Pull up following revision(s) (requested by maxv in ticket #1432):
sys/netinet/if_arp.c: 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 #1432):
sys/netinet/if_arp.c: 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)

Apply patch (requested by maxv in ticket #1431):
suword: Don't allow 4 bytes to overflow beyond the userland space.

(snj)

Apply patch (requested by maxv in ticket #1431):
suword: Don't allow 4 bytes to overflow beyond the userland space.

(snj)

Apply patch (requested by maxv in ticket #1431):
suword: Don't allow 4 bytes to overflow beyond the userland space.

(snj)

Apply patch (requested by spz in ticket #1355):
Fix CVE-2017-3731.

(snj)

Apply patch (requested by spz in ticket #1355):
Fix CVE-2017-3731.

(snj)

Apply patch (requested by spz in ticket #1355):
Fix CVE-2017-3731.

(snj)

Pull up following revision(s) (requested by maxv in ticket #1429):
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 #1429):
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 #1429):
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)

merge for tcpdump 4.9.0 and document

(spz)

uh, this is actually called 20170201 :-)

(mrg)

Add the *_size() variants to the .Nm list so it will be found by
apropos(1).

While here, reorder the function lists to improve readability.

(pgoyette)

remove debugging printf.

(christos)

Remove trailing space.

(wiz)

Quote - to make it a minus.

(wiz)

Use just the right amount of commas.

(wiz)

Factor out reading from a can socket, and move to a helper file.

(bouyer)

regen

(christos)

fix broken entries that prevented compilation.

(christos)

tag variables as __unused to appease compiler.

(mlelstv)

Fix operand width

(christos)

Don't error out for set but unused variables for now.

(christos)

Mark reboot() and _rtt() as __dead.
Use __builtin_unreachable() to ensure Panic really is marked dead.

Thanks to joerg@

(roy)

merge expat-2.2.0 and document

(spz)

More usb_syncmem tweaks.  The main one is sync'ing the entire last TD
returned from ohci_reset_std_chain in control transfers.

(skrll)

fix the prototype for bozo_set_pref().  from cfkoch

(mrg)

Fiddle the (new) fdflags implementation:

Remove some unnecessary cuteness that limited error reporting.
Permit just one -s arg to fdflags
Be deterministic in the case of fdflags -s +cloexec,-cloexec 0
(and similar) - use the last specified, always.
Allow:
FD_0_FLAGS=$( fdflags -v 0 )
# do stuff, manipulating the flags
fdflags -s "FD_0_FLAGS" 0
to save/restore flags for a fd.
Correctly mask result of fcntl(fd, F_GETFD) with FD_CLOEXEC as the
specs require before deciding close on exec is set.

Improve portability as a tool, don't assume strtoi(), nor __arraycount()
and avoid needlessly requiring recent C versions (ie: there's no need to
sprinkle declarations in the middle of the code, it just makes them hard
to find, and benefits nothing.)

Still to do:  As currently implemented, both user, and shell internal fds
are reported, and can be manipulated.  Allowing users to touch the shell's
internal fds is bogus, and providing this easy way to allow users to
discover which values they have is poor.  Fixing this means getting rid
of the use of fcntl(F_MAXFD) and replacing it with a shell maintained
memory of what fds the user (script) has allocated.  The shell's fd
manipulation really still needs major work (including properly fixing
bin/48875)

(kre)

New versions of linux on arm and others have both socketcall and separate
syscalls...

(christos)

use the netid value because it could be udp6...

(christos)

Add posix_madvise(2) in the NAME section.

(abhinav)

Insert comma between two Nm entries.
Remove Pp before Bd

(abhinav)

Don't check hardware precision in autoconfig as some drivers e.g vs(4) the
hardware precision is 4 bits.

Addresses PR/kern 51879: vs(4) audio attach failed.

(nat)

some bugs in chipsfb_putchar_aa():
- fix 64bit padding for scanlines
- clean out latch for every scanline
now fonts with odd widths look right

(macallan)

Improved drain function for vc[0].
No need to drain vc[0] for drivers written with start_output.

(nat)

Fix build with clang.

(roy)

Fix build with clang.

(roy)

add sendmmsg and recvmmsg

(christos)

regen

(christos)

add unimple sendmmsg secvmmsg

(christos)

regen

(christos)

add missing unimpl and sendmmsg and recvmmsg

(christos)

regen

(christos)

and send and recv mmsg and unimpl

(christos)

rows is a uint ... it will always be greater than or equal to 0.

(roy)

regen

(christos)

add send recv mmsg and unimpl
XXX: fix base?

(christos)

regen

(christos)

add sendmmsg and recvmmsg

(christos)

regen

(christos)

add send/recv mmsg and comment on the unimpl ones.

(christos)

expose sendmsg_so and recvmsg_so.

(christos)

update for addition (a while ago) of features2 page

(plunky)

handle accept4 for i386

(christos)

Do not compile accept4 on i386, there is no such syscall in Linux.

(martin)

Wrap a long line

(skrll)

Remove comma after the last Nm entry.

