martijn [Mon, 6 Nov 2023 09:46:04 +0000 (09:46 +0000)]
Mark a couple of tests as broken, now that the sysORTable is gone/empty.
Relying on the live snmpd(8) metrics is brittle and this regress test
should probably move to a custom agent in a similar fashion to the
custom manager from the snmpd(8) regress.
Pointed out by anton@
stsp [Mon, 6 Nov 2023 08:34:41 +0000 (08:34 +0000)]
avoid using MCS9 in iwm(4) Tx rate selection if 40 MHz is disabled
MCS9 is not valid on 20HMz so we must not use it in 20MHz-only mode.
Fixes firmware panic 0x00002078, which indicates the driver sets a rate
with a bad number of antennas, bad number of Tx streams, or bad bandwidth.
Observed by dlg@ with the miktrotik AP for which iwm(4) firmware requires
our 20MHz-only workaround to interoperate in 11ac mode.
ok dlg@
espie [Mon, 6 Nov 2023 08:22:59 +0000 (08:22 +0000)]
actually, I still need to exit() if infocmp can't be run
espie [Mon, 6 Nov 2023 08:21:43 +0000 (08:21 +0000)]
system + exit -> exec
as noticed by afresh1@
(this should be all for now)
espie [Mon, 6 Nov 2023 08:20:51 +0000 (08:20 +0000)]
since we're always running info_cmp in the end, remove the whole
"in case of emergency run info_cmp block" (more efficient: we only
parse the entry we need). Keep the small part that creates a dumb terminal
on VMS, even though we're not running that.
Add an extra state check in the automaton to avoid skipping a tmp_term
espie [Mon, 6 Nov 2023 08:14:36 +0000 (08:14 +0000)]
simplify logic for $entry in the main termcap loop
espie [Mon, 6 Nov 2023 08:13:01 +0000 (08:13 +0000)]
always try to run info_cmp at the end if we haven't got what we wanted
This allows terminals with only terminfo capabilities to show up in
Term::Cap (like kitty) !
espie [Mon, 6 Nov 2023 08:11:11 +0000 (08:11 +0000)]
simplify $ENV{TERMCAP} handling: don't bother setting anything if
the variable doesn't exist
espie [Mon, 6 Nov 2023 08:08:49 +0000 (08:08 +0000)]
instead of a max number of iterations, record terms that have been seen
in a hash instead.
Side effect: we no longer parse the same tc again and again, leading
to a smaller result with the same capabilities but no duplicates.
(instead of parsing 36 entries for xterm, we end up with 10 entries, as
the "tc=" stuff is actually a tree instead of a list: xterm references
several intermediate tc that all lead to the same basic list !)
espie [Mon, 6 Nov 2023 08:05:45 +0000 (08:05 +0000)]
use modern file open instead of GLOBS
espie [Mon, 6 Nov 2023 08:04:38 +0000 (08:04 +0000)]
get rid of useless eval
use \Q\E to quote stuff around tmp_term, instead of rolling our own
in termpat (get rid of termpat)
espie [Mon, 6 Nov 2023 07:59:22 +0000 (07:59 +0000)]
remove upstream files we don't need
espie [Mon, 6 Nov 2023 07:58:47 +0000 (07:58 +0000)]
reapply max=64 for now
espie [Mon, 6 Nov 2023 07:57:38 +0000 (07:57 +0000)]
update to Term::Cap 1.18, so that we can patch it further
as discussed with afresh1@
miod [Sun, 5 Nov 2023 16:33:50 +0000 (16:33 +0000)]
Provide simpler cpu_number, CPU_INFO_UNIT and CPU_IS_XXX macros in the
non-MULTIPROCESSOR case, as done on all other platforms.
ok deraadt@ kettenis@
miod [Sun, 5 Nov 2023 16:27:18 +0000 (16:27 +0000)]
Add cdXX.iso to MDEXT, for it to be included in SHA256; reported by `petcat30'
on bugs@.
espie [Sun, 5 Nov 2023 11:22:59 +0000 (11:22 +0000)]
current code no longer uses Lst_Replace not Lst_Requeue
so don't bother compiling them.
op [Sun, 5 Nov 2023 07:39:16 +0000 (07:39 +0000)]
fix bad path allocation in makeinfo
In one branch of the image handling, the wrong malloc() call was
copied-pasted making the allocation possibly smaller than needed.
