OpenSSL's RSA API is poorly designed and does not have a single place to
properly initialize the key. See
https://github.com/openssl/openssl/issues/5158.
To workaround this flaw, we must lazily instantiate pre-computed
Montgomery bits with locking. This is a ton of complexity. More
importantly, it makes it very difficult to implement RSA without side
channels. The correct in-memory representation of d, dmp1, and dmq1
depend on n, p, and q, respectively. (Those values have private
magnitudes and must be sized relative to the respective moduli.)
08805fe279 attempted to fix up the various
widths under lock, when we set up BN_MONT_CTX. However, this introduces
threading issues because other threads may access those exposed
components (RSA_get0_* also count as exposed for these purposes because
they are get0 functions), while a private key operation is in progress.
Instead, we do the following:
- There is no actual need to minimize n, p, and q, but we have minimized
copies in the BN_MONT_CTXs, so use those.
- Store additional copies of d, dmp1, and dmq1, at the cost of more
memory used. These copies have the correct width and are private,
unlike d, dmp1, and dmq1 which are sadly exposed. Fix private key
operations to use them.
- Move the frozen bit out of rsa->flags, as that too was historically
accessible without locking.
(Serialization still uses the original BIGNUMs, but the RSAPrivateKey
serialization format already inherently leaks the magnitude, so this
doesn't matter.)
Change-Id: Ia3a9b0629f8efef23abb30bfed110d247d1db42f
Reviewed-on: https://boringssl-review.googlesource.com/25824
Commit-Queue: David Benjamin <davidben@google.com>
CQ-Verified: CQ bot account: commit-bot@chromium.org <commit-bot@chromium.org>
Reviewed-by: Adam Langley <agl@google.com>
This is a bit easier to read than BN_less_than_consttime when we must do
>= or <=, about as much work to compute, and lots of code calls BN_cmp
on secret data. This also, by extension, makes BN_cmp_word
constant-time.
BN_equal_consttime is probably a little more efficient and is perfectly
readable, so leave that one around.
Change-Id: Id2e07fe312f01cb6fd10a1306dcbf6397990cf13
Reviewed-on: https://boringssl-review.googlesource.com/25444
Commit-Queue: David Benjamin <davidben@google.com>
CQ-Verified: CQ bot account: commit-bot@chromium.org <commit-bot@chromium.org>
Reviewed-by: Adam Langley <agl@google.com>
There is no more need for the "constant-time" reading beyond bn->top. We
can write the bytes out naively because RSA computations no longer call
bn_correct_top/bn_set_minimal_width.
Specifically, the final computation is a BN_mod_mul_montgomery to remove
the blinding, and that keeps the sizes correct.
Bug: 237
Change-Id: I6e90d81c323b644e179d899f411479ea16deab98
Reviewed-on: https://boringssl-review.googlesource.com/25324
Commit-Queue: David Benjamin <davidben@google.com>
CQ-Verified: CQ bot account: commit-bot@chromium.org <commit-bot@chromium.org>
Reviewed-by: Adam Langley <agl@google.com>
Alas, the existence of RSA keys with q > p is obnoxious, but we can
canonicalize it away. To my knowledge, the remaining leaks in RSA are:
- Key generation. This is kind of hopelessly non-constant-time but
perhaps deserves a more careful ponder. Though hopefully it does not
come in at a measurable point for practical purposes.
- Private key serialization. RSAPrivateKey inherently leaks the
magnitudes of d, dmp1, dmq1, and iqmp. This is unavoidable but
hopefully does not come in at a measurable point for practical
purposes.
- If p and q have different word widths, we currently fall back to the
variable-time BN_mod rather than Montgomery reduction at the start of
CRT. I can think of ways to apply Montgomery reduction, but it's
probably better to deny CRT to such keys, if not reject them outright.
- bn_mul_fixed and bn_sqr_fixed which affect the Montgomery
multiplication bn_mul_mont-less configurations, as well as the final
CRT multiplication. We should fix this.
Bug: 233
Change-Id: I8c2ecf8f8ec104e9f26299b66ac8cbb0cad04616
Reviewed-on: https://boringssl-review.googlesource.com/25263
Commit-Queue: David Benjamin <davidben@google.com>
CQ-Verified: CQ bot account: commit-bot@chromium.org <commit-bot@chromium.org>
Reviewed-by: Adam Langley <agl@google.com>
As the EC code will ultimately want to use these in "words" form by way
of EC_FELEM, and because it's much easier, I've implement these as
low-level words-based functions that require all inputs have the same
width. The BIGNUM versions which RSA and, for now, EC calls are
implemented on top of that.
Unfortunately, doing such things in constant-time and accounting for
undersized inputs requires some scratch space, and these functions don't
take BN_CTX. So I've added internal bn_mod_*_quick_ctx functions that
take a BN_CTX and the old functions now allocate a bit unnecessarily.
