bsd.own.mk: rename GCC_NO_* to CC_WNO_*

Rename compiler-warning-disable variables from
GCC_NO_warning
to
CC_WNO_warning
where warning is the full warning name as used by the compiler.

GCC_NO_IMPLICIT_FALLTHRU is CC_WNO_IMPLICIT_FALLTHROUGH

Using the convention CC_compilerflag, where compilerflag
is based on the full compiler flag name.

introduce some common variables for use in GCC warning disables:

GCC_NO_FORMAT_TRUNCATION -Wno-format-truncation (GCC 7/8)
GCC_NO_STRINGOP_TRUNCATION -Wno-stringop-truncation (GCC 8)
GCC_NO_STRINGOP_OVERFLOW -Wno-stringop-overflow (GCC 8)
GCC_NO_CAST_FUNCTION_TYPE -Wno-cast-function-type (GCC 8)

use these to turn off warnings for most GCC-8 complaints. many
of these are false positives, most of the real bugs are already
commited, or are yet to come.


we plan to introduce versions of (some?) of these that use the
"-Wno-error=" form, which still displays the warnings but does
not make it an error, and all of the above will be re-considered
as either being "fix me" (warning still displayed) or "warning
is wrong."

branches: 1.25.42;
main.c uses non-literal format strings.

document non-literal strings

Add new Makefile knob, USE_FORT, which extends USE_SSP by turning on the
FORTIFY_SOURCE feature of libssp, thus checking the size of arguments to
various string and memory copy and set functions (as well as a few system
calls and other miscellany) where known at function entry. RedHat has
evidently built all "core system packages" with this option for some time.

This option should be used at the top of Makefiles (or Makefile.inc where
this is used for subdirectories) but after any setting of LIB.

This is only useful for userland code, and cannot be used in libc or in
any code which includes the libc internals, because it overrides certain
libc functions with macros. Some effort has been made to make USE_FORT=yes
work correctly for a full-system build by having the bsd.sys.mk logic
disable the feature where it should not be used (libc, libssp iteself,
the kernel) but no attempt has been made to build the entire system with
USE_FORT and doing so will doubtless expose numerous bugs and misfeatures.

Adjust the system build so that all programs and libraries that are setuid,
directly handle network data (including serial comm data), perform
authentication, or appear likely to have (or have a history of having)
data-driven bugs (e.g. file(1)) are built with USE_FORT=yes by default,
with the exception of libc, which cannot use USE_FORT and thus uses
only USE_SSP by default. Tested on i386 with no ill results; USE_FORT=no
per-directory or in a system build will disable if desired.

Use pidfile(3).

Oops remove WARNS=1

Resolve conflicts.

Avoid interraction between pppd and routed, by trying to discover new
interfaces every time there is a new route, irrespective of timeout period.

Use bsd.subdir.mk as appropriate.

- resolve conflicts.
- remove md5 stuff; it is in libc.
- define MCAST_PPP_BUG, until we fix if_ppp.c

Add a manual page link from routed(8) to rdisc(8) as suggested by
Giles Lean <giles@nemeton.com.au> on the tech-net@NetBSD.ORG mailing
list. The rationale for this is that folks may be used to rdisc(8)
or in.rdisc(8) on other systems, and may not realize that NetBSD
implements ICMP Router Discovery in the routed(8) program.

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

- Support rip version 2.
- Add ansi prototypes.
- Be less internet centric.
- Convert queues to use <sys/queue.h> so we don't need -lcompat

Took away use of vax inline functions. Didn't work anyway.

convert to new RCS Id conventions; reduce my headache

specify man pages the new way.

don't do query and trace

Clean up import.

don't need -lutil

Undo that last change. It was 100%, beyond a doubt, completely bogus.
afswitch and af_max were still used, but since they was declared 'common'
by virtue of their declaration in af.h, the symbols were defined at link-
time, and their contents were considered BSS -- init to zero.
Now, you and i know that jumping to zero causes a core dump...
Therefore: af.c has been resurrected, so the procedure pointers actually
point to procedures, tables.c has had it's change nuked, and af.h
has been changed so afswitch and af_max are declared as 'extern'.
It's obvious that that 'fix' wasn't even tested.

Move inet_default into tables.c, and remove all of af.c, as it is no longer
used.

remove query and trace from subdir lists; they're for debugging only.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

RIP address families are still 2 bytes

fix a stupid bug; wrong size was being used for sin_len.

- Support rip version 2.
- Add ansi prototypes.
- Be less internet centric.
- Convert queues to use <sys/queue.h> so we don't need -lcompat

Format police.

Fix bug in af_format where the formatted address was returned in the same
static buffer. When af_format is used more than once in the same printf
statement, the second call will overwrite the first. Changed af_format to
take a buffer and the buffer size where to place the formatted string.

convert to new RCS Id conventions; reduce my headache

some -Wall changes, from Jim Jegers, improved (in some cases) by me.
still needs a bit of work, but closer.

Clean up import.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

- Support rip version 2.
- Add ansi prototypes.
- Be less internet centric.
- Convert queues to use <sys/queue.h> so we don't need -lcompat

convert to new RCS Id conventions; reduce my headache

Clean up import.

Undo that last change. It was 100%, beyond a doubt, completely bogus.
afswitch and af_max were still used, but since they was declared 'common'
by virtue of their declaration in af.h, the symbols were defined at link-
time, and their contents were considered BSS -- init to zero.
Now, you and i know that jumping to zero causes a core dump...
Therefore: af.c has been resurrected, so the procedure pointers actually
point to procedures, tables.c has had it's change nuked, and af.h
has been changed so afswitch and af_max are declared as 'extern'.
It's obvious that that 'fix' wasn't even tested.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

unifdef sgi. NFCI

Remove duplicate RS_LOCAL. From cppcheck via Henning Petersen in PR 46000.

branches: 1.26.2;
Use __dead

merge 2.32 which is actually 2.31. Most important change is:
- use multicast over point-to-point (GRE) links as proposed by Dan Lukes.

