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: 1.57.4;
add LIBKRB5_LDADD/LIBKRB5_DPADD and their static counterparts
to bsd.prog.mk. use them instead of hard coding various lists
of libraries for krb5.

this fixes static builds.

Remove redundant WARNS=5.

branches: 1.55.2;
no more KRB5_DEPRECATED

remove obsolete comment.

make krb5 compile again. XXX: one function left that is deprecated, what's
the new equivalent?

branches: 1.52.4; 1.52.6;
We no longer need -I/usr/include/krb5.

PR/42540: Ed Ravin: /usr/bin/login does not log normal logins, and does not
log ip addresses.
- Factor out the common code in login.c and login_pam.c into common.c
- Always log a login event
- Check passed in sockaddr against the one from getpeername(2).

Enable WARNS=4 by default for usr.bin, except for:
awk bdes checknr compile_et error gss hxtool kgetcred kinit
klist ldd less lex locale login m4 man menuc mk_cmds
mklocale msgc openssl rpcgen rpcinfo sdiff spell ssh
string2key telnet tn3270 verify_krb5_conf xlint

branches: 1.49.4;
Use ${TOOL_AWK} instead of ${AWK} or plain "awk" in make commands.
Pass AWK=${TOOL_AWK:Q} to shell scripts that use awk.

Be consistent about making USE_PAM!=no be exclusive to USE_SKEY!=no.

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.

Goodbye KerberosIV

Introduce PAM_STATIC_LDADD and PAM_STATIC_DPADD. When compiling
with MKPIC=no, possibly because the target does not support shared
libraries, these include libraries required to resolve all symbols
which end up referenced from PAM-using applications. The libraries
presently required are -lcrypt, -lrpcsvc and -lutil.

Add use of these variables which are currently set up to use PAM,
so that they compile when MKPIC=no.

Also, in the telnetd case, reorder the order of the libraries, so
that libtelnet.a comes before -ltermcap and -lutil, again to fix
link error when MKPIC=no.

Discussed with thorpej and christos.

Add PAM support to login(8)

Kill __P(), ANSIfy, remove main() prototype, whitespace; WARNS=2

Generate the copyright string from sys/conf/copyright rather than
replicating it here. Idea from Simon Burge.

Fix the checkflist for builds without Kerberos 4 (MKKERBEROS4=no)
and without Kerberos 4 & 5 (MKKERBEROS=no). Previously checkflist
complained of missing files.

* move kerberos- and kerberos 4-only files into new flists,
distrib/sets/lists/*/krb.*

* make the flist generators grok MKKERBEROS{,4} variables

* fix Makefiles which treat MKKERBEROS=no as MKKERBEROS5=no.
9 out of 10 experts agree that it is ludicrous to build w/
KERBEROS4 and w/o KERBEROS5.

* fix header files, also, which treat MKKERBEROS=no as MKKERBEROS5=no.

* omit some Kerberos-only subdirectories from the build as
MKKERBEROS{,4} indicate

(I acknowledge the sentiment that flists are the wrong way to go,
and that the makefiles should produce the metalog directly. That
sounds to me like the right way to go, but I am not prepared to do
revamp all the makefiles. While my approach is expedient, it fits
painlessly within the current build architecture until we are
delivered from flist purgatory, and it does not postpone our
delivery. Fair enough?)

libkrb depends on libdes, patch in private mail from
Harold Gutch logix at foobar franken de

split MKKERBEROS4 from MKKERBEROS. based on work by lha at stacken.kth.se
(build confirmed with both MKKERBEROS4=yes and MKKERBEROS4=no)

add DPADD.

make this compile without SUPPORT_UTMPX

Factor out the utmp/wtmp/lastlog updating and add utmpx/wtmpx/lastlogx updating.
Both are turned on for now.

Split the notion of building Hesiod, Kerberos, S/key, and YP
infrastructure and using that infrastructure in programs.

* MKHESIOD, MKKERBEROS, MKSKEY, and MKYP control building
of the infratsructure (libraries, support programs, etc.)

* USE_HESIOD, USE_KERBEROS, USE_SKEY, and USE_YP control
building of support for using the corresponding API
in various libraries/programs that can use it.

