tb [Mon, 27 May 2024 16:11:16 +0000 (16:11 +0000)]
openssl: avoid shadowed pkeys in x509.c
ok job jsing
krw [Mon, 27 May 2024 14:46:26 +0000 (14:46 +0000)]
Revert NVME_GONE. Literal 0xffffffff is the preferred idiom here.
Requested by deraadt@
krw [Mon, 27 May 2024 14:18:02 +0000 (14:18 +0000)]
Replace repeated uses of magic number 0xffffffff with nice #define NVME_GONE.
Name taken from FreeBSD.
Add separate #define NVME_TIMO_PT for passthrough command poll timeout. Same
value as the currently reused NVME_TIMO_QOP.
No functional change.
deraadt [Mon, 27 May 2024 13:27:06 +0000 (13:27 +0000)]
sync
mglocker [Mon, 27 May 2024 10:27:58 +0000 (10:27 +0000)]
Use SCSI xfer timeout value for doorbell register status polling instead of
an own introduced timeout value. This fixes an SCSI SYNC command timeout
seen during suspend.
kettenis [Mon, 27 May 2024 09:31:58 +0000 (09:31 +0000)]
Set sc_state to ACPI_STATE_S0 at the end of resume instead of at the start
such that ACPI resume code can see what state we're resuming from.
ok deraadt@
jsg [Mon, 27 May 2024 09:12:31 +0000 (09:12 +0000)]
remove unused typedefs with structs that were removed
ENGINE, SSL and SSL_CTX remain even though the structs in the typedefs
don't exist as they are used as incomplete types.
feedback, ports bulk build and ok tb@
kettenis [Mon, 27 May 2024 06:20:59 +0000 (06:20 +0000)]
Decode remaining ID_AA64ISAR1_EL1 features.
ok jsg@
jsg [Mon, 27 May 2024 04:58:43 +0000 (04:58 +0000)]
remove unused xmphyreg.h
jsg [Mon, 27 May 2024 03:56:59 +0000 (03:56 +0000)]
include net/if_var.h to make this build
djm [Mon, 27 May 2024 01:52:26 +0000 (01:52 +0000)]
don't need sys/queue.h here
jsg [Mon, 27 May 2024 00:54:00 +0000 (00:54 +0000)]
drm/amdgpu: Fix possible NULL dereference in amdgpu_ras_query_error_status_helper()
From Srinivasan Shanmugam
c364e7a34c85c2154fb2e47561965d5b5a0b69b1 in linux-6.6.y/6.6.32
b8d55a90fd55b767c25687747e2b24abd1ef8680 in mainline linux
jsg [Mon, 27 May 2024 00:51:31 +0000 (00:51 +0000)]
drm/amd/display: Fix division by zero in setup_dsc_config
From Jose Fernandez
7e4f50dfc98c49b3dc6875a35c3112522fb25639 in linux-6.6.y/6.6.32
130afc8a886183a94cf6eab7d24f300014ff87ba in mainline linux
kettenis [Sun, 26 May 2024 22:04:52 +0000 (22:04 +0000)]
Revert bits that got accidentally committed.
spotted by patrick@
naddy [Sun, 26 May 2024 20:35:12 +0000 (20:35 +0000)]
remove references to SSH1 and DSA server keys
mglocker [Sun, 26 May 2024 20:06:27 +0000 (20:06 +0000)]
Fix suspend/resume for ums(4) and umt(4); Calling hidm*_attach not at the
end of the device driver attach function can cause accessops functions to
be called by wscons(4) *before* the device has completed its attach
procedure. In this case this has led the device driver to ignore
interrupts after resume, because during resume ums_disable/umt_disable
were called last, *after* ums_enable/umt_enable.
