. Fewer complaints from gcc -Wall.
The same was done to the NetBSD version in 2011 by tsutui.
Remove two more recent load_disklabel() comments implying that
search_label() checks MBR partitions. It doesn't and never has
despite the enticing names/types of the above unused variables.
No functional change.
dlg [Tue, 30 May 2023 23:55:42 +0000 (23:55 +0000)]
add net_tq_barriers
this waits once for something to end in all the net tqs.
ok claudio@
bluhm [Tue, 30 May 2023 19:32:57 +0000 (19:32 +0000)]
Use generic checksum calculation for TCP SYN+ACK packets.
Our syn cache did checksum calculation by hand, instead of the
established mechanism in ip output. The software-checksummed counter
increased once per incoming TCP connection.
Just set the flag M_TCP_CSUM_OUT in syn_cache_respond() and let
in_proto_cksum_out() do the work later. Then hardware checksumming
is used where available. Also remove redundant code. The unhandled
af case is handled in the first switch statement of the function.
tested by Hrvoje Popovski; OK mvs@
op [Tue, 30 May 2023 16:44:16 +0000 (16:44 +0000)]
fail when an argument is given on 'id -R'
from Lucas (lucas [at] sexy [dot] is).
ok millert@
job [Tue, 30 May 2023 16:02:28 +0000 (16:02 +0000)]
Fixup file modification timestamps to optimize failover from RRDP to RSYNC
In the RSYNC protocol a file's last modification time and its size are
used to determine whether sending a (partial) copy over the wire is needed.
Previously, when RRDP data structures are serialized to disk, the mtime of
files in DIR_VALID ended up being UTIME_NOW.
Thus, the mtimes of files obtained through RRDP will never match the mtimes
of the same files available through RSYNC - causing each and every file to
be added to the file transfer list.
Instead, use the internal timestamps of RPKI files as the last modified
timestamp. Specifically, for Signed Objects (ROAs, MFTs, GBRs, TAKs, ASPAs)
the CMS signing-time, for .cer files the X.509 notBefore, and for .crl files
the CRL lastUpdate. This results in a surprising optimization for the number
files which have to be transfered.
OK claudio@
job [Tue, 30 May 2023 15:56:47 +0000 (15:56 +0000)]
Add json.c to linker instructions
krw [Tue, 30 May 2023 15:10:39 +0000 (15:10 +0000)]
sparc64 disklabel(8) must be compiled with SUN_CYLCHECK and SUN_AAT0.
sparc64 auto allocation during install does not use fdisk(8).
Add *.sparc64.ok files that reflect sparc64 reality.
Should fix disklabel(8) regression on sparc64.
Requested by & ok bluhm@
aisha [Tue, 30 May 2023 14:04:53 +0000 (14:04 +0000)]
fix markups in character classes
ok jmc@
claudio [Tue, 30 May 2023 12:14:48 +0000 (12:14 +0000)]
Revert commitid ANSBO0rBvIUtTi45:
Make rpki-client choose the verification time of the time it is invoked
rather than always getting the current system time for every certificate
verification. This will result in output that is not variable on run-time.
Using the time of invocation does not work well with fast publishing CAs. It
can take a few minutes to reach a repo and that CA may have issued certificates
that are not yet valid if that startup time of rpki-client is used to validate.
This still keeps the -P option to specify a fixed validation time.
OK beck@ job@ tb@
claudio [Tue, 30 May 2023 12:12:06 +0000 (12:12 +0000)]
Split cleanup into cleanup and repository cleanup and show how many files
are kept / removed in the repository temporary storage.
After a discussion with tb@ and job@
claudio [Tue, 30 May 2023 12:02:22 +0000 (12:02 +0000)]
Convert all of filemode to use the json API as well.
Output is mostly the same apart from some space differences.
OK tb@ job@
tb [Tue, 30 May 2023 11:09:08 +0000 (11:09 +0000)]
Use error check to ensure we have SignedData in CMS
CMS_get0_SignerInfos() only returns a non-NULL pointer if the CMS object
contains SignedData. The subsequent assert can trigger if we parse an
object that is not of this type. Nothing ensures this up to this point,
so we have no way of knowing that the assertion is actually true. If we
get a CMS object without SignedData, we should ignore it, not abort the
rpki-client run. With this check in place it is also clear that we
actually check point 1a of the list of things to check in RFC 6488,
section 3.
ok claudio job
claudio [Tue, 30 May 2023 08:41:15 +0000 (08:41 +0000)]
Replace the one use of ibuf_prepend() using a similar ibuf_new() + ibuf_cat()
method but instead of overwriting ibuf internals replace the buf a level up.
