CVS log for src/sbin/ifconfig/ifconfig.c

Up to [cvs.NetBSD.org] / src / sbin / ifconfig

Request diff between arbitrary revisions

Keyword substitution: kv
Default branch: MAIN

Pull up following revision(s) (requested by ozaki-r in ticket #811):

	tests/net/if_shmif/t_shmif.sh: revision 1.1
	sbin/ifconfig/ifconfig.c: revision 1.251
	sbin/ifconfig/ifconfig.8: revision 1.130
	sys/rump/net/lib/libshmif/if_shmem.c: revision 1.85
	sys/rump/net/lib/libshmif/if_shmem.c: revision 1.86
	sys/rump/net/lib/libshmif/if_shmem.c: revision 1.87
	etc/mtree/NetBSD.dist.tests: revision 1.206
	distrib/sets/lists/tests/mi: revision 1.1333
	tests/net/if_shmif/Makefile: revision 1.1
	tests/net/Makefile: revision 1.42

shmif: change behaviors about link states

- Change the link state to UP on ifconfig linkstr
  - This behavior emulates physical devices
- Change the link state to UNKNOWN on ifconfig -linkstr just in case
- Reject sending/receiving packets if the link state is DOWN
  - Permit to send/receive packets on UNKNOWN, which is required
    to unbreak some ATF tests written in C

shmif: support media

It enables to link-down shmif by ifconfig media none and link-up
again by media auto.

ifconfig: show link state on -v

We could guess it through "media" or "status" output, however, we
sometimes want to know it directly for debugging or testing.

It is shown only if the -v option is specified.
tests: add tests for shmif

The test file is placed under tests/net, not tests/rump/rumpnet,
to leverage utility functions provided for tests in there.
shmem(4): Fix typo in comment: AFT -> ATF.

Also fix grammar (if I understood correctly what this meant: rump
servers written in C, rather than set up via shell scripts around
rump_server invoking ifconfig).

No functional change intended.

ifconfig: show link state on -v

We could guess it through "media" or "status" output, however, we
sometimes want to know it directly for debugging or testing.

It is shown only if the -v option is specified.

s/addreseses/addresses/ in comments (and one missing whitespace).

ifconfig(8) support unnumbered flag.

ifconfig: go back to using direct ioctls rather than data from getifaddrs

AF_LINK may not be the first address returned for the interface.
Technically, it *might* not even exist on the interface even though
other families do.
This is likely a driver bug if this really is the case though.

As such it's just easier to use direct ioctls rather than thump around
getifaddrs results. As it stands, the code makes a lot of getifaddrs
calls anyway, so an extra ioctl or two won't break the bank.

ifconfig: Just look at if_data->ifi_link_state for carrier

It should be the sole source of truth.
if_data is also carried in ifa_data from getifaddrs(3) which saves
more ioctl calls.

ifconfig: We already have struct if_data in ifa_data.

So let's not bother with an ioctl we don't need.

ifconfig: test cases want LINK_STATE_UNKNOWN to be treated as up.

ifconfig: teach carrier test about ifi_link_state from SIOCGIFDATA

Because not all interfaces support media.

ifconfig: Report link state even if media is not supported

For AF_LINK addrs from getifaddrs(2), ifa_data is struct if_data.
This in turn holds ifi_link_state which we can use to report
link status if the interface does not support media where it's normally
reported.

Based on OpenBSD.

Update for proplib(3) API changes.

Mostly merge changes from HEAD upto 20200411

Merge changes from current as of 20200406

fix printing capabilities and enabled flags.

no need to skip a prefix "0x" of output of snprintb.
ifconfig expect the result of snprintb has "0x" prefix, but snprintb(3) specification has changed
so that if 0 is passed, it prints "0" instead of "0x0". (common/lib/libutil/snprintb.c r1.22)

 Whitespace fix.

Add descr,description/-descr,-description commands to ifconfig(8) to handle a description.

From t-kusaba@IIJ

Sync with HEAD

Sync with HEAD, resolve a few conflicts

 Add SIOCSETHERCAP. It's used to change ec_capenable.

Sync with HEAD, resolve a couple of conflicts

ifconfig: add linkstr/-linkstr to the help message

-W seconds will wait for the detached flag to clear on addresses on
interfaces marked up to allow time for the carrier to appear on the
interface.

This does not extend the -w option duration.

Don't divide flags output

If there are many enabled flags, ifconfig divides flags output into
multiple formatted strings due to snprintb_m, e.g.,
  wm0: flags=8b43<UP,BROADCAST,RUNNING,PROMISC,ALLMULTI,SIMPLEX>wm0: flags=8b43<MULTICAST> mtu 1500

This behavior is probably unexpected. The change always outputs enabled
flags at once like this:
  wm0: flags=8b43<UP,BROADCAST,RUNNING,PROMISC,ALLMULTI,SIMPLEX,MULTICAST> mtu 1500

Move the INET6 specific code for wait_dad_exec() into af_inet6
by using a new afswtch hook af_addr_tentative.

Initialize timespec "end" for the benefit of some gcc versions getting
"may be used uninitialized" warnings wrong.

Remove rtsol(8) and rtsold(8) as their functionality is in dhcpcd(8).
Remove rtsol(8) from rc.d/network.
Add -w seconds command to ifconfig to wait for N seconds for until DAD
has finished on all addresses.
Use ifconfig -w in rc.d/network instead of a forced sleep.

