This will help ensure that the change following this
one to support OpenSSL 3 is sane.
MFC after: 1 week
Reviewed by: emaste
Differential Revision: https://reviews.freebsd.org/D40297
To help grepping kdump output by pid or tid it makes sence to print
some numbers in decimal format always. Eg, process or thread identifier
at least, as they already printed in decimal format.
Switch to print pid/tid arguments of some Linux signal related syscalls
in decimal format.
Reviewed by: jhb
Differential Revision: https://reviews.freebsd.org/D40099
MFC after: 1 week
OPENSSL_API_COMPAT can be used to specify the OpenSSL API version in
use for the purpose of hiding deprecated interfaces and enabling
the appropriate deprecation notices.
This change is a NFC while we're still using OpenSSL 1.1.1 but will
avoid deprecation warnings upon the switch to OpenSSL 3.0.
A future change should migrate to use the OpenSSL 3.0 APIs.
PR: 271615
Reviewed by: emaste
Event: Kitchener-Waterloo Hackathon 202305
Sponsored by: The FreeBSD Foundation
* Understand "-" to mean stdout as per convention.
* Check that the output is a regular file and ignore -u if not, otherwise we might try to rm /dev/stdout.
* As a bonus, if -u was specified but the output file does not exist, proceed as if -u had not been specified instead of erroring out.
MFC after: 1 week
Sponsored by: Klara, Inc.
Reviewed by: cracauer, debdrup
Differential Revision: https://reviews.freebsd.org/D40237
The SPDX folks have obsoleted the BSD-2-Clause-NetBSD identifier. Catch
up to that fact and revert to their recommended match of BSD-2-Clause.
Discussed with: pfg
MFC After: 3 days
Sponsored by: Netflix
The SPDX folks have obsoleted the BSD-2-Clause-FreeBSD identifier. Catch
up to that fact and revert to their recommended match of BSD-2-Clause.
Discussed with: pfg
MFC After: 3 days
Sponsored by: Netflix
Also fix an indentation error I introduced in the previous commit.
Fixes: cb46f47c79
MFC after: 1 week
Sponsored by: Klara, Inc.
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D40042
For DIRDEPS_BUILD we need Makefile.depend.options to
force libegacy to be built on older FreeBSD and non-FreeBSD hosts.
Add readpassphrase to libegacy to avoid the need for libbsd on Linux
src.opts.mk disable TESTS for host if MK_host_egacy is yes
Reviewed by: imp
Sponsored by: Juniper Networks, Inc.
Differential Revision: https://reviews.freebsd.org/D39766
Add Makefile.depend.options to libnetbsd, mkimg and makefs
to ensure libegacy is build if needed.
targets/pseudo/stage/Makefile avoid the need for mtree
the staging process creates target dirs as needed anyway.
Reviewed by: stevek
Sponsored by: Juniper Networks, Inc.
Differential Revision: https://reviews.freebsd.org/D39757
A few Makefiles overrode LINKS and MLINKS when WITH_LLVM_BINUTILS was
set, which caused some llvm- prefixed tools to disappear. One such case
was llvm-ranlib, which some ports invoke explicitly.
Use += when adding to LINKS and MLINKS under WITH_LLVM_BINUTILS.
PR: 270955, 270956, 270959
Submitted by: jbeich
Reviewed by: arichardson
Fixes: 021385aba5 ("Add WITH_LLVM_BINUTILS to install LLVM binuti...")
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D39747
protopr does not support reading from a core anymore.
So don't state that it can.
Reviewed by: glebius, rscheff, rrs
MFC after: 1 week
Sponsored by: Netflix, Inc.
Differential Revision: https://reviews.freebsd.org/D39688
This change touches both kernel and netstat(1), but either of the changes
will fix printing pcb addresses with -A.
The thing is that historically netstat(1) treated TCP differently, and
printed tcpcb address instead of inpcb address. This is not documented
anywhere! With e68b379244 these two addresses became the same. It is
highly likely they will be the same for a long time, but it might be they
will start to differ again in a far future. My proposal is to stop
treating TCP differently with netstat(1) and right now is a good opportunity
to do that, since there will be no behavior change at all. The kernel
change to tcp_inptoxtp() will go into stable/14 to make it compatible with
netstat(1) binary from stable/13. We can drop it later, probably together
with in_ppcb pointer from inpcb. The in_ppcb in xinpcb will stay for size
compatibility.