Move WARNS=3 to the Makefile.inc, and add a little const to the remaining
programs that did not compile before.

constification lossage (hi gimpy)

sync with routed 2.27

fix -Wshadow warnings

sync with version 2.24. Vernon says:
The most significant [fix] involves so called "remote" interfaces
configured in the kludge file to with what appear to be colliding
networks. Edward Mascarenhas <eddiem@vihar.engr.sgi.com> found
the problem and the fix, and I think has tested it in the SGI
network.

specify interface by using interface index (set/getsockopt with 0.0.0.0/8).

Resolve conflicts.

Update to 980910 from ftp.rhyolite.com

Merge the 980602 RIPv2/Router Discovery routed. Fixes floods of host routes
generated when one of the interfaces on the network does not support
broadcast (e.g. HIPPI or ATM).

From Vern Schryver <vjs@rhyolite.com>

* cleanup for WARNS=1
* deprecate register
* bcmp, bcopy, bzero, rindex -> memcmp, memmove, memset, strrchr
* correct use of .Nm

- resolve conflicts.
- remove md5 stuff; it is in libc.
- define MCAST_PPP_BUG, until we fix if_ppp.c

- resolve conflicts
- try to follow rcsid style more closely
- fix rn_walktree callback routine signatures.

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

- Support rip version 2.
- Add ansi prototypes.
- Be less internet centric.
- Convert queues to use <sys/queue.h> so we don't need -lcompat

Update to use timer{add,sub}().

convert to new RCS Id conventions; reduce my headache

some -Wall changes, from Jim Jegers, improved (in some cases) by me.
still needs a bit of work, but closer.

Remove some more uses of obsolete functions.

Clean up import.

light comment change

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

remove an extra whitespace before else keyword.

unifdef sgi. NFCI

PR/44647: commit fix proposed by Takahiro HAYASHI
verified to be working in -current and also helps NetBSD-6 BETA2

branches: 1.29.6; 1.29.8;
Memcpy into local variable to ensure proper alignement.
Fixes PR port-sparc64/43965.

merge 2.32 which is actually 2.31. Most important change is:
- use multicast over point-to-point (GRE) links as proposed by Dan Lukes.

fix printf formats.

branches: 1.26.10; 1.26.16; 1.26.20;
Don't nul-terminate the interface name in a sockaddr_dl that is
part of an ifa_msghdr returned by sysctl(3), because that could
overwrite the ifam_len field of the following ifa_msgher. Instead,
copy the interface name out of the sockaddr_dl, and then nul-terminate.
Fixes the bug reported by Frank Kardel on current-users@. Thanks
Frank for reporting, and for testing the fix.

Coverity CID 2671: Fix use of ifp after free.

sync with routed 2.27

fix -Wshadow warnings

sync with version 2.24. Vernon says:
The most significant [fix] involves so called "remote" interfaces
configured in the kludge file to with what appear to be colliding
networks. Edward Mascarenhas <eddiem@vihar.engr.sgi.com> found
the problem and the fix, and I think has tested it in the SGI
network.

specify interface by using interface index (set/getsockopt with 0.0.0.0/8).

Revert 1.18 -> 1.19; causes infinite loops on machines that use ppp and
do active routing.

branches: 1.19.2;
remove the once hack; vjs says it is not necessary with the monotonic
time kludge. We'll see.

Ignore RTM_OIFINFO messages (send by a COMPAT_14 kernel).

branches: 1.17.6;
Resolve conflicts.

Update to 980910 from ftp.rhyolite.com

Avoid interraction between pppd and routed, by trying to discover new
interfaces every time there is a new route, irrespective of timeout period.

Merge the 980602 RIPv2/Router Discovery routed. Fixes floods of host routes
generated when one of the interfaces on the network does not support
broadcast (e.g. HIPPI or ATM).

From Vern Schryver <vjs@rhyolite.com>

use memmove, memcmp, memset instead of bcopy, bcmp, bzero

* cleanup for WARNS=1
* deprecate register
* bcmp, bcopy, bzero, rindex -> memcmp, memmove, memset, strrchr
* correct use of .Nm

- resolve conflicts.
- remove md5 stuff; it is in libc.
- define MCAST_PPP_BUG, until we fix if_ppp.c

- resolve conflicts
- try to follow rcsid style more closely
- fix rn_walktree callback routine signatures.

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

- Support rip version 2.
- Add ansi prototypes.
- Be less internet centric.
- Convert queues to use <sys/queue.h> so we don't need -lcompat

convert to new RCS Id conventions; reduce my headache

Remove some more uses of obsolete functions.

Clean up import.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

- Support rip version 2.
- Add ansi prototypes.
- Be less internet centric.
- Convert queues to use <sys/queue.h> so we don't need -lcompat

convert to new RCS Id conventions; reduce my headache

some -Wall changes, from Jim Jegers, improved (in some cases) by me.
still needs a bit of work, but closer.

Fix `maskof' function which was returning `netof', per Brad Parker.

branches: 1.5.2;
Clean up import.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

merge 2.32 which is actually 2.31. Most important change is:
- use multicast over point-to-point (GRE) links as proposed by Dan Lukes.

change (mostly) int to socklen_t. GCC 4 doesn't like that int and
socklen_t are different signness.