Spotted while building the gcc11 port, libgccjit.info exercise that
branch and the write out of bounds was discovered by malloc().
Convert the malloc + strlcpy/cat dance to an asprintf, and while here
also fix the leak of the two paths.
ok tb@
claudio [Sat, 4 Nov 2023 11:17:51 +0000 (11:17 +0000)]
Use proper length in session_newmsg() for the RFC9072 case.
Right now OpenBGPD can not produce so many capabilities that this
case is reached but still better to fix it now.
OK tb@
otto [Sat, 4 Nov 2023 11:02:35 +0000 (11:02 +0000)]
KNF plus fixed a few signed vs unsigned compares (that we actually
not real problems)
martijn [Sat, 4 Nov 2023 09:42:17 +0000 (09:42 +0000)]
With the move of SNMPv2-SMI::snmpV2 into application_internal.c, it's no
longer possible to register objects underneath this region. This
effectively made a few snmp(1) regress tests unusable, which essentially
don't belong here. Simply remove these tests.
martijn [Sat, 4 Nov 2023 09:38:47 +0000 (09:38 +0000)]
Move SNMPv2-MIB::system from mib.c into the new application_internal.c.
This was the last consumer of mib.c, so send it to the great bitbucket
in the sky.
This move removes the sysORTable, but it was basically non-functioning
anyway and removal of previous entries from the table didn't caused
any fallout that reached my ears.
I hope to add proper sysORTable/agent-capabilities support, which
is compatible with AgentX's {Add,Remove}AgentCaps, soon.
OK tb@
martijn [Sat, 4 Nov 2023 09:30:28 +0000 (09:30 +0000)]
Move SNMPv2-SMI::snmpV2 from mib.c into the new application_internal.c
OK tb@
martijn [Sat, 4 Nov 2023 09:28:04 +0000 (09:28 +0000)]
Move SNMPv2-MIB::snmp from mib.c into the new application_internal.c
OK tb@
martijn [Sat, 4 Nov 2023 09:22:52 +0000 (09:22 +0000)]
Introduce application_internal.c. This backend is meant to replace
application_legacy.c, mps.c, and mib.c. This commit just introduces the
backend. The existing MIBs inside mib.c will be copied over in
subsequent commits.
OK tb@
anton [Fri, 3 Nov 2023 19:32:28 +0000 (19:32 +0000)]
Check if a signal already has been received before entering a blocking
read as opposed of only doing it afterwards. Issue uncovered by the mail
regress tests.
With claudio@ and ok millert@
cheloha [Fri, 3 Nov 2023 19:16:31 +0000 (19:16 +0000)]
timeout(1): align execvp(3) failure statuses with GNU timeout
Align our exit statuses with those of GNU timeout in the execvp(3)
failure case. Exit with 127 if the utility is not found. Exit with
126 if we cannot execute the utility for any other reason.
While here, the child should _exit(2) instead of calling exit(3) via
err(3).
Update the manpage accordingly.
With input from millert@ and deraadt@.