Reviewed by: tuexen, rrs
Differential Revision: https://reviews.freebsd.org/D39736
Several makefile depend on tools built for host.
At least when using DIRDEPS_BUILD we can build these for the
pseudo machine "host" to facilitate building on older host versions.
Ideally we would build these tools in their own directories to avoid
building more than needed.
For now, setting an appropriate default for BTOOLSPATH will suffice
Reviewed by: stevek
Sponsored by: Juniper Networks, Inc.
Differential Revision: https://reviews.freebsd.org/D39708
The command function is defined to always take a void *. Functions
which accept a pointer to an array of pointers use a local temporary
'argv' assigned from the void *arg.
Reviewed by: zlei
Differential Revision: https://reviews.freebsd.org/D39527
This has an active upstream so will presumably be fixed upstream at
some point.
Reviewed by: emaste
Differential Revision: https://reviews.freebsd.org/D39525
These will hopefully be fixed upstream eventually, but silence the
warnings until then.
Reviewed by: emaste
Differential Revision: https://reviews.freebsd.org/D39518
4aeb63826e got it almost correct (we can't use strcmp() here as
current argument isn't guaranteed to be NUL-terminated), but we also
need to check that current argument length is equal to that of eofstr.
PR: 270867
Reviewed by: bapt
Differential Revision: https://reviews.freebsd.org/D39583
RLIMIT_CPU applies to CPU time, not real (wall-clock) time.
This test failed in AWS, where the real time was 5-7 seconds.
Sum the user and system CPU time used, and validate that.
While I'm here, don't bother specifying -s exit:0 or -e empty,
since those are checked by default.
MFC after: 1 week
Sponsored by: Dell EMC Isilon
This does not remove LLVM_TARGET_MIPS. Note that the only
MACHINE_ARCH values ending in 'hf' were all MIPS architectures, hence
removing the pattern matches for 'hf'.
Reviewed by: emaste
Differential Revision: https://reviews.freebsd.org/D39331
This reverts commit ab80f0b21f. The intent
of this change was to avoid possible compilation errors when certain
.inc files were not regenerated, but the method turns out to cause way
more rebuilds than anticipated. Another method will have to be found,
and in the mean time, WITH_CLEAN is the solution that always works.
Fixes: ab80f0b21f
This flag ensures that the tblgen tools do not actually touch the
produced .inc file, if there are no changes to the contents. In turn,
this may prevent a number of rebuilds of files that include such .inc
files, saving build time.
While here, ensure that the shell invocations to locate the used tblgen
binary do not show unnecessary error messages.
Reported by: des
MFC after: 1 week
Make userland tools such as netstat, route, arp and ndp use
either netlink or rtsock interfaces based on the NETLINK_SUPPORT
options.
Both NETLINK and NETLINK_SUPPORT options are turned on by default.
Reviewed By: eugen
Differential Revision: https://reviews.freebsd.org/D39148
lorder has not been used by the base system since 0e1e341b48 and
hasn't really been necessary for a very long time.
PR: 259254 [exp-run]
Reviewed by: arichardson (earlier)
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D30080
-f may be specified zero, one, or two times. Make this clear in the
usage.
Reviewed by: markj, kib
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D34733
A number of improvements has commited to snl(3) recently.
A notable one is snl(3) build-in parsers for all of the objects
exported by the kernel.
This change updates netlink handling code to the latest available snl(3)
API.
Fix d20d655018 where getlocalbase()
wasn't used to fill out the printf(3) format of _PATH_INCLUDE_LOCAL.
Differential Revision: https://reviews.freebsd.org/D38975
Reviewed by: imp
Most ATM support was removed prior to FreeBSD 12. The netgraph support
was kept as it was less intrusive, but it is presumed to be unused.
