Garbage-collect the remaining vestiges of netisr.

Ensure that all queues passed to ifq_enqueue2() have a valid ifq_lock.

Consistently use IFQ_SET_MAXLEN(), rather than open-coding it. NFC.

net/if_ppp.c: Avoid user-controlled overrun in PPPIOCSCOMPRESS.

Reported-by: syzbot+2c7bda7dc2b6c0d4f279@syzkaller.appspotmail.com

net/if_ppp.c: Sprinkle KNF. No functional change intended.

Adopt <net/if_stats.h>.

branches: 1.166.2;
dedup

Simplify "LIST_HEAD();" to make the code more understandable.
No functional change.

Add __cacheline_aligned to ppp softc list and its mutex just in case.

Fix missing splx in ppp_inproc().

Fix missing mutex_exit in ppp_create().

branches: 1.161.2;
Implement the BPF direction filter (BIOC[GS]DIRECTION). It provides backward
compatibility with BIOC[GS]SEESENT ioctl. The userland interface is the same
as FreeBSD.

This change also fixes a bug that the direction is misunderstand on some
environment by passing the direction to bpf_mtap*() instead of checking
m->m_pkthdr.rcvif.

Remove duplicated inclusion of net/bpf.h.

branches: 1.159.2;
Add one more not supported error

branches: 1.158.8;
MFREE -> m_free

modularize some more drivers and merge the module glue

Destroy the mutex when detaching ppp. Otherwise on a re-attach (ie,
module reload) we can end up with a panic "lock already initialized"

make strip and slip modular, and cosmetic for ppp.

Change the internal name of the module to match its external (file
system) name. Otherwise "bad things" can happen, such as modload(8)
being able to load a second copy!

Modularize the ppp driver, and adjust dependencies of the compressor
modules.

For now, this is still included as a built-in module in GENERIC kernels.

branches: 1.152.2;
Introduce m_set_rcvif and m_reset_rcvif

The API is used to set (or reset) a received interface of a mbuf.
They are counterpart of m_get_rcvif, which will come in another
commit, hide internal of rcvif operation, and reduce the diff of
the upcoming change.

No functional change.

Constify rtentry of if_output

We no longer need to change rtentry below if_output.

The change makes it clear where rtentries are changed (or not)
and helps forthcoming locking (os psrefing) rtentries.

IFQ_ENQUEUE refactor (3/3) : eliminate pktattr argument from IFQ_ENQUEUE caller

sprinkle _KERNEL_OPT

Honor pseudo attach decl generated by config(1).

Introduce p2p_rtrequest() so that IFF_POINTOPOINT interfaces can work
with RTF_LOCAL.
Fixes PR kern/49829.

branches: 1.146.4;
KNF. No functional change.

Cleanup ppp_inproc

- Remove unnecessary variable isr
- Use pktq instead of rv to switch between inet/inet6 and other protocols

ok msaitoh@ and rmind@

- Implement pktqueue interface for lockless IP input queue.
- Replace ipintrq and ip6intrq with the pktqueue mechanism.
- Eliminate kernel-lock from ipintr() and ip6intr().
- Some preparation work to push softnet_lock out of ipintr().

Discussed on tech-net.

ppp_inproc: handle non-IP protocols correctly (hi msaitoh); PR/48813.

Save a NETISR_* value in a variable and call schednetisr() after enqueue
a packet for readability and future modification.

branches: 1.141.2;
Add bpf_filter_ext() to use with BPF COP, restore bpf_filter() as it was
originally to preserve compatibility. Similarly, add bpf_validate_ext()
which takes bpf_ctx_t.

bpf_filter: add a custom argument which can be passed to coprocessor routine.

Implement BPF_COP/BPF_COPX instructions in the misc category (BPF_MISC)
which add a capability to call external functions in a predetermined way.

It can be thought as a BPF "coprocessor" -- a generic mechanism to offload
more complex packet inspection operations. There is no default coprocessor
and this functionality is not targeted to the /dev/bpf. This is primarily
targeted to the kernel subsystems, therefore there is no way to set a custom
coprocessor at the userlevel.

Discussed on: tech-net@
OK: core@

branches: 1.138.2;
Don't check mp for NULL twice. From Michael W. Bomardieri <mb@il.net>
via tech-net@NetBSD.org. Thanks!

PR/47058: Antti Kantee: If the ipv6 flow code modifies the mbuf, pass the
change up to the caller.

branches: 1.136.2; 1.136.8; 1.136.12;
For these interfaces, the implementation of SIOCSIFDSTADDR is identical
to SIOCINITIFADDR, and SIOCSIFDSTADDR callers always fall back to
SIOCINITIFADDR, so just get rid of the SIOCSIFDSTADDR case.

Don't kauth-orize SIOCSIFMTU in pppsioctl() and stf_ioctl(), ifioctl()
has already done that for us.

Convert ppp_list_lock to mutex(9).

Fix misplaced parenthesis. From henning.petersen@t-online.de, thanks.

branches: 1.132.2;
Update the rest of the kernel to conform to the module subsystem's new
locking protocol.

Push the bpf_ops usage back into bpf.h. Push the common ifp->if_bpf
check into the inline functions as well the fourth argument for
bpf_attach.

branches: 1.130.2; 1.130.4;
Redefine bpf linkage through an always present op vector, i.e.
#if NBPFILTER is no longer required in the client. This change
doesn't yet add support for loading bpf as a module, since drivers
can register before bpf is attached. However, callers of bpf can
now be modularized.

Dynamically loadable bpf could probably be done fairly easily with
coordination from the stub driver and the real driver by registering
attachments in the stub before the real driver is loaded and doing
a handoff. ... and I'm not going to ponder the depths of unload
here.

Tested with i386/MONOLITHIC, modified MONOLITHIC without bpf and rump.

Remove a few KAUTH_GENERIC_ISSUSER in favor of more descriptive
alternatives.

Discussed on tech-kern:

http://mail-index.netbsd.org/tech-kern/2009/04/11/msg004798.html

Input from ad@, christos@, dyoung@, tsutsui@.

Okay ad@.

branches: 1.128.2;
ppp_get_compressor: take module_lock when trying to load a module. PR/40428

kill MALLOC and FREE macros.

Fix handling of ppp compressor modules, from Andrew Doran's input.
- ref count each compressor
- allow {un,}registration of several modules at once
- une RUN_ONCE to make sure the mutex is initialised, because
unfortunately built-in (and bootloader-loaded) modules init functions
are run before pseudo-devices attach (reported by Nick Hudson).

Rework the way PPP compmressors are handled and allow them to be
automatically loaded when needed.

*** Summary ***

When a link-layer address changes (e.g., ifconfig ex0 link
02:de:ad:be:ef:02 active), send a gratuitous ARP and/or a Neighbor
Advertisement to update the network-/link-layer address bindings
on our LAN peers.

Refuse a change of ethernet address to the address 00:00:00:00:00:00
or to any multicast/broadcast address. (Thanks matt@.)

Reorder ifnet ioctl operations so that driver ioctls may inherit
the functions of their "class"---ether_ioctl(), fddi_ioctl(), et
cetera---and the class ioctls may inherit from the generic ioctl,
ifioctl_common(), but both driver- and class-ioctls may override
the generic behavior. Make network drivers share more code.

Distinguish a "factory" link-layer address from others for the
purposes of both protecting that address from deletion and computing
EUI64.

Return consistent, appropriate error codes from network drivers.

Improve readability. KNF.

*** Details ***

In if_attach(), always initialize the interface ioctl routine,
ifnet->if_ioctl, if the driver has not already initialized it.
Delete if_ioctl == NULL tests everywhere else, because it cannot
happen.

In the ioctl routines of network interfaces, inherit common ioctl
behaviors by calling either ifioctl_common() or whichever ioctl
routine is appropriate for the class of interface---e.g., ether_ioctl()
for ethernets.

Stop (ab)using SIOCSIFADDR and start to use SIOCINITIFADDR. In
the user->kernel interface, SIOCSIFADDR's argument was an ifreq,
but on the protocol->ifnet interface, SIOCSIFADDR's argument was
an ifaddr. That was confusing, and it would work against me as I
make it possible for a network interface to overload most ioctls.
On the protocol->ifnet interface, replace SIOCSIFADDR with
SIOCINITIFADDR. In ifioctl(), return EPERM if userland tries to
invoke SIOCINITIFADDR.

In ifioctl(), give the interface the first shot at handling most
interface ioctls, and give the protocol the second shot, instead
of the other way around. Finally, let compatibility code (COMPAT_OSOCK)
take a shot.

Pull device initialization out of switch statements under
SIOCINITIFADDR. For example, pull ..._init() out of any switch
statement that looks like this:

switch (...->sa_family) {
case ...:
..._init();
...
break;
...
default:
..._init();
...
break;
}

Rewrite many if-else clauses that handle all permutations of IFF_UP
and IFF_RUNNING to use a switch statement,

switch (x & (IFF_UP|IFF_RUNNING)) {
case 0:
...
break;
case IFF_RUNNING:
...
break;
case IFF_UP:
...
break;
case IFF_UP|IFF_RUNNING:
...
break;
}

unifdef lots of code containing #ifdef FreeBSD, #ifdef NetBSD, and
#ifdef SIOCSIFMTU, especially in fwip(4) and in ndis(4).