Fix compilation problems on non-i386.

sync with routed 2.27

sync with version 2.24. Vernon says:
The most significant [fix] involves so called "remote" interfaces
configured in the kludge file to with what appear to be colliding
networks. Edward Mascarenhas <eddiem@vihar.engr.sgi.com> found
the problem and the fix, and I think has tested it in the SGI
network.

add a comment on about how tracing is secured, but still keep it disabled.

Resolve conflicts.

Update to 980910 from ftp.rhyolite.com

Disable and log RIP_TRACEON and RIP_TRACEOFF. They are insecure.

Merge the 980602 RIPv2/Router Discovery routed. Fixes floods of host routes
generated when one of the interfaces on the network does not support
broadcast (e.g. HIPPI or ATM).

From Vern Schryver <vjs@rhyolite.com>

branches: 1.21.2;
use memmove, memcmp, memset instead of bcopy, bcmp, bzero

* cleanup for WARNS=1
* deprecate register
* bcmp, bcopy, bzero, rindex -> memcmp, memmove, memset, strrchr
* correct use of .Nm

- resolve conflicts.
- remove md5 stuff; it is in libc.
- define MCAST_PPP_BUG, until we fix if_ppp.c

- resolve conflicts
- try to follow rcsid style more closely
- fix rn_walktree callback routine signatures.

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

Fix bugs cgd pointed out.

- Support rip version 2.
- Add ansi prototypes.
- Be less internet centric.
- Convert queues to use <sys/queue.h> so we don't need -lcompat

<sys/errno.h> -> <errno.h>
<sys/signal.h> -> <signal.h>
<sys/syslog.h> -> <syslog.h>

s/an active/a passive/

Fix bug in af_format where the formatted address was returned in the same
static buffer. When af_format is used more than once in the same printf
statement, the second call will overwrite the first. Changed af_format to
take a buffer and the buffer size where to place the formatted string.

fix arguments to tracing functions and/or cast, to avoid warnings.

Update to use timer{add,sub}().

convert to new RCS Id conventions; reduce my headache

some -Wall changes, from Jim Jegers, improved (in some cases) by me.
still needs a bit of work, but closer.

Remove some more uses of obsolete functions.

Eliminate uses of some obsolete functions.

Clean up import.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

- Support rip version 2.
- Add ansi prototypes.
- Be less internet centric.
- Convert queues to use <sys/queue.h> so we don't need -lcompat

convert to new RCS Id conventions; reduce my headache

Clean up import.

fix from Brad Parker <brad@fcr.com> to curb routed inflation when
using MULTICAST. it wasn't properly ignoring the new interface flags.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

fix various typos in comments.

unifdef sgi. NFCI

merge 2.32 which is actually 2.31. Most important change is:
- use multicast over point-to-point (GRE) links as proposed by Dan Lukes.

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

branches: 1.39.26; 1.39.30;
usage: -F was missing the /mask part. Based on PR 26063
by Kouichirou Hiratsuka.

Remove unimplemented and undescribed -p option from usage and man page
SYNOPSIS. Noted by Jeff Ito.

Remove option letter from getopt string, since it does not
have a case handler below. From aaron@openbsd.

From Vernon Schryver:

Wolfgang Solfrank has explained the problem with router discovery
in `routed` in a way I can understand.

Let's assume that the configured preference of the interface is 5.
This gets converted to 0x80000005 through the use of the UNSIGN_PREF
macro. Later on, this value gets put into the PREF macro, which
compares it against the interface metric(s) (let's assume those
values to be 0 for now). Of course the 0x80000005, cast to int,
is much less than 0, so the clamping rule is triggered, which
gives us a value of 1. This is then converted via SIGN_PREF into
0x80000001 and put into the message. Certainly, this isn't what
was intended.

Initialize an uninitialized variable.

use arc4random

avoid FD_SET overrun. from openbsd

strcpy/sprintf cleanup. from openbsd

sync with routed 2.27

branches: 1.30.2;
fix -Wshadow warnings

va_{start,end} audit:
Make sure that each va_start has one and only one matching va_end,
especially in error cases.
If the va_list is used multiple times, do multiple va_starts/va_ends.
If a function gets va_list as argument, don't let it use va_end (since
it's the callers responsibility).

Improved by comments from enami and christos -- thanks!

Heimdal/krb4/KAME changes already fed back, rest to follow.

Inspired by, but not not based on, OpenBSD.

sync with version 2.24. Vernon says:
The most significant [fix] involves so called "remote" interfaces
configured in the kludge file to with what appear to be colliding
networks. Edward Mascarenhas <eddiem@vihar.engr.sgi.com> found
the problem and the fix, and I think has tested it in the SGI
network.

specify interface by using interface index (set/getsockopt with 0.0.0.0/8).

deprecate LOG_ODELAY; it's the default

Format string cleanups by Bill Sommerfeld.

branches: 1.24.4;
add a usec_fudge to make sure that time increases monotonically.
printf format changes.

Use pidfile(3).

branches: 1.22.2;
LP64 tweak in printf(3) instance

Resolve conflicts.

Update to 980910 from ftp.rhyolite.com

- use an array MAXHOSTNAMELEN+1 size to hold hostnames
- ensure hostname from gethostname() is nul-terminated in all cases
- minor KNF
- use MAXHOSTNAMELEN over various other values/defines
- be safe will buffers that hold hostnames

Merge the 980602 RIPv2/Router Discovery routed. Fixes floods of host routes
generated when one of the interfaces on the network does not support
broadcast (e.g. HIPPI or ATM).