Reviewed by: manu
Relnotes: yes
Sponsored by: DARPA
Differential Revision: https://reviews.freebsd.org/D38879
When updating byacc to 20230201, I enabled the backtracking extension as it seemed useful and harmless. With a little more experience, it appears to have more impact than I initially thought and I'm no longer sure how useful it really is, so turn it back off.
PR: 244149, 269425
Sponsored by: Klara, Inc.
Reviewed by: imp, jkim
Differential Revision: https://reviews.freebsd.org/D38931
When doing a fully incremental build (with WITHOUT_CLEAN enabled), from
a commit before llvm 15 was merged (3264f6b88f), to a commit after
that, a number of .inc files were not regenerated. This could lead to
unexpected compilation errors when these .inc files were included from
llvm-project sources, similar to:
In file included from /usr/src/contrib/llvm-project/clang/lib/CodeGen/CGBuiltin.cpp:8268:
/usr/obj/usr/src/amd64.amd64/lib/clang/libclang/clang/Basic/arm_mve_builtin_cg.inc:279:18: error: no matching constructor for initialization of 'clang::CodeGen::Address'
Address Val2 = Address(Val1, CharUnits::fromQuantity(2));
^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Work around this by making the .inc files dependent on the tblgen binary
used for generating them. E.g., we can relatively safely assume that if
the binary gets updated, the .inc files must also be updated. (Although
this is not 100% optimal, the gain by complicating things even more is
probaby not worth the effort.)
MFC after: 3 days
Reviewed by: emaste
Differential Revision: https://reviews.freebsd.org/D38770
Previously strtol was used and the result was directly cast to an int
without checking for an overflow. Use strtonum instead since it is
safer and tells us what went wrong.
Reviewed by: imp
Pull Request: https://github.com/freebsd/freebsd-src/pull/643
Summary:
The LTO issue has been fixed. While -flto for some reason is commented out,
since it wasn't completely removed, it may be expected to be reenabled.
Reviewers: se
Approved by: se
MFC after: 3 days
Subscribers: imp
Differential Revision: https://reviews.freebsd.org/D38755
We already knew how to decode readv(2)/writev(2). Add the versions with
an offset.
Reviewed by: asomers
Differential Revision: https://reviews.freebsd.org/D27531
MFC after: 2 weeks
Note that this enables the backtracking extension, which we had previously left disabled.
PR: 244149, 269425
Sponsored by: Klara, Inc.
Differential Revision: https://reviews.freebsd.org/D38421
In the manual page for zmore(1) and zless(1) it is said that zless(1) is
equivalent to zmore(1) except that it uses less(1) as a pager. However
zmore(1) is able to handle uncompressed files transparently while zless(1)
is not.
Add another case to the switch in lesspipe.sh to handle non-compressed files.
PR: 196437
Reported by: marquis@roble.com
Approved by: mjg@
Differential Revision: https://reviews.freebsd.org/D38674
kevent returns EINTR when I suspend / resume. This causes tail -f
and tail -F to exit with interrupt system call. Ignore this error
and try kevent again.
Sponsored by: Netflix
netstat(1) allows to specify both -i (all interfaces) and -I <ifname>.
However, when both are specified, -I always overrides -i.
Add a comment where appropriate the same way we do in rm(1) for -f and -i.
PR: 202708
Reported by: demon@
Approved by: manpages (pauamma@)
Differential Revision: https://reviews.freebsd.org/D38654
The patch introduced a behavior change coming from getopt_long, namely
the entire argument array would be parsed looking for opts, which
affected uses spawning new processes without passing '--' to delimit
where cpuset arguments end. [1]
Apart from that turns out the change had a bug: using CPU_LEVEL_ROOT
instead of CPU_LEVEL_WHICH, returning a different cpuset than intended.
[2]
The intended functionality is provided with nproc(1), which landed after
this change.
As such, revert for the time being.
This reverts commit d9d5f2c042.
Reported by: Mark Millard [1]
Reported by: jbeich [2]
The read buffer in cnt() is 64 kB, which is a bit excessive for a stack variable. MAXBSIZE has grown since this code was originally written, and it might grow again in the future. Since the program is single-threaded and cnt() does not recurse, we can safely make the buffer static.
