Update for proplib(3) API changes.

(simonb)

Add SRBDS_CTRL bit.

(msaitoh)

Protect using mipsNN_foo() functions with an #ifdef MIPSNN.

Fixes non-MIPSNN kernel compiles pointed out by tsutsui@.

(simonb)

proplib API catchup
NFCI, tested on v210

(macallan)

Build and install llvm-symbolizer for MKLLVM=yes

llvm-symbolizer is an alternative for GNU addr2line(1), heavily used by the
LLVM sanitizers.

Do not install it as tools as it is not necessary as of today in that
stage.

(kamil)

FQA GCHQ

(nia)

Fix comment.

(tsutsui)

Specify -fno-unwind-tables to shrink binary.  Saves ~3.5%.

(tsutsui)

Suppress "options `COMPAT_xx' is not defined" warnings.

(tsutsui)

Fix inverted Y-axis value passed to wsmouse_input(9).

Note this is what src/sys/arch/macppc/dev/ams.c does as
"upward (moving the mouse forward) is positive" as noted in wsmouse(9):
http://cvsweb.netbsd.org/bsdweb.cgi/src/sys/arch/macppc/dev/ams.c#rev1.4

Reported and tested by @NonameInsect on his SE/30 and Xorg server.
Should be pulled up to netbsd-8 and netbsd-9.

(tsutsui)

If we are on a SiByte or Cavium CPU with an FPU, report as "built-in FPU"
instead of saying it's an unknown FPU type.

XXX - add any other CPUs to this list?

(simonb)

Dynamically build the wayname strings instead of using a sparse array
populated with random N-ways entries.  Not only results in smaller code,
but also handles cases with 39-way caches without extra work.

(simonb)

Add braces.

(msaitoh)

Remove debug printf.

(msaitoh)

Update wrong and obsolete comments.

... when you change the set of syscalls.conf keywords this thing
recognizes, updating the documentation too (which AFAIK is limited to
the comment in here) is always helpful ...

(dholland)

Add test case for :Or

(sjg)

Add fork/vfork/posix_spawn tests for processes within pgrp

New tests:
- fork_setpgid
- vfork_setpgid
- posix_spawn_setpgid
- unrelated_tracer_fork_setpgid
- unrelated_tracer_vfork_setpgid
- unrelated_tracer_posix_spawn_setpgid

These tests trigger a kernel assert for pg_jobc going negative.

The tests are temporarily skipped.

(kamil)

When we add "entries", "index", and "sector_size" values to the dictionary,
add them as signed valaues, rather than unsigned (which is how we keep them
in memory).  This causes them be serialized in base-10 (rather than base-16,
which is the default for unsigned).  This behavior is documented in
prop_number(3).  Fixes t_gpt::backup_2part unit test.

(thorpej)

Fix a paste-o that caused prop_data_create_copy() to be intolerant
of creating empty data objects.  Fixes t_ifconfig::ifconfig_description
unit test.

(thorpej)

Ensure copy_data_create_copy() is tolerant of creating empty data objects.

(thorpej)

Update for proplib(3) API changes.

(thorpej)

Update for proplib(3) API changes.

(thorpej)

Update for proplib(3) API changes.

(thorpej)

Use Dx macro

(sevan)

Update for proplib(3) API changes.

(thorpej)

Update for proplib(3) API changes.

(thorpej)

usr.bin/make: fix test failure

The two files include-sub.mk and include-subsub.mk were never intended to
be test cases on their own.  They belong to include-main.mk.

Fixes PR bin/55360.

(rillig)

Pp before Bl

(sevan)

white space

(sevan)

install fault.h

(maxv)

Fix minor problems detected by GCC9 -Wstringop-truncation.

(rin)

Add smscphy(4)

(jmcneill)

Fix wrong divider setting.

(hkenken)

Remove t_proplib.debug from debug/shl.mi, which is built regardless of MKPIC,
and has been moved into debug/mi.