As discussed on tech-net@

Rebase to HEAD as of a few days ago.

sync with head.

for a reference, the tree before this commit was tagged
as yamt-pagecache-tag8.

this commit was splitted into small chunks to avoid
a limitation of cvs.  ("Protocol error: too many arguments")

use the new scopeid functions

Allow -v to be used with other commands (list scan for example) as documented.

resync with head

Abort on error from prop_dictionary_augment.

sync with (a bit old) head

Resync to 2012-11-19 00:00:00 UTC

Use snprintb_m(3) to split flags/capabilities/enabled across multiple
output lines.

As discussed on current-users

sync with head

When ifconfig is invoked without any arguments, display the list of
all network interfaces (equivalent to ifconfig -a), instead of displaying
the usage.  (This is consistent with the behaviour on FreeBSD and OpenBSD.)

Use __dead

A few weeks ago when i added "linkstr" support I didn't know that
the string parser gives a proplib object which isn't a terminated
string.  Fix that mistake now.

Convert from the .ifdef RUMP_ACTION stuff to RUMPPRG.

Implement ifconfig linkstr as proposed on tech-net.

adapt to new rumpclient_init() signature

Update RUMP_ACTION to use the newstyle rump client.

sync to netbsd-5

Add option -N.  -N is just the opposite of option -n in netstat(8)
or route(8): it tells ifconfig(8) to try to resolve numbers to
hosts and service names.

This default ifconfig behavior stays the same as it always was.

Add compile mode which executes rump system calls.  This is useful
when figuring out Where In The Kernel Is Carmen Sandiego's ioctl
for an ifconfig command line, since we can simply single-step into
the kernel.

Activated by "make RUMP_ACTION=1".  No changes to normal case.

Sync with HEAD.

Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html

Pull up following revision(s) (requested by dyoung in ticket #730):
	sbin/ifconfig/af_link.c: revisions 1.4 - 1.6
	sbin/ifconfig/util.h: revision 1.7
	sbin/ifconfig/util.c: revisions 1.10, 1.11
	lib/libc/net/getifaddrs.c: revision 1.12
	lib/libc/net/getifaddrs.3: revision 1.10
	sbin/ifconfig/ifconfig.c: revisions 1.216 - 1.218
Fix indentation: change spaces to tabs.
Use getnameinfo(3) to render a human-readable link-layer address in the
'address: ' line, just as we do in the 'link xx:xx:...:xx' line.
There's no use casting a socket address to sockaddr_dl, only to cast it
back to sockaddr, so don't do it.
Cosmetic: add some whitespace for my ease of reading.
To make sure that we always print the active link-layer address in the
'address: ' field, don't treat the first address as the active address,
but search the link-layer addresses for the ones flagged IFLR_ACTIVE,
and print those.  Extract a subroutine, print_link_addresses(), for
printing link-layer addresses.
For non-AF_LINK ifaddrs, ifa_data is NULL.  AFAICT, this has always been
so.  Say so in the documentation.
Bring getifaddrs(3) behavior in line with the documentation: the
ifa_data member of every AF_LINK struct ifaddrs points at the
corresponding struct if_data.  In ifconfig(8), do not try to suppress
duplicate AF_LINK ifaddrs by checking for a NULL ifa_data.
Don't copy out two AF_LINK struct ifaddrs for each active link-layer
address. getifaddrs(3) used to copy out one ifaddrs for the kernel's
RTM_IFINFO message, and one more for the kernel's RTM_NEWADDR message.
I suppress the first duplicate with a highly conservative change that
wastes a little bit of ifaddrs storage.  The storage is not leaked.

To make sure that we always print the active link-layer address in the
'address: ' field, don't treat the first address as the active address,
but search the link-layer addresses for the ones flagged IFLR_ACTIVE,
and print those.  Extract a subroutine, print_link_addresses(), for
printing link-layer addresses.

Cosmetic: add some whitespace for my ease of reading.

There's no use casting a socket address to sockaddr_dl, only to cast it
back to sockaddr, so don't do it.

Pull up following revision(s) (requested by plunky in ticket #665):
	sbin/ifconfig/ifconfig.c: revision 1.215
Because the IFF_NOARP flag is a negative option it needs to be negated
when being applied from the "arp" option which itself is positive.
problem demonstrated by
# ifconfig tap3 create
# ifconfig tap3
tap3: flags=8802<BROADCAST,SIMPLEX,MULTICAST> mtu 1500
        address: f2:0b:a4:7f:44:03
        media: Ethernet autoselect
# ifconfig tap3 arp
# ifconfig tap3
tap3: flags=8882<BROADCAST,NOARP,SIMPLEX,MULTICAST> mtu 1500
        address: f2:0b:a4:7f:44:03
        media: Ethernet autoselect
# ifconfig tap3 -arp
# ifconfig tap3
tap3: flags=8802<BROADCAST,SIMPLEX,MULTICAST> mtu 1500
        address: f2:0b:a4:7f:44:03
        media: Ethernet autoselect
# ifconfig tap3 destroy
#

Because the IFF_NOARP flag is a negative option it needs to be negated
when being applied from the "arp" option which itself is positive.

problem demonstrated by

# ifconfig tap3 create
# ifconfig tap3
tap3: flags=8802<BROADCAST,SIMPLEX,MULTICAST> mtu 1500
	address: f2:0b:a4:7f:44:03
	media: Ethernet autoselect
# ifconfig tap3 arp
# ifconfig tap3
tap3: flags=8882<BROADCAST,NOARP,SIMPLEX,MULTICAST> mtu 1500
	address: f2:0b:a4:7f:44:03
	media: Ethernet autoselect
# ifconfig tap3 -arp
# ifconfig tap3
tap3: flags=8802<BROADCAST,SIMPLEX,MULTICAST> mtu 1500
	address: f2:0b:a4:7f:44:03
	media: Ethernet autoselect
# ifconfig tap3 destroy
#

fix -Wsign-compare issues

Sync with HEAD.

