RND(4) | Device Drivers Manual | RND(4) |
rnd
—
Applications should read from
/dev/urandom, or the
sysctl(7) variable
kern.arandom
, when they need randomly generated
data, e.g. key material for cryptography or seeds for simulations. (The
sysctl(7) variable
kern.arandom
is limited to 256 bytes per read, but
is otherwise equivalent to reading from /dev/urandom
and always works even in a
chroot(8) environment without
requiring a populated /dev tree and without opening
a file descriptor, so kern.arandom
may be preferable
to use in libraries.)
Systems should be engineered to judiciously read at least once from /dev/random at boot before running any services that talk to the internet or otherwise require cryptography, in order to avoid generating keys predictably. /dev/random may block at any time, so programs that read from it must be prepared to handle blocking. Interactive programs that block due to reads from /dev/random can be especially frustrating.
If interrupted by a signal, reads from either /dev/random or /dev/urandom may return short, so programs that handle signals must be prepared to retry reads.
Writing to either /dev/random or /dev/urandom influences subsequent output of both devices, guaranteed to take effect at next open. If you have a coin in your pocket, you can flip it 256 times and feed the outputs to /dev/random to guarantee your system is in a state that nobody but you and the bored security guard watching the surveillance camera in your office can guess:
% echo tthhhhhthhhththtthhhhthtththttth... > /dev/random
(Sequence generated from a genuine US quarter dollar, guaranteed random.)
rnd
subsystem provides the following security
properties against two different classes of attackers, provided that there is
enough entropy from entropy sources not seen by attackers:
One ‘output’ means a single read, no matter how short it is.
‘Cannot predict’ means it is conjectured of the cryptography in /dev/random that any computationally bounded attacker who tries to distinguish outputs from uniform random cannot do more than negligibly better than uniform random guessing.
An attacker may be able to guess with nonnegligible chance of success what your last keystroke was, but guessing every observation the operating system may have made is more difficult. The difficulty of the best strategy at guessing a random variable is analyzed as the -log_2 of the highest probability of any outcome, measured in bits, and called its min-entropy, or entropy for short in cryptography. For example:
Note that entropy is a property of an observable physical process, like a coin toss, or of a state of knowledge about that physical process; it is not a property of a specific sample obtained by observing it, like the string ‘tthhhhht’. There are also kinds of entropy in information theory other than min-entropy, including the more well-known Shannon entropy, but they are not relevant here.
Hardware devices that the operating system monitors for observations are called entropy sources, and the observations are combined into an entropy pool. The rndctl(8) command queries information about entropy sources and the entropy pool, and can control which entropy sources the operating system uses or ignores.
256 bits of entropy is typically considered intractable to guess with classical computers and with current models of the capabilities of quantum computers.
Systems with nonvolatile storage should store a secret from /dev/urandom on disk during installation or shutdown, and feed it back during boot, so that the work the operating system has done to gather entropy — including the work its operator may have done to flip a coin! — can be saved from one boot to the next, and so that newly installed systems are not vulnerable to generating cryptographic keys predictably.
The boot loaders in some NetBSD ports
support a command to load a seed from disk before the kernel has started.
For those that don't, the
rndctl(8) command can do it
once userland has started, for example by setting
“random_seed=YES
” in
/etc/rc.conf, which is enabled by default; see
rc.conf(5).
But if an attacker has seen the entire state of your machine, refreshing entropy is probably the least of your worries, so we do not address that threat model here.
The rnd
subsystem does
not automatically defend against hardware colluding with
an attacker to influence entropy sources based on the state of the operating
system.
For example, a PCI device or CPU instruction for random number generation which has no side channel to an attacker other than the /dev/urandom device could be bugged to observe all other entropy sources, and to carefully craft ‘observations’ that cause a certain number of bits of /dev/urandom output to be ciphertext that either is predictable to an attacker or conveys a message to an attacker.
No amount of scrutiny by the system's operator could detect this. The only way to prevent this attack would be for the operator to disable all entropy sources that may be colluding with an attacker. If you're not sure which ones are not, you can always disable all of them and fall back to the coin in your pocket.
<sys/rndio.h>
header file, for
querying and controlling the entropy pool.
