openbsd
19 months agoFix a number of out of bound reads in DNS response parsing.
millert [Wed, 15 Mar 2023 22:12:00 +0000 (22:12 +0000)]
Fix a number of out of bound reads in DNS response parsing.
Originally from djm@.  OK deraadt@ florian@ bluhm@

19 months agostate that binary updates are not available for all platforms/architectures.
benno [Wed, 15 Mar 2023 21:42:23 +0000 (21:42 +0000)]
state that binary updates are not available for all platforms/architectures.
triggered by a bug report by 'kodcode', thanks.
wording adjusted by jmc@, ok phessler@

19 months agoopenssh-9.3
djm [Wed, 15 Mar 2023 21:19:57 +0000 (21:19 +0000)]
openssh-9.3

19 months agoslightly earlier, and tweak some dates
deraadt [Wed, 15 Mar 2023 19:54:32 +0000 (19:54 +0000)]
slightly earlier, and tweak some dates

19 months agoDo not leak screen in popups, GitHub issue 3492.
nicm [Wed, 15 Mar 2023 19:23:22 +0000 (19:23 +0000)]
Do not leak screen in popups, GitHub issue 3492.

19 months agoFix the length check when computing a fake challenge for users not
millert [Wed, 15 Mar 2023 17:01:35 +0000 (17:01 +0000)]
Fix the length check when computing a fake challenge for users not
in the S/Key database.  If the system hostname is longer than 126
characters this could result in NUL bytes being written past the
end of a stack buffer.  There is no impact on systems with a hostname
126 characters or less.  Found by Qualys.  OK deraadt@

19 months agoAdd missing nameref resolving for the _DEP() package elements that indicate
kettenis [Wed, 15 Mar 2023 13:01:40 +0000 (13:01 +0000)]
Add missing nameref resolving for the _DEP() package elements that indicate
dependencies between ACPI devices.

ok dlg@, deraadt@

19 months agoShow MFTs as part of the signature path
job [Wed, 15 Mar 2023 11:09:34 +0000 (11:09 +0000)]
Show MFTs as part of the signature path

no objection tb@

19 months agoFor now pass AF_UNSPEC to printsource() to ensure that both inet and inet6
claudio [Wed, 15 Mar 2023 08:43:51 +0000 (08:43 +0000)]
For now pass AF_UNSPEC to printsource() to ensure that both inet and inet6
addrs are fetched from the kernel. This needs more cleanup.

19 months agoMake route sourceaddr always print the used addresses for inet and inet6.
claudio [Wed, 15 Mar 2023 08:42:14 +0000 (08:42 +0000)]
Make route sourceaddr always print the used addresses for inet and inet6.
Print 'default' if no sourceaddr is set and the default algorithm is used.
With and OK kn@ deraadt@

19 months agodrm/amdgpu/vcn: Disable indirect SRAM on Vangogh broken BIOSes
jsg [Wed, 15 Mar 2023 08:36:33 +0000 (08:36 +0000)]
drm/amdgpu/vcn: Disable indirect SRAM on Vangogh broken BIOSes

From Guilherme G. Piccoli in amd-staging-drm-next
https://gitlab.freedesktop.org/drm/amd/-/issues/2385

This fixes amdgpu failing to init on Steam Deck after the drm 6.1 update:

[drm] failed to load ucode VCN0_RAM(0x3A) [drm] psp gfx command LOAD_IP_FW(0x6)
failed and response status is (0xFFFF0000)
[drm] *ERROR* ring vcn_dec_0 test failed (-60)
[drm] *ERROR* hw_init of IP block <vcn_v3_0> failed -60
drm:pid0:amdgpu_device_init *ERROR* amdgpu_device_ip_init failed
drm:pid0:amdgpu_attachhook *ERROR* Fatal error during GPU init

reported and tested by bentley@ on:
bios0: vendor Valve version "F7A0113" date 11/04/2022
bios0: Valve Jupiter

19 months agohandle dmi_get_system_info(DMI_BIOS_VERSION)
jsg [Wed, 15 Mar 2023 08:24:56 +0000 (08:24 +0000)]
handle dmi_get_system_info(DMI_BIOS_VERSION)

19 months agostore bios version, needed for amdgpu Steam Deck quirk
jsg [Wed, 15 Mar 2023 08:20:52 +0000 (08:20 +0000)]
store bios version, needed for amdgpu Steam Deck quirk

19 months agoFix command prompt not to always append argument but only if there has
nicm [Wed, 15 Mar 2023 08:15:39 +0000 (08:15 +0000)]
Fix command prompt not to always append argument but only if there has
actually been expansion. GitHub issue 3493.