In ipw(4), remove an if_set_sadl() call that is out of place.

In nfe(4), reuse the jumbo MTU logic in ether_ioctl().

Let ethernets register a callback for setting h/w state such as
promiscuous mode and the multicast filter in accord with a change
in the if_flags: ether_set_ifflags_cb() registers a callback that
returns ENETRESET if the caller should reset the ethernet by calling
if_init(), 0 on success, != 0 on failure. Pull common code from
ex(4), gem(4), nfe(4), sip(4), tlp(4), vge(4) into ether_ioctl(),
and register if_flags callbacks for those drivers.

Return ENOTTY instead of EINVAL for inappropriate ioctls. In
zyd(4), use ENXIO instead of ENOTTY to indicate that the device is
not any longer attached.

Add to if_set_sadl() a boolean 'factory' argument that indicates
whether a link-layer address was assigned by the factory or some
other source. In a comment, recommend using the factory address
for generating an EUI64, and update in6_get_hw_ifid() to prefer a
factory address to any other link-layer address.

Add a routing message, RTM_LLINFO_UPD, that tells protocols to
update the binding of network-layer addresses to link-layer addresses.
Implement this message in IPv4 and IPv6 by sending a gratuitous
ARP or a neighbor advertisement, respectively. Generate RTM_LLINFO_UPD
messages on a change of an interface's link-layer address.

In ether_ioctl(), do not let SIOCALIFADDR set a link-layer address
that is broadcast/multicast or equal to 00:00:00:00:00:00.

Make ether_ioctl() call ifioctl_common() to handle ioctls that it
does not understand.

In gif(4), initialize if_softc and use it, instead of assuming that
the gif_softc and ifp overlap.

Let ifioctl_common() handle SIOCGIFADDR.

Sprinkle rtcache_invariants(), which checks on DIAGNOSTIC kernels
that certain invariants on a struct route are satisfied.

In agr(4), rewrite agr_ioctl_filter() to be a bit more explicit
about the ioctls that we do not allow on an agr(4) member interface.

bzero -> memset. Delete unnecessary casts to void *. Use
sockaddr_in_init() and sockaddr_in6_init(). Compare pointers with
NULL instead of "testing truth". Replace some instances of (type
*)0 with NULL. Change some K&R prototypes to ANSI C, and join
lines.

branches: 1.123.2; 1.123.4;
- add if_alloc (ours just mallocs), and if_initname and use them (from FreeBSD)
- kill memsets where M_ZERO can be used.

branches: 1.122.2; 1.122.4; 1.122.6;
Merge the socket locking patch:

- Socket layer becomes MP safe.
- Unix protocols become MP safe.
- Allows protocol processing interrupts to safely block on locks.
- Fixes a number of race conditions.

With much feedback from matt@ and plunky@.

branches: 1.121.6; 1.121.8;
Start patching up the kernel so that a network driver always has
the opportunity to handle an ioctl before generic ifioctl handling
occurs. This will ease extending the kernel and sharing of code
between drivers.

First steps: Make the signature of ifioctl_common() match struct
ifinet->if_ioctl. Convert SIOCSIFCAP and SIOCSIFMTU to the new
ifioctl() regime, throughout the kernel.

Start detangling lock.h from intr.h. This is likely to cause short term
breakage, but the mess of dependencies has been regularly breaking the
build recently anyhow.

branches: 1.119.2; 1.119.8;
machine/{bus,cpu,intr}.h -> sys/{bus,cpu,intr}.h

branches: 1.118.2;
Use the softint API.

branches: 1.117.2;
Use ifreq_setaddr(), ifreq_getaddr(), sockaddr_in_init(), and
sockaddr_copy(). Constify. Compare pointers with NULL, not 0.
Don't "test truth" of pointers, but compare with NULL.

branches: 1.116.2; 1.116.6; 1.116.8;
Generic soft interrupts are mandatory.

Merge some of the less invasive changes from the vmlocking branch:

- kthread, callout, devsw API changes
- select()/poll() improvements
- miscellaneous MT safety improvements

branches: 1.114.2; 1.114.4;
Add IPv6 Fast Forward - the IPv4 counterpart:

If ip6_forward successfully forwards a packet, a cache, in this case a
ip6flow struct entry, will be created. ether_input and friends will
then be able to call ip6flow_fastforward with the packet which will then
be passed to if_output (unless an issue is found - in that case the packet
is passed back to ip6_input).

ok matt@ christos@ dyoung@ and joerg@

Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

KNF: de-__P, bzero -> memset, bcmp -> memcmp. Remove extraneous
parentheses in return statements.

Cosmetic: don't open-code TAILQ_FOREACH().

Cosmetic: change types of variables to avoid oodles of casts: in
in6_src.c, avoid casts by changing several route_in6 pointers
to struct route pointers. Remove unnecessary casts to caddr_t
elsewhere.

Pave the way for eliminating address family-specific route caches:
soon, struct route will not embed a sockaddr, but it will hold
a reference to an external sockaddr, instead. We will set the
destination sockaddr using rtcache_setdst(). (I created a stub
for it, but it isn't used anywhere, yet.) rtcache_free() will
free the sockaddr. I have extracted from rtcache_free() a helper
subroutine, rtcache_clear(). rtcache_clear() will "forget" a
cached route, but it will not forget the destination by releasing
the sockaddr. I use rtcache_clear() instead of rtcache_free()
in rtcache_update(), because rtcache_update() is not supposed
to forget the destination.

Constify:

1 Introduce const accessor for route->ro_dst, rtcache_getdst().

2 Constify the 'dst' argument to ifnet->if_output(). This
led me to constify a lot of code called by output routines.

3 Constify the sockaddr argument to protosw->pr_ctlinput. This
led me to constify a lot of code called by ctlinput routines.

4 Introduce const macros for converting from a generic sockaddr
to family-specific sockaddrs, e.g., sockaddr_in: satocsin6,
satocsin, et cetera.

branches: 1.111.4;
__unused removal on arguments; approved by core.

Kill some KAUTH_GENERIC_ISSUSER uses.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

branches: 1.108.4; 1.108.6;
Use the LWP cached credentials where sane.

merge FreeBSD timecounters from branch simonb-timecounters
- struct timeval time is gone
time.tv_sec -> time_second
- struct timeval mono_time is gone
mono_time.tv_sec -> time_uptime
- access to time via
{get,}{micro,nano,bin}time()
get* versions are fast but less precise
- support NTP nanokernel implementation (NTP API 4)
- further reading:
Timecounter Paper: http://phk.freebsd.dk/pubs/timecounter.pdf
NTP Nanokernel: http://www.eecis.udel.edu/~mills/ntp/html/kern.html

branches: 1.106.2;
integrate kauth.

branches: 1.105.2; 1.105.4; 1.105.6; 1.105.8; 1.105.10;
ppp_dequeue: fix a mbuf leak/packet loss introduced by rev.1.104.

branches: 1.104.2;
PR/5901: Felix A. Croes: PPP fast queue blocks traffic at normal priority.
Applied fix, similar to the one suggested in the PR. We use a counter to
limit the number of consecutive packets accepted from the fast queue. This
number can be set via ioctl, but this has not been implemented. Since there
are only 2 queues other proposed solutions such as ALTQ are overkill and
they have not been implemented in the past 7 years. Now LCP echos can be
used to detect that the line is up.

ANSI function decls and application of static.

Overhaul how TTY line disciplines are handled:
- Replace references to linesw[0] with a ttyldisc_default() function
that returns the default ("termios") line discipline.
- The linesw[] array is gone, replaced by a linked list.
- ttyldisc_add() and ttyldisc_remove() have been replaced by
ttyldisc_attach() and ttyldisc_detach().
- Things that provide line disciplines are now responsible for
registering those disciplines with the system. The linesw
structures are no longer declared in tty_conf.c
- Line disciplines are now refcounted; a lookup causes a reference to
be held. ttyldisc_release() releases the reference. Attempts to
detach an in-use line discipline result in EBUSY.
- Fix function signature lossage in if_sl.c, if_strip.c, and tty_tb.c
that was masked by the old tty_conf.c
- tty_init() is no longer necessary; delete it and its call from main().

branches: 1.101.2; 1.101.8;
- sprinkle const
- remove unneeded casts
- use more mem*() instead of b*() funcs.

Yes, it was a cool trick >20 years ago to use "0123456789abcdef"[a] to
implement, xtoa(), but I think defining the samestring 50 times is a bit
too much. Defined HEXDIGITS and hexdigits in subr_prf.c and use it...

no point in assigning to ifq twice.

Fix error with ifq not being set before use. Explicitly set it to NULL just before it may be set to the fastq, since if this becomes a loop (and in one case already is) this will always work. ifq_enqueue2() is designed to handle this case.

factor out the interface queueing code into two functions. One used by
the non point-to-point interfaces that has one queue, and one used by
the point to point interfaces that has two queues. No functional changes.
XXX: The ALTQ stuff makes the code ugly.
XXX: More cleanup to come

nuke trailing whitespace

branches: 1.95.4; 1.95.6;
pasto: simple_lock -> simple_unlock.