From Vern Schryver <vjs@rhyolite.com>

* cleanup for WARNS=1
* deprecate register
* bcmp, bcopy, bzero, rindex -> memcmp, memmove, memset, strrchr
* correct use of .Nm

- resolve conflicts.
- remove md5 stuff; it is in libc.
- define MCAST_PPP_BUG, until we fix if_ppp.c

- resolve conflicts
- try to follow rcsid style more closely
- fix rn_walktree callback routine signatures.

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

- Support rip version 2.
- Add ansi prototypes.
- Be less internet centric.
- Convert queues to use <sys/queue.h> so we don't need -lcompat

<sys/errno.h> -> <errno.h>
<sys/signal.h> -> <signal.h>
<sys/syslog.h> -> <syslog.h>

Fix typo in previous.

Use POSIX signals.

Fix typo.

Update to use timer{add,sub}().

convert to new RCS Id conventions; reduce my headache

some -Wall changes, from Jim Jegers, improved (in some cases) by me.
still needs a bit of work, but closer.

Clean up import.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

- resolve conflicts.
- remove md5 stuff; it is in libc.
- define MCAST_PPP_BUG, until we fix if_ppp.c

branches: 1.1.1;
Initial revision

merge 2.32 which is actually 2.31. Most important change is:
- use multicast over point-to-point (GRE) links as proposed by Dan Lukes.

Coverity CID 2742: Avoid NULL deref.

Coverity CID 1079: Possible NULL deref.

sync with routed 2.27

fix -Wshadow warnings

sync with version 2.24. Vernon says:
The most significant [fix] involves so called "remote" interfaces
configured in the kludge file to with what appear to be colliding
networks. Edward Mascarenhas <eddiem@vihar.engr.sgi.com> found
the problem and the fix, and I think has tested it in the SGI
network.

specify interface by using interface index (set/getsockopt with 0.0.0.0/8).

In walk_supply(), test for IS_PASSIVE in ifp->int_state, not
ifp->int_if_flags (IFF_* bits).

branches: 1.17.10;
Resolve conflicts.

Update to 980910 from ftp.rhyolite.com

Merge the 980602 RIPv2/Router Discovery routed. Fixes floods of host routes
generated when one of the interfaces on the network does not support
broadcast (e.g. HIPPI or ATM).

From Vern Schryver <vjs@rhyolite.com>

use memmove, memcmp, memset instead of bcopy, bcmp, bzero

* cleanup for WARNS=1
* deprecate register
* bcmp, bcopy, bzero, rindex -> memcmp, memmove, memset, strrchr
* correct use of .Nm

- resolve conflicts.
- remove md5 stuff; it is in libc.
- define MCAST_PPP_BUG, until we fix if_ppp.c

- resolve conflicts
- try to follow rcsid style more closely
- fix rn_walktree callback routine signatures.

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

- Support rip version 2.
- Add ansi prototypes.
- Be less internet centric.
- Convert queues to use <sys/queue.h> so we don't need -lcompat

convert to new RCS Id conventions; reduce my headache

some -Wall changes, from Jim Jegers, improved (in some cases) by me.
still needs a bit of work, but closer.

Eliminate uses of some obsolete functions.

Clean up import.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

PR/45842: Henning Petersen: compare fgets with NULL not 0

branches: 1.25.4;
Use bitwise operators, logical and with a constant doesn't make sense

merge 2.32 which is actually 2.31. Most important change is:
- use multicast over point-to-point (GRE) links as proposed by Dan Lukes.

Coverity CID 2307: Possible memory leak.

Coverity CID 2306: Possible memory leak.

Coverity CID 1080: Possible NULL deref.

strcat -> strlcat

strcpy/sprintf cleanup. from openbsd

use snprintf. there are more...

sync with routed 2.27

bitmask operation mistake audit (s/&&/&/). from openbsd

sync with version 2.24. Vernon says:
The most significant [fix] involves so called "remote" interfaces
configured in the kludge file to with what appear to be colliding
networks. Edward Mascarenhas <eddiem@vihar.engr.sgi.com> found
the problem and the fix, and I think has tested it in the SGI
network.

Make sure IS_PASSIVE gets set on a parm's int_state when the
passive option is used in an if=xxx clause.

branches: 1.13.10;
Resolve conflicts.

Update to 980910 from ftp.rhyolite.com

Merge the 980602 RIPv2/Router Discovery routed. Fixes floods of host routes
generated when one of the interfaces on the network does not support
broadcast (e.g. HIPPI or ATM).

From Vern Schryver <vjs@rhyolite.com>

use static int, not just static

y2k fixes: use strptime...

Fix uninitialized variable warning.

use memmove, memcmp, memset instead of bcopy, bcmp, bzero

* cleanup for WARNS=1
* deprecate register
* bcmp, bcopy, bzero, rindex -> memcmp, memmove, memset, strrchr
* correct use of .Nm

- resolve conflicts.
- remove md5 stuff; it is in libc.
- define MCAST_PPP_BUG, until we fix if_ppp.c

Shifting by 32 on a 32-bit machine is undefined
(and in fact -1 << 32 results in -1 on e.g. x86)

- resolve conflicts
- try to follow rcsid style more closely
- fix rn_walktree callback routine signatures.

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

branches: 1.1.1;
Initial revision

sync with version 2.24. Vernon says:
The most significant [fix] involves so called "remote" interfaces
configured in the kludge file to with what appear to be colliding
networks. Edward Mascarenhas <eddiem@vihar.engr.sgi.com> found
the problem and the fix, and I think has tested it in the SGI
network.

Resolve conflicts.

Update to 980910 from ftp.rhyolite.com

- resolve conflicts
- try to follow rcsid style more closely
- fix rn_walktree callback routine signatures.