Sync with wrstuden-revivesa-base-2.

Extract common code.  Create subroutine paddr_prefix_size().

Remove the \n and tabs from the __COPYRIGHT() strings.
(Tweak some to use a consistent format.)

Cosmetic: s/xenv/oenv/

Make usage() static.

Only describe flags -L and -m in the usage if they are available.
That helps me get rid of some conditional compilation (INET6) in
ifconfig.

Let each protocol/feature-module print its own usage, so that the
ifconfig usage reflects the modules that are actually compiled-in.

Write usage information for carp(4) options.

Let us add/remove features from ifconfig, such as support for
various address families (inet, inet6, iso, atalk) and protocols
(802.11, 802.3ad, CARP), simply by trimming the list of sources in
the Makefile.  This helps one customize ifconfig for an embedded
device or for install media, and it eliminates a lot of grotty
#ifdef'age.  Now, the ifconfig syntax and semantics are finalized
at run-time using the constructor routines in each address-family/protocol
module.

(In principle, ifconfig could load virtually all of its syntax from
shared objects.)

Extract a lot of common code into subroutines, in order to shrink
the ifconfig binary a bit.  Make all of the address families share
code for address addition/replacement/removal, and delete "legacy"
code for manipulating addresses.  That may have broken atalk and
iso, despite my best efforts.

Extract an include file, Makefile.inc, containing the make-fu that
both ifconfig and x_ifconfig share.

Sprinkle static.  Change some int's to bool's.  Constify.

Add RCS Ids to carp.c and env.c.  Move media code to a new file,
media.c.  Delete several unneeded header files.

Set, reset, and display the IEEE 802.11 attribute, 'dot11RTSThreshold'.

Bug fix: do not require both a interface address and a destination
address for point-to-point interfaces, but accept a interface
address by itself.

Sync with HEAD.

Sync w/ -current. 34 merge conflicts to follow.

sync with head.

sync with head

Remove one more struct afswtch.af_getprefix() remnant.

Remove remnants of struct afswtch.af_getprefix().

Sync with HEAD.

Fix two bugs:

1 Let us associate a keyword with either a signed or an unsigned
  number.  Associate the '-chan' keyword with an unsigned number.
  This fixes the defect that crashed 'ifconfig wi0 -chan', reported
  by Michael van Elst.

2 Do not try to interpret names as ISO addresses unless the ISO
  address family was specified.  iso_addr(3) does not report errors,
  so it appears to interpret *everything* as an ISO address.  This
  is a stopgap fix for the defect in 'ifconfig lo0 inet6' on a
  !INET6 kernel that deletes the first IPv4 address on lo0.
  (Reported by Scott Ellis.)

sync with head.

Per discussion at
<http://mail-index.netbsd.org/tech-net/2008/04/08/msg000371.html>,
let us add, delete, and activate link-layer addresses with ifconfig:

# ifconfig sip0 link 02:00:00:00:00:01                   [add address]
# ifconfig sip0 link 02:00:00:00:00:02                   [add address]
# ifconfig sip0 link 02:00:00:00:00:02 active            [activate address]
# ifconfig sip0 link 02:00:00:00:00:01 delete            [remove address]

Let the address family modules zero initialize their own ifreqs,
instead of zeroing them in commit_address().

Switch to in6_commit_address() from in6_getprefix() and in6_getaddr().

Temporarily add some debugging code to setia6eui64_impl().

Move IPv4/IPv6-specific code from commit_address() to in_commit_address()
and in6_commit_address().  Fixes the USE_INET6=no build.

Factor out common code, creating direct_ioctl() and indirect_ioctl()
for the two most prevalent styles of ioctl(2) calls in ifconfig(8).

Move IPv6 syntax to af_inet6.c from ifconfig.c.  Move IFKW() macro
from ifconfig.c to parse.h for reuse.

Move ISO, AppleTalk, carp(4) syntax from ifconfig.c to af_iso.c,
af_atalk.c, carp.c, respectively.

Sprinkle 'static const' on keyword objects.

Move vlan(4) syntax from ifconfig.c into vlan.c.

Move tunnel syntax into tunnel.c.  Move net80211 'list' command
syntax into ieee80211.c.

Move net80211 and agr(4) syntax out of ifconfig.c and into ieee80211.c
and agr.c, respectively.

Fix a bug that I introduced in rev 1.88, where the default IPv6
prefix length changed from 64 to 128.  While I am here, allow an
address to be given with prefix length /0, but interpret it as
/128.

Use the right key, "ifcaps" not "ifcap", in do_setifcaps().  assert()
that the object associated with that key is the right size.

Extract subroutine do_setifcaps() and fix the bug that uwe@ reported
in capabilities-setting.

Use prop_dictionary_util(3).

Rename the parser snpaoffset to parse_snpaoffset.

