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: 1.18.46;
Enable WARNS=4 by default, except for:
cpuctl dumplfs hprop ipf iprop-log kadmin kcm kdc kdigest
kimpersonate kstash ktutil makefs ndbootd ntp pppd quot
racoon racoonctl rtadvd sntp sup tcpdchk tcpdmatch tcpdump
traceroute traceroute6 user veriexecgen wsmoused zic
(Mostly third-party applications)

branches: 1.17.20;
- Compile in with debugging by default.
- General cleanup.

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).

use CPPFLAGS instead of CFLAGS

branches: 1.13.2;
enable WARNS=1 by default, but disable in unclean 3rd party code

use WARNS?=1 not WARNS=1 , cleanup manpage typo

Teensy change to use WARNS.

* use _yp_invalid_domain() (from libc) to validate the domainnames
* cleanup for -Wall
* use __RCSID

the following were inspired by openbsd:
* only allow connections from reserved ports
* implement -insecure, which turns off the above restriction (required
for sunos 3.x and ultrix)
* prevent more than 100 domains from being bound at once, preventing
a denial of service attacks

-DDEAMON is no longer used, so remove it from CFLAGS.

RCS id police.

specify man pages the new way.

man page from Jason Thorpe

from wolfgang, various further cleanup, and
mpliment the "clients are asking via RPC, therefore that might mean
the server is possibly dead" heuristic I bailed on earlier.

kill bindirs

Add RCS identifiers.

use BINDIR, rather than DESTDIR to specify binary's location.
the former is correct, the latter is not. actually, this makefile
probably should be hacked a bit more, but i'm not up to it now

initial rev

Remove clause 3 and 4 from TNF licenses

branches: 1.2.44;
Fix files using old TNF copyright notice

Add Luke Mewburn's support for a .ypservers file, which contains a list
of YP servers a client should bind to, mostly verbatim, but slightly
modified for better semantics when nagging servers if a ypset has been
issued. Default to broadcast mode if no .ypservers file is present.

Documentation changes to match, slightly tweaked by Scott Reynolds and
myself.

Closes PR #1759.

Remove workaround for ancient HTML generation code.

Add missing .El. From Henning Petersen in PR 49227.

Use more markup.

Document exponential backoff behavior and SIGHUP support, plus a couple
other minor edits.

branches: 1.18.6; 1.18.22; 1.18.26; 1.18.38;
Convert TNF licenses to new 2 clause variant

branches: 1.17.18;
Improve diagnostics section. Inspired by PR 32217 from Jeff Ito.

Sort SEE ALSO (first section, then name); new sentence, new line; bump date for previous.

YP -> NIS

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

Improve markup in once place, sort SEE ALSO, use standard headers.

Whitespace nits.

note that messages are send to syslog daemon using facility LOG_DAEMON.

* -insecure is for binding to insecure *servers*, not insecure *clients*.
noted by Bill Studenmund <skippy@macro.stanford.edu>.
* use .Tn YP instead of YP

branches: 1.9.2;
clarify that "domainname" in various pathnames is the actual domain name

use WARNS?=1 not WARNS=1 , cleanup manpage typo

Fix files using old TNF copyright notice

* use _yp_invalid_domain() (from libc) to validate the domainnames
* cleanup for -Wall
* use __RCSID

the following were inspired by openbsd:
* only allow connections from reserved ports
* implement -insecure, which turns off the above restriction (required
for sunos 3.x and ultrix)
* prevent more than 100 domains from being bound at once, preventing
a denial of service attacks

Change name of ypservers file to /var/yp/binding/domain.ypservers.
This is a compromise between the Solaris 2.5 naming of this file
(/var/yp/binding/domain/ypservers and /var/yp/binding/domain/vers)
and traditional NetBSD naming (/var/yp/binding/domain.vers).
Changing to the Solaris naming makes the C library incompatible
with ypbind. While it's easy to change the C library, I don't feel
right about introducing a gratuitous incompatibility.

The current name (/var/yp/domain/ypservers) conflicts with the name
of the YP map listing slave servers for a domain.

Per discussion w/ Chris Demetriou and Luke Mewburn.