While there, constify p since it is only used to read.
Sponsored by: Klara, Inc.
Reviewed by: emaste
Differential Revision: https://reviews.freebsd.org/D38608
* Add a test that causes mbrtowc() to return 0.
* Add a test that causes mbrtowc() to return -2.
Sponsored by: Klara, Inc.
Reviewed by: emaste
Differential Revision: https://reviews.freebsd.org/D38605
* Drop <err.h>, which is unnecessary since we use libxo.
* As per POSIX, report an error if output fails.
* Fix some type mismatches.
* Use bool instead of int where appropriate.
* Avoid repeatedly checking for a null filename.
* Miscellaneous other tidying.
* Add tests (partly derived from work performed by SHENG-YI HONG <i19780219111@kimo.com>).
Sponsored by: Klara, Inc.
Reviewed by: kevans
Differential Revision: https://reviews.freebsd.org/D38496
It now reports stats separately for both ports and luns.
Also, prohibit using both "-p" and "-P" at the same time.
MFC after: 1 week
Sponsored by: Axcient
Reviewed by: mav
Differential Revision: https://reviews.freebsd.org/D38500
The 'pid' order was added, but the equivalent comparator was not back in
8b800d3af2 ("Add the PID column to the list of sort keys.").
`ordernames` indices need to match `compares` indices in order for this
to work.
A future commit will restructure all of this to make it less error prone
going forward (see D37083).
MFC after: 3 days
Reviewed by: pstef (D37083)
When a file is missing, patch(1) will prompt for a filename to try and
patch it. If we're doing a dry-run, we'll output that the patch to the
source file was either ignored/failed. If you ignore the prompt in a
dry-run (i.e. just hit enter), we'll output:
X out of X hunks ignored while patching (null)
Let's improve the aesthetics a bit and just omit the last part if the
prompt was ignored:
X out of X hunks ignored
Unfortunately we can't really test this without expect(1) because both
force and batch mode will use the first best guess, which is wiped out
by the "File to patch:" prompt. We could record the initially derived
bestguess there and use *that*, but given that this is only possible in
an interactive session I think it's fine to just omit the filename
rather than adding a fair amount of complexity (which could also break
other scenarios I haven't considered yet)..
Reviewed by: des
Sponsored by: Klara, Inc.
Differential Revision: https://reviews.freebsd.org/D38179
This program prints the number of CPU threads it can run on, while
respecting cpusets (or not, depending on switches).
It aims to be compatible with nproc as found in GNU coreutils.
Reviewed by: des
Reviewed by: pstef
Differential Revision: https://reviews.freebsd.org/D38386
It contain all the binaries and libs from the elftoolchain contrib
project except for libelf which is used everywhere.
All of those tools are never used by the average user.
Sponsored by: Beckhoff Automation GmbH & Co. KG
Differential Revision: https://reviews.freebsd.org/D38224
This update contains only documentation changes (new main repository
URL and changed mail address of the program author) and changes to
the build system that do not affect the FreeBSD base system build.
MFC after: 3 days
Can be used to count the number of hardware threads in the cpu set.
For example:
$ cpuset -g -p $$
pid 2440 mask: 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16,
17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34,
35, 36, 37, 38, 39
pid 2440 domain policy: first-touch mask: 0, 1
$ cpuset -g --count -p $$
40
The intent is to replace calls to sysctl hw.ncpu and kern.smp.cpus which
can be found in the tree, which are not adequate given existence of
cpusets.
Right now only -g -p combination is supported to reduce complexity.
As anything else errors out, this can be expanded later as needed.
Differential Revision: https://reviews.freebsd.org/D36351
timeout(1) is used by /etc/rc.d/zfskeys. Unfortunately, having
timeout(1) installed in /usr/bin causes problems when /usr is an
encrypted ZFS partition.
Implementing timeout(1) in sh(1) is not trivial. A more elegant solution
is to move timeout(1) to /bin so that it is available to early services
in the boot process.
PR: 265221
Reviewed by: allanjude, des, imp
Approved by: allanjude, des, imp
Reported by: Ivan <r4@sovserv.ru>
Fixes: 33ff39796f Add zfskeys rc.d script for auto-loading encryption keys
MFC after: 1 week
Relnotes: yes
Sponsored by: Modirum MDPay
Sponsored by: Klara Inc.