19 months agoReturn the signature length after successful signing operation
tb [Wed, 15 Mar 2023 06:34:07 +0000 (06:34 +0000)]
Return the signature length after successful signing operation

This is required behavior of the EVP_DigestSign() API, but seemingly
almost nothing uses this. Well, turns out ldns does.

Reported by Stephane. Helpful comments by sthen.

ok jsing

19 months agoAdd comments that explain why things are done in this strange order.
tb [Wed, 15 Mar 2023 06:30:21 +0000 (06:30 +0000)]
Add comments that explain why things are done in this strange order.

There's some method to this madness.

ok jsing

19 months agoPush calloc() of ndef_aux down as far as possible and
tb [Wed, 15 Mar 2023 06:28:55 +0000 (06:28 +0000)]
Push calloc() of ndef_aux down as far as possible and
pull the setting of the ex_arg up, so we can do error
checking.

ok jsing

19 months agoError check BIO_asn1_set_{prefix,suffix}() calls
tb [Wed, 15 Mar 2023 06:22:42 +0000 (06:22 +0000)]
Error check BIO_asn1_set_{prefix,suffix}() calls

ok jsing

19 months agoStreaming BIOs assume they can write to NULL BIOs
tb [Wed, 15 Mar 2023 06:14:02 +0000 (06:14 +0000)]
Streaming BIOs assume they can write to NULL BIOs

At least SMIME_text() relies on this. Pushing an error on the stack trips
PKCS7 regress in py-cryptography, so indicate nothing was written instead
of throwing an error.

Reported by Alex Gaynor a while back

ok jsing

19 months agoEnsure negative input to BN_mod_exp_mont_consttime() is correctly reduced.
jsing [Wed, 15 Mar 2023 04:30:20 +0000 (04:30 +0000)]
Ensure negative input to BN_mod_exp_mont_consttime() is correctly reduced.

A negative input to BN_mod_exp_mont_consttime() is not correctly reduced,
remaining negative (when it should be in the range [0, m)). Fix this by
unconditionally calling BN_nnmod() on the input.

Fixes ossfuzz #55997.

ok tb@

19 months agoInclude tests with negative values in BN_mod_exp* regress.
jsing [Wed, 15 Mar 2023 04:26:23 +0000 (04:26 +0000)]
Include tests with negative values in BN_mod_exp* regress.

This currently fails.

19 months agobn_mod_exp_zero: rename result into got
tb [Wed, 15 Mar 2023 00:41:04 +0000 (00:41 +0000)]
bn_mod_exp_zero: rename result into got

19 months agoPrep for rpki-client 8.3
claudio [Tue, 14 Mar 2023 13:31:16 +0000 (13:31 +0000)]
Prep for rpki-client 8.3

19 months agoWEP in bfwm(4) never worked, sync with reality
kn [Tue, 14 Mar 2023 11:25:22 +0000 (11:25 +0000)]
WEP in bfwm(4) never worked, sync with reality

"ifconfig bwfm0 [-]nwkey ..." does not work and thus the installer no longer
offers WEP, so unmention support while leaving the recommendation against it.

19 months agoFree KRL itself in addition to its contents. From Coverity CID 291841,
dtucker [Tue, 14 Mar 2023 07:28:47 +0000 (07:28 +0000)]
Free KRL itself in addition to its contents. From Coverity CID 291841,
ok djm@

19 months agoCheck pointer for NULL before attempting to deref. None of the existing
dtucker [Tue, 14 Mar 2023 07:26:25 +0000 (07:26 +0000)]
Check pointer for NULL before attempting to deref.  None of the existing
callers seem to do that, but it's worth checking.  From Coverity CID
291834, ok djm@

19 months agorpki-client: disallow AIA in self-signed certs
tb [Tue, 14 Mar 2023 07:09:11 +0000 (07:09 +0000)]
rpki-client: disallow AIA in self-signed certs

Per RFC 6487, 4.8.7, self-signed certificates must not have an Authority
Info Access extension. In normal operation this is ensured by ta_parse()
and cert_parse(), respectively. In filemode, only partial checks are
performed, so this is not guaranteed.