(rin)

For /usr/libdata/debug/usr/tests/lib/libprop
- Obsolete t_basic.debug and
- Add t_proplib.debug

(rin)

make find_cpus() static.

(mrg)

ftp: exit if lostpeer invoked by a signal

lostpeer() calls too many async-unsafe functions (both directly
and indirectly) to close and cleanup the remote connections,
so just exit after the cleanup if invoked by a signal.

Reported in private mail by Qi Hou.
May also resolve a crash reported by Thomas Klausner.

(lukem)

Fix tyop in a comment.  Thanks riastradh@.

(simonb)

libexec/httpd: Fix the possible -Werror=stringop-truncation

Error was reported when build.sh was run with MKLIBCSANITIZER=yes flag.

Reviewed by: kamil@, mrg@

(fox)

external/mpl/dhcp: Suppress -Werror=stringop-truncation error

This logic correctly uses strncpy(3) to fully initialize a fixed-width field, and also ensures
NUL-termination on the next line as other users of the field expect.

Add -Werror=stringop-truncation to prevent build failure, when run with MKSANITIZER=yes.

Error was reported when build.sh was run with MKSANITIZER=yes flag.

Reviewed by: kamil@

(fox)

external/cddl/osnet: Suppress -Werror=stringop-truncation error

This logic correctly uses strncpy(3) to fully initialize a fixed-width field, and also ensures
NUL-termination on the next line as other users of the field expect.

Add -Werror=stringop-truncation to prevent build failure, when run with MKSANITIZER=yes.

Error was reported when build.sh was run with MKSANITIZER=yes flag.

Reviewed by: kamil@

(fox)

tests/lib/libarchive: Suppress -Werror=stringop-truncation error

This logic correctly uses strncpy(3) to fully initialize a fixed-width field, and also ensures
NUL-termination on the next line as other users of the field expect.

Add -Werror=stringop-truncation to prevent build failure, when run with MKSANITIZER=yes.

Error was reported when build.sh was run with MKSANITIZER=yes flag.

Reviewed by: kamil@

(fox)

libexec/lfs_cleanerd: Suppress -Werror=stringop-truncation error

This logic correctly uses strncpy(3) to fully initialize a fixed-width field, and also ensures
NUL-termination on the next line as other users of the field expect.

Add -Werror=stringop-truncation to prevent build failure, when run with MKSANITIZER=yes.

Error was reported when build.sh was run with MKSANITIZER=yes flag.

Reviewed by: kamil@, riastradh@

(fox)

make this compile.

(christos)

external/bsd/atf: Suppress -Werror=stringop-truncation error

This logic correctly uses strncpy(3) to fully initialize a fixed-width field, and also ensures
NUL-termination on the next line as other users of the field expect.

Add -Werror=stringop-truncation to prevent build failure, when run with MKSANITIZER=yes.

Error was reported when build.sh was run with MKSANITIZER=yes flag.

Reviewed by: kamil@

(fox)

Sort option descriptions.

(wiz)

Tickets #938 - #951

(martin)

Pull up following revision(s) (requested by skrll in ticket #951):

sys/arch/arm/include/arm32/param.h: revision 1.30

Bump the amount of KMEM allowed to 1/3rd of KVM.  _ARCH_ARM_6 has a split
of 2:2 and others have a split of 3:1.

Tested by martin@  Thanks

(martin)

Pull up following revision(s) (requested by isaki in ticket #950):

sys/dev/audio/audio.c: revision 1.71
sys/dev/audio/audio.c: revision 1.73
sys/dev/audio/audio.c: revision 1.74
sys/dev/audio/audio.c: revision 1.75
sys/dev/audio/audiovar.h: revision 1.12

audio: Fix logic for resuming when the device is in use.

audio_[r/p]mixer_start should never be called when the device is
marked busy.