Users of ikev2_msg_send() are not allowed to hold and reuse a pointer to
msg_data (which is another footgun to disarm at some point).
OK tb@
jsg [Tue, 30 May 2023 08:30:00 +0000 (08:30 +0000)]
spelling
ok jmc@ guenther@ tb@
op [Tue, 30 May 2023 07:37:34 +0000 (07:37 +0000)]
fix some nits on previous
- move a sentence out of a Bd block
- add some .Pp for spacing
- avoid a double colon on a sentence and the usage of second person
- mark STORE_CTX with .Vt
- change one Vt -> Dv (done after this has been ok'd by beck)
ok beck@
espie [Tue, 30 May 2023 04:42:21 +0000 (04:42 +0000)]
Due to the way make is designed, not being able to read a makefile
is basically silent.
Record errors due to missing permissions and other oddities, and display them
when we error out due to lack of targets, as a quality-of-life diagnostic.
Based on a remark from sthen@, with some feedback and tweaks from op@
okay op@, kn@
guenther [Tue, 30 May 2023 02:02:00 +0000 (02:02 +0000)]
Add IBT support to the retpoline+znow PLTs
ok kettenis@
aisha [Mon, 29 May 2023 21:13:24 +0000 (21:13 +0000)]
fix markups and reference sections properly
from Josiah Frentsos <jfrent AT tilde DOT team>
OK op@
beck [Mon, 29 May 2023 15:52:46 +0000 (15:52 +0000)]
Correct test that was pasto'ed incorrectly
This now tests what the comment says it does
beck [Mon, 29 May 2023 14:12:36 +0000 (14:12 +0000)]
Correctly catch all return values from X509_NAME_get_index_by_NID
And some comment requests, from jsing@
ok jsing@
beck [Mon, 29 May 2023 11:57:23 +0000 (11:57 +0000)]
Oops, Fa -> .Fa
beck [Mon, 29 May 2023 11:54:50 +0000 (11:54 +0000)]
Make X509_NAME_get_text_by[NID|OBJ] safer.
This is an un-revert with nits of the previously landed change
to do this which broke libtls. libtls has now been changed to
not use this function.
This change ensures that if something is returned it is "text"
(UTF-8) and a C string not containing a NUL byte. Historically
callers to this function assume the result is text and a C string
however the OpenSSL version simply hands them the bytes from an
ASN1_STRING and expects them to know bad things can happen which
they almost universally do not check for. Partly inspired by
goings on in boringssl.
ok jsing@ tb@
beck [Mon, 29 May 2023 11:14:19 +0000 (11:14 +0000)]
Stop suggesting that children play with loaded revolvers.
This takes much of the language that boring uses to document
the verify callback, and corrects the historical horror that
OpenSSL introduced years ago by suggesting people ignore expiry
dates using the callback instead of the verify flags.
nits by jsg@ and tb@
ok tb@
espie [Mon, 29 May 2023 09:05:24 +0000 (09:05 +0000)]
renamed log file
sf [Mon, 29 May 2023 08:13:35 +0000 (08:13 +0000)]
virtio: Set DRIVER_OK earlier
The DRIVER_OK bit must be set before using any virt-queues. To allow
virtio device drivers to use the virt-queues in their attach functions,
set the bit there and not in the virtio transport attach function. Only
vioscsi and viogpu really need this, but let's only have one standard
way to do this.
Noticed because of hangs with vioscsi on qemu/windows and in the Oracle
cloud. With much debugging help by Aaron Mason.
Also revert vioscsi.c 1.31 "Temporarily workaround double calls into
vioscsi_req_done()"
ok krw@
espie [Mon, 29 May 2023 07:35:39 +0000 (07:35 +0000)]
use v5.36, this one is somewhat trivial
deraadt [Mon, 29 May 2023 04:24:39 +0000 (04:24 +0000)]
some hackery to make arm64 ramdisk compile again
jsg [Mon, 29 May 2023 02:26:14 +0000 (02:26 +0000)]
recognise Cortex-A520 (Hayes), Cortex-A720 (Hunter), Cortex-X4 (Hunter-ELP)
jsing [Sun, 28 May 2023 17:42:30 +0000 (17:42 +0000)]
Provide optimised bn_mulw_{addw,addw_addw,addtw}() for aarch64.