Fix syntax error in last.

Cosmetic: use return rc ? 0 : -1;

Make sure to release a prop_object_t that I am no longer using.

Use prop_dictionary_util(3) some more.

When we read interface flags and capabilities from the kernel, take
care not to record them in our current environment (env), but record
them in the output environment (oenv), instead.  This helps us get
interface capabilities and flags right.

Rename in_addr_commit() to commit_address(), and refactor slightly
to support IPv6 as well as IPv4 (a work in progress).

Make the second argument of af_status() a bool instead of an int.

Exit early with an error if the operator specifies an unsupported
address family on the command line.  The change should help rc
scripts to detect that IPv6 support is missing from the kernel,
with 'ifconfig lo0 inet6'.

Start using prop_dictionary_util(3).

Overhaul ifconfig.  Use fewer global variables.  Take a leap toward
improved modularity and extensibility.

In the new architecture, a directed graph of argument-matching
objects (match objects) expresses the set of feasible ifconfig
statements.  Match objects are labelled by subroutines that provide
the statement semantics.

Many IPv4, IPv6, 802.11, tunnel, and media configurations have been
tested.

AppleTalk, ISO, carp(4), agr(4), and vlan(4) configuration need
testing.

Remove clause 3 and 4 from TNF licenses

Change int to bool in a couple of places.

Straggler from last: call fewer protocol family-specific initialization
routines from main(), since the protocol-family modules are taking
care of it.

Cosmetic: delete dead code.

Cosmetic: use NULL instead of 0 for pointers.  Shorten staircase.

imported Mobile IPv6 code developed by the SHISA project
(http://www.mobileip.jp/).

sync with HEAD

Sync with HEAD.

Add support for the command list scan in ifconfig (this command permits to scan
access point in the neighbourhood).

Complete list of channel attributes and list of management information element
payload.

While here, use estrlcpy instead of strncpy.

From FreeBSD ifconfig and net80211

Pull up following revision(s) (requested by msaitoh in ticket #1801):
	sbin/ifconfig/ifconfig.c: revision 1.173 via patch

        If the kernel does not set the interface up after we set an address,
        do it explicitly, unless we (previously) set it down on user request
        (i.e. "ifconfig <name> x.x.x.x down").

        This will allow the network drivers to be "fixed" while keeping visible
        behaviour the same. Part of fixing PR 30694.

Let us adjust the 802.11 fragmentation threshold with ifconfig ath0
frag 256, for example.

implement ipv6 TSO.
partly from Matthias Scheler.  tested by him.

Add a source-address selection policy mechanism to the kernel.

Also, add ioctls SIOCGIFADDRPREF/SIOCSIFADDRPREF to get/set preference
numbers for addresses.  Make ifconfig(8) set/display preference
numbers.

To activate source-address selection policies in your kernel, add
'options IPSELSRC' to your kernel configuration.

Miscellaneous changes in support of source-address selection:

        1 Factor out some common code, producing rt_replace_ifa().

        2 Abbreviate a for-loop with TAILQ_FOREACH().

        3 Add the predicates on IPv4 addresses IN_LINKLOCAL() and
          IN_PRIVATE(), that are true for link-local unicast
          (169.254/16) and RFC1918 private addresses, respectively.
          Add the predicate IN_ANY_LOCAL() that is true for link-local
          unicast and multicast.

        4 Add IPv4-specific interface attach/detach routines,
          in_domifattach and in_domifdetach, which build #ifdef
          IPSELSRC.

See in_getifa(9) for a more thorough description of source-address
selection policy.

- get rid of p_cfunc2
- add missing initializers

Programs that use efun.

Nuke a leftover XNS fragment.

Remove XNS support.

Trying the old "ifconfig <name> x.x.x.x down" issue again:
if the kernel does not set the interface up after we set an address,
do it explicitly, unless we (previously) set it down on user request.

This will allow the network drivers to be "fixed" while keeping visible
behaviour the same. Part of fixing PR 30694.

fix incorrect usage of strncpy() to (an internal implementation of)
estrlcpy().

okay christos

Integrate Common Address Redundancy Procotol (CARP) from OpenBSD

'pseudo-device	carp'

Thanks to: joerg@ christos@ riz@ and others who tested
Ok: core@

Add two new options from FreeBSD:
	hidessid | -hidessid:	SSID broadcast control
	apbridge | -apbridge:	how to pass packets between wifi clients

Pull up following revision(s) (requested by rpaulo in ticket #1219):
	sbin/ifconfig/ifconfig.c: revision 1.169
Added a new ifconfig flag, -h, allowing, in conjunction with -v, the display
of interface I/O bytes in human-format.
Reviewed by Christos Zoulas and Jason Thorpe.

Added a new ifconfig flag, -h, allowing, in conjunction with -v, the display
of interface I/O bytes in human-format.

Reviewed by Christos Zoulas and Jason Thorpe.

Pull up revision 1.147 via patch (requested by riz in ticket #5518):
Functions for converting media types to/from text are now in libutil.

Pull up revision 1.142 (requested by riz in ticket #5517):
The change in revision 1.38 (in 1997!) was wrong.  If we don't want the kernel
to automatically bring up interfaces, then we should move that code into
ifconfig(8) and out of the kernel, not cycle the interface up and down.  Fixes
PR 9400 -- ifconfig(8) was not working as documented.

split IFCAP_CSUM_xxx to IFCAP_CSUM_xxx_Rx and IFCAP_CSUM_xxx_Tx.