Issue flagged by and ok job

19 months agosync the boot options list, and some macro fixes;
jmc [Tue, 14 Mar 2023 06:55:31 +0000 (06:55 +0000)]
sync the boot options list, and some macro fixes;
ok miod

19 months agosync
deraadt [Tue, 14 Mar 2023 04:51:34 +0000 (04:51 +0000)]
sync

19 months agoTo avoid misunderstanding, keep variables for tcp keepalive in
yasuoka [Tue, 14 Mar 2023 00:24:05 +0000 (00:24 +0000)]
To avoid misunderstanding, keep variables for tcp keepalive in
milliseconds, which is the same unit of tcp_now().  However, keep the
unit of sysctl variables in seconds and convert their unit in
tcp_sysctl().  Additionally revert TCPTV_SRTTDFLT back to 3 seconds,
which was mistakenly changed to 1.5 seconds by tcp_timer.h 1.19.

ok claudio

19 months agoclockintr: add missing newline to ddb(4) printout
cheloha [Tue, 14 Mar 2023 00:11:58 +0000 (00:11 +0000)]
clockintr: add missing newline to ddb(4) printout

Let's also print the state (running, pending, or established) of each
clockintr.

19 months agoDon't lookup the signature path's expiry for trust anchor certs
job [Mon, 13 Mar 2023 21:01:06 +0000 (21:01 +0000)]
Don't lookup the signature path's expiry for trust anchor certs

19 months agoIn filemode, also print details for .cer certificates
job [Mon, 13 Mar 2023 20:55:14 +0000 (20:55 +0000)]
In filemode, also print details for .cer certificates

19 months agoPoint people to boot(8) rather than duplicating information.
miod [Mon, 13 Mar 2023 20:32:28 +0000 (20:32 +0000)]
Point people to boot(8) rather than duplicating information.

19 months agoSwitch alpha to new boot blocks.
miod [Mon, 13 Mar 2023 20:20:59 +0000 (20:20 +0000)]
Switch alpha to new boot blocks.

19 months agoAdd manual page.
miod [Mon, 13 Mar 2023 20:20:13 +0000 (20:20 +0000)]
Add manual page.

19 months ago"ask name" option letter is 'n' instead of 'a' on pmax and alpha for historical
miod [Mon, 13 Mar 2023 20:19:22 +0000 (20:19 +0000)]
"ask name" option letter is 'n' instead of 'a' on pmax and alpha for historical
(and firmware) reasons.

19 months agoCheck that the CMS signing-time is not after the MFT's nextUpdate
job [Mon, 13 Mar 2023 19:54:36 +0000 (19:54 +0000)]
Check that the CMS signing-time is not after the MFT's nextUpdate

Issuing manifests whose validity window is entirely in the past, as
observed from the purported CMS signing-time would be a form of backdating.

OK tb@

19 months agoIn filemode, display the moment the signature path will expire
job [Mon, 13 Mar 2023 19:51:49 +0000 (19:51 +0000)]
In filemode, display the moment the signature path will expire

Previously this was only shown for ROA+ASPA. Now also show for
GBR, Geofeed, Certs, RSC, and TAK.

OK tb@

19 months agoCheck that the CMS signing-time isn't after the X.509 notAfter
job [Mon, 13 Mar 2023 19:46:55 +0000 (19:46 +0000)]
Check that the CMS signing-time isn't after the X.509 notAfter

The CMS signing-time is the purported 'now' from the perspective of the
issuer. It doesn't make sense for an issuer to sign objects that have a
validity window that falls entirely in the past (from the perspective of
the signer). Although CMS signing-time is not a trusted timestamp, it
should never be after X.509 notAfter.

OK tb@

19 months agotypo fix; ok claudio
jmc [Mon, 13 Mar 2023 18:31:54 +0000 (18:31 +0000)]
typo fix; ok claudio

19 months agovmd(8): backout ns8250 changes.
dv [Mon, 13 Mar 2023 18:09:41 +0000 (18:09 +0000)]
vmd(8): backout ns8250 changes.

This backs out commit h1IJwTVsYWfnRKWy which intended to fix interrupt
issues on slower host hardware. Unfortunately, this causes a race
condition on much faster host hardware. It still requires investigation.