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

convert to new RCS Id conventions; reduce my headache

Clean up import.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

s/prefering/preferring/ in comment.

Fix typos, reported by Alexey Dobriyan ("Gathered from Linux"),
forwarded by jmc@openbsd.

Avoid strict alias warnings.

sync with version 2.24. Vernon says:
The most significant [fix] involves so called "remote" interfaces
configured in the kludge file to with what appear to be colliding
networks. Edward Mascarenhas <eddiem@vihar.engr.sgi.com> found
the problem and the fix, and I think has tested it in the SGI
network.

add an unsigned cast; looks safe and innocent, but it might not be...

Revert cast to (int) that caused core-dumps.

Resolve conflicts.

Update to 980910 from ftp.rhyolite.com

If One splits a line for cosmetic reasons, One should ensure that the
preceeding line has a \ if it's part of a #define...

use memmove, memcmp, memset instead of bcopy, bcmp, bzero

* cleanup for WARNS=1
* deprecate register
* bcmp, bcopy, bzero, rindex -> memcmp, memmove, memset, strrchr
* correct use of .Nm

- resolve conflicts
- try to follow rcsid style more closely
- fix rn_walktree callback routine signatures.

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

branches: 1.1.1;
Initial revision

routed: remove extra semicolon at the end of a function-like macro

No binary change.

branches: 1.6.104;
Kill __P().

Update to 980910 from ftp.rhyolite.com

* cleanup for WARNS=1
* deprecate register
* bcmp, bcopy, bzero, rindex -> memcmp, memmove, memset, strrchr
* correct use of .Nm

RCSid Police

- resolve conflicts
- try to follow rcsid style more closely
- fix rn_walktree callback routine signatures.

branches: 1.1.1;
Initial revision

change (mostly) int to socklen_t. GCC 4 doesn't like that int and
socklen_t are different signness.

Coverity CID 2432: Dereference before NULL check.

From Vernon Schryver:

Wolfgang Solfrank has explained the problem with router discovery
in `routed` in a way I can understand.

Let's assume that the configured preference of the interface is 5.
This gets converted to 0x80000005 through the use of the UNSIGN_PREF
macro. Later on, this value gets put into the PREF macro, which
compares it against the interface metric(s) (let's assume those
values to be 0 for now). Of course the 0x80000005, cast to int,
is much less than 0, so the clamping rule is triggered, which
gives us a value of 1. This is then converted via SIGN_PREF into
0x80000001 and put into the message. Certainly, this isn't what
was intended.

sync with routed 2.27

branches: 1.13.2;
fix -Wshadow warnings

sync with version 2.24. Vernon says:
The most significant [fix] involves so called "remote" interfaces
configured in the kludge file to with what appear to be colliding
networks. Edward Mascarenhas <eddiem@vihar.engr.sgi.com> found
the problem and the fix, and I think has tested it in the SGI
network.

specify interface by using interface index (set/getsockopt with 0.0.0.0/8).

branches: 1.10.4;
protability cast.

Resolve conflicts.

Update to 980910 from ftp.rhyolite.com

Merge the 980602 RIPv2/Router Discovery routed. Fixes floods of host routes
generated when one of the interfaces on the network does not support
broadcast (e.g. HIPPI or ATM).

From Vern Schryver <vjs@rhyolite.com>

* cleanup for WARNS=1
* deprecate register
* bcmp, bcopy, bzero, rindex -> memcmp, memmove, memset, strrchr
* correct use of .Nm

- resolve conflicts.
- remove md5 stuff; it is in libc.
- define MCAST_PPP_BUG, until we fix if_ppp.c

- resolve conflicts
- try to follow rcsid style more closely
- fix rn_walktree callback routine signatures.

Make sure to fill in the whole sockaddr_in when sending a discovery packet.

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

branches: 1.1.1;
Initial revision

Remove workaround for ancient HTML generation code.

Remove unimplemented and undescribed -p option from usage and man page
SYNOPSIS. Noted by Jeff Ito.

From Vernon Schryver:

Wolfgang Solfrank has explained the problem with router discovery
in `routed` in a way I can understand.

Let's assume that the configured preference of the interface is 5.
This gets converted to 0x80000005 through the use of the UNSIGN_PREF
macro. Later on, this value gets put into the PREF macro, which
compares it against the interface metric(s) (let's assume those
values to be 0 for now). Of course the 0x80000005, cast to int,
is much less than 0, so the clamping rule is triggered, which
gives us a value of 1. This is then converted via SIGN_PREF into
0x80000001 and put into the message. Certainly, this isn't what
was intended.

Consistently use 'RFC 1234' instead of 'RFC1234' or 'RFC-1234'.
From jmc@openbsd.

.Nm does not need a dummy argument ("") before punctuation or
for correct formatting of the SYNOPSIS any longer.

Sort sections, new sentences begin on new lines, add a missing line,
fix a typo.

sync with routed 2.27

New sentence, new line. From Robert Elz.

branches: 1.37.2;
Spelling fixes, from Sergey Svishchev in kern/16650.

Generate <>& symbolically. I'm avoiding .../dist/... directories for now.

Use mdoc macros instead of man ones.

Punctuation fix, sort sections, lose a .Pp.

"kernel switch ipforwarding" -> "sysctl net.inet.ip.forwarding".

Drop arguments of .Os.

Improve usage of `comprise' according to Webster and misc/10893
by Jim Bernard.