.ypservers -> ypservers, to match Solaris 2.5

Add Luke Mewburn's support for a .ypservers file, which contains a list
of YP servers a client should bind to, mostly verbatim, but slightly
modified for better semantics when nagging servers if a ypset has been
issued. Default to broadcast mode if no .ypservers file is present.

Documentation changes to match, slightly tweaked by Scott Reynolds and
myself.

Closes PR #1759.

Copyright assigned to The NetBSD Foundation.

man page from Jason Thorpe

s/serivce/service/ in comments.

Initialize ypbind_resp before first possible error return condition.

branches: 1.99.2;
Make use of the defined YP errors when ypbind encounters one rather than
just returning an empty (successful) RPC result.

Don't exit when allocating memory for a new bound domain received via a RPC.

Log state transitions.

Add a SIGHUP handler; upon SIGHUP do an extra nag_servers on any
domain that's in DEAD state. This lets you explicitly rescue ypbind
from its exponential backoff when you know the world's back up.

Instead of using magic numbers in what looks like a boolean
(dom_alive), create a state enumeration (domainstates) and use it
instead.

Instead of three states (new, alive, and, effectively, 'troubled') go
to five: new, alive, pinging, lost, and dead.

Domains start in the NEW state. When we get a reply from a server, the
state goes to ALIVE. The state is set to PINGING when we ping the
server (once a minute normally) and if the ping times out, it goes to
LOST. If we stay lost for a minute, go to DEAD, and in DEAD, do
exponential backoff of nag_servers calls.

Getting rid of the broken logic attached to the 'troubled' state fixes
PR 15355 (ypbind defeats disk idle spindown) -- it will now only
rewrite the binding file when the binding changes.

Also, fix the HEURISTIC code so it doesn't trigger except in ALIVE
state. I think this was the source of a lot of the spamming behavior
seen in PR 32519, which is now fixed.

Might also fix PR 23135 (broadcast ypbind sometimes fails to find
servers).

Load up with comments.

Don't rake up the default domain until after processing arguments.
Processing arguments just sets flags -- may as well do it first, and
this way detection of silly errors isn't contingent on having things
fully configured and operating.

While there are times it's appropriate to call a state variable
"evil", this isn't one of them. Since the logic involved is to wait
until the default domain binds before backgrounding, call the variable
"started" instead.

Factor out some rpc validation code.

Don't store the default domain name in a global. While running we
really don't care which domain is the system's default domain.

branches: 1.90.4; 1.90.8; 1.90.18;
Apply casts to cases where xdrproc_t is expected but is not
strictly passed, for example because the second argument is
a different kind of pointer.

Use __dead.

Don't print uninitialized values off the stack, even as debug messages.

When checking if /var/yp/binding/foo.ypservers exists, use the domain
name of the domain being checked instead of the default domain name. Duh.
Necessary for PR 43900.

Remove two entirely unused members of struct domain (previously struct
_dom_binding). I guess these were there because it was cutpasted at
some point from struct dom_binding in <rpcsvc/yp_prot.h>.

Merge a couple more minor improvements from Wolfgang Stukenbrock's
patch in PR 43900.

The FILE* kept open for /var/yp/binding/foo.ypservers needs to be
per-domain. Also, use the file for the domain we're servicing instead
of always the file for the default domain.

Partly from Wolfgang Stukenbrock's patch in PR 43900.

The been_ypset flag really needs to be per-domain also. Do that.

Partly from Wolfgang Stukenbrock's patch in PR 43900.

The ypbindmode really needs to be per-domain, so adjust accordingly.

Partly from Wolfgang Stukenbrock's patch in PR 43900.

Allowing ypset is a global permission flag, not an operating mode like
broadcast vs. configured/direct. Don't conflate the logic. As I proposed
in PR 43900.

Fix initialization of struct domain; initialize all fields in all
cases, merge duplicate code, make sure every domain gets an xid
assigned.

Partly from Wolfgang Stukenbrock's patch in PR 43900.

The name field of a domain should be dom->dom_name, not dom->dom_domain.
(This and the previous patch make the code far more readable.)

The default name for a "domain" variable should be "dom", not "ypdb".
Change them all, and make a couple other related adjustments.

ypbindlist -> domains

xid2ypdb() -> domain_find()

Rename struct _dom_binding -> struct domain.