This results in bn_mul_comba4() and bn_mul_comba8() requiring ~30% less
instructions than they did previously.
jsing [Sun, 28 May 2023 17:22:04 +0000 (17:22 +0000)]
Provide optimised bn_addw_addw()/bn_subw_subw() for aarch64.
jsing [Sun, 28 May 2023 14:54:37 +0000 (14:54 +0000)]
Sprinkle some style(9).
jsing [Sun, 28 May 2023 14:49:21 +0000 (14:49 +0000)]
Expand occurrences of HASH_CTX that were previously missed.
No change in generated assembly.
jsing [Sun, 28 May 2023 14:14:33 +0000 (14:14 +0000)]
Reorder functions.
No intended functional change.
jsing [Sun, 28 May 2023 13:57:27 +0000 (13:57 +0000)]
Clean up includes.
jsing [Sun, 28 May 2023 13:55:55 +0000 (13:55 +0000)]
Remove now unnecessary do {} while(0);
jsing [Sun, 28 May 2023 13:53:08 +0000 (13:53 +0000)]
Inline HASH_MAKE_STRING for SHA256.
No change to generated assembly.
jsing [Sun, 28 May 2023 10:34:17 +0000 (10:34 +0000)]
Rewrite BN_{asc,dec,hex}2bn() using CBS.
This gives us more readable and safer code. There are two intentional
changes to behaviour - firstly, all three functions zero any BN that was
passed in, prior to doing any further processing. This means that a passed
BN is always in a known state, regardless of what happens later. Secondly,
BN_asc2bn() now fails on NULL input, rather than crashing. This brings its
behaviour inline with BN_dec2bn() and BN_hex2bn().
ok tb@
beck [Sun, 28 May 2023 09:06:34 +0000 (09:06 +0000)]
correct comment, spotted by tb@
beck [Sun, 28 May 2023 09:02:01 +0000 (09:02 +0000)]
Refactor tls_check_common_name to use lower level API.
X509_NAME_get_text_by_NID is kind of a bad interface that
we wish to make safer, and does not give us the visibility
we really want here to detect hostile things.
Instead call the lower level functions to do some better
checking that should be done by X509_NAME_get_text_by_NID,
but is not in the OpenSSL version. Specifically we will treat
the input as hostile and fail if:
1) The certificate contains more than one CN in the subject.
2) The CN does not decode as UTF-8
3) The CN is of invalid length (must be between 1 and 64 bytes)
4) The CN contains a 0 byte
4) matches the existing logic, 1 and 2, and 3 are new checks.
ok tb@
asou [Sun, 28 May 2023 05:28:50 +0000 (05:28 +0000)]
Used number of bytes instead of number of 512 byte sectors.
ok dv@
tb [Sun, 28 May 2023 05:25:24 +0000 (05:25 +0000)]
Merge X509_VERIFY_PARAM_ID into X509_VERIFY_PARAM
Back in the day when essentially every struct was open to all applications,
X509_VERIFY_PARAM_ID provided a modicum of opacity. This indirection is now
no longer needed with X509_VERIFY_PARAM being opaque itself, so stop using
X509_VERIFY_PARAM_ID and merge it into X509_VERIFY_PARAM. This is a first
small step towards cleaning up the X509_VERIFY_PARAM mess.
ok jsing
kettenis [Sat, 27 May 2023 19:35:55 +0000 (19:35 +0000)]
Implement battery charge control.
ok patrick@, tobhe@
jsing [Sat, 27 May 2023 18:39:03 +0000 (18:39 +0000)]
Implement SHA256_{Update,Transform,Final}() directly in sha256.c.
m32_common.h is a typical OpenSSL macro horror show - copy the update,
transform and final functions from md32_common.h, manually expanding the
macros for SHA256. This will allow for further clean up to occur.
