openbsd/src/lib/libc/stdlib/random.3, branch OPENBSD_7_9 A mirror of https://github.com/libressl/openbsd.git https://git.lua4.win/openbsd/atom?h=OPENBSD_7_9 2021-02-12T17:03:51+00:00 Some people still argue that rand(3) and random(3) have suitable deterministic 2021-02-12T17:03:51+00:00 deraadt 2021-02-12T17:03:51+00:00 urn:sha1:96694adb26fa7287b876515de64c63e787de5007 use cases, so explain the situation a bit more. Since the 80's, I estimate around 5 algorithm changes, so any chosen seed is unrepeatable UB. +The deterministic sequence algorithm changed a number of times since +original development, is underspecified, and should not be relied upon to +remain consistent between platforms and over time. ok jmc kettenis Xr and Ox fixes; 2014-12-09T21:55:39+00:00 jmc 2014-12-09T21:55:39+00:00 urn:sha1:7b4389f21ce09e7d34ad8e9cedcf015c105d2bb1 Change rand(), random(), drand48(), lrand48(), mrand48(), and srand48() 2014-12-08T21:45:20+00:00 deraadt 2014-12-08T21:45:20+00:00 urn:sha1:9736a7f2a9e739ee9fc431040def5680fbb8794a to returning strong random by default, source from arc4random(3). Parameters to the seeding functions are ignored, and the subsystems remain in strong random mode. If you wish the standardized deterministic mode, call srand_deterministic(), srandom_determistic(), srand48_deterministic(), seed48_deterministic() or lcong48_deterministic() instead. The re-entrant functions rand_r(), erand48(), nrand48(), jrand48() are unaffected by this change and remain in deterministic mode (for now). Verified as a good roadmap forward by auditing 8800 pieces of software. Roughly 60 pieces of software will need adaptation to request the deterministic mode. Violates POSIX and C89, which violate best practice in this century. ok guenther tedu millert Don't describe random() as "better". 2014-11-25T17:40:38+00:00 millert 2014-11-25T17:40:38+00:00 urn:sha1:832147399a6fb1461d52edf96b19c8487ca88e44 Remove the bug about rand() being faster. Add a bug about historical implementations seeding very poorly. zap trailing whitespace; 2014-07-18T06:20:36+00:00 jmc 2014-07-18T06:20:36+00:00 urn:sha1:4f75fccbdbb9d94e0781f2337037b8261eba7f5e it is 2014, and we still need to encourage people away from srand() 2014-07-17T23:12:28+00:00 deraadt 2014-07-17T23:12:28+00:00 urn:sha1:60721d2b6012c2aa79de24c38ab93b24f8a8d6ab and random(). Sigh. oops, i deleted the wrong word 2014-07-13T14:25:00+00:00 tedu 2014-07-13T14:25:00+00:00 urn:sha1:bd5f6d197f463a2a9be53a9d5f8d34cdb7f522c4 remove a lie spotted by guenther 2014-07-13T14:24:23+00:00 tedu 2014-07-13T14:24:23+00:00 urn:sha1:0850040c6d7e12aeb922c3b0f6499d972fbfd4ac Use the fancy .In macro for includes. From Jan Klemkow. ok jmc schwarze 2013-06-05T03:39:23+00:00 tedu 2013-06-05T03:39:23+00:00 urn:sha1:2b7c3ac1a78656e1ad7e45f24ecbff6f63f289e4 Improve standards conformance: ecvt(), fcvt(), gcvt(), mktemp(), ttyslot(), 2012-06-01T01:01:57+00:00 guenther 2012-06-01T01:01:57+00:00 urn:sha1:bda61de216b7a7f245c8e3e25e5d70f4dd73ae5c and valloc() are not in the current version, while posix_memalign() mkstemp(), and mkdtemp() are, and setstate()'s argument has lost a bogus 'const'. ok millert@ jmc@ espie@ kettenis@; ports build testing by naddy@