IFFBITS and IFCAPBITS are now defined in <net/if.h>

Split the IPv4 support into its own file.

Move a comment.

Split IPv6 support out into its own file.

Split OSI support off into its own file.

Split XNS support into its own file.

Split off AppleTalk support into its own file.

const'ify afswtch

Split off 802.11 interface handling into its own file.

Don't hare-code inet/inet6 in tunnel_status(), lookup the address family
name instead.

Split tunnel support into its own file.

Move extern decls to extern.h

Split VLAN configuration bits out into a separate file.

update usage() for agr(4).  pointed by Thomas Klausner.

add agr(4), a pseudo network device driver for link aggregation.

Add ifconfig support for TCP Segment Offload

remove obsolete "register" declarations.

Fix the cast for the getnameinfo() call in the status() function, which
should be "const struct" not "struct" for sockaddr.

WARNS=3

ifconfig -v prints 802.11 statistics.  ifconfig -v prints and zeroes
802.11 statistics.

NI_WITHSCOPEID was not picked up by IETF standardization process.

Functions for converting media types to/from text are now in libutil.

More isdigit() calls that need (unsigned char) cast

De __P

Remove all the caddr_t - mainly casts of the 3rd arg to ioctl()

err -> errx where appropriate.

The change in revision 1.38 (in 1997!) was wrong.  If we don't want the kernel
to automatically bring up interfaces, then we should move that code into
ifconfig(8) and out of the kernel, not cycle the interface up and down.  Fixes
PR 9400 -- ifconfig(8) was not working as documented.

Very minimal changes to make "ssid" and "nwid" aliases, and make the
ssid print out as "ssid", not "nwid".
XXX Really, someone should go through and change around all the
variable names etc. to ssid, but not today.

check strlcpy error.  idea from Ryan McBride

Add 'mode' option to change PHY media for 802.11 devices.  Also,
display PHY mode. E.g., run an 802.11a (5GHz, 54Mbps max) access
point:

> ifconfig ath0 nwid my_access_point mode 11a mediaopt hostap
> ifconfig -m ath0
ath0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
        nwid my_access_point
        powersave off
        bssid 00:05:5d:84:cc:6e chan 3
        address: 00:05:5d:84:cc:6e
        media: IEEE802.11 autoselect mode 11a hostap
        status: active
        supported IEEE802.11 media:
                media autoselect
                media autoselect mediaopt adhoc
                media autoselect mediaopt hostap
                media autoselect mediaopt monitor
                media autoselect mode 11a
                media autoselect mode 11a mediaopt adhoc
                media autoselect mode 11a mediaopt hostap
                media autoselect mode 11a mediaopt monitor
*snip snip*

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22308, verified by myself.

PR/21860: Robert Elz: ifconfig doesn't allow addr/N notation for INET6

use strlcpy

Change -b from skipping  POINTTOPOINT|LOOPBACK  to skipping  ! BROADCAST.
Fixes my [bin/21604]

Add an option (-z) like -v except that it clears the input/output
stats after printing them.
(Basically this just calls SIOCZIFDATA instead of SIOCGIFDATA).
Submitted by Jean-Edouard BABIN in PR bin/21078
However, he suggested using -c, and cgd (reasonably) noted that -z for
"zero" was probably more reasonable.
I've also changed the code very slightly.

Hide bssid/chan for 802.11 if not specified nor not associated.
Allow "-chan" to allow any channel to associate, it doesn't affect
create channel for IBSS/AP mode.

Add missing newline.

New commands:
* bssid xx:xx:xx:xx:xx:xx -- set the desired BSSID of an 802.11
  interface.
* -bssid -- unset the desired BSSID of an 802.11 interface, so
  the interface will choose automatically (default).
* channel x -- set the channel (radio frequency) of an 802.11 interface.

Current BSSID and channel are now reported in the 802.11 status
display, if supported by the interface.

Above changes from David Young <dyoung@ojctech.com>, with some slight
changes by me (use ethers(3) functions rather than hand-parsing/printing
the 802.11 address).

Document bssid/-bssid/channel, and clean up markup of parentheticals
in the manual page.

use snprintb()

sweep of errx/warnx, remove unnecessary trailing \n

grab eui64 only from the same interface.

support "eui64" option (grab interface ID from link-local addr).
from: ww@styx.org

init variable before strtoul

metric and mtu are u_long, not int.

Add support to print interface counters (via SIOCGIFDATA) when the [new]
option -v is provided.

Don't bother testing if a u_short is > 0xffff -- that test is always false.

use in_aliasreq for inet address manipulation, not ifaliasreq.  they have
different alignment constraint.  [tests on 64bit arch wanted]

fix many -Wshadow warnings

Clean up the code a bit so that the ioctl() return value is always
compared to -1 (instead of <0 or !=0) and always to call err() with
EXIT_FAILURE instead of sometimes using 1.

Use getnameinfo() to format AF_LINK addresses again.

Revert last change.  getnameinfo() AF_LINK support is going away until I can
make it lint-clean.

Use getnameinfo() to format AF_LINK addresses rather than doing it ourselves.

remove duplicated line in cmds[].

Add tcp4csum-rx and udp4csum-rx commands for interfaces that only
support TCP/UDP checksums on the in-bound direction.

snprintf return value audit.  from deraadt

Correct style of a comment, and bad else {} scoping from my previous
commit.