OK phessler@
kettenis [Sun, 26 May 2024 18:06:21 +0000 (18:06 +0000)]
Clear interrupt status to prevent power down after resume initiated by
pressing the power button.
ok patrick@
kettenis [Sun, 26 May 2024 13:40:54 +0000 (13:40 +0000)]
Add support for configuring sleep voltage settings based on device tree
properties for the RK809.
ok patrick@
kettenis [Sun, 26 May 2024 13:37:31 +0000 (13:37 +0000)]
Implement wakeup interrupts on amd64. Provide a dummy implementation for
i386 such that we can call the necessary hooks in the suspend/resume code
without adding #ifdefs. Tweak the arm64 implementation such that we can
call the hooks earlier as this is necessary to mask MSI and MSI-X
interrupts on arm64.
ok deraadt@, mlarkin@
jsg [Sun, 26 May 2024 10:01:01 +0000 (10:01 +0000)]
remove unneeded includes
tb [Sun, 26 May 2024 09:54:16 +0000 (09:54 +0000)]
Remove documentation of optional md in one-step hashes
This functionality will be removed, so stop documenting it. Instead
mention that another implementation still supports this.
jsg [Sun, 26 May 2024 08:46:28 +0000 (08:46 +0000)]
remove unneeded includes
jsg [Sun, 26 May 2024 03:28:54 +0000 (03:28 +0000)]
correct an argument to fix build, broken by rev 1.22 in 2013
tb [Sat, 25 May 2024 18:59:03 +0000 (18:59 +0000)]
Eliminate last timegm() correctly this time
Also add a test case with a generalized time representing the moment
one second past the 32-bit epoch wrap.
tb [Sat, 25 May 2024 15:14:26 +0000 (15:14 +0000)]
sync inclusion of <stdlib.h> from libcrypto
tb [Sat, 25 May 2024 15:12:47 +0000 (15:12 +0000)]
Include <stdint.h> in the bytestring .c files
They currently depend on bytestring.h pulling that in.
discussed with jsing
tb [Sat, 25 May 2024 12:47:25 +0000 (12:47 +0000)]
Revert previous
It wasn't quite right, but I also think the test is bogus.
tb [Sat, 25 May 2024 06:42:15 +0000 (06:42 +0000)]
asn1time: another use of gmtime was hiding here
mglocker [Fri, 24 May 2024 20:34:06 +0000 (20:34 +0000)]
Enable UFS "Auto-Hibernation". From the documentation:
Auto-hibernate allows the host controller to put UniPro link into Hibernate
state autonomously.
tb [Fri, 24 May 2024 19:21:58 +0000 (19:21 +0000)]
des_local.h: Remove some unused macros
tb [Fri, 24 May 2024 19:18:07 +0000 (19:18 +0000)]
Remove documentation of DES_enc_{read,write} and DES_rw_mode
ok jsing
tb [Fri, 24 May 2024 19:16:53 +0000 (19:16 +0000)]
Stub out DES_enc_{read,write}(3)
The most terrible code in OpenSSL has its roots in libdes, which came
before SSLeay. Hello, LHASH. Hello speed app. Hello DES (obviously).
There are some diary-style changelog comments dating all the way back
to 1990.
/* This has some uglies in it but it works - even over sockets. */
Well, kind of:
* - This code cannot handle non-blocking sockets.
Also:
/* >output is a multiple of 8 byes, if len < rnum
* >we must be careful. The user must be aware that this
* >routine will write more bytes than he asked for.
* >The length of the buffer must be correct.
* FIXED - Should be ok now 18-9-90 - eay */
Or
/* This is really a bad error - very bad
* It will stuff-up both ends. */
Or
#ifdef _LIBC
extern unsigned long time();
extern int write();
#endif
I can't even...
Delete, delete, delete.
ok jsing
deraadt [Fri, 24 May 2024 18:15:12 +0000 (18:15 +0000)]
sync
miod [Fri, 24 May 2024 16:44:12 +0000 (16:44 +0000)]
Remove target getting in the way of config.status and causing compilation
failure when using make -j; this used to be hidden because of the objective-C
generated files, which are no longer built. No change for regular make.
deraadt [Fri, 24 May 2024 16:35:12 +0000 (16:35 +0000)]
sync
ratchov [Fri, 24 May 2024 15:21:35 +0000 (15:21 +0000)]
sndiod: Set the display string of all server.device entries
with help from edd@ and armani@
ratchov [Fri, 24 May 2024 15:16:09 +0000 (15:16 +0000)]
sndiod: Propagate the controls' display string to clients.
with help from edd@ and armani@
ratchov [Fri, 24 May 2024 15:10:26 +0000 (15:10 +0000)]
sndio: Add a display string to the sioctl_open API
For hardware devices, add a server.device control
with a single item and the device name as display
string.