Move a bunch of messy sunrpc-related initialization stuff into its own
function instead of blatting it all into main().

Assimilate the open/flock combination used several times into its own
function.

Fix up calls to yp_log. They cannot use %m, because if the -d option
is used the messages are sent via printf instead of syslog(3).

Also, make sure that none of the calls include a trailing newline and
issue a trailing newline in the -d case so they actually print properly.
(This was noted by Wolfgang Stukenbrock in PR 43900.)

And finally, fix some cases that could in some circumstances print
uninitialized errnos.

Abstract out some of the handling of the ypservers file.

Check that the domain name is valid up front, instead of doing it
after opening sockets and registering services and whatnot.

Remove now-unnecessary extra forward decls; sprinkle a little more static.

Sort contents of file.

sort includes

Don't use _t as an identifier suffix meaning "variable containing a time".

Introduce DPRINTF() macro, remove most #ifdef DEBUG usages.
No change to compiler output.

u_int32_t -> uint32_t and similar

Don't cast return value of malloc

use __printflike()

Whitespace.

make this compile.

in the case where ypbind is serving more than one domain (i.e.
not just the yp_get_default_domain() domain), we must remove
any old binding files from /var/yp/binding (BINDINGDIR) from
previous runs, or the non-yp_get_default_domain()'s will not
bind properly. add a purge_bindingdir() function that basically
does "rm BINDINGDIR/*.[0-9]" at ypbind startup time.

example case of where this is an issue: bind a second (non-default)
domain. ypbind will create and flock a /var/yp/binding/xxx.2
file for it. stop and restart ypbind. the old /var/yp/binding/xxx.2
file will remain from the previous run. since it is not flock()'d
by the new instance of ypbind, libc functions like yp_master() will
fail without even bothering to talk to ypbind itself. (and ypbind
is totally unaware of the old file...)

fix -Wsign-compare issues

branches: 1.57.18;
- Compile in with debugging by default.
- General cleanup.

Ensure alignment of the structure used for RPC requests. This stop
"ypbind" from segfaulting under NetBSD-sparc64 when "ypwhich" is
used to display the current NIS server.

Fix misleading error message (from PR/32519).

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

Add (unsigned char) cast to ctype functions

Replaced strncpy with strlcpy. Thanks to Peter Postma who
pointed them our in PR #25762. Approved by christos@NetBSD.org.

Homogenize usage messages: make the 'usage' word all lowercase, as this seems
to be the most common practice in our tree.

Modify the licences of code written by Theo De Raadt from a 4-clause
to a 2-clause licence (retaining UCB clauses (1) and (2)), per PR
22409 from Joel Baker, approved by Theo de Raadt, and ratified by
myself - the only discrepancy being the handling of the original
clause 3 in src/usr.sbin/yppoll/yppoll.c.

Apply patch proposted to tech-userlevel on Sun, 12 Oct 2003:
Remove dom_server_port from struct _dom_binding. It's never initialised
and used in only one place, where dom_server_addr.sin_port was probably
intended.

width needs to be updated before each loop, because svc_maxfd may have
changed.

ANSIfy, remove __P(), unifdef __STDC__.

branches: 1.46.2;
convert to use getprogname()

use LOG_WARNING for warnings and LOG_ERR for fatal errors

__CONCAT does token pasting, not string concatnation. if something like:
__CONCAT("foo","bar");
actually works to concantate strings, it's because the preprocessor expands
it into "foo""bar" as separate strings, and then ANSI string concatenation
is performed on that. It's more straightforward to just use ANSI string
concatenation directly, and newer GCCs complain (rightly) about misuse
of token pasting.

More format string cleanups by sommerfeld.

branches: 1.42.4;
use getifaddrs, to avoid complicated alignment constraint in SIOCGIFCONF.
(do we need to keep SIOCGIFCONF code in #ifdef?)

avoid 64bit alignment issue with SIOCGIFCONF (packed struct access).
From: =?iso-8859-1?Q?G=F6ran_Bengtson?= <goeran@cdg.chalmers.se>

Spell "privilege" correctly (correct spelling from Jonathan Stone).