Link: https://marc.info/?l=openbsd-tech&m=169739592322978&w=2
ok millert@
tb [Fri, 3 Nov 2023 15:02:06 +0000 (15:02 +0000)]
typo: ignorning -> ignoring
From Laurie Tratt, ok florian
anton [Fri, 3 Nov 2023 14:05:30 +0000 (14:05 +0000)]
Make netcat tests less flaky by ensuring the server has terminated before
examining the outcome.
ok bluhm@
op [Fri, 3 Nov 2023 13:40:07 +0000 (13:40 +0000)]
initialize `format' earlier
It's easier to see that it's never used un-initialized.
ok tb@
op [Fri, 3 Nov 2023 13:38:28 +0000 (13:38 +0000)]
cast tv_usec to long for *printf
mostly for -portable; on darwin suseconds_t is an int.
ok tb@
espie [Fri, 3 Nov 2023 13:03:02 +0000 (13:03 +0000)]
Be explicit in saying what a shell pattern is.
I'm not a big fan of idiosyncratic lua patterns, especially in
that context, but pointing towards them and not mentionning glob(7)
is thoroughly confusing. Don't push lua patterns down people throats !
At least you know what the incredibly fuzzy "shell patterns" refers to
now !
okay millert@, jmc@
dv [Fri, 3 Nov 2023 11:16:43 +0000 (11:16 +0000)]
vmd(8): fix block size calculation for vioscsi devices.
During the multi-process rework, I changed the way disk sizes are
calculated, switching to storing bytes and not 512-byte blocks. The
size calculation from reading an iso did not receive the change in
scaling, resulting in the virtio scsi emulation to think iso's are
larger than reality.
This fixes booting the latest Ubuntu amd64 iso's as apparently
there's some program that tries to arbitrarily read the tail end
of the iso into memory.
ok mlarkin@
jsg [Fri, 3 Nov 2023 03:29:36 +0000 (03:29 +0000)]
drm/dp_mst: Fix NULL deref in get_mst_branch_device_by_guid_helper()
From Lukasz Majczak
6c668e2f338bde76279aeaf4d4c035c4196a479b in linux-6.1.y/6.1.61
3d887d512494d678b17c57b835c32f4e48d34f26 in mainline linux
jsg [Fri, 3 Nov 2023 03:26:51 +0000 (03:26 +0000)]
drm/amd: Disable ASPM for VI w/ all Intel systems
From Mario Limonciello
024251bab25daf97a3599c985dd04117b9c60d6b in linux-6.1.y/6.1.61
64ffd2f1d00c6235dabe9704bbb0d9ce3e28147f in mainline linux
jsg [Fri, 3 Nov 2023 03:24:23 +0000 (03:24 +0000)]
drm/i915/pmu: Check if pmu is closed before stopping event
From Umesh Nerlige Ramappa
eded5f5261b8ff14417f9441934a1865b379b7db in linux-6.1.y/6.1.61
4cbed7702eb775cca22fff6827a549092cb59f61 in mainline linux
sthen [Thu, 2 Nov 2023 20:47:31 +0000 (20:47 +0000)]
Use pfctl_fopen (which checks to ensure the given file isn't a directory)
for pfctl -f /path/to/pf.conf. Otherwise, if you accidentally use a dir
instead of a file (e.g. if you have an /etc/pf directory and don't notice
the name when tab-completing), you successfully install an empty ruleset.
ok sashan@
tb [Thu, 2 Nov 2023 20:25:48 +0000 (20:25 +0000)]
Fix a few bizarre line wraps in x509.h
paco [Thu, 2 Nov 2023 16:14:46 +0000 (16:14 +0000)]
mention codeberg and kde templates on bsd.port.mk(5) DIST_TUPLE
ok op@
+1 rsadowski
nicm [Thu, 2 Nov 2023 10:38:14 +0000 (10:38 +0000)]
next-prompt can have 1 argument.
tb [Wed, 1 Nov 2023 20:41:12 +0000 (20:41 +0000)]
Use X509_ALGOR_set0_by_nid() in X509_ALGOR_set_evp_md()
ok jsing
tb [Wed, 1 Nov 2023 20:37:42 +0000 (20:37 +0000)]
Add X509_ALGOR_set0_by_nid()
X509_ALGOR_set0() is annoyingly unergonomic since it takes an ASN1_OBJECT
rather than a nid. This means that almost all callers call OBJ_obj2nid()
and they often do this inline without error checking so that the resulting
X509_ALGOR object is corrupted and may lead to incorrect encodings.
