freebsd-dev/sys/dev/random/random_adaptors.c

484 lines
13 KiB
C
Raw Normal View History

/*-
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
* Copyright (c) 2013 Mark R V Murray
* Copyright (c) 2013 Arthur Mesh <arthurmesh@gmail.com>
* Copyright (c) 2013 David E. O'Brien <obrien@NUXI.org>
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer
* in this position and unchanged.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
* IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#include <sys/param.h>
__FBSDID("$FreeBSD$");
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
#include "opt_random.h"
#include <sys/systm.h>
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
#include <sys/conf.h>
#include <sys/fcntl.h>
#include <sys/kernel.h>
2013-08-24 13:54:56 +00:00
#include <sys/kthread.h>
#include <sys/libkern.h>
#include <sys/lock.h>
#include <sys/malloc.h>
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
#include <sys/module.h>
#include <sys/mutex.h>
#include <sys/poll.h>
#include <sys/queue.h>
2013-08-24 13:54:56 +00:00
#include <sys/random.h>
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
#include <sys/sbuf.h>
#include <sys/selinfo.h>
#include <sys/sx.h>
#include <sys/sysctl.h>
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
#include <sys/uio.h>
2013-08-24 13:54:56 +00:00
#include <sys/unistd.h>
#include <dev/random/randomdev.h>
2013-08-24 13:54:56 +00:00
#include <dev/random/random_adaptors.h>
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
#include <dev/random/live_entropy_sources.h>
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
/* The random_adaptors_lock protects random_adaptors_list and friends and random_adaptor.
* We need a sleepable lock for uiomove/block/poll/sbuf/sysctl.
*/
static struct sx random_adaptors_lock;
LIST_HEAD(adaptors_head, random_adaptors);
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
static struct adaptors_head random_adaptors_list = LIST_HEAD_INITIALIZER(random_adaptors_list);
static struct random_adaptor *random_adaptor = NULL; /* Currently active adaptor */
/* End of data items requiring random_adaptors_lock protection */
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
/* The random_readrate_mtx mutex protects the read-rate estimator.
*/
static struct mtx random_read_rate_mtx;
static int random_adaptor_read_rate_cache;
/* End of data items requiring random_readrate_mtx mutex protection */
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
static struct selinfo rsel;
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
/* Utility routine to change active adaptor when the random_adaptors_list
* gets modified.
*
* Walk a list of registered random(4) adaptors and pick either a requested
* one or the highest priority one, whichever comes first. Panic on failure
* as the fallback must always be the "dummy" adaptor.
*/
static void
random_adaptor_choose(void)
{
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
char rngs[128], *token, *cp;
struct random_adaptors *rra, *rrai;
struct random_adaptor *random_adaptor_previous;
int primax;
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
/* We are going to be messing with random_adaptor.
* Exclusive lock is mandatory.
*/
sx_assert(&random_adaptors_lock, SA_XLOCKED);
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
random_adaptor_previous = random_adaptor;
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
random_adaptor = NULL;
if (TUNABLE_STR_FETCH("kern.random.active_adaptor", rngs, sizeof(rngs))) {
cp = rngs;
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
/* XXX: FIX!! (DES):
* - fetch tunable once, at boot
* - make sysctl r/w
* - when fetching tunable or processing a sysctl
* write, parse into list of strings so we don't
* have to do it here again and again
* - sysctl read should return a reconstructed string
*/
while ((token = strsep(&cp, ",")) != NULL) {
LIST_FOREACH(rra, &random_adaptors_list, rra_entries)
if (strcmp(rra->rra_name, token) == 0) {
random_adaptor = rra->rra_ra;
break;
}
if (random_adaptor != NULL) {
printf("random: selecting requested adaptor <%s>\n",
random_adaptor->ra_ident);
break;
}
else
printf("random: requested adaptor <%s> not available\n",
token);
}
}
primax = 0;
if (random_adaptor == NULL) {
/*
* Fall back to the highest priority item on the available
* RNG list.