Don't xref gated(8) and htable(8), which we don't have.

sync with version 2.24. Vernon says:
The most significant [fix] involves so called "remote" interfaces
configured in the kludge file to with what appear to be colliding
networks. Edward Mascarenhas <eddiem@vihar.engr.sgi.com> found
the problem and the fix, and I think has tested it in the SGI
network.

fix up various .Nm abuses:
- keep the case consistent between the actual name and what's referenced.
e.g, if it's `foo', don't use '.Nm Foo' at the start of a sentence.
- remove unnecessary `.Nm foo' after the first occurrence (except for
using `.Nm ""' if there's stuff following, or for the 2nd and so on
occurrences in a SYNOPSIS
- use Sx, Ic, Li, Em, Sq, and Xr as appropriate

insure -> ensure

remove extra period in SEE ALL section

branches: 1.25.4;
typos and clarifications

Resolve conflicts.

Fix typo 'with routing loop' to 'with a routing loop'

Update to 980910 from ftp.rhyolite.com

spell daemon correctly in the opening paragraph...

Merge the 980602 RIPv2/Router Discovery routed. Fixes floods of host routes
generated when one of the interfaces on the network does not support
broadcast (e.g. HIPPI or ATM).

From Vern Schryver <vjs@rhyolite.com>

fix bad .Xr references

macroize BSD, NetBSD, FreeBSD and misc cleanup

add missing .Nm sections.

branches: 1.16.2;
* cleanup for WARNS=1
* deprecate register
* bcmp, bcopy, bzero, rindex -> memcmp, memmove, memset, strrchr
* correct use of .Nm

Fix formatting directives; from SAITOH Masanobu <msaitoh@spa.is.uec.ac.jp>
in PR misc/3795.

fix 3128 from SAITOH Masanobu.
this consisted of nuking a duplicated paragraph; I did it slightly
differently than suggested.

alternate -> alternative, per PR 2643

- resolve conflicts.
- remove md5 stuff; it is in libc.
- define MCAST_PPP_BUG, until we fix if_ppp.c

fix some nits from SAITOH Masanobu, pr-3149

eliminate duplicate paragraph and fix typos; PR misc/3128

- resolve conflicts
- try to follow rcsid style more closely
- fix rn_walktree callback routine signatures.

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

Grammar police, noted by Peter Seebach <seebs@solon.com>; closes PR #1704.

convert to new RCS Id conventions; reduce my headache

Clean up import.

Add RCS indentifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

Remove the redundant copy of this file.
Suggested by David Carrel <carrel@ipsec.com>, PR #2872.

explicitly enumerate the ripv1 fields.

- resolve conflicts
- try to follow rcsid style more closely
- fix rn_walktree callback routine signatures.

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

branches: 1.1.1;
Initial revision

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

- Support rip version 2.
- Add ansi prototypes.
- Be less internet centric.
- Convert queues to use <sys/queue.h> so we don't need -lcompat

Use inet_aton(), not inet_addr().

fix arguments to tracing functions and/or cast, to avoid warnings.

convert to new RCS Id conventions; reduce my headache

kill some redundant includes.

some -Wall changes, from Jim Jegers, improved (in some cases) by me.
still needs a bit of work, but closer.

pull in helpful includes.

Eliminate uses of some obsolete functions.

Clean up import.

fix from Brad Parker <brad@fcr.com> to curb routed inflation when
using MULTICAST. it wasn't properly ignoring the new interface flags.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

update for GCC 6:

- read_rt() has a missing {} issue.

Fix that routed deletes local routes

routed previousely ignored local routes, which have RTF_LOCAL flag, because
such routes have RTF_LLINFO and routed ignored routes having the flag. When
we obsoleted RTF_LLINFO, we removed the ignoring logic from routed, then
routed started removing local routes unexpectedly.

Fix this behavior by teaching local routes to routed to ignore them.

kardel@ reported the issue and helped testing, thanks!

Shuffle pointer magic to not depend on temporary overaligned pointers
into a packed structure.

branches: 1.25.2;
Separate nexthop caches from the routing table

By this change, nexthop caches (IP-MAC address pair) are not stored
in the routing table anymore. Instead nexthop caches are stored in
each network interface; we already have lltable/llentry data structure
for this purpose. This change also obsoletes the concept of cloning/cloned
routes. Cloned routes no longer exist while cloning routes still exist
with renamed to connected routes.

Noticeable changes are:
- Nexthop caches aren't listed in route show/netstat -r
- sysctl(NET_RT_DUMP) doesn't return them
- If RTF_LLDATA is specified, it returns nexthop caches
- Several definitions of routing flags and messages are removed
- RTF_CLONING, RTF_XRESOLVE, RTF_LLINFO, RTF_CLONED and RTM_RESOLVE
- RTF_CONNECTED is added
- It has the same value of RTF_CLONING for backward compatibility
- route's -xresolve, -[no]cloned and -llinfo options are removed
- -[no]cloning remains because it seems there are users
- -[no]connected is introduced and recommended
to be used instead of -[no]cloning
- route show/netstat -r drops some flags
- 'L' and 'c' are not seen anymore
- 'C' now indicates a connected route
- Gateway value of a route of an interface address is now not
a L2 address but "link#N" like a connected (cloning) route
- Proxy ARP: "arp -s ... pub" doesn't create a route

You can know details of behavior changes by seeing diffs under tests/.