With an "I told you so" about touching ns8250 from mlarkin@

19 months agoIn filemode for ROAs/ASPAs display when the Signature path will expire
job [Mon, 13 Mar 2023 18:02:58 +0000 (18:02 +0000)]
In filemode for ROAs/ASPAs display when the Signature path will expire

Also rename 'certification path' to 'signature path' for alignment

OK tb@

19 months agoExtend bgplgd to handle leaked and invalid options to show rib.
claudio [Mon, 13 Mar 2023 17:31:28 +0000 (17:31 +0000)]
Extend bgplgd to handle leaked and invalid options to show rib.
OK tb@

19 months agoExtend bgpctl show rib with two new options:
claudio [Mon, 13 Mar 2023 16:59:22 +0000 (16:59 +0000)]
Extend bgpctl show rib with two new options:
 - invalid: show only routes that are not valid
 - leaked: show only routes where a route leak was detected
OK tb@

19 months agoAdd F_CTL_LEAKED and F_CTL_INELIGIBLE flags for bgpctl to show leaked
claudio [Mon, 13 Mar 2023 16:52:41 +0000 (16:52 +0000)]
Add F_CTL_LEAKED and F_CTL_INELIGIBLE flags for bgpctl to show leaked
and ineligible paths.
While there rename F_PREF_OTC_LOOP to F_PREF_OTC_LEAK since this indicates
that a route leak was detected.
OK tb@

19 months agoTry to explain that the neighor role is required to enable both
claudio [Mon, 13 Mar 2023 16:39:29 +0000 (16:39 +0000)]
Try to explain that the neighor role is required to enable both
ASPA validation and RFC9234 OTC leak detection and role capability.
The role on iBGP peers is forced to 'none' since it makes no sense there.
After input from benno@

19 months agoMerge calling print_certification_path() into adjacent else branch
job [Mon, 13 Mar 2023 15:50:40 +0000 (15:50 +0000)]
Merge calling print_certification_path() into adjacent else branch

OK tb@

19 months agoFix potential underflow when printing STP data.
mbuhl [Mon, 13 Mar 2023 14:58:01 +0000 (14:58 +0000)]
Fix potential underflow when printing STP data.
Reported by Peter J. Philipp
ok claudio

19 months agoRefactor whats printed when in filemode
job [Mon, 13 Mar 2023 13:42:01 +0000 (13:42 +0000)]
Refactor whats printed when in filemode

OK tb@

19 months agoFix usbpcap handler to properly detect overflows using TCHECK macros.
claudio [Mon, 13 Mar 2023 13:36:56 +0000 (13:36 +0000)]
Fix usbpcap handler to properly detect overflows using TCHECK macros.
OK mpi@ mbuhl@

19 months agoAdd "machine poweroff" command on luna88k bootloader.
aoyama [Mon, 13 Mar 2023 11:59:39 +0000 (11:59 +0000)]
Add "machine poweroff" command on luna88k bootloader.

ok miod@

19 months agoRevert accidental change from explicit to implicit tagging in r1.86.
tb [Mon, 13 Mar 2023 10:39:03 +0000 (10:39 +0000)]
Revert accidental change from explicit to implicit tagging in r1.86.

ok job

19 months agoRename some data fields to closer match their provenance
job [Mon, 13 Mar 2023 09:24:37 +0000 (09:24 +0000)]
Rename some data fields to closer match their provenance

OK tb@

19 months agoStop confusing out and asn_bio in BIO_new_NDEF()
tb [Mon, 13 Mar 2023 07:31:09 +0000 (07:31 +0000)]
Stop confusing out and asn_bio in BIO_new_NDEF()

BIO_new_NDEF() sets up an ASN.1 BIO to the output chain and then adds even
more BIOs. Since BIO_push(bio, new_tail) returns bio on success, after the
if ((out = BIO_push(asn_bio, out)) != NULL) the 'out' BIO and the 'asn_bio'
are the same. The code then goes on and uses one or the other. This is very
confusing. Simply stop using out once it's appended to asn_bio.

ok jsing

19 months agoregen
kettenis [Mon, 13 Mar 2023 07:30:44 +0000 (07:30 +0000)]
regen

19 months agoAdd Kingston NV2 NVMe drive.
kettenis [Mon, 13 Mar 2023 07:30:11 +0000 (07:30 +0000)]
Add Kingston NV2 NVMe drive.