Don't forget to call bpfdetach in the clone destroy function.
While here, add a missing static and change some spaces to tabs.

Fix an obvious typo: scf -> sc. Discovered while compiling for x68k.

void in arg prototype.

Make ppp a cloning device. Based on the work of Quentin Garnier.

I changed pppoutput to use M_PREPEND. pppoutput was duplicating
the functionality of M_PREPEND, but with a bug: m_pkthdr.len was
not updated in pppoutput as it is in M_PREPEND.

Also, replace the loop that measures the length of the mbuf chain
with a call to m_length.

This fixes a PR from an anonymous bug reporter. Thank you, anonymous
bug reporter. Thanks, Itojun, for bringing the anonymous bug report
to my attention.

kill sprintf, use snprintf

Also, if we're going to bail, we should free the memory we just allocated...

Previous patch created a dead break.

Fix uninitialized variable warnings

Add a new ioctl PPPIOCGRAWIN to get the last characters we got from the
remote site.

branches: 1.84.2;
KNF

PR/20844: Iain Hibbert: PPP Compressors cannot be loaded as LKM

Remove variable that is only assigned too but not referenced.

backout previous two - if you use ppp* interface, kernel panics instantly.
it is apparent that the change was untested, and severety is high.

Remove unused variable so the file compiles again.

Keep m_pkthdr.len updated correctly and use it rather than a loop to find
out the total length of the packet.

new copyright boilerplate from CMU. from openbsd

branches: 1.77.2;
Make ppp_softc[] extern and declare in if_ppp.c

Convert ioctl code to use EPASSTHROUGH instead of -1 or ENOTTY for
indicating an unhandled "command". ERESTART is -1, which can lead to
confusion. ERESTART has been moved to -3 and EPASSTHROUGH has been
placed at -4. No ioctl code should now return -1 anywhere. The
ioctl() system call is now properly restartable.

bring in latest ALTQ from kjc. ALTQify some of the drivers.

Include <machine/intr.h> unconditionally, instead of only doing so if
__HAVE_GENERIC_SOFT_INTERRUPTS and relying on <sys/param.h> to provide it
otherwise; pointed out by Aymeric Vincent.

remove unnecessary #if NFOO > 0 .... #endif wrappers

add RCSIDs

use unsigned variable types as appropriate

bzero -> memset

branches: 1.69.2;
change the meaning of ifnet.if_lastchange to meet RFC1573 ifLastChange.
follows BSD/OS practice and ucd-snmp code (FreeBSD does it for specific
interfaces only).

was: if_lastchange get updated on every packet transmission/receipt.
now: if_lastchange get updated when IFF_UP is changed.

Remove the use of splimp() from the NetBSD kernel. splnet()
and only splnet() is allowed for the protection of data structures
used by network devices.

branches: 1.67.2;
Fix a rather annoying problem where the sockaddr_dl which holds
the link level name for the interface (ifp->if_sadl) is allocated
before ifp->if_addrlen is initialized, which could lead to allocating
too little space for the link level address.

Do this by splitting allocation of the link level name out of
if_attach() and into if_alloc_sadl(), which is normally called
by functions like ether_ifattach(). Network interfaces which
don't have a link-specific attach routine must call if_alloc_sadl()
themselves (example: gif).

Link level names are freed by if_free_sadl(), which can be called
from e.g. ether_ifdetach(). Drivers never need call if_free_sadl()
themselves as if_detach() will do it if it is not already done.

While here, add the ability to pass an AF_LINK address to
SIOCSIFADDR in ether_ioctl() (this is what caused me to notice
the problem that the above fixes).

For SLIP/STRIP/PPP, use generic soft interrupts, if available.

Fill in if_dlt.

Add ALTQ support.

Adapt to bpfattach() changes, and further centralize the bpfattach()
and bpfdetach() calls into link-type subroutines where possible.

Fix typo (thinko?), which didn't allow MRU to be set below the default
value (instead of the minimum value). Patch supplied by Stephen Degler
in PR #9945, and reviewed by Ignatios Souvatzis.

unique #include opt_inet.h

need opt_inet.h for #ifdef INET

enable VJC only with INET

branches: 1.58.4;
Kill some more register declarations.

New callout mechanism with two major improvements over the old
timeout()/untimeout() API:
- Clients supply callout handle storage, thus eliminating problems of
resource allocation.
- Insertion and removal of callouts is constant time, important as
this facility is used quite a lot in the kernel.

The old timeout()/untimeout() API has been removed from the kernel.

Fix typo introduced in rev. 1.55.
It caused IP6 packets to be sent as AF_UNSPEC instead of AF_INET6.

branches: 1.55.2; 1.55.8;
remove reference to in6_systm.h (file itself will be removed afterwords)

IPv6 kernel code, based on KAME/NetBSD 1.4, SNAP kit 19990628.
(Sorry for a big commit, I can't separate this into several pieces...)
Pls check sys/netinet6/TODO and sys/netinet6/IMPLEMENTATION for details.

- sys/kern: do not assume single mbuf, accept chained mbuf on passing
data from userland to kernel (or other way round).
- "midway" ATM card: ATM PVC pseudo device support, like those done in ALTQ
package (ftp://ftp.csl.sony.co.jp/pub/kjc/).
- sys/netinet/tcp*: IPv4/v6 dual stack tcp support.
- sys/netinet/{ip6,icmp6}.h, sys/net/pfkeyv2.h: IETF document assumes those
file to be there so we patch it up.
- sys/netinet: IPsec additions are here and there.
- sys/netinet6/*: most of IPv6 code sits here.
- sys/netkey: IPsec key management code
- dev/pci/pcidevs: regen

In my understanding no code here is subject to export control so it
should be safe.

Fix compilation problem caused by recent changes to filtering code.

Decouple inbound and outbound filters. Now instead of using "active-filter"
and "pass-filter" and "inbound" and "outbound" qualifiers in the filter
expression, use new "active-filter-in", "active-filter-out", "pass-filter-in",
and "pass-filter-out" without these qualifiers.

This is necessary due to the horrible, awful way "inbound" and "outbound"
were specified for the filter programs when a packet was passed through them.
Basically, the "address" byte in the serial PPP header was overwritten with
a value to indicate the direction. However, the "address" byte doesn't even
exist on PPP headers for all other PPP encaps! So, this old method worked
only for serial encaps, and corrupted packets for all others (PPPoE, ATM, etc.)

* Start out with a data link type of DLT_NULL. When we change an interface
to serial encap, change its data link type to DLT_PPP_SERIAL.
* Work around some serious bogosity in the filtering code which utterly
breaks proper functioning of BPF. The PPP code and pppd(8) WILL be changed
to fix this.

branches: 1.50.4; 1.50.6;
Use M_LINK{0,1} for our own mbuf flags, rather than arbitrarily picking
2 bits.

Revert IPX changes that I committed accidentally.

defopt

branches: 1.47.4;
PR/5414: Ronald Khoo: tcpdump ppp does not respect inbound/outbound qualifiers.

Fix PR5898: ppp delays last packet.

Glue in fast forwarding.

Only run pppasyncstart (sc->sc_start) from the netisr handler.
This allows pppoutput to be called from splimp (e.g., when ipflow is
in use.) without requiring pppasyncstart to run at splimp.
This is believed to fix PR5624.

use #ifdef INET so this compiles again

defopt INET, NETATALK.

Merge changes from pppd-2.3.4; adds ppp-deflate-draft stuff and updates
zlib. Maybe we can merge our other copy of zlib with this one now and
avoid having two copies?

From Paul Mackerras: use sl_compress_setup, not sl_compress_init

Update to ppp-2.3b5

Made pppoutput() public again on behalf of Martin Husemann (PR 3455).
Apparently, the BISDN package uses this function.

Update to ppp-2.3b4; from Paul Mackerras

Use splimp() to block interrupts, not splhigh().

branches: 1.35.4;
backout previous kprintf change

- printf -> kprintf, sprintf -> ksprintf

no need for a local implementation of SIOCGIFMTU; delete it.

Pass a proc pointer down to the usrreq and pcbbind functions for PRU_ATTACH, PRU_BIND and
PRU_CONTROL. The usrreq interface really needs to be split up, but this will have to wait.
Remove SS_PRIV completely.

branches: 1.31.4;
Changed struct ifnet to have a pointer to the softc of the underlying
device and a printable "external name" (name + unit number), thus eliminating
if_name and if_unit. Updated interface to (*if_watchdog)() and (*if_reset)()
to take a struct ifnet *, rather than a unit number.

Make inclusion of the PPP BSD-Compress and Deflate compressors
dependent on the PPP_BSDCOMP and PPP_DEFLATE kernel configuration
options, respectively.

Added packet filtering, support for "PPP Deflate" packet compression,
trivial multicast support, and support for xon/xoff output flow
control to the PPP subsystem. Fixed several bugs, including making
the accumulation and resetting of statistics more consistent. State
for the VJ compressor is now dynamically allocated.