Thanks to enami tsugutomo for pointing both out.

Cause multiple uses of alias, -alias, and delete to produce an error,
and exit.

Previously, combinations would produce unintended results, such as
deleting the primary IP on an interface, instead of deleting an specified
alias.

embed scopeid on scoped ipv6 address. sync with kame

do not go over the end of memory region we are supposed to touch,
overrun on "prefixlen" operation.  there's no SEGV possibility.

Add support of persistent key of IEEE 802.11 wireless LAN card.
Currently, only Aironet ("an") driver/card can be used.
     nwkey persist   (IEEE 802.11 devices only) Enable WEP encryption for IEEE
                     802.11-based wireless network interfaces with the persis-
                     tent key written in the network card.
     nwkey persist:key
                     (IEEE 802.11 devices only) Write the key to the persis-
                     tent memory of the network card, and enable WEP encryp-
                     tion for IEEE 802.11-based wireless network interfaces
                     with the key.

Implement support for IP/TCP/UDP checksum offloading provided by
network interfaces.  This works by pre-computing the pseudo-header
checksum and caching it, delaying the actual checksum to ip_output()
if the hardware cannot perform the sum for us.  In-bound checksums
can either be fully-checked by hardware, or summed up for final
verification by software.  This method was modeled after how this
is done in FreeBSD, although the code is significantly different in
most places.

We don't delay checksums for IPv6/TCP, but we do take advantage of the
cached pseudo-header checksum.

Note: hardware-assisted checksumming defaults to "off".  It is
enabled with ifconfig(8).  See the manual page for details.

Implement hardware-assisted checksumming on the DP83820 Gigabit Ethernet,
3c90xB/3c90xC 10/100 Ethernet, and Alteon Tigon/Tigon2 Gigabit Ethernet.

Pull up revision 1.107 (via patch, requested by itojun):
  Make ``-A'' the default, so that IPv4 aliases are printed.
  Always print mac address.

make -A a default.  present MAC adderss on all ifconfig operations.

assume the presense of getifaddrs(3).

sync usage() more to the reality.  document alias/-alias.

allow ifconfig to take "deprecated" and "-deprecated" for IPv6.  sync with kame

sync usage() with reality.  take "up" and "down" outside of "[af ...],
as "up" and "down" are independent from interface address configuration.

whitespace at EOL

on "tunnel" subcommand, use specified address family.

PR/12275:Kimmo Suominen: show -C in usage

add SIOC[SG]LIFPHYADDR ioctl.  greatly simplify tunnel address settings.
sync with kame.  old ioctls are supplied but not recommended for new code.

convert to use getprogname()

fix nested externs

Pull up revision 1.94 (requested by itojun):
  Initialize an uninitialized pointer. PR11702.

Fix the usage string so it displays the -b and s options to ifconfig -l

unbroken slash notation (10.0.0.0/8) for IPv4.

support "prefixlen n" for IPv4 too.

uninitialized pointer.  PR 11702

Pull up revision 1.89 (requested by bouyer):
	Add support for 802.1Q virtual LANs.

Pull up revision 1.81, 1.87 (requested by bouyer):
  Support cloning of network pseudo-interfaces.

Add powersave/powersavesleep into usage.
Fold usage output (still 14 lines...)

Add powersave, -powersave, and powersavesleep to manipulate 802.11
power management parameters.

show IPv6 interface address status "deprecated" (preferred lifetime = 0).
RFC2462.  sync with kame.

In tunnel_status(), if we have an IPv6 kernel, but a driver doesn't
support IPv6 tunnel endpoint addresses, fall back on the sockaddr
version of the ioctls.

Add vlan(4) support.  From Andy Doran <ad@netbsd.org>.

Pullup 802.11 stuff (approved by jhawk)
- add support for nwkey to ifconfig
	basesrc/sbin/ifconfig/ifconfig.c		1.88
	basesrc/sbin/ifconfig/ifconfig.8		1.39
	syssrc/sys/dev/ic/awi.c				1.26
	syssrc/sys/dev/ic/awi_wep.c			1.3
	syssrc/sys/dev/ic/awivar.h			1.12
	syssrc/sys/dev/pcmcia/if_wi.c			1.26
	syssrc/sys/net/if.c				1.69
	syssrc/sys/net/if_ieee80211.h			1.5

Pullups 802.11 stuff (approved by jhawk)
- allow non-string nwid settings
	basesrc/sbin/ifconfig/ifconfig.c		1.82-1.86
	basesrc/sbin/ifconfig/ifconfig.8		1.37
	syssrc/sys/dev/ic/awi.c				1.21
	syssrc/sys/dev/pcmcia/if_ray.c			1.21
	syssrc/sys/dev/pcmcia/if_wi.c			1.23
	syssrc/sys/dev/pcmcia/if_wivar.h		1.10
	syssrc/sys/net/if_ieee80211.h			1.4

Add nwkey and -nwkey keywords, to enable/disable WEP,
and to set WEP encryption key for IEEE802.11 wireless network interfaces.

Add a -C flag, to list all available network interface cloners.