As discussed on tech-toolchain.

Remove INSTALLFLAGS=-fschg, as per change to usr.bin/ssh/ssh/Makefile.

remove -lvers, it's not used

update build infrastructure for heimdal 0.3a

Fix library order.

Add MK... variables to enable/disable various aspects of building
crypto support into the system. See share/mk/bsd.README for more
a full description.

branches: 1.29.2;
Merge a bunch of things from crypto-us and crypto-intl into basesrc,
adding support for Heimdal/KTH Kerberos where easy to do so. Eliminate
bsd.crypto.mk.

There is still a bunch more work to do, but crypto is now more-or-less
fully merged into the base NetBSD distribution.

branches: 1.28.2;
Clean up changes a bit.

Re-enable login_cap processing, now that a non-existing login.conf
won't make it crash. Also make it less noisy in that case.

Ack. Disable login caps, until problems are fixed.

Add login.conf capability setting.

optionally include CRYPTOPATH Makefile.frag files.

Use bsd.crypto.mk.

Kerberos5 changes to login -- now supports forwarded TGTs.

Build domestic login on domestic systems.
Incorporated (minor) diff from domestic man page.

Remove the crypto-related bits until such time as we have a fully-
integrated source tree. Export-controlled versions of these are now
built during the domestic build process.

don't define WARNS=1 here

SRCS must be defined *before* bsd.prog.mk is included...

Minor changes.

Minor changes.

Pull in bsd.own.mk for SKEY, KERBEROS, KERBEROS5.

CFLAGS->CPPFLAGS; Conditionalize SKEY

Various changes to keep up with krb5, mostly addition of the kcontext
structure. From PR 3826, by Chris Jones.

* add functionality to `force password change at next login'. to use,
set the pw_change field of the user to -1 (defined in <pwd.h> as
_PASSWORD_CHGNOW). based on [bin/936] by Simon Gerraty <sjg@quick.com.au>
* clean up for WARNS?=1

Install statically linked, as suggested by mrg@eterna.com.au in [bin/1715].

Merged with 4.4lite.
Changed to conform to NetBSD's new RCS Id convention.

Add support for Kerberos5 authentication.

add s/key support

some changes to make Kerberos a bit easier to use. from
Michael Graff <explorer@vorpal.com>, with some work by myself...

always use libcrypt

Add RCS indentifiers.

update to new version from uunet

changed to use new libcrypt scheme.

added support for using real crypt

branches: 1.1.1;
Initial revision

Fixed off-by-one in decode_ss (CID 977426)

OK kamil@

Added namelen initialization in getpeername(3) call (CID 979631)

OK kamil@

- use __dead consistently.
- more clang build.

dead for sigint

make krb5 compile again. XXX: one function left that is deprecated, what's
the new equivalent?

branches: 1.3.6; 1.3.8;
don't repeat the word login twice (syslog prepends the program name login:)

kill the space in address formatting to match what other daemons do.

PR/42540: Ed Ravin: /usr/bin/login does not log normal logins, and does not
log ip addresses.
- Factor out the common code in login.c and login_pam.c into common.c
- Always log a login event
- Check passed in sockaddr against the one from getpeername(2).

- use __dead consistently.
- more clang build.

Add some noreturn attributes for clang.

make krb5 compile again. XXX: one function left that is deprecated, what's
the new equivalent?

branches: 1.1.6; 1.1.8;
PR/42540: Ed Ravin: /usr/bin/login does not log normal logins, and does not
log ip addresses.
- Factor out the common code in login.c and login_pam.c into common.c
- Always log a login event
- Check passed in sockaddr against the one from getpeername(2).

no more KRB5_DEPRECATED

last commit before I nuke the old code.

move more of the compat code in the compat block.

make kerberos work again:
1. make notickets external
2. don't use the tty as part of the credential cache, since pts/1 will not work.
3. Attempt to use the newer functions, but punt for now since it does not work yet.

centralize error function processing.

make krb5 compile again. XXX: one function left that is deprecated, what's
the new equivalent?

branches: 1.27.44; 1.27.46;
Get rid of more Kerberos 4 code.