Resolves a panic on resume when audio is playing, PR kern/55301

audio: remove comment that is no longer valid

audio: Only restart recording mixer on resume if it's already been started

Fix suspend/resume.
- Revert temporary usage of sc_[pr]busy during suspend.  These indicate
  whether the mixer needs to be restarted or not.
- Avoid timeout error when about to suspend.

(martin)

Tickets #1554 - #1556

(martin)

Pull up following revision(s) (requested by thorpej in ticket #949):

sys/kern/subr_pcu.c: revision 1.22

Relax the KASSERT() in pcu_discard_all() to allow non-curlwp if it is
in LSIDL state, which can happen if the new LWP is exiting before it's
ever run, e.g. if an error occurs in _lwp_create(2).

(martin)

Pull up following revision(s) (requested by thorpej in ticket #949):

sys/kern/subr_pcu.c: revision 1.22

Relax the KASSERT() in pcu_discard_all() to allow non-curlwp if it is
in LSIDL state, which can happen if the new LWP is exiting before it's
ever run, e.g. if an error occurs in _lwp_create(2).

(martin)

Pull up following revision(s) (requested by skrll in ticket #948):

sys/dev/usb/ugensa.c: revision 1.42
sys/dev/usb/u3g.c: revision 1.41

Remove duplicated USB_VENDOR_SIERRA USB_PRODUCT_SIERRA_USB305 entry
PR kern/55329

Replace calls to usbd_device2interface_handle in u3g_match() and
u3g_attach() with the interface already passed in uiaa->uiaa_iface and
store the interface in sc->sc_iface for later use by u3g_open()/

Also skip Direct IP interfaces in u3g_match() to avoid potential side
effects.

PR kern/55330

(martin)

Pull up following revision(s) (requested by tsutsui in ticket #947):
sys/arch/arc/stand/boot/Makefile: revision 1.18
Explicitly strip debug info on calling objdump for ecoff boot binaries.
As noted by martin@ on tech-toolchain@, release binaries are built
with MKDEBUG=yes so standalone binaries should explicitly remove
debug info generated by -g option.
https://mail-index.netbsd.org/tech-toolchain/2020/06/02/msg003809.html
Should be pulled up to netbsd-9.
(ARC BIOS gets confused by debug info sections in ecoff binaries
and boot binary in 9.0 doesn't work)

(martin)

Tickets #1732 and #1733

(martin)

Pull up following revision(s) (requested by riastradh in ticket #1733):

sys/dev/sysmon/sysmon_envsys.c: revision 1.145

Don't queue sysmon refresh until the rndsource is attached.

Using the rndsource, as refreshing the sensors will do, is not
allowed until _after_ rnd_source_attach.

XXX pullup-7
XXX pullup-8
XXX pullup-9

(martin)

Pull up following revision(s) (requested by riastradh in ticket #1555):

sys/dev/sysmon/sysmon_envsys.c: revision 1.145

Don't queue sysmon refresh until the rndsource is attached.

Using the rndsource, as refreshing the sensors will do, is not
allowed until _after_ rnd_source_attach.

XXX pullup-7
XXX pullup-8
XXX pullup-9

(martin)

Pull up following revision(s) (requested by riastradh in ticket #946):

sys/dev/sysmon/sysmon_envsys.c: revision 1.145

Don't queue sysmon refresh until the rndsource is attached.

Using the rndsource, as refreshing the sensors will do, is not
allowed until _after_ rnd_source_attach.

XXX pullup-7
XXX pullup-8
XXX pullup-9

(martin)

Pull up following revision(s) (requested by jmcneill in ticket #945):

sys/arch/arm/ti/ti_motg.c: revision 1.2

PR# port-evbarm/55263: BeagleBone Black too many interrupts on CPU when
using 5V DC power cable

Handle vbus status change events, and don't busy spin in the hard intr
handler when we see a vbus error (this will always be set when VBUS is not
present).