19 months agosync
deraadt [Mon, 13 Mar 2023 04:15:44 +0000 (04:15 +0000)]
sync

19 months agosh, landisk: set HZ=64 again
cheloha [Sun, 12 Mar 2023 22:18:58 +0000 (22:18 +0000)]
sh, landisk: set HZ=64 again

The addition of HZ to sys/kernel.h in v1.26 overrides the default
definition of HZ in sh/clock.c, changing landisk from HZ=64 to HZ=100.

Explicitly set HZ=64 in the GENERIC and RAMDISK config(8) files to can
change it from 100 back to 64.

Not sure if this is the best thing, but it does fix the problem.

Problem confirmed by, and fix tested by, miod@.

ok miod@

19 months agocorrect macro request (.nm -> .Nm);
jmc [Sun, 12 Mar 2023 20:14:41 +0000 (20:14 +0000)]
correct macro request (.nm -> .Nm);

19 months agorkcomphy(4)
kettenis [Sun, 12 Mar 2023 19:02:34 +0000 (19:02 +0000)]
rkcomphy(4)

19 months agopk7_cb() and cms_cb()
tb [Sun, 12 Mar 2023 17:29:02 +0000 (17:29 +0000)]
pk7_cb() and cms_cb()

Add and fix FALLTHROUGH statement. I was confused for way too long since
I hadn't noticed that this case fell through to the next. Also add and
move some empty lines in the cms_cb() to make this resemble KNF more.

19 months agoAdd rkcomphy(3), a driver for the "naneng" combo PHY found on the RK356x
kettenis [Sun, 12 Mar 2023 14:29:50 +0000 (14:29 +0000)]
Add rkcomphy(3), a driver for the "naneng" combo PHY found on the RK356x
(and RK3588).  This is a PIPE PHY with support for PCIe, SATA, USB3, SGMII
and QSGMII.  For now only PCIe, SATA and USB3 support are implemented.
SATA support has not been tested.

Also add the refernce clocks needed by the PHYs to rkclock(4).

ok mlarkin@

19 months agozap a tab
tb [Sun, 12 Mar 2023 13:31:39 +0000 (13:31 +0000)]
zap a tab

19 months agoAlphabetically sort various bits and pieces - mechanical change
job [Sun, 12 Mar 2023 12:11:45 +0000 (12:11 +0000)]
Alphabetically sort various bits and pieces - mechanical change

Might make later refactors easier to follow

OK tb@

19 months agoAdd crl.c in various places to gain access to the new helper
job [Sun, 12 Mar 2023 11:55:34 +0000 (11:55 +0000)]
Add crl.c in various places to gain access to the new helper

19 months agoRefactor expiration calculation
job [Sun, 12 Mar 2023 11:54:56 +0000 (11:54 +0000)]
Refactor expiration calculation

Unify common code paths which find the exact expiry moment into a new
helper function. Additionally, the new helper offers more accuracy by
checking more applicable CRLs whether their 'nextupdate' is 'sooner'.

tb@ noted: The helper adds a multiplier of log(#crls), but that's
certainly acceptable as it is still very cheap.

OK tb@

19 months agoAvoid an 1 byte out-of-bounds read in ASN1_PRINTABLE_type()
tb [Sun, 12 Mar 2023 11:49:02 +0000 (11:49 +0000)]
Avoid an 1 byte out-of-bounds read in ASN1_PRINTABLE_type()

In case the input is not NUL terminated, the reversed check for length
and terminating NUL results in a one-byte overread. The documentation
says that the input should be a string, but in ASN.1 land you never
know...

Reported by Guido Vranken a while back

ok beck

19 months agoInitialize the local signtime variable in the individual *_parse()
tb [Sun, 12 Mar 2023 11:46:35 +0000 (11:46 +0000)]
Initialize the local signtime variable in the individual *_parse()
functions

ok job

19 months agoInvalidate the signtime output parameter in cms_parse_validate_internal()
tb [Sun, 12 Mar 2023 11:45:52 +0000 (11:45 +0000)]
Invalidate the signtime output parameter in cms_parse_validate_internal()
and in cms_get_signtime().

ok job

19 months agoEnable mvtemp(4)
jmatthew [Sun, 12 Mar 2023 10:50:06 +0000 (10:50 +0000)]
Enable mvtemp(4)

19 months agoUpdate supported hardware list and fix spelling of Marvell while I'm here
jmatthew [Sun, 12 Mar 2023 10:49:05 +0000 (10:49 +0000)]
Update supported hardware list and fix spelling of Marvell while I'm here

