If the segment list is full, print a warning on the console and launch the
system with the available segments.

High memory systems may have more than VM_PHYSSEG_MAX segments; it is
better to truncate the memory and allow the system to work rather than
just panicking. The user can still increase VM_PHYSSEG_MAX (or ask us to).

Fixes issues such as PR/47093.

Note: the warning is logged but does not appear in dmesg, this too needs
to be fixed for the rest of the bootstrap procedure.

(maxv)

Use macros instead of hard-coded constants. By the way, I don't think this
code is correct, but whatever.

(maxv)

Import iomem_ex locally.

(maxv)

speed limit is 80 (missed this one)

(maya)

strncpy->strlcpy. use sizeof instead of numbers.

(maya)

ixg(4): Support TX/RX multiqueue

(msaitoh)

Change the Interrupt Moderation flag from global variable to per device.

(msaitoh)

- Both mbuf_defrag_failed and m_defrag_failed existed. Remove m_defrag_failed.
- rx_irq, dropped_pkts, morerx, moretx and txloops counters are uncounted, so
  remove them.
- Count handleq in ixv_handle_que().
- Detach event counters correctly.
- Set some per-queue event names correctly.

(msaitoh)

Remove checks for cursor being in scrolling region, ncurses doesn't
do checking, rather just scrolls the scrolling region so we shall
follow suit.  SUSv2 says what is happens when scrl is called with the
cursor outside the scrolling is undefined so we should match ncurses.
This fixes PR#51819 without forcing tin to use terminfo directly.

(blymn)

No functional change:
- Rename ixv_sysctl_attach() to ixv_add_device_sysctls().
- Move sysct and evcnt related functions to match ixgbe.c.

(msaitoh)

kill variable stack allocation

(christos)

Add many entries

(ozaki-r)

detach wm(4) evcnt.

pointed out by msaitoh@n.o, thanks.

(knakahara)

ifdef out some dead code. improves code readability.
all later users of y first assign another value.

using ifdefs to make potential future code syncs easier, as is done
elsewhere.

suggested by coverity, CID 1300929, 1300930.

(maya)

regen

(christos)

revert listen

(christos)

use proper arg functions, remove bogus ones.

(christos)

no more warnings for ld.

(christos)

use mkstemp to prevent linker warning.

(christos)

Appease static analyzers by making all code paths which assign
values into p,q sane. Get rid of redundant assignment. Indent
for legibility. NFC.

This doesn't create a functional difference, as all callers
test number >= 0x40000000 anyway.

To see this, note the following:
- consistently, hx is the high bits of x, lx is the low bits,
  x is the float.
- & 0x7fffffff zeroes the sign bit, as does fabs.

A case where it isn't easy to see that there's no functional
change is y1, which does:

ix = hx & 0x7fffffff (zero signbit of high bits of x)
y = fabs(x) (this has a zeroed signbit but otherwise same as x)
ix >= 0x40000000
  pone(y); qone(y)

qone(x) (also pone) do:
  ix = hx & 0x7fffffff

ix in qone and in the calling function are the same number,
and the comparison applies for both, and ix < 0x40000000 isn't
possible.

(Also, no explosions seem to happen when I feed it random numbers)

(maya)

fix tpyo

(macallan)

Turn off fatal linker warnings for now due to mktemp.

(joerg)

Mark help as static+dead. Don't use __attribute__.
XXX needs a lot more static love. Uses mktemp and doesn't link.

(joerg)

make fallback to PIO actually work properly

(macallan)

If the preloaded modules cannot be mapped with the initial amount of VA,
discard the associated bootinfo entry. Otherwise the machine faults and
reboots immediately.

I spotted this bug more than a year ago, but I recently saw that there is
already PR/42645 (7 years old), so just fix it. The size has been increased
in the meantime, so the limit is unlikely to be reached anyway.

(maxv)

Bump the ram disk and miniroot size slightly for clang.

(joerg)

Switch to the 4.0 release branch and update build glue for r294123.

(joerg)

Mark files not in r294123 as dead.

(joerg)

the tls ifdef was a bit too aggressive.

(christos)

Give clang a bit more space on the RAM disk.

(joerg)

loosen the belt again.

(christos)

put back exect directory; we try to remove it before the files inside it and
we fail.

(christos)

fix build (MKUNBOUND=no)

(szptvlfn)

efi_md::md_virt always uses uint64_t.

(nonaka)

PR kern/51280

This allows srt devices to work for IPv6.  srt still needs work
(particularly #ifdef INET6 but also general effeciency and similar.)

