- Oct 02, 2018
-
-
Nicola Tuveri authored
It appears that, in FIPS mode, `ec_precompute_mont_data()` always failed but the error was ignored until commit e3ab8cc4 from #6810. The actual problem lies in the fact that access to the `mont_data` field of an `EC_GROUP` struct should always be guarded by an `EC_GROUP_VERSION(group)` check to avoid OOB accesses, because `group` might come from the FIPS module, which does not define the `mont_data` field inside the EC_GROUP structure. This commit adds the required check before any access to `group->mont_data` in `EC_GROUP_set_generator()`. Fixes #7127 Reviewed-by:
Tim Hudson <tjh@openssl.org> Reviewed-by:
Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com> (Merged from https://github.com/openssl/openssl/pull/7135)
-
- Sep 26, 2018
-
-
Pauli authored
Manual merge of #7047 to 1.0.2-stable. Reviewed-by:
Bernd Edlinger <bernd.edlinger@hotmail.de> (Merged from https://github.com/openssl/openssl/pull/7300)
-
Manikantan Subramanian authored
Fixes #7228 The function BIO_get_host_ip uses gethostbyname, which is not thread safe and hence we grab a lock. In multi-threaded applications, this lock sometimes causes performance bottlenecks. This patch uses the function gethostbyname_r (thread safe version), when available. Reviewed-by:
Paul Dale <paul.dale@oracle.com> Reviewed-by:
Richard Levitte <levitte@openssl.org> (Merged from https://github.com/openssl/openssl/pull/7250)
-
- Sep 24, 2018
-
-
Daniel Bevenius authored
This commit documents the OPENSSL_VERSION_TEXT which is currently missing in the man page. Reviewed-by:
-
- Sep 21, 2018
-
-
Richard Levitte authored
This module includes bn.h via other headers, so it picks up the definition from there and doesn't need to define them locally (any more?). Worst case scenario, the redefinition may be different and cause all sorts of compile errors. Fixes #7227 Reviewed-by:
-
- Sep 20, 2018
-
-
Dr. Matthias St. Pierre authored
Fixes #7240 In FIPS mode, the default FIPS DRBG uses the drbg_get_entropy() callback to reseed itself, which is provided by the wrapping libcrypto library. This callback in turn uses ssleay_rand_bytes() to generate random bytes. Now ssleay_rand_bytes() calls RAND_poll() once on first call to seed itself, but RAND_poll() is never called again (unless the application calls RAND_poll() explicitely). This implies that whenever the DRBG reseeds itself (which happens every 2^14 generate requests) this happens without obtaining fresh random data from the operating system's entropy sources. This patch forces a reseed from system entropy sources on every call to drbg_get_entropy(). In contrary to the automatic reseeding of the DRBG in master, this reseeding does not break applications running in a chroot() environment (see c7504aeb ), because the SSLEAY PRNG does not maintain an error state. (It does not even check the return value of RAND_poll() on its instantiation.) In the worst case, if no random device is available for reseeding, no fresh entropy will be added to the SSLEAY PRNG but it will happily continue to generate random bytes as 'entropy' input for the DRBG's reseeding, which is just as good (or bad) as before this patch. To prevent ssleay_rand_bytes_from_system() (and hence RAND_poll()) from being called twice during instantiation, a separate drbg_get_nonce() callback has been introduced, which is identical with the previous implementation of drbg_get_entropy(). Reviewed-by:
Ben Kaduk <kaduk@mit.edu> (Merged from https://github.com/openssl/openssl/pull/7259)
-
Richard Levitte authored
These both indicate that the file descriptor we're trying to use as a terminal isn't, in fact, a terminal. Fixes #7271 Reviewed-by:
-
- Sep 11, 2018
-
-
Bernd Edlinger authored
Reviewed-by:
-
Richard Levitte authored
Since the SSL code started using RSA_NO_PADDING, the CAPI engine became unusable. This change fixes that. Fixes #7131 Reviewed-by:
-
- Sep 10, 2018
-
-
Matt Caswell authored
Found by Coverity Reviewed-by:
-
- Sep 04, 2018
-
-
Richard Levitte authored
With the introduction of -pkeyopt, the number of bits may change without |newkey| being updated. Unfortunately, there is no API to retrieve the information from a EVP_PKEY_CTX either, so chances are that we report incorrect information. For the moment, it's better not to try to report the number of bits at all. Fixes #7086 Reviewed-by:
Paul Yang <yang.yang@baishancloud.com> Reviewed-by:
-
Richard Levitte authored
Reviewed-by:
-
Matt Caswell authored