Differential Revision: https://reviews.freebsd.org/D38344
* Parse and export newly-added NL_RTA_WEIGHT attribute, providing path
weight for a non-multipath route. This fixes a number of tests in
sys/net/routing which rely on this data.
* Remove handling of NL_RTA_KNH_ID in multipath routes, as it is
not provided.
* Improve kernel/user nexthop index export. As a result,
for multipath routes:
* nhg-kidx attribute represents kernel nhg index (always provided)
* nhg-uidx attribute represents user-provided nhg index (if set)
for non-multipath routes:
* nhop-kidx attribute represents kernel nhop index (always provided)
* nhop-udx attribute represents user-provided nexthop index (if set)
The sources of this program are in contrib/bc, but built using the
Makefiles touched in this commit, which had to be adapted to comply
with changed made to the build system of this software, which is
not used when building in the base system.
Merge commit 'b3392d84da5bf2162baf937c77e0557f3fd8a52b' into zstd_1.5.2
full changelog: https://github.com/facebook/zstd/compare/v1.4.8...v1.5.2
Updated sys/kern/subr_compressor.c to new API
MFC after: 3 days
Relnotes: yes
Sponsored by: Klara, Inc.
This change switches route listing in netstat to netlink, with fallback to rtsock.
The outputs are mostly identical, with an exception of not showing kernel
nexthop indexes for multipath routes.
Differential Revision: https://reviews.freebsd.org/D36529
Several important base system components are written in C++, and the
WITHOUT_CXX option produced a system that was not fully functional.
Just accept this, and remove the option to build without C++ support.
This reverts commit adc3c128c6.
Reviewed by: brooks, kevans, jhb (earlier)
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D33108
Arguments follow primaries, not the other way around.
MFC after: 1 week
Sponsored by: Klara, Inc.
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D38173
sysctl() provides a count of number of bytes in the buffer. That is the
actual buffer length. Whereas looking for an interrupt entry with an
empty name could terminate too early, or overflow the end of the buffer.
The overflow will occur if the table of interrupt names is full.
Reviewed by: kib
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D36628
These appear to simply be the style of arguments are left untouched and
only local variables are modified. While some may prefer that style
this simply serves to complicate things as they're perfectly writeable.
Reviewed by: kib
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D36628
A similar change was made in svn r223931, but it was incomplete, working
only when the utility was invoked as "ncal". Fix the same issue when
invoking as "cal".
PR: 268936
Reported by: Ray Bellis <ray@bellis.me.uk>
MFC after: 2 weeks
Sponsored by: Axcient
Reviewed by: imp
Differential Revision: https://reviews.freebsd.org/D38045
The previous code would skip as many entries at the end of the
array as there were zombies in the list. While here fix type
of cnt.
Submitted by: Ali Abdallah <ali.abdallah suse.com>
PR: 232702
MFC After: 2 weeks
-z is supposed to use only the NUL byte as EOL, but we were
inadvertently using both newline and NUL due to REG_NEWLINE in cflags.
The odds of anyone relying on this bsdgrep-specific bug are quite low,
so let's just fix it. At least one port in the wild has been reported
to expect the intended behavior.
Reported by: Hill Ma <maahiuzeon@gmail.com>
Triaged by: the self-proclaimed peanut gallery on Discord
Document the environment variables used by grep(1).
Reviewed by: pauamma, kevans
MFC after: 5 days
Differential Revision: https://reviews.freebsd.org/D37726
This subsystem is superseded by modern debugging facilities,
e.g. DTrace probes and TCP black box logging.
We intentionally leave SO_DEBUG in place, as many utilities may
set it on a socket. Also the tcp::debug DTrace probes look at
this flag on a socket.
Reviewed by: gnn, tuexen
Discussed with: rscheff, rrs, jtl
Differential revision: https://reviews.freebsd.org/D37694
This is a bit more readable, and this loop is probably unlikely to gain
any `continue` or `break`s.