(kre)

Make bpf MP-safe

By the change, bpf_mtap can run without any locks as long as its bpf filter
doesn't match a target packet. Pushing data to a bpf buffer still needs
a lock. Removing the lock requires big changes and it's a future work.

Another known issue is that we need to remain some obsolete variables to
avoid breaking kvm(3) users such as netstat and fstat. One problem for
MP-ification is that in order to keep statistic counters of bpf_d we need
to use atomic operations for them. Once we retire the kvm(3) users, we
should make the counters per-CPU and remove the atomic operations.

(ozaki-r)

No, do not just copy code from i386 and expect it to work on amd64. There
are several structural differences. At least two issues here: segment
registers that could fault in kernel mode with userland TLS, and a non-
canonical %eip on iret.

Not even tested, but just obvious. By the way, I believe this function is
still buggy since we don't call cpu_fsgs_reload while %fs/%gs could have
been reloaded.

(maxv)

Restore %ds before swapgs. Movs to segment registers are allowed to fault
in kernel mode but simply cause a signal to be sent to userland. The thing
is, in this case %gs is not restored when entering the trap routine, which
means the kernel uses userland's TLS instead of using its own. Which in
short makes it easy to escalate privileges.

Currently, this bug is triggered only in one place, which I am about to
fix too.

(maxv)

make some locks and condvar static

(knakahara)

Revert the part of mkfs.c 1.36 that "gutted the const" (while
retaining the part that added a different one).

That is, re-constipate makefs (well, just a bit, no real pain here.)

(kre)

Sprinkle in a pinch of const, not too much, just enough
to add a little strength without affecting the overall balance...

(kre)

Supress verbose message "This pci host supports neither MSI nor MSI-X."
on VMware and KVM. OK'd by k-nakahara.

(msaitoh)

remove exect test

(christos)

remove exect, it will not be fixed and it will be removed.

(christos)

regen

(christos)

fix listen

(christos)

resolve conflict

(christos)

Add tests for several bpf ioctls

(ozaki-r)

new openldap

(christos)

merge conflicts

(christos)

new bind

(christos)

Merge 9.10.4-P6
4558.  [bug]          Synthesised CNAME before matching DNAME was still
                        being cached when it should have been.  [RT #44318]

4557.  [security]      Combining dns64 and rpz can result in dereferencing
                        a NULL pointer (read).  (CVE-2017-3135) [RT#44434]

(christos)

Cosmetic changes. No functional changes.

(rin)

Cosmetic changes. No functional changes.

(rin)

Replace ".PATH ${SRCDIR}" before including Makefile.common to ensure the
correct order of search path.

(rin)

Factor out a common part in Makefile's of ifconfig and x_ifconfig.

(rin)

Rename sbin/ifconfig/Makefile.inc to Makefile.common; the former can
potentially conflict with bsd.subdir.mk.

(rin)

Remove ifconfig hack on m68000. I've confirmed that the problem has already
gone with TME.

(rin)

Fix build with MKDEBUG=yes for sun2:

- libgcc_g.a, libgcov_g.a, libobjc_g.a, libstdc++_g.a, and libsupc++_g.a are
  marked "gcc" instead of "gcccmds".

- libbfd_g.a and libh_helper_*_g.a are moved from debug/mi to debug/shl.mi.
  This seems strange at first sight, but they are built only when shared
  libraries are enabled.

(rin)

Do not use the assembler version of AES routines for m68000; they contain
instructions available for 68020 and later.

Fix sun2 build.

(rin)

If we are using a timestamp from the command line, don't pay attention to
the user timezone, use UTC instead (for reproducible builds).

(christos)

stampst.st_ion consistent treatment (non-zero vs zero instead of testing == 1)

(christos)

minor bump libc for accept4

(maya)

Shrink the ramdisk of INSTALL kernel for zaurus in order to fit within 5MB.

Now free spaces of the main kernel and ramdisk are about 62KB and 215KB for
INSTALL, respectively.

OK christos

(rin)

Analagous fix to that just committed to the equiv amd64 sources.
Note: this one has yet to be compile tested, so anything is possible...

(kre)

All bow down before the Great Code Compiler.

Specifically, it has been commanded that unused-functions is a warning,
and, as all should know, all warnings be errors.

Hence, since XEN kernels do not call set_sys_gdt(), that function
may not be included (and the XEN kernel  saves a few tens of bytes.)

An alternate fix would be to just remove "static" - but that would be
the equivalent to just giving the compiler the finger ...  hmm, maybe
that would have been a better fix...

(kre)

Document sets programmatically instead of keeping a list that constantly is
out of date.