19 months agoMatch on the Armada 380 temperature sensor, which works the same as the
jmatthew [Sun, 12 Mar 2023 10:46:27 +0000 (10:46 +0000)]
Match on the Armada 380 temperature sensor, which works the same as the
CP110 sensor.

ok kettenis@

19 months agoPut upper bound on number of entries in SSH2_MSG_EXT_INFO request.
dtucker [Sun, 12 Mar 2023 10:40:39 +0000 (10:40 +0000)]
Put upper bound on number of entries in SSH2_MSG_EXT_INFO request.
This is already constrained by the maximum SSH packet size but this
makes it explicit.  Prompted by Coverity CID 291868, ok djm@ markus@

19 months agocalloc can return NULL but xcalloc cannot. From Coverity CID 291881,
dtucker [Sun, 12 Mar 2023 09:41:18 +0000 (09:41 +0000)]
calloc can return NULL but xcalloc cannot.  From Coverity CID 291881,
ok djm@

19 months agoRemove a few extra spaces
tb [Sat, 11 Mar 2023 21:14:26 +0000 (21:14 +0000)]
Remove a few extra spaces

19 months agoNew alpha secondary boot blocks using the MI code base. Not linked to the
miod [Sat, 11 Mar 2023 20:56:01 +0000 (20:56 +0000)]
New alpha secondary boot blocks using the MI code base. Not linked to the
build yet.

19 months agoRemove long dead "tool".
miod [Sat, 11 Mar 2023 20:52:02 +0000 (20:52 +0000)]
Remove long dead "tool".

19 months agoAllow strip and objcopy to be overrridden, for easier cross-buildability. NFC
miod [Sat, 11 Mar 2023 20:49:33 +0000 (20:49 +0000)]
Allow strip and objcopy to be overrridden, for easier cross-buildability. NFC

19 months agoSwitch an early return into goto err
tb [Sat, 11 Mar 2023 16:29:48 +0000 (16:29 +0000)]
Switch an early return into goto err

19 months agoTiny cleanup for readability
tb [Sat, 11 Mar 2023 16:02:06 +0000 (16:02 +0000)]
Tiny cleanup for readability

Turn a malloc() into calloc() and check two function calls directly
forever instead of a combined check afterward.

19 months agoUse "if (ptr == NULL)" instead of "if (!ptr)"
tb [Sat, 11 Mar 2023 15:56:03 +0000 (15:56 +0000)]
Use "if (ptr == NULL)" instead of "if (!ptr)"

Requested by jsing

19 months agoFix double free after BIO_new_NDEF()
tb [Sat, 11 Mar 2023 15:50:13 +0000 (15:50 +0000)]
Fix double free after BIO_new_NDEF()

Once the asn_bio is prepended to the out chain, and before the asn1_cb()
has done its thing, asn_bio needs to be popped off again on error. Failing
to do this can cause write after frees or double frees when the out BIO
is used after the function returned.

Based on a very complicated diff by Matt Caswell and Viktor Dukhovni.
This was part of the fixes in OpenSSL 1.1.1t.

ok jsing

19 months agoFix an off-by-one in dsa_check_key()
tb [Sat, 11 Mar 2023 15:29:03 +0000 (15:29 +0000)]
Fix an off-by-one in dsa_check_key()

The private key is a random number in [1, q-1], so 1 must be allowed.
Since q is at least an 160-bit prime and 2^159 + 1 is not prime (159
is not a power of 2), the probability that this is hit is < 2^-159,
but a tiny little bit wrong is still wrong.

Found while investigating a report by bluhm
ok jsing

19 months agoCall CRYPTO_cleanup_all_ex_data() from OPENSSL_cleanup().
jsing [Sat, 11 Mar 2023 14:29:02 +0000 (14:29 +0000)]
Call CRYPTO_cleanup_all_ex_data() from OPENSSL_cleanup().

Issue reported by Graham Percival (@gperciva)

ok tb@

19 months agoAdd OPENSSL_cleanup() calls to some regress.
jsing [Sat, 11 Mar 2023 14:27:37 +0000 (14:27 +0000)]
Add OPENSSL_cleanup() calls to some regress.

This gets us some minimal test coverage.

19 months agoAvoid -0 in BN_div_word().
jsing [Sat, 11 Mar 2023 14:14:54 +0000 (14:14 +0000)]
Avoid -0 in BN_div_word().