Net prototypes

wrt. previous change: can't compute `ilen' that early; just do computation
separately when logging.
Notes: consider using mbuf pkthdr length field in PPP code.
condider doing packet log after de-compression.

Init variable before use (PRs 1646 & 2042).

Remove old workaround for a bug.

Add some missing statistics. From Thorsten Lockert.

splnet --> splsoftnet

Latest version of PPP stuff, with packet compression and other
improvements. The PPP kernel code is now split into if_ppp.c,
containing generic PPP support, and ppp_tty.c, which specifically
supports PPP on async tty devices (as a line discipline). This is
so that other devices can be supported without making them look
like ttys.

fixed sized types, where appropriate. when casting pointers to
integers to do math on them, cast to long. ioctl commands are
u_longs.

be more careful with types, also pull in headers where necessary.

bug 367. paulus says the fix is right & critical.

The ppp interface now retries the mbuf allocation when it receives
a character and it doesn't already have enough space allocated.
It still needs cluster mbufs to be able to decompress VJ-compressed
packets. It drops packets if it can't allocate mbufs rather than
taking the interface down.

Due to popular revulsion, the ppp interface now drops packets if

If we can't get mbuf clusters, take the ppp interface down instead
of writing beyond the end of ordinary mbufs.

branches: 1.15.2;
bug #319. fix from <alasdair@wildcat.demon.co.uk>

New RCS ID's, take two. they're more aesthecially pleasant, and use 'NetBSD'

Some restructuring of the PPP packet input procedure to make it easier
to implement PPP over sync lines and PPP compression protocols.

some minor splx-type bug fixes from christos@deshaw.com.

check for escaped char before checking for escape char
so if peer escapes 0x5d we interpret it correctly

MIN -> min, MAX -> max

Update to 4.4-Lite networking code, with a few local changes.

Version from ppp-2.1 release;
this version has been restructured to make more of the code usable
with sync serial drivers

PPP_HEADER_LEN -> PPP_HDRLEN

include <machine/cpu.h> rather than <machine/mtpr.h> -- if the latter
exists at all, it's supposed to be included by <machine/cpu.h>

Canonicalize all #includes.

Removed test (CCOUNT(&sc->sc_ttyp->t_outq) == 0) for whether
to call pppstart or not: now we call pppstart for every packet,
which should aid recovery from lost transmitter interrupts.
Also a fix for 386BSD/FreeBSD which doesn't affect NetBSD.

branches: 1.3.2;
Fixed bug in if_ppp.c so that received IP packets are passed correctly to BPF.

Modified if_ppp.c and if_ppp.h to add priority queueing for "interactive"
traffic (done in a similar fashion to if_sl.c), and BPF support.

ppp from paul mackerras

Ensure that all queues passed to ifq_enqueue2() have a valid ifq_lock.

Consistently use IFQ_SET_MAXLEN(), rather than open-coding it. NFC.

net/if_ppp.c: Avoid user-controlled overrun in PPPIOCSCOMPRESS.

Reported-by: syzbot+2c7bda7dc2b6c0d4f279@syzkaller.appspotmail.com

net/if_ppp.c: Sprinkle KNF. No functional change intended.

Adopt <net/if_stats.h>.

branches: 1.166.2;
dedup

Simplify "LIST_HEAD();" to make the code more understandable.
No functional change.

Add __cacheline_aligned to ppp softc list and its mutex just in case.

Fix missing splx in ppp_inproc().

Fix missing mutex_exit in ppp_create().

branches: 1.161.2;
Implement the BPF direction filter (BIOC[GS]DIRECTION). It provides backward
compatibility with BIOC[GS]SEESENT ioctl. The userland interface is the same
as FreeBSD.

This change also fixes a bug that the direction is misunderstand on some
environment by passing the direction to bpf_mtap*() instead of checking
m->m_pkthdr.rcvif.

Remove duplicated inclusion of net/bpf.h.

branches: 1.159.2;
Add one more not supported error

branches: 1.158.8;
MFREE -> m_free

modularize some more drivers and merge the module glue

Destroy the mutex when detaching ppp. Otherwise on a re-attach (ie,
module reload) we can end up with a panic "lock already initialized"

make strip and slip modular, and cosmetic for ppp.

Change the internal name of the module to match its external (file
system) name. Otherwise "bad things" can happen, such as modload(8)
being able to load a second copy!

Modularize the ppp driver, and adjust dependencies of the compressor
modules.

For now, this is still included as a built-in module in GENERIC kernels.

branches: 1.152.2;
Introduce m_set_rcvif and m_reset_rcvif

The API is used to set (or reset) a received interface of a mbuf.
They are counterpart of m_get_rcvif, which will come in another
commit, hide internal of rcvif operation, and reduce the diff of
the upcoming change.

No functional change.

Constify rtentry of if_output

We no longer need to change rtentry below if_output.

The change makes it clear where rtentries are changed (or not)
and helps forthcoming locking (os psrefing) rtentries.

IFQ_ENQUEUE refactor (3/3) : eliminate pktattr argument from IFQ_ENQUEUE caller

sprinkle _KERNEL_OPT

Honor pseudo attach decl generated by config(1).

Introduce p2p_rtrequest() so that IFF_POINTOPOINT interfaces can work
with RTF_LOCAL.
Fixes PR kern/49829.

branches: 1.146.4;
KNF. No functional change.

Cleanup ppp_inproc

- Remove unnecessary variable isr
- Use pktq instead of rv to switch between inet/inet6 and other protocols

ok msaitoh@ and rmind@

- Implement pktqueue interface for lockless IP input queue.
- Replace ipintrq and ip6intrq with the pktqueue mechanism.
- Eliminate kernel-lock from ipintr() and ip6intr().
- Some preparation work to push softnet_lock out of ipintr().

Discussed on tech-net.

ppp_inproc: handle non-IP protocols correctly (hi msaitoh); PR/48813.

Save a NETISR_* value in a variable and call schednetisr() after enqueue
a packet for readability and future modification.

branches: 1.141.2;
Add bpf_filter_ext() to use with BPF COP, restore bpf_filter() as it was
originally to preserve compatibility. Similarly, add bpf_validate_ext()
which takes bpf_ctx_t.

bpf_filter: add a custom argument which can be passed to coprocessor routine.

Implement BPF_COP/BPF_COPX instructions in the misc category (BPF_MISC)
which add a capability to call external functions in a predetermined way.

It can be thought as a BPF "coprocessor" -- a generic mechanism to offload
more complex packet inspection operations. There is no default coprocessor
and this functionality is not targeted to the /dev/bpf. This is primarily
targeted to the kernel subsystems, therefore there is no way to set a custom
coprocessor at the userlevel.

Discussed on: tech-net@
OK: core@

branches: 1.138.2;
Don't check mp for NULL twice. From Michael W. Bomardieri <mb@il.net>
via tech-net@NetBSD.org. Thanks!

PR/47058: Antti Kantee: If the ipv6 flow code modifies the mbuf, pass the
change up to the caller.

branches: 1.136.2; 1.136.8; 1.136.12;
For these interfaces, the implementation of SIOCSIFDSTADDR is identical
to SIOCINITIFADDR, and SIOCSIFDSTADDR callers always fall back to
SIOCINITIFADDR, so just get rid of the SIOCSIFDSTADDR case.

Don't kauth-orize SIOCSIFMTU in pppsioctl() and stf_ioctl(), ifioctl()
has already done that for us.

Convert ppp_list_lock to mutex(9).

Fix misplaced parenthesis. From henning.petersen@t-online.de, thanks.

branches: 1.132.2;
Update the rest of the kernel to conform to the module subsystem's new
locking protocol.

Push the bpf_ops usage back into bpf.h. Push the common ifp->if_bpf
check into the inline functions as well the fourth argument for
bpf_attach.

branches: 1.130.2; 1.130.4;
Redefine bpf linkage through an always present op vector, i.e.
#if NBPFILTER is no longer required in the client. This change
doesn't yet add support for loading bpf as a module, since drivers
can register before bpf is attached. However, callers of bpf can
now be modularized.

Dynamically loadable bpf could probably be done fairly easily with
coordination from the stub driver and the real driver by registering
attachments in the stub before the real driver is loaded and doing
a handoff. ... and I'm not going to ponder the depths of unload
here.

Tested with i386/MONOLITHIC, modified MONOLITHIC without bpf and rump.

Remove a few KAUTH_GENERIC_ISSUSER in favor of more descriptive
alternatives.

Discussed on tech-kern:

http://mail-index.netbsd.org/tech-kern/2009/04/11/msg004798.html

Input from ad@, christos@, dyoung@, tsutsui@.

Okay ad@.

branches: 1.128.2;
ppp_get_compressor: take module_lock when trying to load a module. PR/40428

kill MALLOC and FREE macros.

Fix handling of ppp compressor modules, from Andrew Doran's input.
- ref count each compressor
- allow {un,}registration of several modules at once
- une RUN_ONCE to make sure the mutex is initialised, because
unfortunately built-in (and bootloader-loaded) modules init functions
are run before pseudo-devices attach (reported by Nick Hudson).

