Fix for the attack described in the paper "Recovering OpenSSL
ECDSA Nonces Using the FLUSH+RELOAD Cache Side-channel Attack"
by Yuval Yarom and Naomi Benger. Details can be obtained from:
http://eprint.iacr.org/2014/140
Thanks to Yuval Yarom and Naomi Benger for discovering this
flaw and to Yuval Yarom for supplying a fix.
knock-on work than expected - they've been extracted into a patch
series that can be completed elsewhere, or in a different branch,
before merging back to HEAD.
deprecate the original (numeric-only) scheme, and replace with the
CRYPTO_THREADID object. This hides the platform-specifics and should reduce
the possibility for programming errors (where failing to explicitly check
both thread ID forms could create subtle, platform-specific bugs).
Thanks to Bodo, for invaluable review and feedback.
to 'unsigned long' (ie. odd platforms/compilers), so a pointer-typed
version was added but it required portable code to check *both* modes to
determine equality. This commit maintains the availability of both thread
ID types, but deprecates the type-specific accessor APIs that invoke the
callbacks - instead a single type-independent API is used. This simplifies
software that calls into this interface, and should also make it less
error-prone - as forgetting to call and compare *both* thread ID accessors
could have led to hard-to-debug/infrequent bugs (that might only affect
certain platforms or thread implementations). As the CHANGES note says,
there were corresponding deprecations and replacements in the
thread-related functions for BN_BLINDING and ERR too.
CRYPTO_get_idptr_callback(), CRYPTO_thread_idptr() for a 'void *' type
thread ID, since the 'unsigned long' type of the existing thread ID
does not always work well.
timing attacks.
BN_FLG_EXP_CONSTTIME requests this algorithm, and this done by default for
RSA/DSA/DH private key computations unless
RSA_FLAG_NO_EXP_CONSTTIME/DSA_FLAG_NO_EXP_CONSTTIME/
DH_FLAG_NO_EXP_CONSTTIME is set.
Submitted by: Matthew D Wood
Reviewed by: Bodo Moeller
- possibility of re-creation of the blinding parameters after a
fixed number of uses (suggested by Bodo)
- calculatition of the rsa::e in case it's absent and p and q
are present (see bug report #785)
- improve the performance when if one rsa structure is shared by
more than a thread (see bug report #555)
- fix the problem described in bug report #827
- hide the definition ot the BN_BLINDING structure in bn_blind.c
- Remove unused and unuseful debug cruft.
- Remove unnecessary 'top' fudging from BN_copy().
- Fix a potential memory leak and simplify the expansion logic in
BN_bin2bn().
Submitted by: Nils Larsch
Reviewed by: Geoff Thorpe
corresponding headers are only required for API functions or structure
details. This now includes the bignum types and BUF_MEM. Subsequent commits
will remove various dependencies on bn.h and buffer.h and update the
makefile dependencies.
- Remove some unnecessary "+1"-like fudges. Sizes should be handled
exactly, as enlarging size parameters causes needless bloat and may just
make bugs less likely rather than fixing them: bn_expand() macro,
bn_expand_internal(), and BN_sqr().
- Deprecate bn_dup_expand() - it's new since 0.9.7, unused, and not that
useful.
- Remove unnecessary zeroing of unused bytes in bn_expand2().
- Rewrite BN_set_word() - it should be much simpler, the previous
complexities probably date from old mismatched type issues.
- Add missing bn_check_top() macros in bn_word.c
- Improve some degenerate case handling in BN_[add|sub]_word(), add
comments, and avoid a bignum expansion if an overflow isn't possible.
bignums are passed in and out of functions and APIs in a consistent form
has highlighted that zero-valued bignums don't need any allocated word
data. The use of BN_set_word() to initialise a bignum to zero causes
needless allocation and gives it a return value that must be checked. This
change converts BN_zero() to a self-contained macro that has no
return/expression value and does not cause any expansion of bignum data.
Note, it would be tempting to rewrite the deprecated version as a
success-valued comma expression, such as;
#define BN_zero(a) ((a)->top = (a)->neg = 0, 1)
However, this evaluates 'a' twice and would confuse initialisation loops
(eg. while(..) { BN_zero(bn++) } ). As such, the deprecated version
continues to use BN_set_word().
once in the source (where it is set for the benefit of no other code
whatsoever). I've deprecated the declaration in the header and likewise
made the use of the flag conditional in bn_lib.c. Note, this change also
NULLs the 'd' pointer in a BIGNUM when it is reset but not deallocated.
(ie. where top may be zero, or it may be one if the corresponding word is
set to zero). Note, this only affects the macros in bn.h, there are probably
similar corrections required in some c files.
Also, clarify the audit-related macros at the top of the header. Mental
note: I must not forget to clean all this out before 0.9.8 is released ...