Currently, the use of BN_div_word() can result in -0 - avoid this by
setting negative again, at the end of the computation.

Should fix oss-fuzz 56667.

ok tb@

19 months agoCorrect sign handling in BN_add_word().
jsing [Sat, 11 Mar 2023 14:13:11 +0000 (14:13 +0000)]
Correct sign handling in BN_add_word().

A sign handling bug was introduced to BN_add_word() in bn_word.c r1.18.
When handling addition to a negative bignum, the BN_sub_word() call can
result in the sign being flipped, which we need to account for. Use the
same code in BN_sub_word() - while not technically needed here it keeps
the code consistent.

Issue discovered by tb@

ok tb@

19 months agoRemove a pesky space.
jsing [Sat, 11 Mar 2023 14:05:02 +0000 (14:05 +0000)]
Remove a pesky space.

19 months agoAdd regress coverage for BN_{add,sub,mul,div,mod}_word().
jsing [Sat, 11 Mar 2023 14:04:21 +0000 (14:04 +0000)]
Add regress coverage for BN_{add,sub,mul,div,mod}_word().

This also provides some indirect coverage for BN_hex2bn(), BN_bn2hex() and
BN_get_word().

Two of these tests are currently failing and will be fixed shortly.

19 months agoMark test table as static const.
jsing [Sat, 11 Mar 2023 14:02:26 +0000 (14:02 +0000)]
Mark test table as static const.

19 months agoAdd missing $OpenBSD$ tag.
aoyama [Sat, 11 Mar 2023 10:33:27 +0000 (10:33 +0000)]
Add missing $OpenBSD$ tag.

19 months agoBump version to 7.8
claudio [Sat, 11 Mar 2023 10:06:17 +0000 (10:06 +0000)]
Bump version to 7.8

19 months agoFactor out the cache reset logic into rtr_reset_cache() and use it
claudio [Sat, 11 Mar 2023 10:04:59 +0000 (10:04 +0000)]
Factor out the cache reset logic into rtr_reset_cache() and use it
consitently in all places where the cache should be reset.

Additionally adjust the FSM to handle a connection close event in the
protocol negotiation phase as a protocol mismatch. Some caches fail
to send an error PDU and this allows them to work. A drawback is that
if the connection is closed because of a different reason the system
will fall back to a lower then necessary version. The RFC and draft
mandates this behaviour.

OK tb@

19 months agobtrace(8) retrieves the types of the arguments of each tracepoint
bluhm [Fri, 10 Mar 2023 23:02:30 +0000 (23:02 +0000)]
btrace(8) retrieves the types of the arguments of each tracepoint
from the kernel.  Add them to a cache when needed, and print them
with btrace -l.  If the argument is of type int, print the value
as %d signed integer.  This gives pretty output for refcnt changes.
All other types still use %ul.
The probe and argument number can be used as array index.  Remove
the qsort and search for probe number to simplify the code.  Use
array index for both probes and argument types.
OK mpi@

19 months agoAdd new dt(4) ioctl DTIOCGARGS to get the types of probe arguments.
bluhm [Fri, 10 Mar 2023 22:14:32 +0000 (22:14 +0000)]
Add new dt(4) ioctl DTIOCGARGS to get the types of probe arguments.
They are already tracked as strings in the kernel.  Export them to
userland using one ioctl(2) for all arguments of each probe.
OK mpi@

19 months agoRemove unused make variables FSTYPE (=install360) and FSDISKTYPE
krw [Fri, 10 Mar 2023 20:16:53 +0000 (20:16 +0000)]
Remove unused make variables FSTYPE (=install360) and FSDISKTYPE
(=mini34).  Preparatory to removing some obsolete/unused disktab
entries.

19 months agoFix possible leak of spibuf and flowbuf in error case.
tobhe [Fri, 10 Mar 2023 19:26:06 +0000 (19:26 +0000)]
Fix possible leak of spibuf and flowbuf in error case.

ok patrick@ bluhm@

19 months agosync
tb [Fri, 10 Mar 2023 16:47:36 +0000 (16:47 +0000)]
sync

19 months agoCrankl libcrypto/libssl/libtls minors after symbol addition
tb [Fri, 10 Mar 2023 16:45:50 +0000 (16:45 +0000)]
Crankl libcrypto/libssl/libtls minors after symbol addition