openssl: Stop dragging libpthread into every libcrypto user.

With the new _NETBSD_PTHREAD_CREATE_WEAK, libcrypto can have a call
to pthread_create without linking against libpthread, so it is not
necessary to link all libcrypto users against libpthread too.

Applications that link against libcrypto but not libpthread can't use
threaded OpenSSL functionality, of course -- OSSL_set_max_threads
will always fail in such applications.

This reverts all the makefile churn that I found for unnecessary
libpthread linkage since the openssl 3.5 import.

PR lib/59685: libcrypto should not depend on libpthread

branches: 1.16.2;
crypto needs threads

branches: 1.15.2;
Handle OpenSSL-3.x

branches: 1.14.6;
include bsd.own.mk to be sure NETBSDSRCDIR is set.

Remove MKCRYPTO option.

Originally, MKCRYPTO was introduced because the United States
classified cryptography as a munition and restricted its export. The
export controls were substantially relaxed fifteen years ago, and are
essentially irrelevant for software with published source code.

In the intervening time, nobody bothered to remove the option after
its motivation -- the US export restriction -- was eliminated. I'm
not aware of any other operating system that has a similar option; I
expect it is mainly out of apathy for churn that we still have it.
Today, cryptography is an essential part of modern computing -- you
can't use the internet responsibly without cryptography.

The position of the TNF board of directors is that TNF makes no
representation that MKCRYPTO=no satisfies any country's cryptography
regulations.

My personal position is that the availability of cryptography is a
basic human right; that any local laws restricting it to a privileged
few are fundamentally immoral; and that it is wrong for developers to
spend effort crippling cryptography to work around such laws.

As proposed on tech-crypto, tech-security, and tech-userlevel to no
objections:

https://mail-index.netbsd.org/tech-crypto/2017/05/06/msg000719.html
https://mail-index.netbsd.org/tech-security/2017/05/06/msg000928.html
https://mail-index.netbsd.org/tech-userlevel/2017/05/06/msg010547.html

P.S. Reviewing all the uses of MKCRYPTO in src revealed a lot of
*bad* crypto that was conditional on it, e.g. DES in telnet... That
should probably be removed too, but on the grounds that it is bad,
not on the grounds that it is (nominally) crypto.

Add -lcrypt where -lcrypto is specified.

Separate primes manpage from that of factor. Correct DIAGNOSTICS
information.

use NETBSDSRCDIR as appropriate

make factor work with and without openssl.

Use libcrypto's bignum support to implement a Pollard Rho factoring
algorithm so we can factorise numbers larger than a host long.

use CPPFLAGS instead of CFLAGS

branches: 1.6.2;
Back out the .PATH.c changes. The .depend problem (and others)
will be fixed using the new .NOPATH make feature instead.

Use .PATH.c: ...

merge with Lite, new RCS id conventions, etc.

specify man pages the new way.

Add RCS identifiers.

branches: 1.1.1;
Initial revision

more clarifications.

mention hex numbers

Fix a typo (s/int/in/)

PR/55693: Andreas Gustafsson: factor(6) lists factors in wrong order
Sync with FreeBSD and change their -h (that printed hex) to -x because
we were already using -h.

Add -h option to factor(6): duplicate factors are printed in
"human-readable" form of x^n.

Imported and adapted from FreeBSD svn r272166 and r272207; this fixes
false positives for products of primes larger than 2^16. For example,
before this commit:

$ /usr/games/primes 4295360521 4295360522
4295360521
but
$ /usr/games/factor 4295360521
4295360521: 65539 65539

or
$ /usr/games/primes 3825123056546413049 3825123056546413050
3825123056546413049
yet
$ /usr/games/factor 3825123056546413049
3825123056546413049: 165479 23115459100831

or
$ /usr/games/primes 18446744073709551577
18446744073709551577
although
$ /usr/games/factor 18446744073709551577
18446744073709551577: 139646831 132095686967

Incidentally, the above examples show the smallest and largest cases that
were erroneously stated as prime in the range 2^32 .. 3825123056546413049
.. 2^64; the primes(6) program now stops at 3825123056546413050 as
primality tests on larger integers would be by brute force factorization.

In addition, special to the NetBSD version:
. for -d option, skip first difference when start is >65537 as it is incorrect
. corrected usage to mention both the existing -d as well as the new -h option

For original FreeBSD commit message by Colin Percival, see:
http://svnweb.freebsd.org/base?view=revision&revision=272166

branches: 1.12.24;
Follow the Fundamental Theory of Algebra. Disallow factorising of
numbers less than 2 as it is not
- naturally unique (negative numbers)
- finite (0)
- non-empty (1)

Discussed with the kristaps and wiz

Add backslash before a dash to get a minus.

Sync with reality. Bump date.

Separate primes manpage from that of factor. Correct DIAGNOSTICS
information.

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

Patches provided by Joel Baker in PR 22269, verified by myself.

Lots of minor fixes resulting from reading these man pages in detail.

Whitespace and/or punctuation fixes.

WARNSify

merge with Lite, new RCS id conventions, etc.

Update to use new -mandoc macros.