Add the necessary sndioctl(1) bits to print it.
ratchov [Fri, 24 May 2024 15:03:12 +0000 (15:03 +0000)]
sndiod: Use a 'unsigned char *' for the pointer to the temp sock buffer
ratchov [Fri, 24 May 2024 15:01:53 +0000 (15:01 +0000)]
sndiod: Hide "hardware" device's server.device control.
If sndiod is using another sndiod instance instead of the bare
hardware, then it must disable the underlying server.device
control to avoid conflicts with its own server.device control.
ratchov [Fri, 24 May 2024 15:00:15 +0000 (15:00 +0000)]
sndiod: Don't mark as dirty controls that have not been changed
op [Fri, 24 May 2024 13:32:03 +0000 (13:32 +0000)]
typo: "pwdtries" -> "passwordtries"; ok millert@
tb [Fri, 24 May 2024 12:57:20 +0000 (12:57 +0000)]
Update references to new and not quite so new RFCs
Four warnings now contain "RFC 9582" rather than "RFC 6482bis".
Also update some references to I-Ds.
ok claudio
nicm [Fri, 24 May 2024 12:41:24 +0000 (12:41 +0000)]
Do not escape $ unless DQ is set, that is the only case where we need to
escape it.
nicm [Fri, 24 May 2024 12:39:06 +0000 (12:39 +0000)]
Add N to search backwards in tree modes, from Fadi Afani in GitHub issue
3982.
krw [Fri, 24 May 2024 12:04:07 +0000 (12:04 +0000)]
Add support for NVMe passthrough commands IDENTIFY, GET_LOG_PG and SELFTEST.
Enables suitably inquisitive software (e.g. smartmontools) to get information on
nvme(4) disks.
Based on work with dlg@ at h2k23, various at p2k24 and subsequent improvements
and tests by jmatthew@.
ok dlg@ for more permissive h2k23 version, jmatthew@
op [Fri, 24 May 2024 11:52:45 +0000 (11:52 +0000)]
capitalize the first word of the -S description
jsg [Fri, 24 May 2024 10:05:55 +0000 (10:05 +0000)]
remove unneeded includes
mglocker [Fri, 24 May 2024 09:51:13 +0000 (09:51 +0000)]
Add support for suspend/resume.
ok dlg@
sashan [Fri, 24 May 2024 06:38:41 +0000 (06:38 +0000)]
pfsync must let to progress state for destination peer
The issue has been noticed by matthieu@ when he was chasing
cause of excessive pfsync traffic between firewall boxes.
When comparing content of state tables between primary
and backup firewall the backup firewall showed many
states as follows:
ESTABLISHED:SYN_SENT
FIN_WAIT_2:SYN_SENT
* :SYN_SENT
this is caused by pfsync_upd_tcp() which fails to update
TCP-state for destination connection peer, so it remains
stuck in SYN_SENT.
matthieu@ confirms diff helps with 'stuck-state'. It also
seems to help with excessive pfsync traffic.
ok @dlg
jsg [Fri, 24 May 2024 06:26:47 +0000 (06:26 +0000)]
remove unneeded includes
jsg [Fri, 24 May 2024 06:02:53 +0000 (06:02 +0000)]
remove unneeded includes; ok miod@
jsg [Fri, 24 May 2024 04:36:26 +0000 (04:36 +0000)]
add back var removed in rev 1.7 in 2009, makes puc@cardbus build
jsg [Thu, 23 May 2024 23:47:16 +0000 (23:47 +0000)]
remove unused struct fwd_perm_list, no decl with complete type
ok djm@
op [Thu, 23 May 2024 17:10:00 +0000 (17:10 +0000)]
require an error message on table_proc failure replies
The error message is not really used, it just gets translated to a
TEMPFAIL, but it allows to have mandatory logging of errors instead
of relying on tables to hopefully log something.
To ease the transition, don't make it mandatory in smtpd(8) yet,
but document it as such.
ok millert@, gilles@
op [Thu, 23 May 2024 17:05:45 +0000 (17:05 +0000)]
catch EOFs during table_proc handshake
while here also do a s/fatalx/fatal since getline() sets errno on
failure.
ok millert@, gilles@
stsp [Thu, 23 May 2024 11:19:13 +0000 (11:19 +0000)]
increment CCMP decryption error counter if hw decrypt fails to get PN
This case will only occur if the IV has been stripped by hardware and
the driver has not cleared the protected bit in the frame header as it
should. Incrementing this counter will make the problem more obvious
when looking at netstat -W output.
