2000-10-18 15:00:36 +00:00
|
|
|
/*
|
|
|
|
* Prime generation.
|
|
|
|
*/
|
|
|
|
|
2001-09-22 20:52:21 +00:00
|
|
|
#include <assert.h>
|
2020-02-23 15:29:40 +00:00
|
|
|
#include <math.h>
|
|
|
|
|
2000-10-18 15:00:36 +00:00
|
|
|
#include "ssh.h"
|
Complete rewrite of PuTTY's bignum library.
The old 'Bignum' data type is gone completely, and so is sshbn.c. In
its place is a new thing called 'mp_int', handled by an entirely new
library module mpint.c, with API differences both large and small.
The main aim of this change is that the new library should be free of
timing- and cache-related side channels. I've written the code so that
it _should_ - assuming I haven't made any mistakes - do all of its
work without either control flow or memory addressing depending on the
data words of the input numbers. (Though, being an _arbitrary_
precision library, it does have to at least depend on the sizes of the
numbers - but there's a 'formal' size that can vary separately from
the actual magnitude of the represented integer, so if you want to
keep it secret that your number is actually small, it should work fine
to have a very long mp_int and just happen to store 23 in it.) So I've
done all my conditionalisation by means of computing both answers and
doing bit-masking to swap the right one into place, and all loops over
the words of an mp_int go up to the formal size rather than the actual
size.
I haven't actually tested the constant-time property in any rigorous
way yet (I'm still considering the best way to do it). But this code
is surely at the very least a big improvement on the old version, even
if I later find a few more things to fix.
I've also completely rewritten the low-level elliptic curve arithmetic
from sshecc.c; the new ecc.c is closer to being an adjunct of mpint.c
than it is to the SSH end of the code. The new elliptic curve code
keeps all coordinates in Montgomery-multiplication transformed form to
speed up all the multiplications mod the same prime, and only converts
them back when you ask for the affine coordinates. Also, I adopted
extended coordinates for the Edwards curve implementation.
sshecc.c has also had a near-total rewrite in the course of switching
it over to the new system. While I was there, I've separated ECDSA and
EdDSA more completely - they now have separate vtables, instead of a
single vtable in which nearly every function had a big if statement in
it - and also made the externally exposed types for an ECDSA key and
an ECDH context different.
A minor new feature: since the new arithmetic code includes a modular
square root function, we can now support the compressed point
representation for the NIST curves. We seem to have been getting along
fine without that so far, but it seemed a shame not to put it in,
since it was suddenly easy.
In sshrsa.c, one major change is that I've removed the RSA blinding
step in rsa_privkey_op, in which we randomise the ciphertext before
doing the decryption. The purpose of that was to avoid timing leaks
giving away the plaintext - but the new arithmetic code should take
that in its stride in the course of also being careful enough to avoid
leaking the _private key_, which RSA blinding had no way to do
anything about in any case.
Apart from those specific points, most of the rest of the changes are
more or less mechanical, just changing type names and translating code
into the new API.
2018-12-31 13:53:41 +00:00
|
|
|
#include "mpint.h"
|
2020-02-20 18:02:15 +00:00
|
|
|
#include "mpunsafe.h"
|
2020-02-23 14:08:57 +00:00
|
|
|
#include "sshkeygen.h"
|
2000-10-18 15:00:36 +00:00
|
|
|
|
2000-11-15 15:03:17 +00:00
|
|
|
/*
|
|
|
|
* This prime generation algorithm is pretty much cribbed from
|
|
|
|
* OpenSSL. The algorithm is:
|
2019-09-08 19:29:00 +00:00
|
|
|
*
|
2000-11-15 15:03:17 +00:00
|
|
|
* - invent a B-bit random number and ensure the top and bottom
|
|
|
|
* bits are set (so it's definitely B-bit, and it's definitely
|
|
|
|
* odd)
|
2019-09-08 19:29:00 +00:00
|
|
|
*
|
2000-11-15 15:03:17 +00:00
|
|
|
* - see if it's coprime to all primes below 2^16; increment it by
|
|
|
|
* two until it is (this shouldn't take long in general)
|
2019-09-08 19:29:00 +00:00
|
|
|
*
|
2000-11-15 15:03:17 +00:00
|
|
|
* - perform the Miller-Rabin primality test enough times to
|
|
|
|
* ensure the probability of it being composite is 2^-80 or
|
|
|
|
* less
|
2019-09-08 19:29:00 +00:00
|
|
|
*
|
2000-11-15 15:03:17 +00:00
|
|
|
* - go back to square one if any M-R test fails.