Rework the way PPP compmressors are handled and allow them to be
automatically loaded when needed.

*** Summary ***

When a link-layer address changes (e.g., ifconfig ex0 link
02:de:ad:be:ef:02 active), send a gratuitous ARP and/or a Neighbor
Advertisement to update the network-/link-layer address bindings
on our LAN peers.

Refuse a change of ethernet address to the address 00:00:00:00:00:00
or to any multicast/broadcast address. (Thanks matt@.)

Reorder ifnet ioctl operations so that driver ioctls may inherit
the functions of their "class"---ether_ioctl(), fddi_ioctl(), et
cetera---and the class ioctls may inherit from the generic ioctl,
ifioctl_common(), but both driver- and class-ioctls may override
the generic behavior. Make network drivers share more code.

Distinguish a "factory" link-layer address from others for the
purposes of both protecting that address from deletion and computing
EUI64.

Return consistent, appropriate error codes from network drivers.

Improve readability. KNF.

*** Details ***

In if_attach(), always initialize the interface ioctl routine,
ifnet->if_ioctl, if the driver has not already initialized it.
Delete if_ioctl == NULL tests everywhere else, because it cannot
happen.

In the ioctl routines of network interfaces, inherit common ioctl
behaviors by calling either ifioctl_common() or whichever ioctl
routine is appropriate for the class of interface---e.g., ether_ioctl()
for ethernets.

Stop (ab)using SIOCSIFADDR and start to use SIOCINITIFADDR. In
the user->kernel interface, SIOCSIFADDR's argument was an ifreq,
but on the protocol->ifnet interface, SIOCSIFADDR's argument was
an ifaddr. That was confusing, and it would work against me as I
make it possible for a network interface to overload most ioctls.
On the protocol->ifnet interface, replace SIOCSIFADDR with
SIOCINITIFADDR. In ifioctl(), return EPERM if userland tries to
invoke SIOCINITIFADDR.

In ifioctl(), give the interface the first shot at handling most
interface ioctls, and give the protocol the second shot, instead
of the other way around. Finally, let compatibility code (COMPAT_OSOCK)
take a shot.

Pull device initialization out of switch statements under
SIOCINITIFADDR. For example, pull ..._init() out of any switch
statement that looks like this:

switch (...->sa_family) {
case ...:
..._init();
...
break;
...
default:
..._init();
...
break;
}

Rewrite many if-else clauses that handle all permutations of IFF_UP
and IFF_RUNNING to use a switch statement,

switch (x & (IFF_UP|IFF_RUNNING)) {
case 0:
...
break;
case IFF_RUNNING:
...
break;
case IFF_UP:
...
break;
case IFF_UP|IFF_RUNNING:
...
break;
}

unifdef lots of code containing #ifdef FreeBSD, #ifdef NetBSD, and
#ifdef SIOCSIFMTU, especially in fwip(4) and in ndis(4).

In ipw(4), remove an if_set_sadl() call that is out of place.

In nfe(4), reuse the jumbo MTU logic in ether_ioctl().

Let ethernets register a callback for setting h/w state such as
promiscuous mode and the multicast filter in accord with a change
in the if_flags: ether_set_ifflags_cb() registers a callback that
returns ENETRESET if the caller should reset the ethernet by calling
if_init(), 0 on success, != 0 on failure. Pull common code from
ex(4), gem(4), nfe(4), sip(4), tlp(4), vge(4) into ether_ioctl(),
and register if_flags callbacks for those drivers.

Return ENOTTY instead of EINVAL for inappropriate ioctls. In
zyd(4), use ENXIO instead of ENOTTY to indicate that the device is
not any longer attached.

Add to if_set_sadl() a boolean 'factory' argument that indicates
whether a link-layer address was assigned by the factory or some
other source. In a comment, recommend using the factory address
for generating an EUI64, and update in6_get_hw_ifid() to prefer a
factory address to any other link-layer address.

Add a routing message, RTM_LLINFO_UPD, that tells protocols to
update the binding of network-layer addresses to link-layer addresses.
Implement this message in IPv4 and IPv6 by sending a gratuitous
ARP or a neighbor advertisement, respectively. Generate RTM_LLINFO_UPD
messages on a change of an interface's link-layer address.

In ether_ioctl(), do not let SIOCALIFADDR set a link-layer address
that is broadcast/multicast or equal to 00:00:00:00:00:00.

Make ether_ioctl() call ifioctl_common() to handle ioctls that it
does not understand.

In gif(4), initialize if_softc and use it, instead of assuming that
the gif_softc and ifp overlap.

Let ifioctl_common() handle SIOCGIFADDR.

Sprinkle rtcache_invariants(), which checks on DIAGNOSTIC kernels
that certain invariants on a struct route are satisfied.

In agr(4), rewrite agr_ioctl_filter() to be a bit more explicit
about the ioctls that we do not allow on an agr(4) member interface.

bzero -> memset. Delete unnecessary casts to void *. Use
sockaddr_in_init() and sockaddr_in6_init(). Compare pointers with
NULL instead of "testing truth". Replace some instances of (type
*)0 with NULL. Change some K&R prototypes to ANSI C, and join
lines.

branches: 1.123.2; 1.123.4;
- add if_alloc (ours just mallocs), and if_initname and use them (from FreeBSD)
- kill memsets where M_ZERO can be used.

branches: 1.122.2; 1.122.4; 1.122.6;
Merge the socket locking patch:

- Socket layer becomes MP safe.
- Unix protocols become MP safe.
- Allows protocol processing interrupts to safely block on locks.
- Fixes a number of race conditions.

With much feedback from matt@ and plunky@.

branches: 1.121.6; 1.121.8;
Start patching up the kernel so that a network driver always has
the opportunity to handle an ioctl before generic ifioctl handling
occurs. This will ease extending the kernel and sharing of code
between drivers.

First steps: Make the signature of ifioctl_common() match struct
ifinet->if_ioctl. Convert SIOCSIFCAP and SIOCSIFMTU to the new
ifioctl() regime, throughout the kernel.

Start detangling lock.h from intr.h. This is likely to cause short term
breakage, but the mess of dependencies has been regularly breaking the
build recently anyhow.

branches: 1.119.2; 1.119.8;
machine/{bus,cpu,intr}.h -> sys/{bus,cpu,intr}.h

branches: 1.118.2;
Use the softint API.

branches: 1.117.2;
Use ifreq_setaddr(), ifreq_getaddr(), sockaddr_in_init(), and
sockaddr_copy(). Constify. Compare pointers with NULL, not 0.
Don't "test truth" of pointers, but compare with NULL.

branches: 1.116.2; 1.116.6; 1.116.8;
Generic soft interrupts are mandatory.

Merge some of the less invasive changes from the vmlocking branch:

- kthread, callout, devsw API changes
- select()/poll() improvements
- miscellaneous MT safety improvements

branches: 1.114.2; 1.114.4;
Add IPv6 Fast Forward - the IPv4 counterpart:

If ip6_forward successfully forwards a packet, a cache, in this case a
ip6flow struct entry, will be created. ether_input and friends will
then be able to call ip6flow_fastforward with the packet which will then
be passed to if_output (unless an issue is found - in that case the packet
is passed back to ip6_input).

ok matt@ christos@ dyoung@ and joerg@

Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

KNF: de-__P, bzero -> memset, bcmp -> memcmp. Remove extraneous
parentheses in return statements.

Cosmetic: don't open-code TAILQ_FOREACH().

Cosmetic: change types of variables to avoid oodles of casts: in
in6_src.c, avoid casts by changing several route_in6 pointers
to struct route pointers. Remove unnecessary casts to caddr_t
elsewhere.

Pave the way for eliminating address family-specific route caches:
soon, struct route will not embed a sockaddr, but it will hold
a reference to an external sockaddr, instead. We will set the
destination sockaddr using rtcache_setdst(). (I created a stub
for it, but it isn't used anywhere, yet.) rtcache_free() will
free the sockaddr. I have extracted from rtcache_free() a helper
subroutine, rtcache_clear(). rtcache_clear() will "forget" a
cached route, but it will not forget the destination by releasing
the sockaddr. I use rtcache_clear() instead of rtcache_free()
in rtcache_update(), because rtcache_update() is not supposed
to forget the destination.

Constify:

1 Introduce const accessor for route->ro_dst, rtcache_getdst().

2 Constify the 'dst' argument to ifnet->if_output(). This
led me to constify a lot of code called by output routines.

3 Constify the sockaddr argument to protosw->pr_ctlinput. This
led me to constify a lot of code called by ctlinput routines.

4 Introduce const macros for converting from a generic sockaddr
to family-specific sockaddrs, e.g., sockaddr_in: satocsin6,
satocsin, et cetera.

branches: 1.111.4;
__unused removal on arguments; approved by core.