RSA only needs lshift (for BN_MONT_CTX) and sub (for CRT), but the
generic EC code wants add as well.
The generic EC code isn't even remotely constant-time, and I hope to
ultimately use stack-allocated EC_FELEMs, so I've made the actual
implementations here implemented in "words", which is much simpler
anyway due to not having to take care of widths.
I've also gone ahead and switched the EC code to these functions,
largely as a test of their performance (an earlier iteration made the EC
code noticeably slower). These operations are otherwise not
performance-critical in RSA.
The conversion from BIGNUM to BIGNUM+BN_CTX should be dropped by the
static linker already, and the unused BIGNUM+BN_CTX functions will fall
off when EC_FELEM happens.
Update-Note: BN_mod_*_quick bounce on malloc a bit now, but they're not
really used externally. The one caller I found was wpa_supplicant
which bounces on malloc already. They appear to be implementing
compressed coordinates by hand? We may be able to convince them to
call EC_POINT_set_compressed_coordinates_GFp.
Bug: 233, 236
Change-Id: I2bf361e9c089e0211b97d95523dbc06f1168e12b
Reviewed-on: https://boringssl-review.googlesource.com/25261
Commit-Queue: David Benjamin <davidben@google.com>
CQ-Verified: CQ bot account: commit-bot@chromium.org <commit-bot@chromium.org>
Reviewed-by: Adam Langley <agl@google.com>
d, dmp1, dmq1, and iqmp have private magnitudes. This is awkward because
the RSAPrivateKey serialization leaks the magnitudes. Do the best we can
and fix them up before any RSA operations.
This moves the piecemeal BN_MONT_CTX_set_locked into a common function
where we can do more complex canonicalization on the keys. Ideally this
would be done on key import, but the exposed struct (and OpenSSL 1.1.0's
bad API design) mean there is no single point in time when key import is
finished.
Also document the constraints on RSA_set0_* functions. (These
constraints aren't new. They just were never documented before.)
Update-Note: If someone tried to use an invalid RSA key where d >= n,
dmp1 >= p, dmq1 >= q, or iqmp >= p, this may break. Such keys would not
have passed RSA_check_key, but it's possible to manually assemble
keys that bypass it.
Bug: 232
Change-Id: I421f883128952f892ac0cde0d224873a625f37c5
Reviewed-on: https://boringssl-review.googlesource.com/25259
Commit-Queue: David Benjamin <davidben@google.com>
CQ-Verified: CQ bot account: commit-bot@chromium.org <commit-bot@chromium.org>
Reviewed-by: Adam Langley <agl@google.com>
Give a non-minimal modulus, there are two possible values of R we might
pick: 2^(BN_BITS2 * width) or 2^(BN_BITS2 * bn_minimal_width).
Potentially secret moduli would make the former attractive and things
might even work, but our only secret moduli (RSA) have public bit
widths. It's more cases to test and the usual BIGNUM invariant is that
widths do not affect numerical output.
Thus, settle on minimizing mont->N for now. With the top explicitly made
minimal, computing |lgBigR| is also a little simpler.
This CL also abstracts out the < R check in the RSA code, and implements
it in a width-agnostic way.
Bug: 232
Change-Id: I354643df30530db7866bb7820e34241d7614f3c2
Reviewed-on: https://boringssl-review.googlesource.com/25250
Reviewed-by: Adam Langley <agl@google.com>
This is actually a bit more complicated (the mismatching widths cases
will never actually happen in RSA), but it's easier to think about and
removes more width-sensitive logic.
Bug: 232
Change-Id: I85fe6e706be1f7d14ffaf587958e930f47f85b3c
Reviewed-on: https://boringssl-review.googlesource.com/25246
Reviewed-by: Adam Langley <agl@google.com>
Update-Note: Some RSA_FLAG_* constants are gone. Code search says they
were unused, but they can be easily restored if this breaks anything.
Change-Id: I47f642af5af9f8d80972ca8da0a0c2bd271c20eb
Reviewed-on: https://boringssl-review.googlesource.com/24244
Reviewed-by: Adam Langley <agl@google.com>
The first step of RSA with the CRT optimization is to reduce our input
modulo p and q. We can do this in constant-time[*] with Montgomery
reduction. When p and q are the same size, Montgomery reduction's bounds
hold. We need two rounds of it because the first round gives us an
unwanted R^-1.
This does not appear to have a measurable impact on performance. Also
add a long TODO describing how to make the rest of the function
constant-time[*] which hopefully we'll get to later. RSA blinding should
protect us from it all, but make this constant-time anyway.
Since this and the follow-up work will special-case weird keys, add a
test that we don't break those unintentionally. (Though I am not above
breaking them intentionally someday...)