|
|
|
|
*/
|
|
|
|
|
2020-02-23 15:29:40 +00:00
|
|
|
ProgressPhase primegen_add_progress_phase(ProgressReceiver *prog,
|
|
|
|
unsigned bits)
|
|
|
|
{
|
|
|
|
/*
|
|
|
|
* The density of primes near x is 1/(log x). When x is about 2^b,
|
|
|
|
* that's 1/(b log 2).
|
|
|
|
*
|
|
|
|
* But we're only doing the expensive part of the process (the M-R
|
|
|
|
* checks) for a number that passes the initial winnowing test of
|
|
|
|
* having no factor less than 2^16 (at least, unless the prime is
|
|
|
|
* so small that PrimeCandidateSource gives up on that winnowing).
|
|
|
|
* The density of _those_ numbers is about 1/19.76. So the odds of
|
|
|
|
* hitting a prime per expensive attempt are boosted by a factor
|
|
|
|
* of 19.76.
|
|
|
|
*/
|
|
|
|
const double log_2 = 0.693147180559945309417232121458;
|
|
|
|
double winnow_factor = (bits < 32 ? 1.0 : 19.76);
|
|
|
|
double prob = winnow_factor / (bits * log_2);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Estimate the cost of prime generation as the cost of the M-R
|
|
|
|
* modexps.
|
|
|
|
*/
|
|
|
|
double cost = (miller_rabin_checks_needed(bits) *
|
|
|
|
estimate_modexp_cost(bits));
|
|
|
|
return progress_add_probabilistic(prog, cost, prob);
|
|
|
|
}
|
|
|
|
|
|
|
|
mp_int *primegen(PrimeCandidateSource *pcs, ProgressReceiver *prog)
|
|
|
|
{
|
|
|
|
pcs_ready(pcs);
|
2020-02-29 16:32:16 +00:00
|
|
|
|
2020-02-29 16:44:56 +00:00
|
|
|
while (true) {
|
|
|
|
progress_report_attempt(prog);
|
2020-02-29 16:32:16 +00:00
|
|
|
|
2020-02-29 16:44:56 +00:00
|
|
|
mp_int *p = pcs_generate(pcs);
|
2020-02-29 16:32:16 +00:00
|
|
|
|
2020-02-29 16:44:56 +00:00
|
|
|
MillerRabin *mr = miller_rabin_new(p);
|
|
|
|
bool known_bad = false;
|
|
|
|
unsigned nchecks = miller_rabin_checks_needed(mp_get_nbits(p));
|
|
|
|
for (unsigned check = 0; check < nchecks; check++) {
|
|
|
|
if (!miller_rabin_test_random(mr)) {
|
|
|
|
known_bad = true;
|
|
|
|
break;
|
2020-02-29 16:32:16 +00:00
|
|
|
}
|
|
|
|
}
|
2020-02-29 16:44:56 +00:00
|
|
|
miller_rabin_free(mr);
|
|
|
|
|
|
|
|
if (!known_bad) {
|
|
|
|
/*
|
|
|
|
* We have a prime!
|
|
|
|
*/
|
|
|
|
pcs_free(pcs);
|
|
|
|
return p;
|
|
|
|
}
|
Complete rewrite of PuTTY's bignum library.
The old 'Bignum' data type is gone completely, and so is sshbn.c. In
its place is a new thing called 'mp_int', handled by an entirely new
library module mpint.c, with API differences both large and small.