Kill some KAUTH_GENERIC_ISSUSER uses.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

branches: 1.108.4; 1.108.6;
Use the LWP cached credentials where sane.

merge FreeBSD timecounters from branch simonb-timecounters
- struct timeval time is gone
time.tv_sec -> time_second
- struct timeval mono_time is gone
mono_time.tv_sec -> time_uptime
- access to time via
{get,}{micro,nano,bin}time()
get* versions are fast but less precise
- support NTP nanokernel implementation (NTP API 4)
- further reading:
Timecounter Paper: http://phk.freebsd.dk/pubs/timecounter.pdf
NTP Nanokernel: http://www.eecis.udel.edu/~mills/ntp/html/kern.html

branches: 1.106.2;
integrate kauth.

branches: 1.105.2; 1.105.4; 1.105.6; 1.105.8; 1.105.10;
ppp_dequeue: fix a mbuf leak/packet loss introduced by rev.1.104.

branches: 1.104.2;
PR/5901: Felix A. Croes: PPP fast queue blocks traffic at normal priority.
Applied fix, similar to the one suggested in the PR. We use a counter to
limit the number of consecutive packets accepted from the fast queue. This
number can be set via ioctl, but this has not been implemented. Since there
are only 2 queues other proposed solutions such as ALTQ are overkill and
they have not been implemented in the past 7 years. Now LCP echos can be
used to detect that the line is up.

ANSI function decls and application of static.

Overhaul how TTY line disciplines are handled:
- Replace references to linesw[0] with a ttyldisc_default() function
that returns the default ("termios") line discipline.
- The linesw[] array is gone, replaced by a linked list.
- ttyldisc_add() and ttyldisc_remove() have been replaced by
ttyldisc_attach() and ttyldisc_detach().
- Things that provide line disciplines are now responsible for
registering those disciplines with the system. The linesw
structures are no longer declared in tty_conf.c
- Line disciplines are now refcounted; a lookup causes a reference to
be held. ttyldisc_release() releases the reference. Attempts to
detach an in-use line discipline result in EBUSY.
- Fix function signature lossage in if_sl.c, if_strip.c, and tty_tb.c
that was masked by the old tty_conf.c
- tty_init() is no longer necessary; delete it and its call from main().

branches: 1.101.2; 1.101.8;
- sprinkle const
- remove unneeded casts
- use more mem*() instead of b*() funcs.

Yes, it was a cool trick >20 years ago to use "0123456789abcdef"[a] to
implement, xtoa(), but I think defining the samestring 50 times is a bit
too much. Defined HEXDIGITS and hexdigits in subr_prf.c and use it...

no point in assigning to ifq twice.

Fix error with ifq not being set before use. Explicitly set it to NULL just before it may be set to the fastq, since if this becomes a loop (and in one case already is) this will always work. ifq_enqueue2() is designed to handle this case.

factor out the interface queueing code into two functions. One used by
the non point-to-point interfaces that has one queue, and one used by
the point to point interfaces that has two queues. No functional changes.
XXX: The ALTQ stuff makes the code ugly.
XXX: More cleanup to come

nuke trailing whitespace

branches: 1.95.4; 1.95.6;
pasto: simple_lock -> simple_unlock.

Don't forget to call bpfdetach in the clone destroy function.
While here, add a missing static and change some spaces to tabs.

Fix an obvious typo: scf -> sc. Discovered while compiling for x68k.

void in arg prototype.

Make ppp a cloning device. Based on the work of Quentin Garnier.

I changed pppoutput to use M_PREPEND. pppoutput was duplicating
the functionality of M_PREPEND, but with a bug: m_pkthdr.len was
not updated in pppoutput as it is in M_PREPEND.

Also, replace the loop that measures the length of the mbuf chain
with a call to m_length.

This fixes a PR from an anonymous bug reporter. Thank you, anonymous
bug reporter. Thanks, Itojun, for bringing the anonymous bug report
to my attention.

kill sprintf, use snprintf

Also, if we're going to bail, we should free the memory we just allocated...

Previous patch created a dead break.

Fix uninitialized variable warnings

Add a new ioctl PPPIOCGRAWIN to get the last characters we got from the
remote site.

branches: 1.84.2;
KNF

PR/20844: Iain Hibbert: PPP Compressors cannot be loaded as LKM

Remove variable that is only assigned too but not referenced.

backout previous two - if you use ppp* interface, kernel panics instantly.
it is apparent that the change was untested, and severety is high.

Remove unused variable so the file compiles again.

Keep m_pkthdr.len updated correctly and use it rather than a loop to find
out the total length of the packet.

new copyright boilerplate from CMU. from openbsd

branches: 1.77.2;
Make ppp_softc[] extern and declare in if_ppp.c

Convert ioctl code to use EPASSTHROUGH instead of -1 or ENOTTY for
indicating an unhandled "command". ERESTART is -1, which can lead to
confusion. ERESTART has been moved to -3 and EPASSTHROUGH has been
placed at -4. No ioctl code should now return -1 anywhere. The
ioctl() system call is now properly restartable.

bring in latest ALTQ from kjc. ALTQify some of the drivers.

Include <machine/intr.h> unconditionally, instead of only doing so if
__HAVE_GENERIC_SOFT_INTERRUPTS and relying on <sys/param.h> to provide it
otherwise; pointed out by Aymeric Vincent.

remove unnecessary #if NFOO > 0 .... #endif wrappers

add RCSIDs

use unsigned variable types as appropriate

bzero -> memset

branches: 1.69.2;
change the meaning of ifnet.if_lastchange to meet RFC1573 ifLastChange.
follows BSD/OS practice and ucd-snmp code (FreeBSD does it for specific
interfaces only).

was: if_lastchange get updated on every packet transmission/receipt.
now: if_lastchange get updated when IFF_UP is changed.

Remove the use of splimp() from the NetBSD kernel. splnet()
and only splnet() is allowed for the protection of data structures
used by network devices.

branches: 1.67.2;
Fix a rather annoying problem where the sockaddr_dl which holds
the link level name for the interface (ifp->if_sadl) is allocated
before ifp->if_addrlen is initialized, which could lead to allocating
too little space for the link level address.

Do this by splitting allocation of the link level name out of
if_attach() and into if_alloc_sadl(), which is normally called
by functions like ether_ifattach(). Network interfaces which
don't have a link-specific attach routine must call if_alloc_sadl()
themselves (example: gif).

Link level names are freed by if_free_sadl(), which can be called
from e.g. ether_ifdetach(). Drivers never need call if_free_sadl()
themselves as if_detach() will do it if it is not already done.

While here, add the ability to pass an AF_LINK address to
SIOCSIFADDR in ether_ioctl() (this is what caused me to notice
the problem that the above fixes).

For SLIP/STRIP/PPP, use generic soft interrupts, if available.

Fill in if_dlt.

Add ALTQ support.

Adapt to bpfattach() changes, and further centralize the bpfattach()
and bpfdetach() calls into link-type subroutines where possible.

Fix typo (thinko?), which didn't allow MRU to be set below the default
value (instead of the minimum value). Patch supplied by Stephen Degler
in PR #9945, and reviewed by Ignatios Souvatzis.

unique #include opt_inet.h

need opt_inet.h for #ifdef INET

enable VJC only with INET

branches: 1.58.4;
Kill some more register declarations.

New callout mechanism with two major improvements over the old
timeout()/untimeout() API:
- Clients supply callout handle storage, thus eliminating problems of
resource allocation.
- Insertion and removal of callouts is constant time, important as
this facility is used quite a lot in the kernel.

The old timeout()/untimeout() API has been removed from the kernel.

Fix typo introduced in rev. 1.55.
It caused IP6 packets to be sent as AF_UNSPEC instead of AF_INET6.

branches: 1.55.2; 1.55.8;
remove reference to in6_systm.h (file itself will be removed afterwords)

IPv6 kernel code, based on KAME/NetBSD 1.4, SNAP kit 19990628.
(Sorry for a big commit, I can't separate this into several pieces...)
Pls check sys/netinet6/TODO and sys/netinet6/IMPLEMENTATION for details.

- sys/kern: do not assume single mbuf, accept chained mbuf on passing
data from userland to kernel (or other way round).
- "midway" ATM card: ATM PVC pseudo device support, like those done in ALTQ
package (ftp://ftp.csl.sony.co.jp/pub/kjc/).
- sys/netinet/tcp*: IPv4/v6 dual stack tcp support.
- sys/netinet/{ip6,icmp6}.h, sys/net/pfkeyv2.h: IETF document assumes those
file to be there so we patch it up.
- sys/netinet: IPsec additions are here and there.
- sys/netinet6/*: most of IPv6 code sits here.
- sys/netkey: IPsec key management code
- dev/pci/pcidevs: regen

In my understanding no code here is subject to export control so it
should be safe.

Fix compilation problem caused by recent changes to filtering code.

Decouple inbound and outbound filters. Now instead of using "active-filter"
and "pass-filter" and "inbound" and "outbound" qualifiers in the filter
expression, use new "active-filter-in", "active-filter-out", "pass-filter-in",
and "pass-filter-out" without these qualifiers.

This is necessary due to the horrible, awful way "inbound" and "outbound"
were specified for the filter programs when a packet was passed through them.
Basically, the "address" byte in the serial PPP header was overwritten with
a value to indicate the direction. However, the "address" byte doesn't even
exist on PPP headers for all other PPP encaps! So, this old method worked
only for serial encaps, and corrupted packets for all others (PPPoE, ATM, etc.)