Kill __P(), ANSIfy, remove main() prototype, whitespace; WARNS=2

Only include Kerberos 4 headers when compiling in a Kerberos enviroment

From Jukka Salmi in NetBSD PR misc/28727

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

Patches provided by Joel Baker in PR 22365, verified by myself.

simplify by asprintf/strdup

when creating a v4 ticket file, name it after the users uid, not our
current (should finally fix bin/13040)

update infrastructure for krb4 1.1 and heimdal 0.4e

implement 5to4 (getting v4 tickets from the v5 ones)
from the patch in bin/13040 by lha@stacken.kth.se (slightly modified)

* Use krb5_warn() instead of non-functional com_err().
* knf patrol
* De-__P()

don't need to specify facility LOG_AUTH as it's the default

Don't strcpy the contents of an environment variable into a fixed-size
buffer, use strlcpy instead. Should fix security/11550.

Replace all calls to 'error_message()' with 'krb5_get_err_text()', since
error_message() does not seem to display meaningful information, under
Heimdal.

Check retrieved TGT against local keytab, if it exists.

krb5_get_in_tkt() (called by krb5_get_in_tkt_with_password()) may
eventually call krb5_free_principal() via krb5_free_creds_contents(),
(when it succeeds, in particular). Check for the creds.server
already being freed, and don't free it again.

If neither Kerberos IV or Kerberos V are configured, don't
issue "Warning: no Kerberos tickets issued."

branches: 1.12.2;
Allow krb5 and krb4 to be compiled in to the same login binary at once,
krb5 can request a forwardable TGT,
can get both krb4 and krb5 tickets, if explicitely configured to do so
by the krb5.conf.

branches: 1.11.2;
Patch from assar@netbsd.org to make k5login.c work with both MIT and Heimdal
kerberos trees.

Fixed 'login in free' warning in kdestroy().

Made login continue without kerberos when there is no krb5.conf present
(and KERBEROS5 is defined).

branches: 1.8.4;
make this compile....

Kerberos5 changes to login -- now supports forwarded TGTs.

Get rid of special cases for `s/key' password.

Various changes to keep up with krb5, mostly addition of the kcontext
structure. From PR 3826, by Chris Jones.

* add functionality to `force password change at next login'. to use,
set the pw_change field of the user to -1 (defined in <pwd.h> as
_PASSWORD_CHGNOW). based on [bin/936] by Simon Gerraty <sjg@quick.com.au>
* clean up for WARNS?=1

remove dangerous sprintf calls.

Merged with 4.4lite.
Changed to conform to NetBSD's new RCS Id convention.

Add support for Kerberos5 authentication.

Get rid of more Kerberos 4 code.

Kill __P(), ANSIfy, remove main() prototype, whitespace; WARNS=2

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

Patches provided by Joel Baker in PR 22365, verified by myself.

include des.h, not kerberosIV/des.h

use strlcpy

If neither Kerberos IV or Kerberos V are configured, don't
issue "Warning: no Kerberos tickets issued."

branches: 1.15.2;
Allow krb5 and krb4 to be compiled in to the same login binary at once,
krb5 can request a forwardable TGT,
can get both krb4 and krb5 tickets, if explicitely configured to do so
by the krb5.conf.

branches: 1.14.6;
Make this compile again in an ELF world.

Use S_IS*(), not S_IF*.

s/bzero/memset

Get rid of special cases for `s/key' password.

* add functionality to `force password change at next login'. to use,
set the pw_change field of the user to -1 (defined in <pwd.h> as
_PASSWORD_CHGNOW). based on [bin/936] by Simon Gerraty <sjg@quick.com.au>
* clean up for WARNS?=1

Prototype everything, fix RCS Id

remove dangerous sprintf calls.

use == not =. pr#2449 (aaron)

Use tty as cred file uniquifier. (From Michael Graff)

Merged with 4.4lite.
Changed to conform to NetBSD's new RCS Id convention.

Add support for Kerberos5 authentication.

some changes to make Kerberos a bit easier to use. from
Michael Graff <explorer@vorpal.com>, with some work by myself...

Add RCS identifiers.

branches: 1.1.1;
update to new version from uunet

Add newgrp(1) to SEE ALSO cross-reference list.