The main aim of this change is that the new library should be free of
timing- and cache-related side channels. I've written the code so that
it _should_ - assuming I haven't made any mistakes - do all of its
work without either control flow or memory addressing depending on the
data words of the input numbers. (Though, being an _arbitrary_
precision library, it does have to at least depend on the sizes of the
numbers - but there's a 'formal' size that can vary separately from
the actual magnitude of the represented integer, so if you want to
keep it secret that your number is actually small, it should work fine
to have a very long mp_int and just happen to store 23 in it.) So I've
done all my conditionalisation by means of computing both answers and
doing bit-masking to swap the right one into place, and all loops over
the words of an mp_int go up to the formal size rather than the actual
size.
I haven't actually tested the constant-time property in any rigorous
way yet (I'm still considering the best way to do it). But this code
is surely at the very least a big improvement on the old version, even
if I later find a few more things to fix.
I've also completely rewritten the low-level elliptic curve arithmetic
from sshecc.c; the new ecc.c is closer to being an adjunct of mpint.c
than it is to the SSH end of the code. The new elliptic curve code
keeps all coordinates in Montgomery-multiplication transformed form to
speed up all the multiplications mod the same prime, and only converts
them back when you ask for the affine coordinates. Also, I adopted
extended coordinates for the Edwards curve implementation.
sshecc.c has also had a near-total rewrite in the course of switching
it over to the new system. While I was there, I've separated ECDSA and
EdDSA more completely - they now have separate vtables, instead of a
single vtable in which nearly every function had a big if statement in
it - and also made the externally exposed types for an ECDSA key and
an ECDH context different.
A minor new feature: since the new arithmetic code includes a modular
square root function, we can now support the compressed point
representation for the NIST curves. We seem to have been getting along
fine without that so far, but it seemed a shame not to put it in,
since it was suddenly easy.
In sshrsa.c, one major change is that I've removed the RSA blinding
step in rsa_privkey_op, in which we randomise the ciphertext before
doing the decryption. The purpose of that was to avoid timing leaks
giving away the plaintext - but the new arithmetic code should take
that in its stride in the course of also being careful enough to avoid
leaking the _private key_, which RSA blinding had no way to do
anything about in any case.
Apart from those specific points, most of the rest of the changes are
more or less mechanical, just changing type names and translating code
into the new API.
2018-12-31 13:53:41 +00:00
|
|
|
|
|
|
|
mp_free(p);
|
2000-10-18 15:00:36 +00:00
|
|
|
}
|
|
|
|
}
|
2020-02-23 15:29:40 +00:00
|
|
|
|
2020-02-29 16:44:56 +00:00
|
|
|
|
2020-02-23 15:29:40 +00:00
|
|
|
/* ----------------------------------------------------------------------
|
|
|
|
* Reusable null implementation of the progress-reporting API.
|
|
|
|
*/
|
|
|
|
|
|
|
|
ProgressPhase null_progress_add_probabilistic(
|
|
|
|
ProgressReceiver *prog, double c, double p) {
|
|
|
|
ProgressPhase ph = { .n = 0 };
|
|
|
|
return ph;
|
|
|
|
}
|
|
|
|
void null_progress_ready(ProgressReceiver *prog) {}
|
|
|
|
void null_progress_start_phase(ProgressReceiver *prog, ProgressPhase phase) {}
|
|
|
|
void null_progress_report_attempt(ProgressReceiver *prog) {}
|
|
|
|
void null_progress_report_phase_complete(ProgressReceiver *prog) {}
|
|
|
|
const ProgressReceiverVtable null_progress_vt = {
|
|
|
|
null_progress_add_probabilistic,
|
|
|
|
null_progress_ready,
|
|
|
|
null_progress_start_phase,
|
|
|
|
null_progress_report_attempt,
|
|
|
|
null_progress_report_phase_complete,
|
|
|
|
};
|
|
|
|
|
|
|
|
/* ----------------------------------------------------------------------
|
|
|
|
* Helper function for progress estimation.
|
|
|
|
*/
|
|
|
|
|
|
|
|
double estimate_modexp_cost(unsigned bits)
|
|
|
|
{
|
|
|
|
/*
|
|
|
|
* A modexp of n bits goes roughly like O(n^2.58), on the grounds
|
|
|
|
* that our modmul is O(n^1.58) (Karatsuba) and you need O(n) of
|
|
|
|
* them in a modexp.
|
|
|
|
*/
|
|
|
|
return pow(bits, 2.58);
|
|
|
|
}
|