Provide an internal alternative X509_ALGOR_set0_by_nid() that takes a nid
instead of an ASN1_OBJECT and performs proper error checking. This will be
used to convert callers of X509_ALGOR_set0() in the library.
ok jsing
tb [Wed, 1 Nov 2023 20:26:24 +0000 (20:26 +0000)]
Explain the weird order of doing things in X509_ALGOR_set0()
tb [Wed, 1 Nov 2023 20:25:27 +0000 (20:25 +0000)]
Rename ptype and pval to parameter_type and parameter_value
ok jsing
tb [Wed, 1 Nov 2023 20:22:24 +0000 (20:22 +0000)]
Unindent X509_ALGOR_set0_parameter()
ok jsing
tb [Wed, 1 Nov 2023 20:19:16 +0000 (20:19 +0000)]
Split X509_ALGOR_set0_parameter() out of X509_ALGOR_set0()
ok jsing
tb [Wed, 1 Nov 2023 20:14:51 +0000 (20:14 +0000)]
Split X509_ALGOR_set0_obj() out of X509_ALGOR_set0()
ok jsing
dtucker [Wed, 1 Nov 2023 02:08:38 +0000 (02:08 +0000)]
Specify ssh binary to use instead of relying on installed one. Fixes
test failures in -portable when running tests prior to installation.
djm [Wed, 1 Nov 2023 00:29:46 +0000 (00:29 +0000)]
add some tests of forced commands overriding Subsystem directives
millert [Tue, 31 Oct 2023 20:42:14 +0000 (20:42 +0000)]
cksum: better describe the -c (checklist) option
The -c option can handle entries in normal (BSD) and reverse (GNU)
format. Both hex and base64 entries are supported. OK jmc@
millert [Tue, 31 Oct 2023 19:37:17 +0000 (19:37 +0000)]
cksum -c: support checking base64 digests in reverse mode too.
In reverse mode, the entry does not include the digest algorithm
so the specified algorithm (or the default) is used when checking.
OK deraadt@
claudio [Tue, 31 Oct 2023 08:00:33 +0000 (08:00 +0000)]
Remove 3 expected failures those got fixed in the regress code.
claudio [Tue, 31 Oct 2023 07:59:52 +0000 (07:59 +0000)]
unlink("/") just needs to error. Checking for a specific errno makes
little sense here since there are multiple possible errnos that could
be returned. On OpenBSD this returns EISDIR and not EBUSY.
OK mbuhl@ millert@
claudio [Tue, 31 Oct 2023 07:57:59 +0000 (07:57 +0000)]
When creating a file in a directory the file gid is inherited from
the directory and so checking against getgid() makes no sense.
OK mbuhl@ millert@
claudio [Tue, 31 Oct 2023 07:56:44 +0000 (07:56 +0000)]
Ignore closefrom() failure. This fails normally since fd 4 and up are all
closed.
OK mbuhl@ millert@
jsg [Tue, 31 Oct 2023 05:46:36 +0000 (05:46 +0000)]
remove unneeded includes
ok deraadt@
dtucker [Tue, 31 Oct 2023 04:15:40 +0000 (04:15 +0000)]
Don't try to use sudo inside sshd log wrapper. We still need to
check if we're using sudo since we don't want to chown unecessarily,
as on some platforms this causes an error which pollutes stderr.
We also don't want to unnecessarily invoke sudo, since it's running
in the context of the proxycommand, on *other* platforms it may not
be able to authenticate, and if we're using SUDO then it should
already be privileged.
dtucker [Tue, 31 Oct 2023 02:58:45 +0000 (02:58 +0000)]
Only try to chmod logfile if we have sudo. If we don't have sudo
then we won't need to chmod.
millert [Tue, 31 Oct 2023 01:08:51 +0000 (01:08 +0000)]
Update awk to Oct 30, 2023 version.