- Fix off-by-one error in previous.
- Fix a typo of ioctl name in warning message.

check the length of nwid argument before copy to avoid SEGV.
pointed out by enami-san

cast u_char for ctype (nwid stuff) to fix possible overrun problems.

warn instead of print 802.11 nwid with bad length field.

change the argument of SIOCS80211NWID and SIOCG80211NWID ioctls from
u_int8_t array to struct ieee80211_nwid to prepend length field.
The length field is necessary because IEEE 802.11 spec doesn't prohibit
even '\0' for SSID.
Though the name and the value of SIOC... macro is unchanged, this change
breaks binary compatibility.  The only affected userland program on the
tree is ifconfig(8).
As Jason suggested on tech-net, it is better than live with problems
since there are no releases for this ioctls yet.

Add `create' and `destroy' commands.

- Add `tunnel' and `deletetunnel' commands, which issue the ioctls
  that gifconfig(8) would issue to configure tunnel endpoints.  This
  allows IP tunnel interfaces (`gif' right now, and `gre' later) to
  be configured with ifconfig(8), and via /etc/ifconfig.<interface>.
  Partially taken from similar changes in OpenBSD.
- Const poison the command functions a bit.  We really need to clean
  up the command function interface.

Rationale for pulling into netbsd-1-5 branch: While this looks like
a feature enhancement, it actually fixes a bug -- without this change
or a change in rc.d's network bring-up, it is not possible to configure
IP tunnels as a normal part of bringing up the network.

- Add `tunnel' and `deletetunnel' commands, which issue the ioctls
  that gifconfig(8) would issue to configure tunnel endpoints.  This
  allows IP tunnel interfaces (`gif' right now, and `gre' later) to
  be configured with ifconfig(8), and via /etc/ifconfig.<interface>.
  Partially taken from similar changes in OpenBSD.
- Const poison the command functions a bit.  We really need to clean
  up the command function interface.

Allocate flagreq on stack.

use getifaddrs, not SIOCGIFCONF, to avoid alignment constraints.
TODO: maybe it's better to pass struct ifaddrs * to in{,6}_alias.

clarify in_alias() and in_status().
in_status() handles the whole inet address family.  in_alias() handles
printout for single address (for both the primary address or an alias).
it is more consistent with, for example, the way at_status() is used.

Explicitly pick up ''the first address'' to delete.

Set flagreq.ifr_name before issueing SIOCGIFFLAGS for setifflags.
Withoug this change, flagreq was never initialized and ifconfig w/o addr
always failed with ENXIO.

Fix PR bin/9629.  The global variable ifr was used to save interface
flags even though the media routines tweak it.

Remove unnecessary continue statement.

Remove now unnecessary assignment.  It looks like 4.4lite merge left over.

Don't copy from destination to destination.

Per a discussion on gcc-bugs, port-alpha/9494 is not a compiler bug, but
rather a bug in the code (which is violating the C standard by casting
an unaligned pointer to a pointer with more strict alignment constraints).

Fix the code to not do such things.  Fixes port-alpha/9494.

Salvage the better comment appeared in rev 1.51 but gone in 1.53.

more fix to SIOCGIFCONF alignment.

Fix a buglet that caused media-specific options to *always* be printed, even if
they were not set.

Fix a logic buglet in status printing code.

- Print 802.11 status (i.e. network ID) after mtu, etc. but before
  protcol addresses.
- When searching for status bits, use the currently selected media type,
  not the currently active media type.

Update copyright.

Make display of media status information (e.g. carrier, etc.) table-driven
and easily extensible (by merely adding to the tables in ifmedia.h).

allow setting (and reporting) nwid for ieee 802.11 devices.

Fix a bug in get_media_options() which wasn't caught with global options;
make sure we only return option bits.

implement netmask slash notation for IP and IPv6 addresses

Pull up to last week's -current.

bring in KAME scopeid hack.  always show scopeid if it is scoped.
sync with latest KAME.

fix alignment problem of SIOCGIFCONF, on 64bit arch.

Make the previous actually compile.

cleanup error message

don't allow destination addresses to be set for non-Point-to-point interfaces

Add a new -s option to make it easier to test the link-level status
(i.e., 10baseT carrier/no carrier) of an interface from scripts

ifconfig -s <interface> will exit with a false status if the interface
reports its unconnected.

-s also works in conjunction with -l and -a, filtering out interfaces
which are reporting down.

Also, add -b (which shows only broadcast interfaces with -l and -a).

I find these options useful in network autoconfig scripts for mobile
systems.

make ifconfig IPv6-ready.
TODO: update manpage for new options.

pullup 1.51->1.52 (thorpej)

pullup 1.50->1.51 (thorpej): Don't force "inet" on status display

Fix a slight braino in last.

Don't force `inet' on status display; get all address families that have
configured addresses.  From Jun-ichiro itojun Hagino, PR #7593.

revert the previous changes that removed special cased ioctls for ISO
stuff.

Add support for displaying interface alias information. Based on PR/6109 by
Phil Nelson.

Oops, add the "instance" command to the usage info.

Allow the user to explicitly set the media instance.

Hm, actions might not always be deferred; rename appropriately.

Completely rewrite the engine that sets media and media options.  Previous
code would hit the interface multiple times.  Now, all media and media
options are gathered and processed at once, hitting the interface only
once (except to grab the currently selected media exactly once; there is
no way to avoid that).