*/
LIST_FOREACH(rrai, &random_adaptors_list, rra_entries) {
if (rrai->rra_ra->ra_priority >= primax) {
random_adaptor = rrai->rra_ra;
primax = rrai->rra_ra->ra_priority;
}
}
if (random_adaptor != NULL)
printf("random: selecting highest priority adaptor <%s>\n",
random_adaptor->ra_ident);
}
KASSERT(random_adaptor != NULL, ("adaptor not found"));
/* If we are changing adaptors, deinit the old and init the new. */
if (random_adaptor != random_adaptor_previous) {
#ifdef RANDOM_DEBUG
printf("random: %s - changing from %s to %s\n", __func__,
(random_adaptor_previous == NULL ? "NULL" : random_adaptor_previous->ra_ident),
random_adaptor->ra_ident);
#endif
if (random_adaptor_previous != NULL) {
randomdev_deinit_reader();
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
(random_adaptor_previous->ra_deinit)();
}
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
(random_adaptor->ra_init)();
}
randomdev_init_reader(random_adaptor->ra_read);
}
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
/* XXX: FIX!! Make sure we are not inserting a duplicate */
void
random_adaptor_register(const char *name, struct random_adaptor *ra)
{
struct random_adaptors *rra;
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
KASSERT(name != NULL && ra != NULL, ("invalid input to %s", __func__));
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
rra = malloc(sizeof(*rra), M_ENTROPY, M_WAITOK);
rra->rra_name = name;
rra->rra_ra = ra;
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
sx_xlock(&random_adaptors_lock);
LIST_INSERT_HEAD(&random_adaptors_list, rra, rra_entries);
random_adaptor_choose();
KASSERT(random_adaptor != NULL, ("No active random adaptor in %s", __func__));
sx_xunlock(&random_adaptors_lock);
}
2013-08-24 13:54:56 +00:00
void
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
random_adaptor_deregister(const char *name)
2013-08-24 13:54:56 +00:00
{
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
struct random_adaptors *rra;
2013-08-24 13:54:56 +00:00
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
KASSERT(name != NULL, ("invalid input to %s", __func__));
2013-08-24 13:54:56 +00:00
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
sx_xlock(&random_adaptors_lock);
KASSERT(random_adaptor != NULL, ("No active random adaptor in %s", __func__));
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
LIST_FOREACH(rra, &random_adaptors_list, rra_entries)
if (strcmp(rra->rra_name, name) == 0) {
LIST_REMOVE(rra, rra_entries);
break;
}
random_adaptor_choose();
sx_xunlock(&random_adaptors_lock);
2013-08-24 13:54:56 +00:00
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
free(rra, M_ENTROPY);
}
/* ARGSUSED */
int
random_adaptor_read(struct cdev *dev __unused, struct uio *uio, int flags)
{
void *random_buf;
int c, error;
ssize_t nbytes;
#ifdef RANDOM_DEBUG_VERBOSE
printf("random: %s %ld\n", __func__, uio->uio_resid);
#endif
random_buf = malloc(PAGE_SIZE, M_ENTROPY, M_WAITOK);
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
sx_slock(&random_adaptors_lock);
KASSERT(random_adaptor != NULL, ("No active random adaptor in %s", __func__));
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
/* Let the entropy source do any pre-read setup. */
(random_adaptor->ra_read)(NULL, 0);
/* (Un)Blocking logic */
error = 0;
while (!random_adaptor->ra_seeded() && error == 0) {
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
if (flags & O_NONBLOCK) {
error = EWOULDBLOCK;
break;
}
/* Sleep instead of going into a spin-frenzy */
error = sx_sleep(&random_adaptor, &random_adaptors_lock,
PUSER | PCATCH, "randrd", hz/10);
KASSERT(random_adaptor != NULL, ("No active random adaptor in %s",
__func__));
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
/* keep tapping away at the pre-read until we seed/unblock. */
(random_adaptor->ra_read)(NULL, 0);
2013-08-24 13:54:56 +00:00
}
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
mtx_lock(&random_read_rate_mtx);
2013-08-24 13:54:56 +00:00
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
/* The read-rate stuff is a rough indication of the instantaneous read rate,
* used to increase the use of 'live' entropy sources when lots of reads are done.