(abhinav)

regen

(martin)

Add accept4

(martin)

Sync with HEAD

(skrll)

regen

(martin)

Add accept4

(martin)

Simplify.  No functional change.

(skrll)

Use Xr to refer fnmatch(3).

(abhinav)

Regen for accept4

(martin)

Add accept4 definitions

(martin)

Add test program for PR kern/51762

(martin)

PR kern/51762: add a test program

(martin)

Pass kernend as the start parameter to uvm_page_physload().  This
fixes pmap_steal_memory() as it can now unplug from the start of the
segment.  This might be the right thing to do too as the pages before
kernend are accessed via directly mapped P1, but i need to ask someone
with a clue.  This is also what evbsh3 always did.

Anyway, this at least unbreaks landisk after uvm hotplug fallout.

(uwe)

PR port-mac68k/51923
Bump default MAXTSIZ to 32MB for m68k ports, which allos us to use GCC 5.4 on
GENERIC kernels.
Ok jklos

(rin)

PR port-mac68k/51923
Remove unused SYSPTSIZE and USRPTSIZE from m68k ports.
Ok jklos

(rin)

pmap_steal_memory - a few more s/start/avail_start/ fixes for the
hotplug conversion.

(uwe)

Add missing spaces in split string literals.

(uwe)

pmap_steal_memory - s/start/avail_start/ to fix mistake in the hotplug
conversion.  The kernel now panics (can't unplug memory from the
middle of a segment) instead of triggering a reset.

(uwe)

It's completely bogus to pass physmem (i.e. the memory size) as the
"start" parameter to uvm_page_physload().  I guess it never exploded
only because it's the correct "avail_start" that is used.

(uwe)

add +

(christos)

set 7 and 8 to close on exec.

(christos)

Add fdflags builtin. Discussed with Chet and he has implemented it for
bash too.

(christos)

support anti-aliased fonts, glyphcache etc.

(macallan)

chipsfb needs glyphcache now

(macallan)

Who Ride Wit Us?

(christos)

Fix these comments, we probably won't want to keep them up to date.

(maxv)

Increase KERNTEXTOFF from 1MB to 2MB on amd64. [1MB; 2MB[ is now handled
by UVM, so there is no physical loss.

On amd64 we always remap the kernel text with 2MB pages, and because of the
1MB start address we were forced to map [0MB; 2MB[ inside the first large
page. The problem is, the lower half is used by UVM to allocate physical
pages, and it is possible that some of these could be used by userland. We
could end up with userland-controllable data mapped into the kernel text on
a privileged page, which is far from being a good idea from a security pov.

I am not fixing i386 yet, because the large page size depends on PAE, and
we probably don't want to have a text located at 4MB on low-memory systems.

(note: I didn't introduce this issue, it was already there when I came in)

(maxv)

declare sockaddr_big;

(christos)

The first va should depend on the text offset, not the kernel base. Use
rounddown. Note: this value is still wrong, it should be roundup. But
that's another issue that will be fixed in amd64 soon.

(maxv)

expose copyout_sockname_sb

(christos)

implement accept4

(christos)

regen

(christos)

implement accept4
XXX: mips looks busted (the syscalls file has lots of NOARGS syscalls that
should be STD).

(christos)

Fix a typo : pthread_mutexaddr_init -> pthread_mutexattr_init.

(njoly)

remove WM_CORE_LOCK in deferred start routine.

It is enough to check each txq's txq_stopping with
mutex_enter(not mutex_tryenter).

reviewed by ozaki-r@n.o.

(knakahara)

wlan interfaces make interrupt routine running on softint context.

see http://mail-index.netbsd.org/tech-kern/2016/12/06/msg021281.html

tested device:
* ath at pci: AR5212, AR5424
* athn at pci: AR9287
* ipw at pci: 2100BG
* iwi at pci: 2915ABG
* iwm at pci: 3165, 7260, 8260
* iwn at pci: 4945, 6235
* ral at pci: RT2560
* rtwn at pci: RTL8192CE

(nonaka)

Use __read_mostly on these variables, to reduce the probability of false
sharing.

(maxv)

PR port-sparc64/51925: cosmetic cleanup, no functional change

(martin)

Style / more const.  No functional change intended.

(skrll)

Whitespace

(skrll)

Use lower case

(skrll)

Remove useless check

(skrll)

Sprinkle some const

(skrll)

Add a TAB

(skrll)

Modify comment. Use long form.

(msaitoh)

Fix typo in comment.

(msaitoh)

wpi(4): use MSI if available.

(jakllsch)

iwn(4): use MSI if available.

(nonaka)

Defer some pr_input to workqueue

pr_input is currently called in softint. Some pr_input such as ICMP, ICMPv6
and CARP can add/delete/update IP addresses and routing table entries. For
example, icmp6_redirect_input updates an a routing table entry and
nd6_ra_input may delete an IP address.

Basically such operations shouldn't be done in softint. That aside, we have
a reason to avoid the situation; psz/psref waits cannot be used in softint,
however they are required to work in such pr_input in the MP-safe world.