Since timing between hardware events contributes to the entropy pool, statistics about the entropy pool over time may serve as a side channel for the state of the pool, so access to such statistics is restricted to the super-user and should be used with caution.
Several ioctls are concerned with particular entropy sources, described by the following structure:
typedef struct { char name[16]; /* symbolic name */ uint32_t total; /* estimate of entropy provided */ uint32_t type; /* RND_TYPE_* value */ uint32_t flags; /* RND_FLAG_* mask */ } rndsource_t; #define RND_TYPE_UNKNOWN #define RND_TYPE_DISK /* disk device */ #define RND_TYPE_ENV /* environment sensor (temp, fan, &c.) */ #define RND_TYPE_NET /* network device */ #define RND_TYPE_POWER /* power events */ #define RND_TYPE_RNG /* hardware RNG */ #define RND_TYPE_SKEW /* clock skew */ #define RND_TYPE_TAPE /* tape drive */ #define RND_TYPE_TTY /* tty device */ #define RND_TYPE_VM /* virtual memory faults */ #define RND_TYPE_MAX /* value of highest-numbered type */ #define RND_FLAG_COLLECT_TIME /* use timings of samples */ #define RND_FLAG_COLLECT_VALUE /* use values of samples */ #define RND_FLAG_ESTIMATE_TIME /* estimate entropy of timings */ #define RND_FLAG_ESTIMATE_VALUE /* estimate entropy of values */ #define RND_FLAG_NO_COLLECT /* ignore samples from this */ #define RND_FLAG_NO_ESTIMATE /* do not estimate entropy */
The following ioctls are supported:
RNDGETENTCNT
(uint32_t)RNDGETSRCNUM
(rndstat_t)typedef struct { uint32_t start; uint32_t count; rndsource_t source[RND_MAXSTATCOUNT]; } rndstat_t;
Fill the sources array with information
about up to count entropy sources, starting at
start. The actual number of sources described is
returned in count. At most
RND_MAXSTATCOUNT
sources may be requested at
once.
RNDGETSRCNAME
(rndstat_name_t)typedef struct { char name[16]; rndsource_t source; } rndstat_name_t;
Fill source with information about the
entropy source named name, or fail with
ENOENT
if there is none.
RNDCTL
(rndctl_t)typedef struct { char name[16]; uint32_t type; uint32_t flags; uint32_t mask; } rndctl_t;
For each entropy source of the type
type, or if type is
0xff
then for the entropy source named
name, replace the flags in
mask by flags.
RNDADDDATA
(rnddata_t)typedef struct { uint32_t len; uint32_t entropy; unsigned char data[RND_SAVEWORDS * sizeof(uint32_t)]; } rnddata_t;
Feed len bytes of data to the entropy pool. The sample is expected to have been drawn with at least entropy bits of entropy.
This ioctl can be used only once per boot. It is intended for a system that saves entropy to disk on shutdown and restores it on boot, so that the system can immediately be unpredictable without having to wait to gather entropy.
RNDGETPOOLSTAT
(rndpoolstat_t)typedef struct { uint32_t poolsize; /* size of each LFSR in pool */ uint32_t threshold; /* no. bytes of pool hash returned */ uint32_t maxentropy; /* total size of pool in bits */ uint32_t added; /* no. bits of entropy ever added */ uint32_t curentropy; /* current entropy `balance' */ uint32_t discarded; /* no. bits dropped when pool full */ uint32_t generated; /* no. bits yielded by pool while curentropy is zero */ } rndpoolstat_t;
Return various statistics about entropy.
rnd
can be set by privileged users:
kern.entropy.collection
(bool)RNDADDDATA
ioctl.kern.entropy.depletion
(bool)kern.entropy.consolidate
(int)# sysctl -w
kern.entropy.consolidate=1
The following read-only sysctl(8) variables provide information to privileged users about the state of the entropy pool:
kern.entropy.needed
(unsigned int)kern.entropy.pending
(unsigned int)kern.entropy.consolidate
.kern.entropy.epoch
(unsigned int)kern.entropy.consolidate
, as an
unsigned 32-bit integer. Consulted inside the kernel by subsystems such as
cprng(9) to decide whether to
reseed. Initially set to 2^32 - 1 (i.e.,
(unsigned)-1
) meaning the system has never reached
full entropy and the entropy has never been consolidated; never again set
to 2^32 - 1. Never zero, so applications can initialize a cache of the
epoch to zero to ensure they reseed the next time they check whether it is
different from the stored epoch.rnd
subsystem at the time of writing. It may be
out-of-date by the time you read it, and nothing in here should be construed
as a guarantee about the behaviour of the /dev/random
and /dev/urandom devices.)