No change in generated assembly.
ok beck@ tb@
jsing [Sat, 27 May 2023 18:33:34 +0000 (18:33 +0000)]
Add HASH_NO_UPDATE and HASH_NO_TRANSFORM to md32_common.h
This makes it possible to still use minimal parts of md32_common.h, while
disabling the update and transform functions.
ok beck@ tb@
jsing [Sat, 27 May 2023 15:50:56 +0000 (15:50 +0000)]
Add coverage for calling BN_{dec,hex}2bn() with NULL inputs.
tb [Sat, 27 May 2023 13:54:46 +0000 (13:54 +0000)]
Bump LibreSSL version
espie [Sat, 27 May 2023 10:08:45 +0000 (10:08 +0000)]
the solver needs some love for later
espie [Sat, 27 May 2023 10:08:25 +0000 (10:08 +0000)]
remove the extra local variable (blessing a ref to self DOES work)
accordingly zap the extra blocks that are no longer needed making
this look less funky
espie [Sat, 27 May 2023 10:07:33 +0000 (10:07 +0000)]
comment/cosmetic
espie [Sat, 27 May 2023 10:07:12 +0000 (10:07 +0000)]
more comment
espie [Sat, 27 May 2023 10:06:55 +0000 (10:06 +0000)]
yeah, yeah, those names are horrible
espie [Sat, 27 May 2023 10:06:38 +0000 (10:06 +0000)]
finish making this a singleton class (which I don't always like
and is definitely worthy of a comment)
espie [Sat, 27 May 2023 10:05:50 +0000 (10:05 +0000)]
more descriptive comment
espie [Sat, 27 May 2023 10:04:17 +0000 (10:04 +0000)]
indicate that not having hashes already is just legacy needed
until all old packages are gone
espie [Sat, 27 May 2023 10:03:43 +0000 (10:03 +0000)]
point people towards relevant comments
espie [Sat, 27 May 2023 10:03:21 +0000 (10:03 +0000)]
a few comments
explicitly return undef from check,
which will prevent casual me from 'fixing' it later
espie [Sat, 27 May 2023 10:02:38 +0000 (10:02 +0000)]
prepare for 5.36: pass $code as @code so that we don't have
to decide a default right away (I should apply this
everywhere I go to read a plist actually)
espie [Sat, 27 May 2023 10:01:51 +0000 (10:01 +0000)]
comments
espie [Sat, 27 May 2023 10:01:38 +0000 (10:01 +0000)]
document base method
espie [Sat, 27 May 2023 10:01:21 +0000 (10:01 +0000)]
try/catch to be revisited
espie [Sat, 27 May 2023 10:01:08 +0000 (10:01 +0000)]
comments/cosmetics
espie [Sat, 27 May 2023 10:00:48 +0000 (10:00 +0000)]
restore documentation: explain what ArcCheck does
espie [Sat, 27 May 2023 10:00:23 +0000 (10:00 +0000)]
cosmetic/comments improvement
espie [Sat, 27 May 2023 09:59:51 +0000 (09:59 +0000)]
- make the handling of interactive more regular by
delegating the choice of the class to a separate method
- zap/improve meaningless comments
- mark ntogo as a focus for improvements
espie [Sat, 27 May 2023 09:58:26 +0000 (09:58 +0000)]
prepare for 5.36, document methods in base class
jsing [Sat, 27 May 2023 09:18:17 +0000 (09:18 +0000)]
Clean up alignment handling for SHA-512.
This recommits r1.37 of sha512.c, however uses uint8_t * instead of void *
for the crypto_load_* functions and primarily uses const uint8_t * to track
input, only casting to const SHA_LONG64 * once we know that it is suitably
aligned. This prevents the compiler from implying alignment based on type.
Tested by tb@ and deraadt@ on platforms with gcc and strict alignment.
ok tb@
otto [Sat, 27 May 2023 04:33:00 +0000 (04:33 +0000)]
Remove malloc interposition, a workaround that was once needed for emacs
ok guenther@
claudio [Fri, 26 May 2023 14:57:38 +0000 (14:57 +0000)]
Properly account del_extra_files and add the value to the json header.
OK tb@ (from a larger diff)
tb [Fri, 26 May 2023 13:44:05 +0000 (13:44 +0000)]
Move verified_chain from SSL to SSL_HANDSHAKE
This is a better version of the fix for the missing pointer invalidation
but a bit larger, so errata got the minimal fix.
tested by jcs
ok jsing
kn [Fri, 26 May 2023 12:13:26 +0000 (12:13 +0000)]
Remove net lock from DIOC{S,G}ETLIMIT
Grab the pf lock for pf_pool_limits[] in pfsync such that all access is
covered by the pf lock; document accordingly.