No functional change for people who do not work on wifi drivers.
kevlo [Thu, 23 May 2024 08:06:45 +0000 (08:06 +0000)]
regen
kevlo [Thu, 23 May 2024 08:06:22 +0000 (08:06 +0000)]
Add support for Quectel EM060K.
The Quectel EM060K is set by default to MBIM mode.
Tested by Shawn Chiou
ok jsg@
tb [Thu, 23 May 2024 06:49:55 +0000 (06:49 +0000)]
Make signature of SSL_COMP_add_compression_method(3) match reality
jsg [Thu, 23 May 2024 03:21:08 +0000 (03:21 +0000)]
remove unneeded includes; ok mpi@
tb [Thu, 23 May 2024 02:00:38 +0000 (02:00 +0000)]
x509_v3.c: indent labels
tb [Thu, 23 May 2024 01:57:32 +0000 (01:57 +0000)]
x509_v3.c: remove an unnecessary else
tb [Thu, 23 May 2024 01:56:14 +0000 (01:56 +0000)]
x509_v3.c: consistently call STACK_OF(X509_EXTENSIONS) arguments sk
(where it doesn't conflict with a local variable)
tb [Thu, 23 May 2024 01:52:05 +0000 (01:52 +0000)]
x509_v3.c: zap another pointless local variable
tb [Thu, 23 May 2024 01:50:52 +0000 (01:50 +0000)]
x509_v3.c: add a few empty lines
tb [Thu, 23 May 2024 01:43:51 +0000 (01:43 +0000)]
X509v3_get_ext_by_NID: make obj const, test & assign
tb [Thu, 23 May 2024 01:42:23 +0000 (01:42 +0000)]
x509_v3.c: remove a pointless local variable
tb [Thu, 23 May 2024 01:40:23 +0000 (01:40 +0000)]
x509_v3.c: mechanically replace ex with ext and new_ex with new_ext
jsg [Thu, 23 May 2024 00:45:08 +0000 (00:45 +0000)]
playing is a var in main() not a global, remove extern
mglocker [Wed, 22 May 2024 18:10:00 +0000 (18:10 +0000)]
Fix some bogus masking for the HCS (Host Controller Status) register.
martijn [Wed, 22 May 2024 16:25:18 +0000 (16:25 +0000)]
regen
martijn [Wed, 22 May 2024 16:24:59 +0000 (16:24 +0000)]
Add support for MT7922. This brings the MT7922 in line with MT7921.
feedback and OK claudio@
naddy [Wed, 22 May 2024 15:24:55 +0000 (15:24 +0000)]
Do not pass -Werror twice when building with clang.
jsg [Wed, 22 May 2024 14:25:47 +0000 (14:25 +0000)]
remove prototypes with no matching function; ok miod@
jsg [Wed, 22 May 2024 14:22:27 +0000 (14:22 +0000)]
remove extern with no matching var; ok miod@
tb [Wed, 22 May 2024 14:03:24 +0000 (14:03 +0000)]
Exercise EVP_chacha20_poly1305() with in-place decryption
This needs quite a bit of cleanup but let's have some tests rather than
none.
tb [Wed, 22 May 2024 14:02:08 +0000 (14:02 +0000)]
Fix in-place decryption for EVP_chacha20_poly1305()
Take the MAC before clobbering the input value on decryption. Fixes hangs
during the QUIC handshake with HAProxy using TLS_CHACHA20_POLY1305_SHA256.
Found, issue pinpointed, and initial fix tested by Lucas Gabriel Vuotto:
Let me take this opportunity to thank the HAProxy team for going out of
their way to keep supporting LibreSSL. It's much appreciated.
See https://github.com/haproxy/haproxy/issues/2569
tweak/ok jsing
claudio [Wed, 22 May 2024 12:33:07 +0000 (12:33 +0000)]
Add pattern for mwx(4) firmware.