Thanks to Andres Erbsen for discussions on how to do this bit properly.
[*] Ignoring the pervasive bn_correct_top problem for the moment.
Change-Id: Ide099a9db8249cb6549be99c5f8791a39692ea81
Reviewed-on: https://boringssl-review.googlesource.com/24204
Reviewed-by: Adam Langley <agl@google.com>
This rewrites the internals with a "words" variant that can avoid
bn_correct_top. It still ultimately calls bn_correct_top as the calling
convention is sadly still BIGNUM, but we can lift that calling
convention out incrementally.
Performance seems to be comparable, if not faster.
Before:
Did 85000 ECDSA P-256 signing operations in 5030401us (16897.3 ops/sec)
Did 34278 ECDSA P-256 verify operations in 5048029us (6790.4 ops/sec)
After:
Did 85000 ECDSA P-256 signing operations in 5021057us (16928.7 ops/sec)
Did 34086 ECDSA P-256 verify operations in 5010416us (6803.0 ops/sec)
Change-Id: I1159746dfcc00726dc3f28396076a354556e6e7d
Reviewed-on: https://boringssl-review.googlesource.com/23065
Reviewed-by: Adam Langley <agl@google.com>
Generating a 2048-bit RSA key with e = 3 (don't do this), the failure
rate at 5*bits iterations appears to be around 7 failures in 1000 tries.
Bump the limit up to 32*bits. This should give a failure rate of around
2 failures in 10^14 tries.
(The FIPS 186-4 algorithm is meant for saner values of e, like 65537. e
= 3 implies a restrictive GCD requirement: the primes must both be 2 mod
3.)
Change-Id: Icd373f61e2eb90df5afaff9a0fc2b2fbb6ec3f0a
Reviewed-on: https://boringssl-review.googlesource.com/22584
Commit-Queue: David Benjamin <davidben@google.com>
Commit-Queue: Adam Langley <agl@google.com>
Reviewed-by: Adam Langley <agl@google.com>
CQ-Verified: CQ bot account: commit-bot@chromium.org <commit-bot@chromium.org>
crypto/{asn1,x509,x509v3,pem} were skipped as they are still OpenSSL
style.
Change-Id: I3cd9a60e1cb483a981aca325041f3fbce294247c
Reviewed-on: https://boringssl-review.googlesource.com/19504
Reviewed-by: Adam Langley <agl@google.com>
Commit-Queue: David Benjamin <davidben@google.com>
CQ-Verified: CQ bot account: commit-bot@chromium.org <commit-bot@chromium.org>
We've got three versions of DATA_TOO_LARGE and two versions of
DATA_TOO_SMALL with no apparent distinction between them.
Change-Id: I18ca2cb71ffc31b04c8fd0be316c362da4d7daf9
Reviewed-on: https://boringssl-review.googlesource.com/17529
Reviewed-by: David Benjamin <davidben@google.com>
Commit-Queue: David Benjamin <davidben@google.com>
CQ-Verified: CQ bot account: commit-bot@chromium.org <commit-bot@chromium.org>
Public and private RSA keys have the same type in OpenSSL, so it's
probably prudent for us to catch this case with an error rather than
crash. (As we do if you, say, configure RSA-PSS parameters on an Ed25519
EVP_PKEY.) Bindings libraries, in particular, tend to hit this sort of
then when their callers do silly things.
Change-Id: I2555e9bfe716a9f15273abd887a8459c682432dd
Reviewed-on: https://boringssl-review.googlesource.com/17325
Commit-Queue: Adam Langley <agl@google.com>
Reviewed-by: Adam Langley <agl@google.com>
CQ-Verified: CQ bot account: commit-bot@chromium.org <commit-bot@chromium.org>
This change allows blinding to be disabled without also having to remove
|e|, which would disable the CRT and the glitch checks. This is to
support disabling blinding in the FIPS power-on tests.
(Note: the case where |e| isn't set is tested by RSATest.OnlyDGiven.)
Change-Id: I28f18beda33b1687bf145f4cbdfd37ce262dd70f
Reviewed-on: https://boringssl-review.googlesource.com/17146
Commit-Queue: Adam Langley <alangley@gmail.com>
Commit-Queue: Adam Langley <agl@google.com>
Reviewed-by: David Benjamin <davidben@google.com>
Reviewed-by: Adam Langley <agl@google.com>
This has since been done.
Change-Id: I498f845fa4ba3d1c04a5892831be4b07f31536d4
Reviewed-on: https://boringssl-review.googlesource.com/16124
Commit-Queue: David Benjamin <davidben@google.com>
Commit-Queue: Steven Valdez <svaldez@google.com>
Reviewed-by: Steven Valdez <svaldez@google.com>
CQ-Verified: CQ bot account: commit-bot@chromium.org <commit-bot@chromium.org>