*/
nbytes = (uio->uio_resid + 32 - 1)/32; /* Round up to units of 32 */
random_adaptor_read_rate_cache += nbytes*32;
random_adaptor_read_rate_cache = MIN(random_adaptor_read_rate_cache, 32);
mtx_unlock(&random_read_rate_mtx);
if (error == 0) {
nbytes = uio->uio_resid;
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
/* The actual read */
while (uio->uio_resid && !error) {
c = MIN(uio->uio_resid, PAGE_SIZE);
(random_adaptor->ra_read)(random_buf, c);
error = uiomove(random_buf, c, uio);
}
2013-08-24 13:54:56 +00:00
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
/* Let the entropy source do any post-read cleanup. */
(random_adaptor->ra_read)(NULL, 1);
2013-08-24 13:54:56 +00:00
if (nbytes != uio->uio_resid && (error == ERESTART ||
error == EINTR) )
error = 0; /* Return partial read, not error. */
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
}
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
sx_sunlock(&random_adaptors_lock);
free(random_buf, M_ENTROPY);
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
return (error);
2013-08-24 13:54:56 +00:00
}
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
int
random_adaptor_read_rate(void)
{
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
int ret;
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
mtx_lock(&random_read_rate_mtx);
ret = random_adaptor_read_rate_cache;
random_adaptor_read_rate_cache = 1;
mtx_unlock(&random_read_rate_mtx);
return (ret);
}
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
/* ARGSUSED */
int
random_adaptor_write(struct cdev *dev __unused, struct uio *uio, int flags __unused)
{
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
int c, error = 0;
void *random_buf;
ssize_t nbytes;
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
#ifdef RANDOM_DEBUG
printf("random: %s %zd\n", __func__, uio->uio_resid);
#endif
random_buf = malloc(PAGE_SIZE, M_ENTROPY, M_WAITOK);
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
sx_slock(&random_adaptors_lock);
2013-08-24 13:54:56 +00:00
KASSERT(random_adaptor != NULL, ("No active random adaptor in %s", __func__));
2013-08-24 13:54:56 +00:00
nbytes = uio->uio_resid;
while (uio->uio_resid > 0 && error == 0) {
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
c = MIN(uio->uio_resid, PAGE_SIZE);
error = uiomove(random_buf, c, uio);
if (error)
break;
(random_adaptor->ra_write)(random_buf, c);
/* Introduce an annoying delay to stop swamping */
error = sx_sleep(&random_adaptor, &random_adaptors_lock,
PUSER | PCATCH, "randwr", hz/10);
KASSERT(random_adaptor != NULL, ("No active random adaptor in %s",
__func__));
}
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
sx_sunlock(&random_adaptors_lock);
if (nbytes != uio->uio_resid && (error == ERESTART ||
error == EINTR) )
error = 0; /* Partial write, not error. */
free(random_buf, M_ENTROPY);
return (error);
}
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
/* ARGSUSED */
int
random_adaptor_poll(struct cdev *dev __unused, int events, struct thread *td __unused)
2013-08-24 13:54:56 +00:00
{
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
#ifdef RANDOM_DEBUG
printf("random: %s\n", __func__);
#endif
2013-08-24 13:54:56 +00:00
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
sx_slock(&random_adaptors_lock);
2013-08-24 13:54:56 +00:00
KASSERT(random_adaptor != NULL, ("No active random adaptor in %s", __func__));
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
if (events & (POLLIN | POLLRDNORM)) {
if (random_adaptor->ra_seeded())
events &= (POLLIN | POLLRDNORM);
else
selrecord(td, &rsel);
2013-08-24 13:54:56 +00:00
}
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
sx_sunlock(&random_adaptors_lock);
return (events);
}
/* This will be called by the entropy processor when it seeds itself and becomes secure */
void
random_adaptor_unblock(void)
{
selwakeuppri(&rsel, PUSER);