(christos)

__empty -> __nothing

(rin)

oops, do not write the same file twice -- accidentally ran patch twice

(maya)

Document accept4 in accept(2)

(maya)

__empty -> __nothing

(christos)

empty to nothing

(christos)

use __nothing instead since __empty is taken in stl_deque.h

(christos)

Add accept4, a tiny wrapper around paccept.

accept4 is a syscall in Linux, FreeBSD and OpenBSD. It is used in
LLVM, zeromq, and probably others. paccept is a superset of it.

adding it to libc ensures it is used by programs and prevents the
need to define the same wrapper in every program.

(maya)

nsd is optional

(christos)

Use strncpy here because we need to be portable, and we have the +1 byte
to save us :-)

(christos)

Make paccept a weak symbol in preparation for using it in another function

(maya)

switch to __empty

(rin)

remove comment in comment

(christos)

use __empty

(christos)

need FILESBUILD_

(christos)

don't define __empty.

(christos)

__empty is defined in <sys/cdefs.h>

(christos)

add __empty

(christos)

Switch to x_fsck_ffs and x_newfs. Fix floppy image overflow.

(rin)

Add smaller versions of fsck_ffs(8) and newfs(8) for install media, where
support for Endian-Independent FFS and Apple UFS is disabled unless FFS_EI=1
and APPLE_UFS=1 are added to CRUNCHENV, respectively.

This reduces the size of ramdisk image for atari by over 15KB.

Thanks tsutsui and christos for their useful comments.

(rin)

gut const for now.

(christos)

Add new entry in TODO.ptrace

Added:
    add PT_SET_SIGMASK and PT_GET_SIGMASK - used by checkpointing software

Example software: gdb, criu

(kamil)

Time waro (thanks Kamil)

(kre)

Ammend previous, mention compat/amd64/i386/lib/libc as well.

(kre)

Omit duplicate text.

(riastradh)

Note libc update build issue after change from arch/exect.S to exect.c
(The build system really should be able to cope with changes like this
one without obj dir fiddling or clean builds)

(kre)

Remove gdt_reload_cpu. GDTR takes a VA as base, and in our x86
implementation this VA is per-cpu and does not change; there is therefore
no need to remotely reload GDTR.

(maxv)

ixv(4): Add Intel 10G Ethernet virtual function driver.

(msaitoh)

Localify, add a comment and merge some others.

(maxv)

Print CPU number correctly.

(msaitoh)

- Fix aprint_*()s in ixv_allocate_msix().
- Style fix. Reduce diff against ixgbe.c. No functional change.

(msaitoh)

Call IFQ_SET_MAXLEN() even if TX multiqueue is used. ALTQ uses it.

(msaitoh)

Enable TX multiqueue by default on both ixg(4) and ixv(4).

(msaitoh)

Enable multiqueue.

(msaitoh)

Use dmat64 if available. Same as ixgbe.c rev. 1.60.

(msaitoh)

Fix a bug that 1Gbps is printed even if the phyiscal media is at 100Mbps.

(msaitoh)

Don't track /etc/nsd/nsd.conf and /etc/unbound/unbound.conf

(christos)

Fix detach path:
- Call softint_disestablish() and pci_intr_disestablish() for all queues
  correctly.
- Add missing sysctl_teardown() to delete sysctl entries.
- Call pci_intr_release(). Now NETBSD_MSI_OR_MSIX is completely deleted.

(msaitoh)

Don't store random (from ASLR) pointers into the superblock. Should be the
last (famous last words) problem with reproducible builds!

(christos)

Add queue number into interrupt name.

(msaitoh)

pthread_dbg(3): Removed from the base distribution.

(kamil)

Set ifm_active to IFM_NONE when the physical interface's link is down.
Now ifconfig shows "media: Ethernet autoselect '(none)'".

(msaitoh)

libpthread_dbg(3) deletion from the base distribution

libpthread_dbg(3) is a remnant library from the M:N thread model
(pre-NetBSD-5.0) API to introspect threads within a process and for use
of debuggers.

Currently in the 1:1 model it's not used in GDB neither in LLDB and it's
not either planned to be used. It's current function to read pthread_t
structures is realizable within a regular debugger capable to
instrospect objects within a tracee (GDB, LLDB...).

Remaining users of this API can still use this library from
pkgsrc/devel/libpthread_dbg.

Sponsored by <The NetBSD Foundation>

(kamil)

no, this is wrong.

(christos)

zero out the superblock so that it does not contain random stuff in the
spare fields.

(christos)

Detach libpthread_dbg(3) from gdb(1) and gdbtui (gdb.old)

libpthread_dbg(3) is a remnant library from the M:N thread model
(pre-NetBSD-5.0) API to introspect threads within a process and for use
of debuggers.