* Start out with a data link type of DLT_NULL. When we change an interface
to serial encap, change its data link type to DLT_PPP_SERIAL.
* Work around some serious bogosity in the filtering code which utterly
breaks proper functioning of BPF. The PPP code and pppd(8) WILL be changed
to fix this.

branches: 1.50.4; 1.50.6;
Use M_LINK{0,1} for our own mbuf flags, rather than arbitrarily picking
2 bits.

Revert IPX changes that I committed accidentally.

defopt

branches: 1.47.4;
PR/5414: Ronald Khoo: tcpdump ppp does not respect inbound/outbound qualifiers.

Fix PR5898: ppp delays last packet.

Glue in fast forwarding.

Only run pppasyncstart (sc->sc_start) from the netisr handler.
This allows pppoutput to be called from splimp (e.g., when ipflow is
in use.) without requiring pppasyncstart to run at splimp.
This is believed to fix PR5624.

use #ifdef INET so this compiles again

defopt INET, NETATALK.

Merge changes from pppd-2.3.4; adds ppp-deflate-draft stuff and updates
zlib. Maybe we can merge our other copy of zlib with this one now and
avoid having two copies?

From Paul Mackerras: use sl_compress_setup, not sl_compress_init

Update to ppp-2.3b5

Made pppoutput() public again on behalf of Martin Husemann (PR 3455).
Apparently, the BISDN package uses this function.

Update to ppp-2.3b4; from Paul Mackerras

Use splimp() to block interrupts, not splhigh().

branches: 1.35.4;
backout previous kprintf change

- printf -> kprintf, sprintf -> ksprintf

no need for a local implementation of SIOCGIFMTU; delete it.

Pass a proc pointer down to the usrreq and pcbbind functions for PRU_ATTACH, PRU_BIND and
PRU_CONTROL. The usrreq interface really needs to be split up, but this will have to wait.
Remove SS_PRIV completely.

branches: 1.31.4;
Changed struct ifnet to have a pointer to the softc of the underlying
device and a printable "external name" (name + unit number), thus eliminating
if_name and if_unit. Updated interface to (*if_watchdog)() and (*if_reset)()
to take a struct ifnet *, rather than a unit number.

Make inclusion of the PPP BSD-Compress and Deflate compressors
dependent on the PPP_BSDCOMP and PPP_DEFLATE kernel configuration
options, respectively.

Added packet filtering, support for "PPP Deflate" packet compression,
trivial multicast support, and support for xon/xoff output flow
control to the PPP subsystem. Fixed several bugs, including making
the accumulation and resetting of statistics more consistent. State
for the VJ compressor is now dynamically allocated.

Net prototypes

wrt. previous change: can't compute `ilen' that early; just do computation
separately when logging.
Notes: consider using mbuf pkthdr length field in PPP code.
condider doing packet log after de-compression.

Init variable before use (PRs 1646 & 2042).

Remove old workaround for a bug.

Add some missing statistics. From Thorsten Lockert.

splnet --> splsoftnet

Latest version of PPP stuff, with packet compression and other
improvements. The PPP kernel code is now split into if_ppp.c,
containing generic PPP support, and ppp_tty.c, which specifically
supports PPP on async tty devices (as a line discipline). This is
so that other devices can be supported without making them look
like ttys.

fixed sized types, where appropriate. when casting pointers to
integers to do math on them, cast to long. ioctl commands are
u_longs.

be more careful with types, also pull in headers where necessary.

bug 367. paulus says the fix is right & critical.

The ppp interface now retries the mbuf allocation when it receives
a character and it doesn't already have enough space allocated.
It still needs cluster mbufs to be able to decompress VJ-compressed
packets. It drops packets if it can't allocate mbufs rather than
taking the interface down.

Due to popular revulsion, the ppp interface now drops packets if

If we can't get mbuf clusters, take the ppp interface down instead
of writing beyond the end of ordinary mbufs.

branches: 1.15.2;
bug #319. fix from <alasdair@wildcat.demon.co.uk>

New RCS ID's, take two. they're more aesthecially pleasant, and use 'NetBSD'

Some restructuring of the PPP packet input procedure to make it easier
to implement PPP over sync lines and PPP compression protocols.

some minor splx-type bug fixes from christos@deshaw.com.

check for escaped char before checking for escape char
so if peer escapes 0x5d we interpret it correctly

MIN -> min, MAX -> max

Update to 4.4-Lite networking code, with a few local changes.

Version from ppp-2.1 release;
this version has been restructured to make more of the code usable
with sync serial drivers

PPP_HEADER_LEN -> PPP_HDRLEN

include <machine/cpu.h> rather than <machine/mtpr.h> -- if the latter
exists at all, it's supposed to be included by <machine/cpu.h>

Canonicalize all #includes.

Removed test (CCOUNT(&sc->sc_ttyp->t_outq) == 0) for whether
to call pppstart or not: now we call pppstart for every packet,
which should aid recovery from lost transmitter interrupts.
Also a fix for 386BSD/FreeBSD which doesn't affect NetBSD.

branches: 1.3.2;
Fixed bug in if_ppp.c so that received IP packets are passed correctly to BPF.

Modified if_ppp.c and if_ppp.h to add priority queueing for "interactive"
traffic (done in a similar fashion to if_sl.c), and BPF support.

ppp from paul mackerras

net/if_ppp.c: Avoid user-controlled overrun in PPPIOCSCOMPRESS.

Reported-by: syzbot+2c7bda7dc2b6c0d4f279@syzkaller.appspotmail.com

net/if_ppp.c: Sprinkle KNF. No functional change intended.

Adopt <net/if_stats.h>.

branches: 1.166.2;
dedup

Simplify "LIST_HEAD();" to make the code more understandable.
No functional change.

Add __cacheline_aligned to ppp softc list and its mutex just in case.

Fix missing splx in ppp_inproc().

Fix missing mutex_exit in ppp_create().

branches: 1.161.2;
Implement the BPF direction filter (BIOC[GS]DIRECTION). It provides backward
compatibility with BIOC[GS]SEESENT ioctl. The userland interface is the same
as FreeBSD.

This change also fixes a bug that the direction is misunderstand on some
environment by passing the direction to bpf_mtap*() instead of checking
m->m_pkthdr.rcvif.

Remove duplicated inclusion of net/bpf.h.

branches: 1.159.2;
Add one more not supported error

branches: 1.158.8;
MFREE -> m_free

modularize some more drivers and merge the module glue

Destroy the mutex when detaching ppp. Otherwise on a re-attach (ie,
module reload) we can end up with a panic "lock already initialized"

make strip and slip modular, and cosmetic for ppp.

Change the internal name of the module to match its external (file
system) name. Otherwise "bad things" can happen, such as modload(8)
being able to load a second copy!

Modularize the ppp driver, and adjust dependencies of the compressor
modules.

For now, this is still included as a built-in module in GENERIC kernels.

branches: 1.152.2;
Introduce m_set_rcvif and m_reset_rcvif

The API is used to set (or reset) a received interface of a mbuf.
They are counterpart of m_get_rcvif, which will come in another
commit, hide internal of rcvif operation, and reduce the diff of
the upcoming change.

No functional change.

Constify rtentry of if_output

We no longer need to change rtentry below if_output.

The change makes it clear where rtentries are changed (or not)
and helps forthcoming locking (os psrefing) rtentries.

IFQ_ENQUEUE refactor (3/3) : eliminate pktattr argument from IFQ_ENQUEUE caller

sprinkle _KERNEL_OPT

Honor pseudo attach decl generated by config(1).

Introduce p2p_rtrequest() so that IFF_POINTOPOINT interfaces can work
with RTF_LOCAL.
Fixes PR kern/49829.

branches: 1.146.4;
KNF. No functional change.

Cleanup ppp_inproc

- Remove unnecessary variable isr
- Use pktq instead of rv to switch between inet/inet6 and other protocols

ok msaitoh@ and rmind@

- Implement pktqueue interface for lockless IP input queue.
- Replace ipintrq and ip6intrq with the pktqueue mechanism.
- Eliminate kernel-lock from ipintr() and ip6intr().
- Some preparation work to push softnet_lock out of ipintr().

Discussed on tech-net.

ppp_inproc: handle non-IP protocols correctly (hi msaitoh); PR/48813.

Save a NETISR_* value in a variable and call schednetisr() after enqueue
a packet for readability and future modification.

branches: 1.141.2;
Add bpf_filter_ext() to use with BPF COP, restore bpf_filter() as it was
originally to preserve compatibility. Similarly, add bpf_validate_ext()
which takes bpf_ctx_t.

bpf_filter: add a custom argument which can be passed to coprocessor routine.

Implement BPF_COP/BPF_COPX instructions in the misc category (BPF_MISC)
which add a capability to call external functions in a predetermined way.

It can be thought as a BPF "coprocessor" -- a generic mechanism to offload
more complex packet inspection operations. There is no default coprocessor
and this functionality is not targeted to the /dev/bpf. This is primarily
targeted to the kernel subsystems, therefore there is no way to set a custom
coprocessor at the userlevel.