Use Nm instead of Xr to itself, remove reference to uucp(1), fix casing
of Kerberos.

Reference login.conf.

Prefer mdoc markup to roff.

Remove trailing whitespace.

Update documentation for pam, and don't accept flags that we don't support
anymore.

Remove some whitespace.

Add a -a option to pass the host address of the host we logged in from in
addition to the host name. This option is needed by programs suchs as
telnetd, because at the point they invoke login, they already have opened
a pty, and the stdin of login is no longer the socket connected to the
remote host. In addition don't chop the hostname up to the first dot if
the domain matches. These practices are 70's fashion and they only serve
to lose information. These days we have long enough fields in utmpx and
wtmpx to store the full names.

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

Patches provided by Joel Baker in PR 22365, verified by myself.

facilty->facility, from Igor Sobrado in PR 19670

Sort SEE ALSO section, and correct section for kerberos (1 -> 8).

Xr passwd.conf

branches: 1.18.4;
Clean up changes a bit.

Kerberos5 changes to login -- now supports forwarded TGTs.

More and more .Os cleanups. .Os is defined in the tmac.doc-common file,
so we shouldn't override it with versions in the manpages. Many more to
come.

Build domestic login on domestic systems.
Incorporated (minor) diff from domestic man page.

Correct documentation of /etc/nologin to note that it does not
apply to the superuser, per PR#6328.
Correct some nroff nits in the process.

Execute ttyaction on termination of rlogind/telnetd sessions.
Also, say a little bit about ttyaction in the getty and login manpages.

Add .Xr ttyaction 3 per PR#4647

branches: 1.11.2;
fix .Nm usage

Several things:
* Change the semantics of the `-s' option somewhat. If specified, allow
either Kerberos or S/Key login, but not a plain password.
* Eliminate the special `s/key' password; just type it at the prompt.
* Remove the root instance special case. This is a serious security hole
waiting to happen, and no other system works this way.
* Don't force a password change if Kerberos was used. Also, don't call
/bin/passwd at all if the password change isn't required.

Apply [bin/3270] from Simon J. Gerraty <sjg@quick.com.au>, with fixes by me:
* if the user has an s/key, provide a reminder in the password prompt
* if '-s' is given once, force a user that has an s/key to use it
* if '-s' is given more than once, only permit s/key logins

alternate -> alternative, per PR 2643

Sync with 4.4lite2

Fix typo, noted by Masanobu Saitoh in PR #1272

Merged with 4.4lite.
Changed to conform to NetBSD's new RCS Id convention.

add s/key support

Fix spelling errors.

Add RCS indentifiers.

branches: 1.1.1;
Initial revision

Remove the syslogging of a dial out login warning based on the tty name, because
the test is sloppy and doesn't correspond to the current naming anyway.
OK gdt@
PR#377 can remain closed.

branches: 1.104.4;
Fix gcc48 build without Kerberos.

While here, rectify related bug where if gethostname() failed, login
would null-terminate uninitialized stack trash and cheerfully pass it
on to kerberos. In this case, revert to "amnesiac" instead.

branches: 1.103.2;
Add some noreturn attributes for clang.

don't abuse the instance variable

make kerberos work again:
1. make notickets external
2. don't use the tty as part of the credential cache, since pts/1 will not work.
3. Attempt to use the newer functions, but punt for now since it does not work yet.

fix the USE_KERBEROS=no USE_PAM=no build.

make krb5 compile again. XXX: one function left that is deprecated, what's
the new equivalent?

branches: 1.98.2; 1.98.4;
NULL does not need a cast

PR/42540: Ed Ravin: /usr/bin/login does not log normal logins, and does not
log ip addresses.
- Factor out the common code in login.c and login_pam.c into common.c
- Always log a login event
- Check passed in sockaddr against the one from getpeername(2).

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

branches: 1.95.12;
PR/36294: De Zeurkous: login(1) -- inconsistent exit behavior on EOF
- Always EXIT_FAILURE when login fails.
- Cleanup fork code in krb5 (old login).