Proposed on tech-net and tech-kern:
http://mail-index.netbsd.org/tech-net/2016/03/11/msg005701.html

merge 2.32 which is actually 2.31. Most important change is:
- use multicast over point-to-point (GRE) links as proposed by Dan Lukes.

fix printf formats.

branches: 1.22.26; 1.22.38;
PR 9431: if rnh_addradd() fails, total_routes will be wrong.

constification lossage (hi gimpy)

branches: 1.20.2;
strcpy/sprintf cleanup. from openbsd

use snprintf. there are more...

array seems too small

Avoid strict alias warnings.

sync with routed 2.27

fix -Wshadow warnings

sync with version 2.24. Vernon says:
The most significant [fix] involves so called "remote" interfaces
configured in the kludge file to with what appear to be colliding
networks. Edward Mascarenhas <eddiem@vihar.engr.sgi.com> found
the problem and the fix, and I think has tested it in the SGI
network.

make local functions static and portability casts.

PR/9390: Jarkko Torppa: Routed leaks memory when route addition fails.

Ignore RTM_OIFINFO messages (send by a COMPAT_14 kernel).

branches: 1.10.6;
LP64 tweaks.

Resolve conflicts.

Update to 980910 from ftp.rhyolite.com

Merge the 980602 RIPv2/Router Discovery routed. Fixes floods of host routes
generated when one of the interfaces on the network does not support
broadcast (e.g. HIPPI or ATM).

From Vern Schryver <vjs@rhyolite.com>

make these compile on the alpha after WARNS=1.

* cleanup for WARNS=1
* deprecate register
* bcmp, bcopy, bzero, rindex -> memcmp, memmove, memset, strrchr
* correct use of .Nm

- resolve conflicts.
- remove md5 stuff; it is in libc.
- define MCAST_PPP_BUG, until we fix if_ppp.c

- resolve conflicts
- try to follow rcsid style more closely
- fix rn_walktree callback routine signatures.

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

branches: 1.1.1;
Initial revision

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

- Support rip version 2.
- Add ansi prototypes.
- Be less internet centric.
- Convert queues to use <sys/queue.h> so we don't need -lcompat

convert to new RCS Id conventions; reduce my headache

Clean up import.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

- Support rip version 2.
- Add ansi prototypes.
- Be less internet centric.
- Convert queues to use <sys/queue.h> so we don't need -lcompat

<sys/errno.h> -> <errno.h>
<sys/signal.h> -> <signal.h>
<sys/syslog.h> -> <syslog.h>

Fix bug in af_format where the formatted address was returned in the same
static buffer. When af_format is used more than once in the same printf
statement, the second call will overwrite the first. Changed af_format to
take a buffer and the buffer size where to place the formatted string.

convert to new RCS Id conventions; reduce my headache

kill some redundant includes.

some -Wall changes, from Jim Jegers, improved (in some cases) by me.
still needs a bit of work, but closer.

pull in helpful includes.

Eliminate uses of some obsolete functions.

Clean up import.

Undo that last change. It was 100%, beyond a doubt, completely bogus.
afswitch and af_max were still used, but since they was declared 'common'
by virtue of their declaration in af.h, the symbols were defined at link-
time, and their contents were considered BSS -- init to zero.
Now, you and i know that jumping to zero causes a core dump...
Therefore: af.c has been resurrected, so the procedure pointers actually
point to procedures, tables.c has had it's change nuked, and af.h
has been changed so afswitch and af_max are declared as 'extern'.
It's obvious that that 'fix' wasn't even tested.

Move inet_default into tables.c, and remove all of af.c, as it is no longer
used.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

- Support rip version 2.
- Add ansi prototypes.
- Be less internet centric.
- Convert queues to use <sys/queue.h> so we don't need -lcompat

convert to new RCS Id conventions; reduce my headache

some -Wall changes, from Jim Jegers, improved (in some cases) by me.
still needs a bit of work, but closer.

Clean up import.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

unifdef sgi. NFCI

fix sign-compare issue

branches: 1.31.30;
Coverity CID 1301: Return early if we could not open /dev/null

constification lossage (hi gimpy)

sprintf can return negative

strcpy/sprintf cleanup. from openbsd

sync with routed 2.27

va_{start,end} audit:
Make sure that each va_start has one and only one matching va_end,
especially in error cases.
If the va_list is used multiple times, do multiple va_starts/va_ends.
If a function gets va_list as argument, don't let it use va_end (since
it's the callers responsibility).

Improved by comments from enami and christos -- thanks!

Heimdal/krb4/KAME changes already fed back, rest to follow.

Inspired by, but not not based on, OpenBSD.

sync with version 2.24. Vernon says:
The most significant [fix] involves so called "remote" interfaces
configured in the kludge file to with what appear to be colliding
networks. Edward Mascarenhas <eddiem@vihar.engr.sgi.com> found
the problem and the fix, and I think has tested it in the SGI
network.

portability casts and re-indentation of some code.

Resolve conflicts.

Update to 980910 from ftp.rhyolite.com

Merge the 980602 RIPv2/Router Discovery routed. Fixes floods of host routes
generated when one of the interfaces on the network does not support
broadcast (e.g. HIPPI or ATM).

From Vern Schryver <vjs@rhyolite.com>

Use S_IS*(), not S_IF*.

make these compile on the alpha after WARNS=1.

use memmove, memcmp, memset instead of bcopy, bcmp, bzero

* cleanup for WARNS=1
* deprecate register
* bcmp, bcopy, bzero, rindex -> memcmp, memmove, memset, strrchr
* correct use of .Nm

- resolve conflicts.
- remove md5 stuff; it is in libc.
- define MCAST_PPP_BUG, until we fix if_ppp.c

- resolve conflicts
- try to follow rcsid style more closely
- fix rn_walktree callback routine signatures.