(martin)

Recommend using `pkg_admin {fetch-pkg-vulnerabilities,audit}' instead of
{download-vulnerability-list,audit-packages}.

(leot)

Properly quote printf string

Noticed by Ottavio Caruso on #netbsd@Freenode, thanks!

(leot)

Fix bohr bug triggered only once by syzkaller 2,5 months ago.

In sockopt_alloc(), 'sopt' may already have been initialized with
'sopt->sopt_data = sopt->sopt_buf'. If the allocation fails, we
end up with 'sopt->sopt_data = NULL', and later try to free this
NULL pointer in sockopt_destroy().

Fix that by not modifying 'sopt_data' if the allocation failed.

Difficult to reproduce in normal times, but fault(4) makes it
easy.

Reported-by: syzbot+380cb5d518742f063ad2@syzkaller.appspotmail.com

(maxv)

One more libprop-related cat pages fixup.

(thorpej)

Pull up following revision(s) (requested by jmcneill in ticket #944):

sys/arch/arm/ti/ti_dpll_clock.c: revision 1.3

Fix dpll clock setting on am335x

(martin)

Pull up following revision(s) (requested by jmcneill in ticket #943):

sys/arch/arm/ti/ti_cpufreq.c: revision 1.3

Fix SoC revision detection. BeagleBone Black reports 1GHz as a supported
frequency again.

(martin)

Pull up the following, requested by kre in ticket #942:

external/public-domain/tz/dist/Makefile        up to 1.1.1.26
external/public-domain/tz/dist/NEWS            up to 1.1.1.29
external/public-domain/tz/dist/TZDATA_VERSION  up to 1.19
external/public-domain/tz/dist/africa          up to 1.1.1.22
external/public-domain/tz/dist/asia            up to 1.1.1.25
external/public-domain/tz/dist/backward        up to 1.1.1.11
external/public-domain/tz/dist/backzone        up to 1.1.1.18
external/public-domain/tz/dist/europe          up to 1.1.1.26
external/public-domain/tz/dist/leap-seconds.list up to 1.1.1.13
external/public-domain/tz/dist/leapseconds      up to 1.1.1.15
external/public-domain/tz/dist/leapseconds.awk  up to 1.1.1.10
external/public-domain/tz/dist/northamerica    up to 1.1.1.25
external/public-domain/tz/dist/theory.html      up to 1.1.1.10
external/public-domain/tz/dist/version          up to 1.1.1.16
external/public-domain/tz/dist/zone.tab        up to 1.1.1.18
external/public-domain/tz/dist/zone1970.tab    up to 1.1.1.20
doc/3RDPARTY (apply patch)
distrib/sets/lists/base/mi (apply patch)

Import tzdata2020a from ftp://ftp.iana.org/tz/releases/tzdata2020a.tar.gz

Summary of changes in tzdata2020a (2020-04-23 16:03:47 -0700):

    Morocco resumes summer time on 2020-05-31, not 2020-05-24.
    Canada's Yukon advanced to -07 year-round on 2020-03-08 (summer time
      will not end this year)..
    America/Nuuk renamed from America/Godthab (both names now exist).

(martin)

Apply patch, requested by bouyer in ticket #941:

sys/arch/x86/x86/pmap.c (apply patch)

Fix Xen dom0 kernel build with options DIAGNOSTIC.

(martin)

Pull up following revision(s) (requested by kre in ticket #940):

bin/sh/expand.h: revision 1.25
bin/sh/expand.c: revision 1.134
bin/sh/expand.c: revision 1.135
bin/sh/expand.c: revision 1.136
bin/sh/expand.c: revision 1.137

Remove a (completely harmless) duplicate assignment introduced in a
code merge from FreeBSD in 2017.  NFC.