Remove duplicate #includes
From Slava Semushin <slava.semushin@gmail.com>, via private mail

Correct indent. It's important around many #ifdefs.

s/0/NULL/ for execl*'s last argument.

branches: 1.91.2;
Use TABs instead of spaces.

Remove unused variable 'salt'
Found by code inspection in one of my classes, not via Coverity :)

OK'd by christos

Get rid of more Kerberos 4 code.

PR/23616: Christian Biere: login might not back-off as expected

PR/31059: Zafer Aydogan: login too verbose

Don't display a different message for root login on insecure terminal.

Approved by christos@.

PR/32873: Johan Veenhuizen: login(1) does not initialize lastlogx struct
properly

PR/32870: Johan Veenhuizen: login(1) does not obey .hushlogin

Add a colon before %m in calls to syslog(3) for consistency with other
messages in this same file and in many other tools.

Make S/Key prompt compliant with RFC 2289. Patch supplied by Dave Huang
in PR bin/23167.

branches: 1.82.2;
- use F_CLOSEM instead of walking the fd list
- determine pty name properly for ptyfs ptys
- use setproctitle(NULL) to kill our args

Kill __P(), ANSIfy, remove main() prototype, whitespace; WARNS=2

Kill __P() and ANSIfy, ok christos.

Add a -a option to pass the host address of the host we logged in from in
addition to the host name. This option is needed by programs suchs as
telnetd, because at the point they invoke login, they already have opened
a pty, and the stdin of login is no longer the socket connected to the
remote host. In addition don't chop the hostname up to the first dot if
the domain matches. These practices are 70's fashion and they only serve
to lose information. These days we have long enough fields in utmpx and
wtmpx to store the full names.

Use getprogname(); add -F to usage. From Kouichirou Hiratsuka in PR 26222.

branches: 1.77.2; 1.77.4; 1.77.6;
Generate the copyright string from sys/conf/copyright rather than
replicating it here. Idea from Simon Burge.

Welcome to 2004!

snprintf() terminates string with \0, so there's no need for "sizeof(x)-1".

Make getlastlogx have the pathname to the lastlogx database as first
argument, to be consistent with updlastlogx.

Approved by christos, reviewed by kleink.

[The lastlogxname function should not be used any longer.]

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

Patches provided by Joel Baker in PR 22365, verified by myself.

Update copyright notice.

use strlcpy

disallow users from alter log file entries by using "login foo".
from xs@kittenz.org

make sure to use %s on printf format string. xs@kittenz.org

remove debugging message leftover

make this compile without SUPPORT_UTMPX

Factor out the utmp/wtmp/lastlog updating and add utmpx/wtmpx/lastlogx updating.
Both are turned on for now.

branches: 1.65.2;
Happy New Year!

comment or delete text after CPP directives.

LOG_ODELAY is deprecated. use LOG_ERR for fatal errors

Happy new year!

Don't syslog that krb5_init_context() failed if it failed due
to Kerberos not being configured on the system.

If neither Kerberos IV or Kerberos V are configured, don't
issue "Warning: no Kerberos tickets issued."

skey_keyinfo() returns const.

branches: 1.58.2;
Backout login_get_kconf function, because it breaks crypto-intl builds.
Keep the variables for setting default behaviour with krb4 and krb5
compiled in, even though they act like constants, to facilitate adding
another preference mechanism later.

Allow krb5 and krb4 to be compiled in to the same login binary at once,
krb5 can request a forwardable TGT,
can get both krb4 and krb5 tickets, if explicitely configured to do so
by the krb5.conf.

branches: 1.56.2;
- For root, if ignorenologin is set in login.conf, don't call
checknologin (previously, checknologin is called if ignorenologin is set).
- For non-root user, make ignorenologin take effect.

Cosmetic changes.

Move include/kerberosIV/com_err.h to include/com_err.h.

Add login_getpwclass to libutil as convenience function for
programs originally for FreeBSD.
Add parsing of "setenv" parameter which can be used to set
up an initial environment on login.

Removed some fallback cases since that is now done in libutil.

Clean up changes a bit.

Re-enable login_cap processing, now that a non-existing login.conf
won't make it crash. Also make it less noisy in that case.

Ack. Disable login caps, until problems are fixed.