The change implements the workqueue pr_input framework called wqinput which
provides a means to defer pr_input of a protocol to workqueue easily.
Currently icmp_input, icmp6_input, carp_proto_input and carp6_proto_input
are deferred to workqueue by the framework.

Proposed and discussed on tech-kern and tech-net

(ozaki-r)

put back the include of dtrace_bsd.h, it actually is needed.

(chs)

fix this module to depend on the solaris module rather than the dtrace module.
remove an unneeded header.

(chs)

add some missing bits from freebsd r211554:
remove the DOODAD from prochandler().
we are now DOODAD-compatible with current freebsd.

(chs)

implement proc_getlwpstatus() for netbsd using PT_GET_SIGINFO.

(chs)

Use boolean logical and operation on booleans, rather than bitwise and.

From David Binderman in PR kern/51936.

(jakllsch)

Add missing entries in the NAME section.

These functions are described in the body of the man page but not
mentioned in the NAME section. Without an entry in the NAME section
apropos(1) doesn't get this man page in the results when queried for
one of those functions.

ok from christos.

(abhinav)

Since we are avoiding VAR_INTERNAL, allow the variable :_ stores to
to be specified, also allows for multiple stages of modification to
be stashed.

(sjg)

PR/51935: David Binderman: bad call to printf
- fix debugging printfs in general.
- KNF

(christos)

Just have :_ use current ctxt, avoid need to force VAR_INTERNAL.

(sjg)

Not sure what we are trying to achieve here, but there are two issues;
error can be printed while it is not initialized, and if m_pulldown fails
m is freed and reused.

Quickly reviewed by christos and martin

(maxv)

Add comma between two Nm entries in the NAME section.

(abhinav)

Add modff and modfl in the NAME section.

(abhinav)

Fix a sentence.
Correct function name in the DESCRIPTION section.

(abhinav)

Also, fix spelling of described.

(abhinav)

Fix sentence.

(abhinav)

Remove comma after last Nm entry.

(abhinav)

Remove comma after the last Nm entry.

(abhinav)

PR/51933: David Binderman: Boolean confusion

(christos)

match the man page, and explain why.

(christos)

regen

(christos)

Add evbuffer.h.3

(christos)

now we have evbuffer.h.3 and ebuffer.3 so handle it carefully.

(christos)

TX multiqueue. If you want to disable it, enable IXGBE_LEGACY_TX
in ixgbe_netbsd.h

(msaitoh)

Restrict special mode/owner settings to truncate_test.root_owned;
otherwise the Atffile will be generated unreadable for group/others
as well.

(martin)

Update supporting files for components which rely on autoconf to allow systems
introducing since release of software to be recognised. This should hopefully
allow the builds to progress a littles further on systems such as the POWER8
which features a little endian 64-bit PowerPC CPU identified as ppc64le.

(sevan)

Use new wm_gmii_setup_phytype() function to setup sc_phytype and
mii_{read|write}reg. This function is called twie.

To identify PHY type, correct read/write function should be selected. To
select correct read/write function, PCI ID or MAC type are required without
accessing PHY registers.

On the first call of this function, PHY ID is not known yet. Check PCI ID or
MAC type. The list of the PCI ID may not be perfect, so the result might be
incorrect.

In the second call, PHY OUI and model are used to identify PHY type. It might
not be perfpect because of the lack of compared entry, but it would be better
than the first call.

If the detected new result and previous assumption is different, diagnous
message will be printed.

(msaitoh)

Reduce return points

(ozaki-r)

Kill tsleep/wakeup and use cv

(ozaki-r)

Make bpf_gstats percpu

(ozaki-r)

Use pslist(9) instead of queue(9) for psz/psref

As usual some member variables of struct bpf_d and bpf_if remain to avoid
breaking kvm(3) users (netstat and fstat).

(ozaki-r)

Use kmem(9) instead of malloc/free

(ozaki-r)

Make global variables static

(ozaki-r)

Add a test case for BIOCGBLEN and BIOCSBLEN

(ozaki-r)

Be more careful about malloc returning NULL

(skrll)

Linux and FreeBSD defines 0x10f5 as E1000_DEV_ID_ICH9_IGP_M_AMT. In reality,
This is not IGP but BM. Add new case to identify PHY type correctly on this
device. Fixes PR#51924 reported byJarle Greipsland.

This change fix only for 0x10f5. Need more work to be the best.

(msaitoh)

regen.

(msaitoh)

Add some Intel devices.

(msaitoh)

tabify

(knakahara)

fix: wm(4) Rx descriptor is corrupted under high rate.

don't write descriptor which is not set DD bit.

(knakahara)

sync only mapped size.

(nonaka)

Added missing splx(s).

(nonaka)

restore r1.118

(maya)

one more include file

(christos)

need visibility.h

(christos)

fix non _LP64

(christos)

add extra file.

(christos)

new libevent

(christos)

add more man pages

(christos)

regen man pages

(christos)

update for libevent

(christos)