Pointed out by Roland Illig.

prevent sign extension from making expression always false.
remove masking and cast (requested by kre@)

When expanding a here-doc (NXHERE - the type with an unquoted end delim)
the output will not be further processed (at all) so there is no need
to escape magic chars in the output, and doing so leaves stray CTLESC
chars in the here doc text.  Not good.  So don't do that...

To save a strlen() of the result, to determine the size of the here doc,
make rmescapes() return the length of the resulting string (this isn't
needed for other uses, so didn't happen previously).

Reported on current-users@ (2020-02-06) by Jun Ebihara

XXX pullup -9

(martin)

Pull up following revision(s) (requested by tsutsui in ticket #939):

distrib/arc/ramdisk/Makefile: revision 1.29
sys/arch/arc/arc/arcbios.c: revision 1.18
distrib/arc/ramdisk/list: revision 1.24
sys/arch/arc/conf/GENERIC: revision 1.198
sys/arch/arc/conf/RAMDISK: revision 1.83
distrib/arc/ramdisk/Makefile: revision 1.28

bump ramdisk size to 3200k for gcc 8.  yay!

-

Make sure to preserve MIPS_CURLWP (t8) register from ARC BIOS calls.

Fixes silent hangup right after starting a kernel on my
Express5800/230 R4400 PCI (finally I've repaired its PSU and RTC).
Reviewed by soda@.  Maybe other ARC machines are also affected.

Note GXemul was not affected by this problem because it doesn't
have actual ARC BIOS ROM and emulates ARC functions internally.

Should be pulled up to at least netbsd-9.

-

Shrink ramdisk to make RAMDISK kernel smaller for ARC BIOS restriction.

Try to shrink kernels to smaller than ~6MB due to ARC BIOS restriction.

Currently NetBSD/arc kernels are loaded at 0x80200000
(on some machines ARC BIOS uses regions before it) and
ARC BIOS also uses around 0x807f0000, so bootloader fails
on loading >6MB kernels.

Note GXemul doesn't have this restriction because it loads
a kernel directly via a command line argument.

Tested by installing using sysinst on RAMDISK kernel on netbsd-9.
Should be pulled up to netbsd-9.

(martin)

Pull up following revision(s) (requested by rin in ticket #1732):

sys/arch/powerpc/oea/pmap.c: revision 1.96

Stop returning while PMAP_LOCK() (= KERNEL_LOCK(1, NULL)) is held.
Kernel freeze with heavy load is significantly mitigated (fixed?),
in which I could not even enter DDB from console.

XXX
There is still inconsistency in usage of two PVO pools.
I will send-pr later.

XXX
pullup to netbsd-[987]

(martin)

Pull up following revision(s) (requested by rin in ticket #1554):

sys/arch/powerpc/oea/pmap.c: revision 1.96

Stop returning while PMAP_LOCK() (= KERNEL_LOCK(1, NULL)) is held.
Kernel freeze with heavy load is significantly mitigated (fixed?),
in which I could not even enter DDB from console.

XXX
There is still inconsistency in usage of two PVO pools.
I will send-pr later.

XXX
pullup to netbsd-[987]

(martin)

Pull up following revision(s) (requested by rin in ticket #938):

sys/arch/powerpc/oea/pmap.c: revision 1.96

Stop returning while PMAP_LOCK() (= KERNEL_LOCK(1, NULL)) is held.
Kernel freeze with heavy load is significantly mitigated (fixed?),
in which I could not even enter DDB from console.

XXX
There is still inconsistency in usage of two PVO pools.
I will send-pr later.

XXX
pullup to netbsd-[987]

(martin)

Fix debug sets for libprop changes

(martin)

Add fault(4).

(maxv)

NetBSD 9.99.65 -- proplib(3) API additions also impact the kernel ABI.

(thorpej)

Update for proplib(3) API changes.

(thorpej)

Update for proplib(3) API changes.

(thorpej)

Update for proplib(3) API changes.

(thorpej)

Update for proplib(3) API changes.