Use pidfile(3).

branches: 1.38.2;
<sys/fcntl.h> -> <fcntl.h>

- Makes ypbind use syslog() (with LOG_DAEMON) instead of warn()/err().
Messages will still be printed to stderr if ypbind is started in debug mode.
- Don't exist if none of the servers listed in
/var/yp/binding/<domainname>.ypservers are reacheable. It's a temporary
failure that can be recovered from.

minor KNF nits.

revert; we can't break the existing ABI and API

use u_int32_t instead of u_long

branches: 1.33.2;
Fix typo in previous.

Cast pointers to u_long before casting to another integral type.

Teensy change to use WARNS.

* use _yp_invalid_domain() (from libc) to validate the domainnames
* cleanup for -Wall
* use __RCSID

the following were inspired by openbsd:
* only allow connections from reserved ports
* implement -insecure, which turns off the above restriction (required
for sunos 3.x and ultrix)
* prevent more than 100 domains from being bound at once, preventing
a denial of service attacks

Change name of ypservers file to /var/yp/binding/domain.ypservers.
This is a compromise between the Solaris 2.5 naming of this file
(/var/yp/binding/domain/ypservers and /var/yp/binding/domain/vers)
and traditional NetBSD naming (/var/yp/binding/domain.vers).
Changing to the Solaris naming makes the C library incompatible
with ypbind. While it's easy to change the C library, I don't feel
right about introducing a gratuitous incompatibility.

The current name (/var/yp/domain/ypservers) conflicts with the name
of the YP map listing slave servers for a domain.

Per discussion w/ Chris Demetriou and Luke Mewburn.

.ypservers -> ypservers, to match Solaris 2.5

Add Luke Mewburn's support for a .ypservers file, which contains a list
of YP servers a client should bind to, mostly verbatim, but slightly
modified for better semantics when nagging servers if a ypset has been
issued. Default to broadcast mode if no .ypservers file is present.

Documentation changes to match, slightly tweaked by Scott Reynolds and
myself.

Closes PR #1759.

xdr_bool wants a bool_t (which happens to be an int32_t on our port)
and not a char

Fix a misplaced #endif - if DEBUG wasn't defined, ypbind would never
detach from the controlling tty.

Code cleanups.
- check malloc returns
- null terminate strncpy() strings
- use snprintf instead of sprintf
- pass the right arguments to the right functions
- check usage
- use err(3) and warn(3) instead of printfs.

Fix typo that can cause ypbind to assume an error when one doesn't
actually exist if the server takes a dive.

Fix buglet in ypbindproc_domain_2() which caused:

ypbind; sleep 10; ypwhich

to fail (where failure is defined as ypwhich hangs until it gives up,
complaining about not being able to communicate with ypbind). Failure
mode pointed out by Chris G. Demetriou <cgd@cs.cmu.edu>.

Pull down yp/xdr fixes from release branch.

branches: 1.20.4;
apply patch from PR 2085: don't treat XIDs and domain mapping pointers
interchangeably, because this is fragile and doesn't work at all on
the alpha. (get rid of the old hack used on the alpha, and fix RCS
Ids, too.)

really nasty hack for the alpha. bigtime XXX

Daemonify only after the bind has been completed, and use daemon().

Closer to KNF, ANSI, and POSIX.

Recompute `width' before each select(), as the value of `svc_maxfd' can
be changed by the rpc library.

Avoid getdtablesize(3). Also, remove some useless FD_CLR()s.

be a bit more careful with types, and prototype scope.

create ypbind.lock file
fix ypset code

force first binding poll automatically -- someone will need it very soon!

Keep a server even if we lost contact when noone else is available

changes from <ws@tools.de> to do selective pinging of servers.

branches: 1.9.2;
from wolfgang, various further cleanup, and
mpliment the "clients are asking via RPC, therefore that might mean
the server is possibly dead" heuristic I bailed on earlier.

improved rebinding from ws@tools.de

correct copyright

sunos compatible /var/yp/binding/ files.

fix typo in last

ypset should only be permitted from a reserved port.
from Tor Egge <tegge@pvv.unit.no>

#endif in the wrong place

added my copyright.. how self efacing of me

initial rev