They did not make it clear how the memory management works for the |pctx| parameter. Reviewed-by:
-
Jakub Wilk authored
The default input format is PEM, so explicit "-inform DER" is needed to read DER-encoded CRL. CLA: trivial Reviewed-by:
-
Matt Caswell authored
Just remove that statement. It's not been true since 2005. Reviewed-by:
Nicola Tuveri <nic.tuv@gmail.com> (Merged from https://github.com/openssl/openssl/pull/6906)
-
- Sep 03, 2018
-
-
Pauli authored
Reviewed-by:
-
- Sep 02, 2018
-
-
Eric Brown authored
This trivial patch removes a duplicated call to ASN1_INTEGER_set. Fixes Issue #6977 Signed-off-by:
Eric Brown <browne@vmware.com> Reviewed-by:
Andy Polyakov <appro@openssl.org> Reviewed-by:
-
- Aug 28, 2018
-
-
Andy Polyakov authored
(cherry picked from commit 324b9560 ) Reviewed-by:
-
Andy Polyakov authored
(cherry picked from commit e02c519c ) Resolved conflicts: crypto/bn/bn_blind.c Reviewed-by:
-
Andy Polyakov authored
In [most common] case of p and q being of same width, it's possible to replace CRT modulo operations with Montgomery reductions. And those are even fixed-length Montgomery reductions... (cherry picked from commit 41bfd5e7 ) Resolved conflicts: crypto/rsa/rsa_eay.c Reviewed-by:
-
Andy Polyakov authored
Add bn_mul_fixed_top, bn_from_mont_fixed_top, bn_mod_sub_fixed_top. Switch to bn_{mul|sqr}_fixed_top in bn_mul_mont_fixed_top and remove memset in bn_from_montgomery_word. (cherry picked from commit fcc4ee09 ) Resolved conflicts: crypto/bn/bn_mod.c crypto/bn_int.h Reviewed-by:
-
- Aug 27, 2018
-
-
Hubert Kario authored
the option is provided in the -help message of the s_server utility but it is not documented in the man page, this fixes it Reviewed-by:
-
- Aug 17, 2018
-
-
Andy Polyakov authored
Reviewed-by:
Viktor Dukhovni <viktor@openssl.org> (Merged from https://github.com/openssl/openssl/pull/6916) (cherry picked from commit 0da7358b) Resolved conflicts: crypto/x509v3/v3_purp.c
-
- Aug 14, 2018
-
-
Matt Caswell authored
Reviewed-by: -
Matt Caswell authored
Reviewed-by: -
Matt Caswell authored
Reviewed-by: -
Richard Levitte authored
Reviewed-by:
Matt Caswell <matt@openssl.org> (Merged from https://github.com/openssl/openssl/pull/6957)
-
Matt Caswell authored
Reviewed-by:
-
Matt Caswell authored
Reviewed-by:
-
Richard Levitte authored
Just as was done recently for i2d_ASN1_OBJECT, we also make i2d_ASN1_BOOLEAN comply with the documentation. Reviewed-by:
Rich Salz <rsalz@openssl.org> (Merged from https://github.com/openssl/openssl/pull/6943)
-
- Aug 11, 2018
-
-
Richard Levitte authored
Since 0.9.7, all i2d_ functions were documented to allocate an output buffer if the user didn't provide one, under these conditions (from the 1.0.2 documentation): For OpenSSL 0.9.7 and later if B<*out> is B<NULL> memory will be allocated for a buffer and the encoded data written to it. In this case B<*out> is not incremented and it points to the start of the data just written. i2d_ASN1_OBJECT was found not to do this, and would crash if a NULL output buffer was provided. Fixes #6914 Reviewed-by:
-
- Aug 10, 2018
-
-
Andy Polyakov authored
Reviewed-by:
-
Andy Polyakov authored
It was false positive, but one can as well view it as readability issue. Switch even to unsigned indices because % BN_BYTES takes 4-6 instructions with signed dividend vs. 1 (one) with unsigned. Reviewed-by:
-
Andy Polyakov authored
"Computationally constant-time" means that it might still leak information about input's length, but only in cases when input is missing complete BN_ULONG limbs. But even then leak is possible only if attacker can observe memory access pattern with limb granularity. Reviewed-by:
-
- Aug 07, 2018
-
-
Richard Levitte authored
Reviewed-by:
-
- Aug 01, 2018
-
-
Andy Polyakov authored
Reviewed-by:
-
Andy Polyakov authored
(back-ported from commit 37132c97 ) Reviewed-by:
-
Billy Brumley authored
Those even order that do not play nicely with Montgomery arithmetic (back-ported from commit 3a6a4a93 ) Reviewed-by:
-
Andy Polyakov authored
New implementation failed to correctly reset r->neg flag. Spotted by OSSFuzz. Reviewed-by:
-
Andy Polyakov authored
Originally suggested solution for "Return Of the Hidden Number Problem" is arguably too expensive. While it has marginal impact on slower curves, none to ~6%, optimized implementations suffer real penalties. Most notably sign with P-256 went more than 2 times[!] slower. Instead, just implement constant-time BN_mod_add_quick. Reviewed-by:
-