(thorpej)

Update for proplib(3) API changes.

(thorpej)

Use -fno-unwind-tables to shrink boot binaries a bit.

(tsutsui)

Fix "illegal exceptoin" error on loading a secondary boot.

Recent gcc/binutils create more than two problem header sections
so the primary bootxx should handle them accordingly.
See my port-mips@ port for more details:
https://mail-index.netbsd.org/port-mips/2020/06/06/msg000948.html

Should be pulled up to netbsd-9.

(tsutsui)

Update for proplib(3) API changes.

(thorpej)

Update for proplib(3) API changes.

(thorpej)

Update for proplib(3) API changes.

(thorpej)

Update for proplib(3) API changes.

(thorpej)

Update for proplib(3) API changes.

(thorpej)

Update for proplib(3) API changes.

(thorpej)

uvmexp.ncolors being used where uvmexp.colormask wanted

(ad)

Update for proplib(3) API changes.

(thorpej)

Update for proplib(3) API changes.

(thorpej)

Correction to previous - set VI_PAGES correctly.

(ad)

Update for proplib(3) API changes.

(thorpej)

lwp_exit(): add a warning about (l != curlwp)

(ad)

Adjust previous.  In the condvar case the wakeup might already have been
eaten.

(ad)

Correct a deprecation warning.

(thorpej)

sun2 and vax have no native TLS support, so avoid using modern jemalloc
as it has a couple of nasty race conditions when using TSD.

(joerg)

Don't use deprecated proplib APIs (in the old "basic" test case).

(thorpej)

Improvements to the problib(3) API:

==> Provide a much more complete set of setters and getters for different
    value types in the prop_array_util(3) and prop_dictionary_util(3)
    functions.

==> Overhaul the prop_data(3), prop_number(3), and prop_string(3) APIs
    to be easier to use and less awkwardly named,  Deprecate the old
    awkward names, and produce link-time warnings when they are referenced.

==> Deprecate mutable prop_data(3) and prop_string(3) objects.  The old
    APIs that support them still exist, but will now produce link-time
    warnings when used.

==> When the new prop_string(3) API is used, strings are internally
    de-duplicated as a memory footprint optimization.

==> Provide a rich set of bounds-checked gettter functions in and a
    corresponding set of convenience setters in the prop_number(3) API.

==> Add a new prop_bool_value(3) function that is equivalent to
    prop_bool_true(3), but aligned with the new "value" routines in
    prop_data(3), prop_string(3), and prop_number(3).

(thorpej)

Use more markup.

(wiz)

Reduce macro arguments.

(wiz)

Remove trailing whitespace.

(wiz)

Relax the KASSERT() in pcu_discard_all() to allow non-curlwp if it is
in LSIDL state, which can happen if the new LWP is exiting before it's
ever run, e.g. if an error occurs in _lwp_create(2).

(thorpej)

Add a test case to ensure that _lwp_create() fails with the
expected error code when a bad new-lwp-id pointer is passed.

(thorpej)

FTP

(nia)

Make libsa's vsnprintf() work as expected when passed a NULL
destinatino buffer.

(thorpej)

Increase readability by reducing #ifdef using a macro.

(simonb)

Conditionally enable the formerly #if'ed-out code to use the
WRITE_REGION_1 command for writing blocks of data to the device.
This is to address a performance issue where "ifconfig run0 up" would
take as long as 20-30 seconds when using a UHCI or OHCI host
controller due to the large number of control transfers performed by
the driver in combination with the inability of those host controllers
to perform multiple control transfers per USB frame.  Limit the
transfers to 64 bytes as in the corresponding #if'ed-out code in
FreeBSD.  Currently only enabled for mac_ver 0x5390 as it is the only
version tested so far.

(gson)

Fix ews4800mips example.  Its raw partition is rsd0p, not rsd0c.

(tsutsui)

KNF

(skrll)

More __BITS.  NFCI.