Add login.conf capability setting.

copyright year change (thanks cgd) so we emit , 2000 now.

Made login continue without kerberos when there is no krb5.conf present
(and KERBEROS5 is defined).

branches: 1.45.4;
make this compile....

Kerberos5 changes to login -- now supports forwarded TGTs.

Don't declare login here. It is declared in <util.h>

branches: 1.42.2;
Show year of last login.

add copyright 1999

Add { and } to shut up egcs. Reformat the more questionable code.

const poisoning.

do _NOT_ use system(3) in setuid programs. KNF.

- 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

Need <time.h> for ctime() and time() prototypes.

Also save groups before chdir($HOME). This fixes a problem pointed
out by Bernd Ernesti with /home/lusers/joe (being uid joe, gid lusers),
with /home/lusers mode 750 and owner root:lusers.

Happy new year!

Give up special privileges before chdir($HOME) and access(.hushlogin),
fixing PR 4636 by myself with some help from Jason Thorpe.

print TNF copyright, like the kernel does.

branches: 1.31.2;
Undo part of the previous; don't allow logins if we've passed pw_change.
The semantics of this are not well documented. *sigh*

Refuse login only when we've past pw_expire, not pw_change. Check pw_expire
first.

Several things:
* Change the semantics of the `-s' option somewhat. If specified, allow
either Kerberos or S/Key login, but not a plain password.
* Eliminate the special `s/key' password; just type it at the prompt.
* Remove the root instance special case. This is a serious security hole
waiting to happen, and no other system works this way.
* Don't force a password change if Kerberos was used. Also, don't call
/bin/passwd at all if the password change isn't required.

If we compile without SKEY, abort if a -s option is used, rather than silently
failing to enforce it.

Fix uninitialized variable.

Lseek(2) usage cleanup: the use of L_SET/L_INCR/L_XTND is deprecated,
use SEEK_SET/SEEK_CUR/SEEK_END instead.

Various changes to keep up with krb5, mostly addition of the kcontext
structure. From PR 3826, by Chris Jones.

* add functionality to `force password change at next login'. to use,
set the pw_change field of the user to -1 (defined in <pwd.h> as
_PASSWORD_CHGNOW). based on [bin/936] by Simon Gerraty <sjg@quick.com.au>
* clean up for WARNS?=1

add comparisons to NULL implicit in my last changes
compare result of getopt() to -1, not EOF

Don't leak some information (``you have no s/key'').

Only information leaks now are:
* if '-s -s' is used (only allow s/key users, and force s/key use),
then "login incorrect" will be given if a non-s/key user (or
non-existant user) attempts to login; no password will be prompted
for.
XXX: maybe this should be fixed, but further analysis is required.
* an s/key user will be reminded in the "Password" prompt that they
have an s/key. Therefore it would be possible to determine if a user
is active on the machine if they have an s/key.
XXX: maybe an option is required to control this behaviour

use _PASSWORD_WARNDAYS from <pwd.h>

Apply [bin/3270] from Simon J. Gerraty <sjg@quick.com.au>, with fixes by me:
* if the user has an s/key, provide a reminder in the password prompt
* if '-s' is given once, force a user that has an s/key to use it
* if '-s' is given more than once, only permit s/key logins

Move stdio.h before skey.h to get the FILE struct.

include <skey.h> for skey function prototypes, add parens for gcc -Wall

remove dangerous sprintf calls.

Longer login name support: use MAXLOGNAME, not UT_NAMESIZE

After the chown(ttyn, ...), call ttyaction(ttyn, "login", pwd->pw_name)

Don't display skey error on s/key logins... This gives attackers some
information

Updated to use <util.h>.
From Greg Hudson <ghudson@MIT.EDU>.

Merged with 4.4lite.
Changed to conform to NetBSD's new RCS Id convention.

pr#377: dialup line hack is silly

Add support for Kerberos5 authentication.

add s/key support

Oops. Fix typo.

Make sure we check the tty `secure' status even if root has no password.
Suggested by Havard Eidnes <Havard.Eidnes@runit.sintef.no>, but reimplemented
due to bugs in his patch.
Also, don't display `root login refused ...' if the password was mistyped.