Device drivers gather samples from entropy sources and absorb them
into a collection of per-CPU Keccak sponges called ‘entropy
pools’ using the rnd(9)
kernel API. The device driver furnishes an estimate for the entropy of the
sampling process, under the assumption that each sample is independent. When
the estimate of entropy pending among the per-CPU entropy pools reaches a
threshold of 256 bits, the entropy is drawn from the per-CPU pools and
consolidated into a global pool. Keys for
/dev/random, /dev/urandom,
kern.arandom
, and the in-kernel
cprng(9) subsystem are
extracted from the global pool.
Early after boot, before CPUs have been detected, device drivers instead enter directly into the global pool. If anything in the system extracts data from the pool before the threshold has been reached at least once, the system will print a warning to the console and reset the entropy estimate to zero. The reason for resetting the entropy estimate to zero in this case is that an adversary who can witness output from the pool with partial entropy — say, 32 bits — can undergo a feasible brute force search to ascertain the complete state of the pool; as such, the entropy of the adversary's state of knowledge about the pool is zero.
If the operator is confident that the drivers' estimates of the entropy of the sampling processes are too conservative, the operator can issue
# sysctl -w
kern.entropy.consolidate=1
Short reads from /dev/urandom are served by a persistent per-CPU Hash_DRBG instance that is reseeded from the entropy pool after any entropy consolidation. Reads from /dev/random and long reads from /dev/urandom are served by a temporary Hash_DRBG seeded from the entropy pool on each read.
When ‘entropy depletion’ is enabled by setting the
sysctl variable
kern.entropy.depletion
=1
,
every read from /dev/random is limited to 256 bits,
since reading more than that would nearly always block again.
rnd
subsystem may print the following warnings to
the console likely indicating security issues:
The entropy may be low enough that an adversary who sees the output could guess the state of the pool by brute force, so in this event the system resets its estimate of entropy to none.
This message is rate-limited to happen no more often than once
per minute, so if you want to make sure it is gone you should consult
kern.entropy.needed
to confirm it is zero.
The rnd
subsystem may print any of various
messages about obtaining an entropy seed from the bootloader to diagnose
saving and loading seeds on disk:
random_seed=YES
can serve instead.Elaine Barker and John Kelsey, Recommendation for Random Number Generation Using Deterministic Random Bit Generators, National Institute of Standards and Technology, https://csrc.nist.gov/publications/detail/sp/800-90a/rev-1/final, United States Department of Commerce, June 2015, NIST Special Publication 800-90A, Revision 1.
Meltem Sönmez Turan, Elaine Barker, John Kelsey, Kerry A. McKay, Mary L. Baish, and Mike Boyle, Recommendations for the Entropy Sources Used for Random Bit Generation, National Institute of Standards and Technology, https://csrc.nist.gov/publications/detail/sp/800-90b/final, United States Department of Commerce, January 2018, NIST Special Publication 800-90B.
Daniel J. Bernstein, Entropy Attacks!, http://blog.cr.yp.to/20140205-entropy.html, 2014-02-05.
Nadia Heninger, Zakir Durumeric, Eric Wustrow, and J. Alex Halderman, Mining Your Ps and Qs: Detection of Widespread Weak Keys in Network Devices, Proceedings of the 21st USENIX Security Symposium, USENIX, https://www.usenix.org/conference/usenixsecurity12/technical-sessions/presentation/heninger, https://factorable.net/, 205-220, August 2012.
Edwin T. Jaynes, Probability Theory: The Logic of Science, Cambridge University Press, https://bayes.wustl.edu/, 2003.
rnd
subsystem was first implemented by
Michael Graff
<explorer@flame.org>,
was then largely rewritten by Thor Lancelot Simon
<tls@NetBSD.org>, and was
most recently largely rewritten by Taylor R. Campbell
<riastradh@NetBSD.org>.
August 7, 2023 | NetBSD 10.0 |