(skrll)

If the frame is not aligned, leave right away. This place probably needs
to be revisited, because %rbp could easily contain garbage.

Reported-by: syzbot+ecb40cf7f8acc102c29b@syzkaller.appspotmail.com

(maxv)

kMSan: re-set the orig after pool_cache_get_slow(), using the address of
the caller of pool_cache_get_paddr().

Otherwise the orig is just pool_cache_get_paddr(), and that's not really
useful for debugging.

(maxv)

terminfo(5): Numbre -> Number

Thanks to yukonbob

(roy)

Change const unsigned to preprocessor define

Fixes GCC -O0 build with the stack protector.

(kamil)

IME, SCIM

(sevan)

make: add :Or for reverse sort

:Or is more efficient than :O:[-1..1]

Reviewed by: christos

(sjg)

cached_stats: do not conflate stat and lstat

While make uses lstat quite rarely, it does so for reason.
Avoid confusing the results.

Reviewed by: christos

(sjg)

Register eight vHCI buses, and use separate KCOV mailboxes for them.

(maxv)

Add support for npfd_flags

(christos)

fix wrong size addition (Andrew Cagney)
XXX: This file is nearly identical with /usr/src/sys/netipsec/key_debug.c
and should be merged.

(christos)

Check x > 0 *before*, not after, using (x - 1) as index of array.

(rin)

Remove more bogus tests for 64-bit i386 and SuperH.

These are always false.

(nia)

Rework CVMSEG LM usage a litte:
- remove unused LM slots
- use #defines for defining the size of CVMSEG LM users

XXX: Need to dynamically set CVMMEMCTL[LMEMSZ] during startup so we can
    both adapt to any future increase in CVMSEG LM usage and not waste
    any more L2 that we need to.
XXX: Still need to move general IOBDMA conf to a different (new?) header.

(simonb)

Replace calls to usbd_device2interface_handle in u3g_match() and
u3g_attach() with the interface already passed in uiaa->uiaa_iface and
store the interface in sc->sc_iface for later use by u3g_open()/

Also skip Direct IP interfaces in u3g_match() to avoid potential side effects.

PR kern/55330

(skrll)

Remove duplicated USB_VENDOR_SIERRA USB_PRODUCT_SIERRA_USB305 entry

PR kern/55329

(skrll)

Now that we don't zero out the second link in the U-boot bootmem
descripter, use that to configure free memory.

(simonb)

Refactor a little and follow new format of "npfctl list".

Fix the below ATF failures.
    - net/if_ipsec/t_ipsec_natt:ipsecif_natt_transport_null
    - net/if_ipsec/t_ipsec_natt:ipsecif_natt_transport_rijndaelcbc
    - net/ipsec/t_ipsec_natt:ipsec_natt_transport_ipv4_null
    - net/ipsec/t_ipsec_natt:ipsec_natt_transport_ipv4_rijndaelcbc

ok'ed by ozaki-r@n.o, thanks.

(knakahara)

Fix indent.

(hkenken)

Fix KASSERT

(hkenken)

Near complete conversion to __BITS

(skrll)

Add self-tests for 3des and blowfish too.

Nobody should use these on new disks but it is good to have them so
we don't break decrypting old disks.

(riastradh)

Call clts/stts in fpu_kern_enter/leave so they work.

(riastradh)

Consistent whitespace.  Register defines are <tab> indented. Bits defines
for registers are <tab><space> indented.

(skrll)

Tidy up some ugly output from mips_page_physload() when a segment starts
and finishes in the same page.

(simonb)

ACAB (worksafe variant)

(nia)

ACAB

(nia)

dhcpcd: re-obsolete the chroot sub dirs

Ordering matters - dhcpcd needs to be fixed before obsolete for postinstall

(roy)

Add missing comma.