Discussed on: tech-net@
OK: core@

branches: 1.138.2;
Don't check mp for NULL twice. From Michael W. Bomardieri <mb@il.net>
via tech-net@NetBSD.org. Thanks!

PR/47058: Antti Kantee: If the ipv6 flow code modifies the mbuf, pass the
change up to the caller.

branches: 1.136.2; 1.136.8; 1.136.12;
For these interfaces, the implementation of SIOCSIFDSTADDR is identical
to SIOCINITIFADDR, and SIOCSIFDSTADDR callers always fall back to
SIOCINITIFADDR, so just get rid of the SIOCSIFDSTADDR case.

Don't kauth-orize SIOCSIFMTU in pppsioctl() and stf_ioctl(), ifioctl()
has already done that for us.

Convert ppp_list_lock to mutex(9).

Fix misplaced parenthesis. From henning.petersen@t-online.de, thanks.

branches: 1.132.2;
Update the rest of the kernel to conform to the module subsystem's new
locking protocol.

Push the bpf_ops usage back into bpf.h. Push the common ifp->if_bpf
check into the inline functions as well the fourth argument for
bpf_attach.

branches: 1.130.2; 1.130.4;
Redefine bpf linkage through an always present op vector, i.e.
#if NBPFILTER is no longer required in the client. This change
doesn't yet add support for loading bpf as a module, since drivers
can register before bpf is attached. However, callers of bpf can
now be modularized.

Dynamically loadable bpf could probably be done fairly easily with
coordination from the stub driver and the real driver by registering
attachments in the stub before the real driver is loaded and doing
a handoff. ... and I'm not going to ponder the depths of unload
here.

Tested with i386/MONOLITHIC, modified MONOLITHIC without bpf and rump.

Remove a few KAUTH_GENERIC_ISSUSER in favor of more descriptive
alternatives.

Discussed on tech-kern:

http://mail-index.netbsd.org/tech-kern/2009/04/11/msg004798.html

Input from ad@, christos@, dyoung@, tsutsui@.

Okay ad@.

branches: 1.128.2;
ppp_get_compressor: take module_lock when trying to load a module. PR/40428

kill MALLOC and FREE macros.

Fix handling of ppp compressor modules, from Andrew Doran's input.
- ref count each compressor
- allow {un,}registration of several modules at once
- une RUN_ONCE to make sure the mutex is initialised, because
unfortunately built-in (and bootloader-loaded) modules init functions
are run before pseudo-devices attach (reported by Nick Hudson).

Rework the way PPP compmressors are handled and allow them to be
automatically loaded when needed.

*** Summary ***

When a link-layer address changes (e.g., ifconfig ex0 link
02:de:ad:be:ef:02 active), send a gratuitous ARP and/or a Neighbor
Advertisement to update the network-/link-layer address bindings
on our LAN peers.

Refuse a change of ethernet address to the address 00:00:00:00:00:00
or to any multicast/broadcast address. (Thanks matt@.)

Reorder ifnet ioctl operations so that driver ioctls may inherit
the functions of their "class"---ether_ioctl(), fddi_ioctl(), et
cetera---and the class ioctls may inherit from the generic ioctl,
ifioctl_common(), but both driver- and class-ioctls may override
the generic behavior. Make network drivers share more code.

Distinguish a "factory" link-layer address from others for the
purposes of both protecting that address from deletion and computing
EUI64.

Return consistent, appropriate error codes from network drivers.

Improve readability. KNF.

*** Details ***

In if_attach(), always initialize the interface ioctl routine,
ifnet->if_ioctl, if the driver has not already initialized it.
Delete if_ioctl == NULL tests everywhere else, because it cannot
happen.

In the ioctl routines of network interfaces, inherit common ioctl
behaviors by calling either ifioctl_common() or whichever ioctl
routine is appropriate for the class of interface---e.g., ether_ioctl()
for ethernets.

Stop (ab)using SIOCSIFADDR and start to use SIOCINITIFADDR. In
the user->kernel interface, SIOCSIFADDR's argument was an ifreq,
but on the protocol->ifnet interface, SIOCSIFADDR's argument was
an ifaddr. That was confusing, and it would work against me as I
make it possible for a network interface to overload most ioctls.
On the protocol->ifnet interface, replace SIOCSIFADDR with
SIOCINITIFADDR. In ifioctl(), return EPERM if userland tries to
invoke SIOCINITIFADDR.

In ifioctl(), give the interface the first shot at handling most
interface ioctls, and give the protocol the second shot, instead
of the other way around. Finally, let compatibility code (COMPAT_OSOCK)
take a shot.

Pull device initialization out of switch statements under
SIOCINITIFADDR. For example, pull ..._init() out of any switch
statement that looks like this:

switch (...->sa_family) {
case ...:
..._init();
...
break;
...
default:
..._init();
...
break;
}

Rewrite many if-else clauses that handle all permutations of IFF_UP
and IFF_RUNNING to use a switch statement,

switch (x & (IFF_UP|IFF_RUNNING)) {
case 0:
...
break;
case IFF_RUNNING:
...
break;
case IFF_UP:
...
break;
case IFF_UP|IFF_RUNNING:
...
break;
}

unifdef lots of code containing #ifdef FreeBSD, #ifdef NetBSD, and
#ifdef SIOCSIFMTU, especially in fwip(4) and in ndis(4).

In ipw(4), remove an if_set_sadl() call that is out of place.

In nfe(4), reuse the jumbo MTU logic in ether_ioctl().

Let ethernets register a callback for setting h/w state such as
promiscuous mode and the multicast filter in accord with a change
in the if_flags: ether_set_ifflags_cb() registers a callback that
returns ENETRESET if the caller should reset the ethernet by calling
if_init(), 0 on success, != 0 on failure. Pull common code from
ex(4), gem(4), nfe(4), sip(4), tlp(4), vge(4) into ether_ioctl(),
and register if_flags callbacks for those drivers.

Return ENOTTY instead of EINVAL for inappropriate ioctls. In
zyd(4), use ENXIO instead of ENOTTY to indicate that the device is
not any longer attached.

Add to if_set_sadl() a boolean 'factory' argument that indicates
whether a link-layer address was assigned by the factory or some
other source. In a comment, recommend using the factory address
for generating an EUI64, and update in6_get_hw_ifid() to prefer a
factory address to any other link-layer address.

Add a routing message, RTM_LLINFO_UPD, that tells protocols to
update the binding of network-layer addresses to link-layer addresses.
Implement this message in IPv4 and IPv6 by sending a gratuitous
ARP or a neighbor advertisement, respectively. Generate RTM_LLINFO_UPD
messages on a change of an interface's link-layer address.

In ether_ioctl(), do not let SIOCALIFADDR set a link-layer address
that is broadcast/multicast or equal to 00:00:00:00:00:00.

Make ether_ioctl() call ifioctl_common() to handle ioctls that it
does not understand.

In gif(4), initialize if_softc and use it, instead of assuming that
the gif_softc and ifp overlap.

Let ifioctl_common() handle SIOCGIFADDR.

Sprinkle rtcache_invariants(), which checks on DIAGNOSTIC kernels
that certain invariants on a struct route are satisfied.

In agr(4), rewrite agr_ioctl_filter() to be a bit more explicit
about the ioctls that we do not allow on an agr(4) member interface.

bzero -> memset. Delete unnecessary casts to void *. Use
sockaddr_in_init() and sockaddr_in6_init(). Compare pointers with
NULL instead of "testing truth". Replace some instances of (type
*)0 with NULL. Change some K&R prototypes to ANSI C, and join
lines.

branches: 1.123.2; 1.123.4;
- add if_alloc (ours just mallocs), and if_initname and use them (from FreeBSD)
- kill memsets where M_ZERO can be used.

branches: 1.122.2; 1.122.4; 1.122.6;
Merge the socket locking patch:

- Socket layer becomes MP safe.
- Unix protocols become MP safe.
- Allows protocol processing interrupts to safely block on locks.
- Fixes a number of race conditions.

With much feedback from matt@ and plunky@.

branches: 1.121.6; 1.121.8;
Start patching up the kernel so that a network driver always has
the opportunity to handle an ioctl before generic ifioctl handling
occurs. This will ease extending the kernel and sharing of code
between drivers.

First steps: Make the signature of ifioctl_common() match struct
ifinet->if_ioctl. Convert SIOCSIFCAP and SIOCSIFMTU to the new
ifioctl() regime, throughout the kernel.

Start detangling lock.h from intr.h. This is likely to cause short term
breakage, but the mess of dependencies has been regularly breaking the
build recently anyhow.

branches: 1.119.2; 1.119.8;
machine/{bus,cpu,intr}.h -> sys/{bus,cpu,intr}.h

branches: 1.118.2;
Use the softint API.

branches: 1.117.2;
Use ifreq_setaddr(), ifreq_getaddr(), sockaddr_in_init(), and
sockaddr_copy(). Constify. Compare pointers with NULL, not 0.
Don't "test truth" of pointers, but compare with NULL.

branches: 1.116.2; 1.116.6; 1.116.8;
Generic soft interrupts are mandatory.