Hard memory pool limits don't need the net lock for protection, pool(9)s
have their own internal lock and the pf lock fully covers limit values.
(pf_pool_limits[] access in DIOCXCOMMIT remains under pf *and net* lock
until the rest in there gets pulled out of the net lock.)
OK sashan
kn [Fri, 26 May 2023 11:41:50 +0000 (11:41 +0000)]
simplify yes/no answer handling
ask_yn() returns 0/1 on yes/no, so use the shorter '|| skip' found in
install.sub; no functional change.
OK solene
kurt [Thu, 25 May 2023 19:35:58 +0000 (19:35 +0000)]
Disable witness for mutexes created on the stack which allows
the ddb command 'show witness' to succeed without panicking.
Leaving witness enabled on these mutexes saves a pointer to
struct lock_type on the stack which gets clobbered resulting
in a panic in witness_ddb_display_descendants().
okay miod@
kurt [Thu, 25 May 2023 19:32:34 +0000 (19:32 +0000)]
Work around sparc64 WITNESS kernel failing to load by moving large witness
data structures from bss to be allocated in witness_initialize().
Tested on sparc64, amd64, arm64, i386, octeon. okay miod@
claudio [Thu, 25 May 2023 14:20:25 +0000 (14:20 +0000)]
In session_process_msg() recheck the validity of the rbuf before moving
the remaining data around.
There is an improbable case where a NOTIFICATION is received while also
reaching the MSG_PROCESS_LIMIT. In this case rbuf is NULL when breaking
out of the for loop and hitting this code.
sthen@ is the (un)lucky person to hit that improbable case
OK tb@ sthen@
claudio [Thu, 25 May 2023 12:52:56 +0000 (12:52 +0000)]
Update usage add -P epoch
claudio [Thu, 25 May 2023 12:49:39 +0000 (12:49 +0000)]
Fix repo_cleanup_entry() state machine so that the repository lookups
are done when the full repo path is available. Without this all repo
lookups returned NULL and the code did not work as intended.
OK tb@
op [Thu, 25 May 2023 11:37:58 +0000 (11:37 +0000)]
remove two unused defines
last PROC_COUNT use was removed with the switch to fork+exec by eric@ in
2016, CA_FILE with the removal of cert.c two years ago.
ok tb@, kn@
sthen [Thu, 25 May 2023 09:41:46 +0000 (09:41 +0000)]
After RFC 9110, the IANA services registry now lists both udp and tcp
for https (HTTP/3 over QUIC). Add it to /etc/services so that it's included
when /etc/rc populates sysctl net.inet.udp.baddynamic.
suggested by Renauld Allard, ok tb@
op [Thu, 25 May 2023 07:46:21 +0000 (07:46 +0000)]
Forcibly update the EVP_PKEY's internal key
To aid privilege separation, libtls maintains application-specific data
on the key inside the EVP_PKEY abstraction because the EVP API doesn't
provide a way to do that on the EVP_PKEY itself.
OpenSSL 3 changed behavior of EVP_PKEY_get1_RSA() and related functions.
These now return a struct from some cache. Thus, modifying the RSA will
no longer modify the EVP_PKEY like it did previously, which was clearly
implied to be the case in the older documentation.
This is a subtle breaking change that affects several applications.
While this is documented, no real solution is provided. The transition
plan from one OpenSSL major version to the next one tends to involve
many #ifdef in the ecosystem, and the only suggestion provided by the
new documentation is to switch to a completely unrelated, new API.
Instead, forcibly reset the internal key on EVP_PKEY after modification,
this way the change is picked up also by OpenSSL 3.
Fixes issue 1171 in OpenSMTPD-portable
ok tb@, jsing@
claudio [Thu, 25 May 2023 07:45:33 +0000 (07:45 +0000)]
Do not use wakeup_proc() outside of the sleep machinery. Just use
wakeup_one().
OK mpi@
djm [Wed, 24 May 2023 23:01:06 +0000 (23:01 +0000)]
fix AuthorizedPrincipalsCommand when AuthorizedKeysCommand appears
previously in configuration. Reported by John Meyers in bz3574
ok dtucker@