wakeup(&random_adaptor);
printf("random: unblocking device.\n");
/* Do arc4random(9) a favour while we are about it. */
(void)atomic_cmpset_int(&arc4rand_iniseed_state, ARC4_ENTR_NONE, ARC4_ENTR_HAVE);
}
static int
random_sysctl_adaptors_handler(SYSCTL_HANDLER_ARGS)
{
struct random_adaptors *rra;
struct sbuf sbuf;
int error, count;
sx_slock(&random_adaptors_lock);
sbuf_new_for_sysctl(&sbuf, NULL, 64, req);
count = 0;
LIST_FOREACH(rra, &random_adaptors_list, rra_entries)
sbuf_printf(&sbuf, "%s%s(%d)",
(count++ ? "," : ""), rra->rra_name, rra->rra_ra->ra_priority);
error = sbuf_finish(&sbuf);
sbuf_delete(&sbuf);
sx_sunlock(&random_adaptors_lock);
2013-08-24 13:54:56 +00:00
return (error);
}
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
static int
random_sysctl_active_adaptor_handler(SYSCTL_HANDLER_ARGS)
{
struct random_adaptors *rra;
struct sbuf sbuf;
int error;
sx_slock(&random_adaptors_lock);
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
KASSERT(random_adaptor != NULL, ("No active random adaptor in %s", __func__));
sbuf_new_for_sysctl(&sbuf, NULL, 16, req);
LIST_FOREACH(rra, &random_adaptors_list, rra_entries)
if (rra->rra_ra == random_adaptor) {
sbuf_cat(&sbuf, rra->rra_name);
break;
}
error = sbuf_finish(&sbuf);
sbuf_delete(&sbuf);
sx_sunlock(&random_adaptors_lock);
return (error);
}
void
random_adaptors_init(void)
{
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
#ifdef RANDOM_DEBUG
printf("random: %s\n", __func__);
#endif
SYSCTL_PROC(_kern_random, OID_AUTO, adaptors,
CTLTYPE_STRING | CTLFLAG_RD | CTLFLAG_MPSAFE,
NULL, 0, random_sysctl_adaptors_handler, "A",
"Random Number Generator adaptors");
2013-08-24 13:54:56 +00:00
SYSCTL_PROC(_kern_random, OID_AUTO, active_adaptor,
CTLTYPE_STRING | CTLFLAG_RD | CTLFLAG_MPSAFE,
NULL, 0, random_sysctl_active_adaptor_handler, "A",
2013-08-24 13:54:56 +00:00
"Active Random Number Generator Adaptor");
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
sx_init(&random_adaptors_lock, "random_adaptors");
mtx_init(&random_read_rate_mtx, "read rate mutex", NULL, MTX_DEF);
/* The dummy adaptor is not a module by itself, but part of the
* randomdev module.
*/
random_adaptor_register("dummy", &randomdev_dummy);
live_entropy_sources_init();
}
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
void
random_adaptors_deinit(void)
{
#ifdef RANDOM_DEBUG
printf("random: %s\n", __func__);
#endif
live_entropy_sources_deinit();
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
/* Don't do this! Panic will surely follow! */
/* random_adaptor_deregister("dummy"); */
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
mtx_destroy(&random_read_rate_mtx);
sx_destroy(&random_adaptors_lock);
}
/*
* Reseed the active adaptor shortly before starting init(8).
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
*/
/* ARGSUSED */
static void
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
random_adaptors_seed(void *unused __unused)
{
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
sx_slock(&random_adaptors_lock);
This is the much-discussed major upgrade to the random(4) device, known to you all as /dev/random. This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources. The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people. The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway. Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to. My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise. My Nomex pants are on. Let the feedback commence! Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?) Approved by: so(des)
2014-10-30 21:21:53 +00:00
KASSERT(random_adaptor != NULL, ("No active random adaptor in %s", __func__));
random_adaptor->ra_reseed();
sx_sunlock(&random_adaptors_lock);
arc4rand(NULL, 0, 1);
}
SYSINIT(random_seed, SI_SUB_KTHREAD_INIT, SI_ORDER_FIRST,
random_adaptors_seed, NULL);