Add RCS indentifiers.

branches: 1.1.1;
Initial revision

- remove duplicate comment
- flush after printing the number
(from kre@)

Remove is_hex_str() (trying to guess if a number was hex or not). It is not
documented and can lead to unexpected behavior.

From gson@
- don't assume -h is always on for large factors
- fix saved large factors printing when -h

- for readability when the exponent < 10 don't prefix it with 0x (from kre)
- fix usage
- merge printing code.

PR/55695: Andreas Gustafsson: factor(6) -h option doesn't always work
Handle -h for factors greater than the primes table.

revert previous and don't parse octal

factor: usage(): mark __dead

- Accept octal input.
- Don't play with the original string so we can print it.

PR/55693: Andreas Gustafsson: factor(6) lists factors in wrong order
Sync with FreeBSD and change their -h (that printed hex) to -x because
we were already using -h.

fix for OpenSSL-1.1

Add -h option to factor(6): duplicate factors are printed in
"human-readable" form of x^n.

Imported and adapted from FreeBSD svn r272166 and r272207; this fixes
false positives for products of primes larger than 2^16. For example,
before this commit:

$ /usr/games/primes 4295360521 4295360522
4295360521
but
$ /usr/games/factor 4295360521
4295360521: 65539 65539

or
$ /usr/games/primes 3825123056546413049 3825123056546413050
3825123056546413049
yet
$ /usr/games/factor 3825123056546413049
3825123056546413049: 165479 23115459100831

or
$ /usr/games/primes 18446744073709551577
18446744073709551577
although
$ /usr/games/factor 18446744073709551577
18446744073709551577: 139646831 132095686967

Incidentally, the above examples show the smallest and largest cases that
were erroneously stated as prime in the range 2^32 .. 3825123056546413049
.. 2^64; the primes(6) program now stops at 3825123056546413050 as
primality tests on larger integers would be by brute force factorization.

In addition, special to the NetBSD version:
. for -d option, skip first difference when start is >65537 as it is incorrect
. corrected usage to mention both the existing -d as well as the new -h option

For original FreeBSD commit message by Colin Percival, see:
http://svnweb.freebsd.org/base?view=revision&revision=272166

branches: 1.26.18;
remove duplicated #defines (in a usually unused part of the code)

branches: 1.25.4;
Properly print string.

Follow the Fundamental Theory of Algebra. Disallow factorising of
numbers less than 2 as it is not
- naturally unique (negative numbers)
- finite (0)
- non-empty (1)

Discussed with the kristaps and wiz

cast isblank(3)'s argument to unsigned char.

rename pollard_pminus1->pollard_rho for consistency

-Fix an old bug in the "pollard" code: it gets its argument passed
by reference, and changes the value behind the pointer under some
circumstances (basically if it finds more than 2 different factors).
It also calls itself if it finds a factor which is not considered prime
(by openssl's miller-rabin check) and uses the call argument afterwards.
This doesn't work -- we need to copy the argument into its own storage.
-Modify the code to do the "rho" algorithm as was initially announced.
It takes somewhat longer in rare cases, but still works in cases where
the "p-1" algorithm is unusable. This might fix PR misc/43192
by Luiz Henrique de Figueiredo.
-Add some optional debug support, minor cleanup.

fix an obvious flaw in bounds check: the array of precomputed primes
could be overrun if its last entry (65537) was a factor of the input
(this does not affect PR misc/43192 -- the factors are much larger
here: 7742394596501*159455563099482401)

sprinkle static

Remove the \n and tabs from the __COPYRIGHT() strings.

branches: 1.17.6;
convert __attribute__s to applicable cdefs.h macros

branches: 1.16.10;
Fixed a comment that said the factors in the output would be strictly
ascending.

Check large factor for being prime before applying Pollard's
algorithm; fixes "factor 2147483647111311". Correct comment;
algorithm is Pollard p-1, not Pollard rho. Increase base if p-1
algorithm reaches 1; fixes "factor 99999999999991". Testcases from
David A Bagley <bagleyd@tux.org>.

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

Patches provided by Joel Baker in PR 22269, verified by myself.

Provide a BN_dec2bn() shim for the non-openssl case that reports an error
if strtoul() fails.

Fix a logic botch where if a number smaller than the square of the seive
was prime to still called the Pollard Rho function when it didn't have
to. Problem report by Nathan Williams.

Unfortunately this one can't be picked up by a simple regression test
since the broken way still produced the correct output, but just took
far longer...

make factor work with and without openssl.

Use libcrypto's bignum support to implement a Pollard Rho factoring
algorithm so we can factorise numbers larger than a host long.

Add use of `const' where appropriate to the games.

This merges in all such remaining changes from the Linux port of the
NetBSD games, except in hunt (where substantial changes from OpenBSD
need to be looked at).

Some such changes were previously covered in PRs bin/6041, bin/6146,
bin/6148, bin/6150, bin/6151, bin/6580, bin/6660, bin/7993, bin/7994,
bin/8039, bin/8057 and bin/8093.

mark non-returning functions (PR#6144 by Joseph Myers <jsm28@cam.ac.uk>)

WARNSify

Sync to 4.4BSD-Lite2

merge with Lite, new RCS id conventions, etc.

bring in limits.h

Eliminate a compiler warning.

Add RCS identifiers.

branches: 1.1.1;
Initial revision