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

- Support rip version 2.
- Add ansi prototypes.
- Be less internet centric.
- Convert queues to use <sys/queue.h> so we don't need -lcompat

<sys/errno.h> -> <errno.h>
<sys/signal.h> -> <signal.h>
<sys/syslog.h> -> <syslog.h>

fix arguments to tracing functions and/or cast, to avoid warnings.

convert to new RCS Id conventions; reduce my headache

Use S_IS*().

kill some redundant includes.

some -Wall changes, from Jim Jegers, improved (in some cases) by me.
still needs a bit of work, but closer.

Eliminate uses of some obsolete functions.

Clean up import.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

- Support rip version 2.
- Add ansi prototypes.
- Be less internet centric.
- Convert queues to use <sys/queue.h> so we don't need -lcompat

fix arguments to tracing functions and/or cast, to avoid warnings.

convert to new RCS Id conventions; reduce my headache

Clean up import.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

Default to -Wno-sign-compare -Wno-pointer-sign for clang.
Push -Wno-array-bounds down to the cases that depend on it.
Selectively disable warnings for 3rd party software or non-trivial
issues to be reviewed later to get clang -Werror to build most of the
tree.

don't override LDSTATIC if NOPIC is defined.

allow LDSTATIC definition to be overridden by bsd.own.mk/mk.conf

link rtquery against dynamic libs; from Soren S. Jorvang in PR bin/4178

Install `rtquery' in /usr/sbin.

- resolve conflicts.
- remove md5 stuff; it is in libc.
- define MCAST_PPP_BUG, until we fix if_ppp.c

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

branches: 1.1.1;
Initial revision

- resolve conflicts.
- remove md5 stuff; it is in libc.
- define MCAST_PPP_BUG, until we fix if_ppp.c

branches: 1.1.1;
Initial revision

Add xref to pkgsrc/net/gated, so people know where to find gated(8).
Improve markup of RFCs in SEE ALSO.
Sort options.
Bump date.

Correct error in previous, noted by Nick Hudson.

Consistently use 'RFC 1234' instead of 'RFC1234' or 'RFC-1234'.
From jmc@openbsd.

.Nm does not need a dummy argument ("") before punctuation or
for correct formatting of the SYNOPSIS any longer.

Sort SEE ALSO.

Drop argument of .Os.

Resolve conflicts.

s/ and and / and /

Update to 980910 from ftp.rhyolite.com

fix a roff command bug

fix bad .Xr references

- Use .Nm "" instead of .Nm rtquery in SYNOPSIS.
- Don't use captialized wort to refer command name.
- Delete period at the end of ``SEE ALSO'' list.
- Use .Xr instead of .Em to refer routed.

all but the first .Nm in SYNOPSIS need to be spelt out. [bin/4256]

* cleanup for WARNS=1
* deprecate register
* bcmp, bcopy, bzero, rindex -> memcmp, memmove, memset, strrchr
* correct use of .Nm

- resolve conflicts.
- remove md5 stuff; it is in libc.
- define MCAST_PPP_BUG, until we fix if_ppp.c

fix some nits from SAITOH Masanobu, pr-3149

fix .I which should be .Em, from SAITOH Masanobu in pr-3109

- resolve conflicts
- try to follow rcsid style more closely
- fix rn_walktree callback routine signatures.

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

branches: 1.1.1;
Initial revision

don't use sprintf

gcc warns about array bounds limits, but unfortunately the API here requires
flex arrays in a union which is not allowed. So do a trivial pointer assignment
to baffle gcc again.

branches: 1.23.2; 1.23.8;
Use __dead

merge 2.32 which is actually 2.31. Most important change is:
- use multicast over point-to-point (GRE) links as proposed by Dan Lukes.

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

branches: 1.20.4; 1.20.8;
convert __attribute__s to applicable cdefs.h macros

branches: 1.19.2;
__attribute -> __attribute__.

branches: 1.18.10;
quell GCC 4.1 uninitialised variable warnings.

XXX: we should audit the tree for which old ones are no longer needed
after getting the older compilers out of the tree..

change (mostly) int to socklen_t. GCC 4 doesn't like that int and
socklen_t are different signness.

sync with routed 2.27

Negative exit code cleanup: Replace exit(-x) with exit(x).
As seen on tech-userlevel.

sync with version 2.24. Vernon says:
The most significant [fix] involves so called "remote" interfaces
configured in the kludge file to with what appear to be colliding
networks. Edward Mascarenhas <eddiem@vihar.engr.sgi.com> found
the problem and the fix, and I think has tested it in the SGI
network.

Pull in <md5.h> rather than defining our own MD5 structures (!).

portability casts

negative ip octets don't make sense. Now we print byte swapped but the
addresses are at least recognizable without negating :-)

Resolve conflicts.

Update to 980910 from ftp.rhyolite.com

Merge the 980602 RIPv2/Router Discovery routed. Fixes floods of host routes
generated when one of the interfaces on the network does not support
broadcast (e.g. HIPPI or ATM).

From Vern Schryver <vjs@rhyolite.com>

Remove the unrecognized `-v' option from the usage information;
from Geoff C. Wing in PR bin/5521.

add noreturn attribute, and fix the return value of main.

* cleanup for WARNS=1
* deprecate register
* bcmp, bcopy, bzero, rindex -> memcmp, memmove, memset, strrchr
* correct use of .Nm

- resolve conflicts.
- remove md5 stuff; it is in libc.
- define MCAST_PPP_BUG, until we fix if_ppp.c

- resolve conflicts
- try to follow rcsid style more closely
- fix rn_walktree callback routine signatures.

- Finished up merge.

- Add some prototypes and definitions for types as necessary,
wrapped in __NetBSD__ to maintain portability.

- Explicit size types in structures that go out the wire.

- RCS id police.

branches: 1.1.1;
Initial revision