Currently in the 1:1 model it's not used in GDB neither in LLDB and it's
not either planned to be used. It's current function to read pthread_t
structures is realizable within a regular debugger capable to
instrospect objects within a tracee (GDB, LLDB...).

pthread_dbg(3) is going to be removed from the base distribution and moved
to pkgsrc.

Patch reviewed by <christos>

Sponsored by <The NetBSD Foundation>

(kamil)

Detach libpthread_dbg(3) from gdb(1) and gdbtui

libpthread_dbg(3) is a remnant library from the M:N thread model
(pre-NetBSD-5.0) API to introspect threads within a process and for use
of debuggers.

Currently in the 1:1 model it's not used in GDB neither in LLDB and it's
not either planned to be used. It's current function to read pthread_t
structures is realizable within a regular debugger capable to
instrospect objects within a tracee (GDB, LLDB...).

pthread_dbg(3) is going to be removed from the base distribution and moved
to pkgsrc.

Patch reviewed by <christos>

Sponsored by <The NetBSD Foundation>

(kamil)

add config files as examples

(christos)

don't overwrite etc files.

(christos)

add nsd and unbound example dirs

(christos)

add nsd and unbound example dirs

(christos)

PR/51958: Ngie Cooper: ESRCH error description in hcreate(3) contains typo
for action

(christos)

Kernel build objdirs have a kernel name after the objdir. Use a regex for
that. Note that debug-prefix-map and debug-regex-map insert entries in a
linked list in reverse order.

(christos)

Use make -C instead of cd && make; this is done to preserve the logical path
for debugging info DW_AT_comp_dir

(christos)

use the logical path for -C if correct.

(christos)

Mark exect(3) obsolete and bind it to plain execve(2) on all platforms

The original exect(2) from BSD4.2 was enabling bit for tracing
(single-step mode) and calling execve(2). The purpose of it was to generate
a signal for a tracer once the application will change its image to a new
program.

This approach no longer works as:
- exect(2) traces (single-steps) libc and it requires hundreds or
  thousands steps before entering a new image
- it's vax and x86 specific code
- this functionality has been moved to the kernel - once a process is
  traced it will generate SIGTRAP with si_code TRAP_EXEC and route it to
  its debugger
- the side effects and unportability make this interface unusable
- there are no known users of this interface
- it apparently never worked better since day0 of NetBSD ("day0 bug")

Users are requested to move to other execve(2) variants. Calling current
execve(2) as it is the most similar behavior to this one from BSD4.2.

Discussed several times on mailing lists and in PR/51700.

Add warning to exect(3) telling about marking this function obsolete.

This function is prepared to be removed in next libc major bump.

Sponsored by <The NetBSD Foundation>

(kamil)

Define NOMAN before including bsd.own.mk to set MKMAN=no appropriately.

(rin)

Update TODO.ptrace

Note PT_WATCHPOINT change to PT_*ETDBREGS.
Remove GDB and LLDB related entries from generic ptrace(2) TODO.
Note need for TRAP_SCE and TRAP_SCX si_codes in PT_SYSCALL*.
Note removal request of pthread_dbg(3).

Sponsord by <The NetBSD Foundation>

(kamil)

Do not compile in progress.c when SMALLPROG is defined.

(rin)

Document in libc's shlib_version request for exect(3) removal

This functionality has been moved long time to the kernel. Kernel exec()
emits SIGTRAP under a debugger (unless PT_SYSCALL traced).

(kamil)

Implement m68k assembly version of AES.
It's approx 1.4 times faster than the original one.

(isaki)

KNF and trailing whitespace.  No functional change.

(skrll)

Fix non-DIAGNOSTIC build

(skrll)

Set IFM_10G_T on 10G...

(msaitoh)

Fix link status stuff:
- Set get_link_status in ixv_init_locked() to get link status correctly.
- Don't discard checked speed in ixv_config_link()...

(msaitoh)

Fix device attach:
- Even if ixgbevf_negotiate_api_version() failed, continue attaching.
  Same as Linux. Tested on ESXi 6.0.
- Fix argument of pci_msix_alloc_exact().
- Use different intr_xname for each interrupt.
- Add missing pci_intr_release() on error.
- Add missign kcpuset_destroy().
- Calculate MSI-X vectors correctly.

(msaitoh)

Add missing NULL checks for m_get_rcvif

(ozaki-r)

Tweak m_get_rcvif

Call pserialize_read_exit if if_byindex returns NULL in m_get_rcvif.
By changing so, callers need to call m_put_rcvif only if m_get_rcvif
returns non-NULL.