(wiz)

Note import of dhcpcd-9.1.1

(roy)

dhcpcd: kqueue(2) support removed in favour of pollts(2)

(roy)

Sync

(roy)

getaddrinfo.3: order of the struct members is wrong, correct it

(nia)

Remove bogus tests for 64-bit i386 and SuperH

(nia)

G.C. No functional change. OK'd by bouyer@.

(msaitoh)

Nix trailing whitespace.  NFCI.

(riastradh)

Nix trailing whitespace.  NFCI.

(riastradh)

Nix trailing whitespace.  NFCI.

(riastradh)

Only clear BSS up to "end" symbol, not all the way to mips_round_page(end).
U-Boot puts some data immediately after the BSS section cleared in the ELF
segments, and we don't want to overwrite that.

XXX: Stop using mips_round_page(end) on all MIPS?

(simonb)

Nix trailing whitespace.  NFCI.

(riastradh)

If _malloc_thread_cleanup is implement, call it from libpthread.
Provide the hook from modern jemalloc to avoid using TSD for the thread
destruction cleanup as it can result in reentrancy crashes if fork is
called from a thread that never called malloc as it will result in a
late malloc from the pre-fork synchronisation handler.

(joerg)

PR kern/55032 (rump/rumpkern/t_vm:uvmwait test case now fails)

Work around issues with rump's pagedaemon emulation, including one that's
sensitive to timing effects (i.e. scheduler changes).  While here tidy up
some other stuff around the emulation of page alloc/free.

(ad)

Deal with a couple of problems with threads being awoken early due to
timeouts or cancellation where:

- The restarting thread calls _lwp_exit() before another thread gets around
  to waking it with _lwp_unpark(), leading to ESRCH (observed by joerg@).
  (I may have removed a similar check mistakenly over the weekend.)

- The restarting thread considers itself gone off the sleep queue but
  at the same time another thread is part way through waking it, and hasn't
  fully completed that operation yet by setting thread->pt_mutexwait = 0.
  I think that could have potentially lead to the list of waiters getting
  messed up given the right circumstances.

(ad)

Re-enable "set cpufreq at boot" code.

(jmcneill)

PR# port-evbarm/55263: BeagleBone Black too many interrupts on CPU when
using 5V DC power cable

Handle vbus status change events, and don't busy spin in the hard intr
handler when we see a vbus error (this will always be set when VBUS is not
present).

(jmcneill)

Fix dpll clock setting on am335x

(jmcneill)

Don't print an error message on spurious interrupt. Way too noisy for Xen.

(bouyer)

Skip this test on machines where it would cause serious swapping.

(martin)

Fix SoC revision detection. BeagleBone Black reports 1GHz as a supported
frequency again.

(jmcneill)

If we can't find a hwmod name, use the device type and mmio address

(jmcneill)

Add a line break before rnd_attach_source so the "entropy: ready" announcement does not get mixed with device prints

(jmcneill)

Disable setting cpufreq at boot for now

(jmcneill)

Highest speed is first opp, not last

(jmcneill)

Convert ohcireg.h to __BITS.  NFCI.

(skrll)

Catch up to clock changes in last dts import.

(jmcneill)

Add FHRP, GLBP, HSRP

(sevan)

Add Allen Ginsberg

(sevan)

defflag SOC_IMX6DL and SOC_IMX6Q

(jmcneill)

Ensure the dhcpcd log socket is removed.

(roy)

Explicitly strip debug info on calling objdump for ecoff boot binaries.

As noted by martin@ on tech-toolchain@, release binaries are built
with MKDEBUG=yes so standalone binaries should explicitly remove
debug info generated by -g option.
https://mail-index.netbsd.org/tech-toolchain/2020/06/02/msg003809.html

Should be pulled up to netbsd-9.
(ARC BIOS gets confused by debug info sections in ecoff binaries
and boot binary in 9.0 doesn't work)

(tsutsui)

Manifest that we need liblzma only when USE_XZ_SETS is enabled.

(rin)