OK phessler@ martijn@
miod [Wed, 22 May 2024 11:49:36 +0000 (11:49 +0000)]
Do not pass -Werror if building with gcc 3, for asn1.h and bio.h cause
(admittedly bogus) warnings with gcc 3.
mglocker [Wed, 22 May 2024 11:46:06 +0000 (11:46 +0000)]
Make the interrupt status poll function generic, rather than for only one
status type. We'll use it later on to query more interrupt status types.
tb [Wed, 22 May 2024 09:44:10 +0000 (09:44 +0000)]
crib better wording from schwarze's EVP_PKEY_get_attr_by_NID(3)
tb [Wed, 22 May 2024 09:34:40 +0000 (09:34 +0000)]
Fix incorrect X509v3_get_ext_by_NID(3) return values
This error comes from upstream, where it is still wrong.
claudio [Wed, 22 May 2024 09:24:11 +0000 (09:24 +0000)]
When clearing the wait channel also clear the wait message.
There is no reason to keep the wait message in place since it will
never show up even in ddb show proc output.
OK jca@
claudio [Wed, 22 May 2024 09:22:55 +0000 (09:22 +0000)]
In the big p_stat switch in ptsignal do not call return but instead
use one of the gotos. In this case goto out with mask and prop set to 0.
OK jca@
claudio [Wed, 22 May 2024 09:20:22 +0000 (09:20 +0000)]
Just grab the SCHED_LOCK() once in donice() before walking the ps_threads
list. setpriority() is trivial and probably faster than releasing and
relocking SCHED_LOCK().
OK jca@
martijn [Wed, 22 May 2024 08:44:02 +0000 (08:44 +0000)]
When localtime() fails, throw an error instead of returning a
non-sensical 0 filled string.
OK florian@
claudio [Wed, 22 May 2024 08:42:34 +0000 (08:42 +0000)]
Adjust code since remote_bgpid, bgpid and clusterid changed to
host byte order.
OK tb@
claudio [Wed, 22 May 2024 08:41:14 +0000 (08:41 +0000)]
Convert bgpid, remote_bgpid and clusterid to host byte order.
Before the RDE used host byte order for remote_bgpid but all the other
code used network byte order. The reason for that was that bgpid was
initially an IPv4 address but since RFC 6286 in 2011 this is much more
relaxed and so it makes more sense to just treat them as numbers and
so host byte order.
OK tb@
op [Wed, 22 May 2024 08:41:02 +0000 (08:41 +0000)]
align table_proc id printing with the rest of smtpd
ok gilles@
op [Wed, 22 May 2024 08:39:50 +0000 (08:39 +0000)]
improve the documentation for fetch request
point out that it is supposed to wrap around and fix the misleading
example, then highlight that not-found can only be returned by empty
tables.
ok gilles@
martijn [Wed, 22 May 2024 08:38:57 +0000 (08:38 +0000)]
Add firmware and driver own functions. These are needed to bring up the
MT7922.
OK claudio@
jsg [Wed, 22 May 2024 05:51:49 +0000 (05:51 +0000)]
remove prototypes with no matching function and externs with no var
djm [Wed, 22 May 2024 04:20:00 +0000 (04:20 +0000)]
this test has been broken since 2014, and has been testing the same
key exchange algorithm repeatedly instead of testing all of them.
Spotted by nreilly AT blackberry.com in bz3692
Who broke the test? me.
jsg [Tue, 21 May 2024 23:16:06 +0000 (23:16 +0000)]
remove switch_exit() prototypes, replaced by sched_exit()
jan [Tue, 21 May 2024 19:49:06 +0000 (19:49 +0000)]
Refactor vmx(4) receive offloading.
- avoid combination of M_TCP_CSUM_IN_OK and M_UDP_CSUM_IN_OK
- remove useless letoh32() calls
- simplify the IPv4 checksum flag check
ok bluhm@
mglocker [Tue, 21 May 2024 18:19:22 +0000 (18:19 +0000)]
Remove obsolete comment, and fix some spacing.
mglocker [Tue, 21 May 2024 16:09:00 +0000 (16:09 +0000)]
Only enable BUS_DMA_64BIT if the controller supports it.
florian [Tue, 21 May 2024 15:12:25 +0000 (15:12 +0000)]
Inform user land when vltime / pltime changes.
Do not send a RTM_CHGADDRATTR route message when the address is
tentative because we will send one when DAD finishes.
To be used by rad(8) shortly.
OK bluhm
bluhm [Tue, 21 May 2024 11:19:39 +0000 (11:19 +0000)]
Make igc(4) and ix(4) struct names unique.
In ddb show struct gets confused if the same struct name exists
twice. Add a prefix to tx_ring and rx_ring in drivers.
OK miod@ jan@