This is really just a version number bump as we already have the
fixes committed.
djm [Mon, 30 Oct 2023 23:00:25 +0000 (23:00 +0000)]
move PKCS#11 setup code to test-exec.sh so it can be reused elsewhere
millert [Mon, 30 Oct 2023 22:12:02 +0000 (22:12 +0000)]
This is the OpenBSD version of Awk.
millert [Mon, 30 Oct 2023 17:52:54 +0000 (17:52 +0000)]
Minor cosmetic changes to make our awk match my github branch.
djm [Mon, 30 Oct 2023 17:32:00 +0000 (17:32 +0000)]
tidy and refactor PKCS#11 setup code
Replace the use of a perl script to delete the controlling TTY with a
SSH_ASKPASS script to directly load the PIN.
Move PKCS#11 setup code to functions in anticipation of it being used
elsewhere in additional tests.
Reduce stdout spam
tb [Mon, 30 Oct 2023 17:15:21 +0000 (17:15 +0000)]
Add support for OpenSSL 3.1 interop tests
Until OpenSSL 3.1 has replaced OpenSSL 3.0 on most architectures, run
both tests. Installed packages of OpenSSL 3.0 will update automatically
to 3.1, so regress runners should not need to do anything.
millert [Mon, 30 Oct 2023 16:14:51 +0000 (16:14 +0000)]
Include strings.h for the strncasecmp() prototype.
From upstream.
nicm [Mon, 30 Oct 2023 16:05:30 +0000 (16:05 +0000)]
Do not allow combined UTF-8 characters that are too long, GitHub issue
3729.
bluhm [Mon, 30 Oct 2023 13:27:53 +0000 (13:27 +0000)]
Do not truncate MSG_EOR in recvmsg().
The soreceive() code depends on the fact that MSG_EOR is set on the
last mbuf of the chain. In sbappendcontrol() move MSG_EOR to the
end like sbcompress() does it. This fixes MSG_EOR handling for
SOCK_SEQPACKET sockets with control message.
bug reported by Eric Wong
analysed, tested and OK claudio@
mvs [Mon, 30 Oct 2023 12:50:59 +0000 (12:50 +0000)]
Use KERNEL_ASSERT_UNLOCKED() instead of KASSERT(!_kernel_lock_held()).
The first one defined as the second, so no functional changed.
ok claudio
claudio [Mon, 30 Oct 2023 07:13:10 +0000 (07:13 +0000)]
Use ERESTART for any single_thread_set() error in sys_execve().
If single thread is already held by another thread just unwind to userret()
wait there and retry the system call later (if at all).
OK mpi@
claudio [Mon, 30 Oct 2023 07:04:36 +0000 (07:04 +0000)]
Adjust KERNEL_ASSERT_UNLOCKED() to not assert during a panic.
KERNEL_ASSERT_UNLOCKED calls _kernel_lock_held() which returns true
if panicstr || db_active which triggers this assert. Workaround this by
checking them before.
This will alter the following Syzkaller reports:
Reported-by: syzbot+169110a0815838ab5940@syzkaller.appspotmail.com
Reported-by: syzbot+3c2eced405b9de6f79c2@syzkaller.appspotmail.com
OK mpi@
jmc [Mon, 30 Oct 2023 06:11:04 +0000 (06:11 +0000)]
seperate -> separate;
deraadt [Mon, 30 Oct 2023 01:34:13 +0000 (01:34 +0000)]
make special note, right at the start, that DNS is handled seperately
from regular network. All the crummy pledge clones don't do that.