Old m_get_rcvif/m_put_rcvif could forget to call pserialize_read_exit,
which is pointed out by maxv@. The change fixes it.

(ozaki-r)

Use m_get_rcvif_psref instead of m_get_rcvif

Because the critical sections are now sleepable.

Reviewed by knakahara@

(ozaki-r)

Ensure the vc block size is set before call to set defaults.

It should address PR kern/51746 Panics from audio.

(nat)

Use ELFOSABI_LINUX instead of a magic number.

(uwe)

Remove __DATE__! Intel really likes non-reprodible builds!

(christos)

white space

(rin)

+ /usr/bin/nc    (someone please check these are correct)

(kre)

In cpu_mcontext32_validate, allow the registers to have different locations
if the LDT is user-set.

I am intentionally not allowing this in check_sigcontext32, because I don't
think Wine uses it.

(maxv)

Use full month name in Dd.

(wiz)

mention netcat (nc)

(christos)

add nc

(christos)

Add nc.

(christos)

- Disable crypto for now, and functionality that we don't provide.
- Fix warnings

(christos)

Add the USER_LDT sysarch options in netbsd32. We don't translate 'desc',
since if we ever implement USER_LDT we will only allow 8-byte-sized
entries, which have the same layout on amd64 and i386.

(maxv)

Remove unnecessary flag.

(nonaka)

Remove useless checks

(skrll)

Sprinkle const and reduce scope of a couple of variables.

(skrll)

Whitespace

(skrll)

1352

(snj)

1352

(snj)

Apply patch (requested by spz in ticket #1352):
Fix CVE-2016-7056 and CVE-2017-3731.

(snj)

Apply patch (requested by spz in ticket #1352):
Fix CVE-2016-7056 and CVE-2017-3731.

(snj)

Fix misuse of arc4random by using arc4random_uniform.
Noticed by riastradh.

(dholland)

sun4v: update TODO list to reflect the current state - TRAP_SETUP() was adapted to handle both sun4u and sun4v systems in version 1.404 of locore.s

(palle)

more sanity checks on the mbuf we get.

(bouyer)

1355, 1356

(snj)

Pull up following revision(s) (requested by maxv in ticket #1356):
sys/netinet/if_arp.c: revision 1.238, 1.239 via patch
Make sure the protocol address length equals that of IPv4. Also, make sure
the hardware address length equals that of the interface we received the
packet on. Otherwise a packet could easily set them both to zero and make
the kernel read beyond the allocated mbuf, which is terrible.
Note: for the latter we drop the packet instead of replying, since it is
malformed.
Note: I also added an ugly hack in CARP, since it apparently expects at
least six bytes.
--
Add some checks, mostly same as in_arpinput.

(snj)

1355, 1356

(snj)

Pull up following revision(s) (requested by maxv in ticket #1356):
sys/netinet/if_arp.c: revision 1.238, 1.239 via patch
Make sure the protocol address length equals that of IPv4. Also, make sure
the hardware address length equals that of the interface we received the
packet on. Otherwise a packet could easily set them both to zero and make
the kernel read beyond the allocated mbuf, which is terrible.
Note: for the latter we drop the packet instead of replying, since it is
malformed.
Note: I also added an ugly hack in CARP, since it apparently expects at
least six bytes.
--
Add some checks, mostly same as in_arpinput.

(snj)

Pull up following revision(s) (requested by maxv in ticket #1355):
sys/net/if_arcsubr.c: revision 1.76 via patch
sys/net/if_ecosubr.c: revision 1.50 via patch
sys/net/if_ethersubr.c: revision 1.236 via patch
sys/net/if_fddisubr.c: revision 1.104 via patch
sys/net/if_tokensubr.c: revision 1.80 via patch
Don't forget to free the mbuf when we decide not to reply to an ARP
request. This obviously is a terrible bug, since it allows a remote sender
to DoS the system with specially-crafted requests sent in a loop.

(snj)

Pull up following revision(s) (requested by maxv in ticket #1355):
sys/net/if_arcsubr.c: revision 1.76 via patch
sys/net/if_ecosubr.c: revision 1.50 via patch
sys/net/if_ethersubr.c: revision 1.236 via patch
sys/net/if_fddisubr.c: revision 1.104 via patch
sys/net/if_tokensubr.c: revision 1.80 via patch
Don't forget to free the mbuf when we decide not to reply to an ARP
request. This obviously is a terrible bug, since it allows a remote sender
to DoS the system with specially-crafted requests sent in a loop.

(snj)

Introduce can_ifattach(), for common setups of can interfaces.
Convert output to a ifq and change canloop_output to canloop_ifstart

(bouyer)