Suggested by: pstef
Differential Revision: https://reviews.freebsd.org/D37676
The previous logic conflated some things... in this block:
- j: input characters rendered so far
- nc: number of characters in the line
- col: columns rendered so far
- hw: column width ((h)ard (w)idth?)
Comparing j to hw or col to nc are naturally wrong, as col and hw are
limits on their respective counters and nc is already brought down to hw
if the input line should be truncated to start with.
Right now, we end up easily truncating lines with tabs in them as we
count each tab for $tabwidth lines in the input line, but we really
should only be accounting for them in the column count. The problem is
most easily demonstrated by the two input files added for the tests,
the two tabbed lines lose at least a word or two even though there's
plenty of space left in the row for each side.
Reviewed by: bapt, pstef
Sponsored by: Klara, Inc.
Differential Revision: https://reviews.freebsd.org/D37676
Instead of providing no /usr/bin/objdump when LLVM_BINUTILS is false.
PR: 267854 [exp-run]
Reviewed by: dim
MFC after: 2 weeks
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D37445
For the TCP protocol inpcb storage specify allocation size that would
provide space to most of the data a TCP connection needs, embedding
into struct tcpcb several structures, that previously were allocated
separately.
The most import one is the inpcb itself. With embedding we can provide
strong guarantee that with a valid TCP inpcb the tcpcb is always valid
and vice versa. Also we reduce number of allocs/frees per connection.
The embedded inpcb is placed in the beginning of the struct tcpcb,
since in_pcballoc() requires that. However, later we may want to move
it around for cache line efficiency, and this can be done with a little
effort. The new intotcpcb() macro is ready for such move.
The congestion algorithm data, the TCP timers and osd(9) data are
also embedded into tcpcb, and temprorary struct tcpcb_mem goes away.
There was no extra allocation here, but we went through extra pointer
every time we accessed this data.
One interesting side effect is that now TCP data is allocated from
SMR-protected zone. Potentially this allows the TCP stacks or other
TCP related modules to utilize that for their own synchronization.
Large part of the change was done with sed script:
s/tp->ccv->/tp->t_ccv./g
s/tp->ccv/\&tp->t_ccv/g
s/tp->cc_algo/tp->t_cc/g
s/tp->t_timers->tt_/tp->tt_/g
s/CCV\(ccv, osd\)/\&CCV(ccv, t_osd)/g
Dependency side effect is that code that needs to know struct tcpcb
should also know struct inpcb, that added several <netinet/in_pcb.h>.
Differential revision: https://reviews.freebsd.org/D37127
mtflag is used to add pthread mutex locking around operations to make
them thread-safe. Setting the state to _SERVED is not conditional on
locking.
Reviewed by: imp, emaste
Differential Revision: https://reviews.freebsd.org/D37541
GCC 12 defaults to C++17 which removes (not just deprecates)
std::auto_ptr<>. Trying to use CXXSTD of c++03 doesn't work with
libc++ headers, but c++11 does.
Reviewed by: brooks, imp, emaste
Differential Revision: https://reviews.freebsd.org/D37531
trpt(8) was utility to pull TCP debugging data from the kernel
originating back from 4.2BSD. It is not used nowadays by TCP
developers. We have more powerful debugging facilities, e.g.
the Dtrace probing, the TCP black box logging and siftr.
Discussed with: rscheff, tuexen, rrs, jtl and others
I think this was probably just a typo. initifstat() continues around
a similar loop if the mib data fails to fetch, and fetch_ifstat() was
already using a FOREACH_SAFE loop here so expected to keep going.
Calling clearifstat() from the fetch routine also seems wrong, and the
sort_interface_list() call triggered by the existing needsort = 1 will
itself set needclear to trigger a future clearifstat().
Reported by: GCC 12 -Wuse-after-free
Differential Revision: https://reviews.freebsd.org/D36823
A small reduction in build infrastructure complexity; when we had both
Clang and GCC in the tree it was useful to have both built, and choose
one or the other to install as /usr/bin/cc. Now only Clang is in the
tree, and there is no point in building and installing base Clang but
not providing it as cc (and c++, cpp).
Reviewed by: imp
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D37075