ok millert
jsg [Mon, 30 Oct 2023 01:02:30 +0000 (01:02 +0000)]
drm/amd/pm: add unique_id for gc 11.0.3
From Kenneth Feng
2a2df4f74f3746b614af686a9cbaa390d3828b09 in linux-6.1.y/6.1.60
4953856f280b2b606089a72a93a1e9212a3adaca in mainline linux
jsg [Mon, 30 Oct 2023 00:59:56 +0000 (00:59 +0000)]
drm: panel-orientation-quirks: Add quirk for One Mix 2S
From Kai Uwe Broulik
c3657e5fa7b6310a485f59ca965d12a618e0c8df in linux-6.1.y/6.1.60
cbb7eb2dbd9472816e42a1b0fdb51af49abbf812 in mainline linux
jsg [Mon, 30 Oct 2023 00:56:47 +0000 (00:56 +0000)]
drm/i915: Retry gtt fault when out of fence registers
From Ville Syrjala
add2eeda1d6e3ead1d61db70e37615e5ef6c5e75 in linux-6.1.y/6.1.60
e339c6d628fe66c9b64bf31040a55770952aec57 in mainline linux
cheloha [Sun, 29 Oct 2023 23:32:52 +0000 (23:32 +0000)]
libc: Makefile.inc: remove "sparc" from quad support list; ok deraadt@
millert [Sun, 29 Oct 2023 23:31:15 +0000 (23:31 +0000)]
Make gnu99 the default for gcc 3.3.6 and 4.2.1, not gnu89.
While these versions of gcc don't have full C99 support, it is
better than defaulting to C89 when building modern software.
OK deraadt@
millert [Sun, 29 Oct 2023 14:26:13 +0000 (14:26 +0000)]
Enable ISO C11 APIs when building libc, even with an older compiler.
Otherwise, the prototypes for timespec_get() and aligned_alloc()
are not visible. OK guenther@
millert [Sun, 29 Oct 2023 14:23:04 +0000 (14:23 +0000)]
Use clock_gettime(), not timespec_get() like other parts of ifconfig.
This makes is possible to build the base system with a C99 compiler.
OK deraadt@ mvs@
tb [Sun, 29 Oct 2023 13:22:37 +0000 (13:22 +0000)]
Fix an error exit in X509v3_addr_validate_path()
If the topmost cert is invalid, this should result in a validation failure.
Do the same dance as elsewhere permitting the verify callback to intercept
the error but ensuring that we throw an error.
ok jsing
kn [Sun, 29 Oct 2023 11:27:11 +0000 (11:27 +0000)]
Unmention/don't explain SSL, drop 9y old "ssl" keyword/deprecation warning
Switch "ssl" to "tls" in relayd.conf(5) if you haven't done so in the last
ten years, "ssl" is now an error.
Say "TLS" not "SSL/TLS" and drop the primer in the TLS RELAYS section.
OK benno
martijn [Sun, 29 Oct 2023 11:26:12 +0000 (11:26 +0000)]
Add a few tests to make sure that getnext requests on/under a registered
instance with a registered region above cause a recursion in the tree.
martijn [Sun, 29 Oct 2023 11:20:06 +0000 (11:20 +0000)]
When doing a get{next,bulk} below an instance we must move the OID to
the next sibling. Not simply copying over the value of the next to use
region, since that might be the parent and we would walk backwards in
the tree.
OK tb@
martijn [Sun, 29 Oct 2023 11:15:09 +0000 (11:15 +0000)]
Add a regress test to make sure that agentx responses equal to
searchrange.end are rejected.
martijn [Sun, 29 Oct 2023 11:12:57 +0000 (11:12 +0000)]
searchrange.end is non-inclusive. Adjust the tests for that.
OK tb@
martijn [Sun, 29 Oct 2023 11:10:07 +0000 (11:10 +0000)]
Add a test to the end of agentx_varbind_finalize(), after completing the
full OID, that checks if we're >= searchrange.end. If so, just make it
an endOfMIBView.