Add RCS identifiers.

update to new version from uunet

changed to use new libcrypt scheme.

Cleanup for GCC 2 and make copyright notice look better.

fixed wfj's trampling on UCB copyright notices.

branches: 1.1.1;
Initial revision

Fix various typos, mainly in comments.

s/begining/beginning/ in comments and messages.

usr.bin: remove unnecessary lint comment CONSTCOND

Since 2021-01-31, lint no longer warns about 'do ... while (0)'.

No functional change.

Avoid use-after-free bug in PAM environment

Traditional BSD putenv(3) was creating an internal copy of the passed
argument. Unfortunately this was causing memory leaks and was changed by
POSIX to not allocate.

Adapt the putenv(3) usage to modern POSIX (and NetBSD) semantics.

branches: 1.25.16;
- Added error checks for initgroups(3) and setgroups(2).
- Reorder functions in privilege regain - setgroups(2) should be called after
seteuid(2).

OK christos@

Remove the syslogging of a dial out login warning based on the tty name, because
the test is sloppy and doesn't correspond to the current naming anyway.
OK gdt@
PR#377 can remain closed.

branches: 1.23.4;
fix unused variable warnings

branches: 1.22.2;
trimusername (which does not exist) probably was meant to be called
trimloginname (which does exist, at least)

make krb5 compile again. XXX: one function left that is deprecated, what's
the new equivalent?

branches: 1.20.6; 1.20.8;
PR/42540: Ed Ravin: /usr/bin/login does not log normal logins, and does not
log ip addresses.
- Factor out the common code in login.c and login_pam.c into common.c
- Always log a login event
- Check passed in sockaddr against the one from getpeername(2).

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

branches: 1.18.12;
PR/36294: De Zeurkous: login(1) -- inconsistent exit behavior on EOF
- Always EXIT_FAILURE when login fails.
- Cleanup fork code in krb5 (old login).

GC old signal handling and merge it in 4 places:
1. Immediately ignore all the signals we want to ignore and set the alarm.
2. Before we exit on error, restore all signals we modified
3. Before we exec a shell, restore all signals we modified and ignore TSTP.
Job control aware shells know how to deal with this.
4. Temporarily handle SIGINT while we read motd.

PR/33128: Martin Husemann: When login in on the console, SIGABRT is ignored
Both SIGINT and SIGABRT should be restored before executing the shell.

Fix Coverity run 5, issue 1931 -- memory leak. This was noted as being a
bug by cgilbert.
Approved by christos@.

PR/23616: Christian Biere: login might not back-off as expected

PR/30923: Zafer Aydogan: pam too verbose

Don't display a different message for root logins on insecure terminals.

Approved by christos@.

Propagate change from login.c where if getpeername succeeded, then we have
the sockaddr from the remote peer.

PR/32870: Johan Veenhuizen: login(1) does not obey .hushlogin

Update documentation for pam, and don't accept flags that we don't support
anymore.

use getgrnam_r; from Jon Nemeth

Terminate variable argument list with NULL, not 0.

Safety boots: don't depend upon getpwnam_r() to set pwd to NULL on all
failures, especially as we're ignoring the return result.

Don't report that we don't know the username. We don't want to give out
more information that we need to.

- use getpwnam_r
- KNF

branches: 1.4.2;
1. Restore missing tty action.
2. Pass the nested user name to pam.
3. always try to fill in sockaddr.

- deal with ptyfs pty names.
- put back chown pty code.
- use setproctitle(NULL) to kill our argument display.
- don't wait for any pid, just the one we forked.

Add setting of PAM_SOCKADDR.

Add PAM support to login(8)

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

Patches provided by Joel Baker in PR 22365, verified by myself.

* add functionality to `force password change at next login'. to use,
set the pw_change field of the user to -1 (defined in <pwd.h> as
_PASSWORD_CHGNOW). based on [bin/936] by Simon Gerraty <sjg@quick.com.au>
* clean up for WARNS?=1

Merged with 4.4lite.
Changed to conform to NetBSD's new RCS Id convention.

Add support for Kerberos5 authentication.

Add RCS identifiers.

branches: 1.1.1;
Initial revision