Implement a suggestion from Wolfgang Solfrank on the "supported media"
display.  The display now looks like this:

bishop:thorpej 35$ obj.alpha/ifconfig -m fxp0
fxp0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
        media: Ethernet autoselect (100baseTX full-duplex)
        status: active
        supported Ethernet media:
                media autoselect
                media 100baseTX mediaopt full-duplex
                media 100baseTX
                media 10baseT mediaopt full-duplex
                media 10baseT
                media 100baseTX mediaopt loopback
        inet 129.99.50.41 netmask 0xffffff00 broadcast 129.99.50.255
bishop:thorpej 36$

The intent here is to allow cut'n'pasting of the output to a command
line, or allow easier automatic extraction from scripts, etc.

XXX Still at issue is what to do about non-0 media instances.

Print the media instance if it's not zero, a reasonable compromise such
that no one will have to see those ugly instance numbers if they have
an interface with only one PHY (most of them out there).

Completely rewrite the way media descriptions are represented.  The same
data structure is used, but a much saner matching mechanism is used, one
which allows greater ease in adding new types.

Also, rewrite the way media words are displayed.  The status display now
looks like this:

bishop:thorpej 137$ ifconfig -m fxp0
fxp0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
        media: Ethernet autoselect (10baseT)
        status: active
        supported media:
                Ethernet autoselect
                Ethernet 100baseTX <full-duplex>
                Ethernet 100baseTX
                Ethernet 10baseT <full-duplex>
                Ethernet 10baseT
                Ethernet 100baseTX <loopback>
        inet 129.99.50.41 netmask 0xffffff00 broadcast 129.99.50.255
bishop:thorpej 138$

This makes it more clear which media type (e.g. Ethernet) is currently
in-use, handy for devices such as the TI ThunderLAN which can also have
both Ethernet and Token Ring PHYs attached.

Copyright assigned to The NetBSD Foundation, Inc.

Don't declare optind.

* cleanup for WARNS=1
* deprecate register
* use .Nm correctly

note that we have set ifflags so that we can reset them again after
setting the interface address.  this is necessary as we may want to
configure an interface `down' by default, but, SIOCAIFADDR turns the
interface back on for us.

ifr_metric vs. ifr_mtu: don't confuse people. Its a union anyway.

Don't define the address of an interface to 0.0.0.0 if it's just being
brought "up" (without an address); don't give it an address at all.
From Koji Imada <koji@math.human.nagoya-u.ac.jp> in [bin/3477]

Force the ifaliasreq "addreq" to have 4-byte alignment, so that casting
one of its sockaddrs to a sockaddr_in doesn't cause an unaligned access
fault on the Alpha.

Implemented two modifier flags for -a and -l: -d (only show "down" interfaces)
and -u (only show "up" interfaces)

Add mtu n parameter. Part of code by Matthias Pfaller (PR 1648).

- netatalk additions
- KNF
- prototype fixes
- ifconfig -a now prints all the address family parameters for all configured
  interfaces.

Add a "-l" flag, which lists available network interfaces, with no
additional information.

Print the station address of interfaces that have one.  From
Matt Thomas <matt@3am-software.com>

Display the correct interface name in the event SIOCGIFMEDIA "succeeds"
but doesn't return any media types.

Fix a slight bogon in printing active/current media, from Matt Thomas.

Fix lookup_media_word() to properly distinguish between the "autoselect"
media type and error.  (oops!)

Do that last change somewhat differently; We need to know the
currently selected media, and thus cannot allow E2BIG, since the
top-level kernel ioctl code won't perform the copyout of the
ifmediareq if it sees an "error".

Ignore E2BIG in domediaopt() like we do in setmedia().  Oversight pointed
out by Andrew Gallatin <gallatin@CS.Duke.EDU>.

Add support for BSD/OS-style media selection:
	- new "media" keyword specified media to select:
		ifconfig de0 media utp
	- new "mediaopt/-mediaopt" keywords for setting/clearing
	  media options (such as full-duplex)
	- new "-m" flag to display all supported media types for
	  a given interface.
Also, some generial cleanup of argument parsing while I was there.
Manual page updates from Jeremy Cooper <jeremy@netbsd.org>.

From trunk:
Eliminate SS_PRIV; instead, pass down a proc pointer to the usrreq methods
that need it.
Fix numerous memory leaks and bogus return values.

Update to match kernel changes.

Agree with kernel on format of SIOCGIFCONF results.

Add support for -DINET_ONLY to make this smaller in a ramdisk.

Use inet_aton(), not inet_addr().  Suggested by John Hawkinson.

convert to new RCS Id conventions; reduce my headache

 fix usage message to include -a.  fix command line handling for same

spacing nit.

more cleanups from Jim Jegers, passed over by me.

ioctl types.

Eliminate uses of some obsolete functions.

Merge with 4.4-Lite version.

Import original 4.4-Lite version.

(1) update usage info
(2) fix every instance of fprintf/exit except usage to use err()/errx()
(3) make netmask print out with a leading 0x, so that you can paste it back.

add MULTICAST recognition

LLC -> LINK, documentation

Add RCS identifiers.

removed aui/bnc hack from ifconfig. Use -llc0/llc0 instead.

fix ifconfig once again. multiple protocols really cooked hte code i wrote
before for -a.

Remove ifdefs hiding ISO and XNS functionality
(from Havard Eidnes <Havard.Eidnes@runit.sintef.no>)

Added another sunos-ism: ifconfig -a

changed "Id" to "Header" for rcsids

added rcs ids to all files

after 0.2.2 "stable" patches applied

initial import of 386bsd-0.1 sources

Initial revision