OK tb@
dtucker [Sun, 29 Oct 2023 06:22:07 +0000 (06:22 +0000)]
Only try to chown logfiles that exist to prevent spurious errors.
millert [Sat, 28 Oct 2023 22:38:22 +0000 (22:38 +0000)]
substr: fix buffer overflow with utf-8 strings
We need to use u8_strlen(), not strlen(), to compute the length.
Otherwise, there may be an out of bounds write when writing the NUL
terminator to set the length of the substring.
https://github.com/onetrueawk/awk/pull/205
millert [Sat, 28 Oct 2023 00:25:18 +0000 (00:25 +0000)]
Back out change to make the default -std=gnu99 for now.
There has been some unexpected fallout. Requested by deraadt@.
jan [Fri, 27 Oct 2023 20:56:47 +0000 (20:56 +0000)]
Forward TCP LRO disabling to parent devices.
Also disable TCP LRO on bridged vlan(4) and default for bpe(4), nvgre(4) and
vxlan(4).
ok bluhm@
mpi [Fri, 27 Oct 2023 19:18:53 +0000 (19:18 +0000)]
Make out-of-swap checks more robust.
Consider that the swap space is full when 99% of it is filled with pages that
are no longer present in memory. This prevents deadlocks when out-of-swap if
some swap ranges had I/O errors and have been marked as 'bad', or if some
pages are unreachable by the pagedaemon and still holding some slots.
Also introduce uvm_swapisfilled() to check if there are some free slots in
the swap. Note that we consider the swap space completly filled if it is not
possible to write a full cluster. This prevents deadlocks if a few slots are
never allocated.
ok miod@
mpi [Fri, 27 Oct 2023 19:13:51 +0000 (19:13 +0000)]
Do not decrement the swap counter if the anon is associated to a "bad" slot.
When such anon is freed its content is obviously not living in swap.
ok miod@
claudio [Fri, 27 Oct 2023 14:22:43 +0000 (14:22 +0000)]
Simplify the handling of optparamlen and RFC9072 support.
Use the ibuf_size() of the optional params (opb) when writing the various
headers. Only use optparamlen to write the Opt Parm Len field of the OPEN
message. Since there the size could be forced to 255 in case of a RFC9072
exended header. Also use size_t for optparamlen since the new ibuf API
no longer depends on using the right sized type.
OK tb@
martijn [Fri, 27 Oct 2023 10:32:11 +0000 (10:32 +0000)]
If an invalid varbindlist was returned by a backend we would call
appl_request_upstream_resolve() twice where the first call would already
return a reply and free the upstream request, leading to a use after
free.
Make appl_request_downstream_free() call appl_request_upstream_resolve()
unconditionally and remove the call from appl_response().
OK tb@
martijn [Fri, 27 Oct 2023 10:26:20 +0000 (10:26 +0000)]
Add a test that would cause a use after free previously by sending a
protocol error-ed message and immediately followed up by a connection
close.
OK tb@
martijn [Fri, 27 Oct 2023 10:23:58 +0000 (10:23 +0000)]
Use event_add(), instead of calling appl_agentx_send() directly. If an
error occurs the connection will be freed and if the caller uses the
connection afterwards it will lead to a use after free.
OK tb@
claudio [Fri, 27 Oct 2023 09:40:27 +0000 (09:40 +0000)]
The holdtime is added to the ibuf with ibuf_add_n16() so no need to
htons() the value before.
OK tb@
anton [Fri, 27 Oct 2023 07:33:06 +0000 (07:33 +0000)]
Include wait(2) status in error message, in the hopes of providing clues
on why this occasionally fails.
naddy [Thu, 26 Oct 2023 21:30:52 +0000 (21:30 +0000)]
sync
naddy [Thu, 26 Oct 2023 19:28:30 +0000 (19:28 +0000)]
do not create /usr/local/share/nls and subdirectories by default
The share/nls/<locale> paths are unused.
ok miod@ deraadt@