1. rnd_reference was a duplication of the work the the engine
framework does, and wasn't ever checked. Removed.
2. use the NO_ macros to disable appropriate algorithms.
3. Only implement the RNG stuff if AEPRAND is defined (default: not
defined, because the AEP people plan on having boards without it.
I'll see if I can device a more dynamic way of disabling this).
4. aep_finish() now closes all connections, and if that worked, does a
proper finalize.
5. proper AEP types are used to conform to the AEP definitions of
their own functions.
6. remake the use of thread locks. The use of CRYPTO_LOCK_DYNLOCK was
definitely inappropriate, and for random generator stuff, it's
better to use CRYPTO_LOCK_RAND.
Also, I applied certain changes that were provided by the AEP people.
Among others, BN_CTX_new() is not used to initialise a BN context
(this was never done before, and may have made things slower or not
working at all.
patches taken from Red Hat Linux 7.2. Original code from Broadcom with
patches and backport by Nalin, more backport to fix warnings and const
changes by Mark
Submitted by: Mark Cox
Reviewed by:
PR:
for acceleration only at the moment, but full key management is being
worked on for the future. This code has been compiled cross-platform but
not extensively tested
Submitted by: Mark Cox, Baltimore Technologies
Reviewed by: Mark Cox
PR:
7.2 and been given extensive testing; it also compiles okay on our selection
of random machines (including 64-bit)
Submitted by: AEP, Mark Cox
Reviewed by: Mark Cox
PR:
implement it for nCipher hardware. The interface in itself should be
clear enough, but the nCipher implementation is currently not the
best when it comes to getting a passphrase from the user. However,
getting it better is a little hard until a better user interaction
method is create.
Also, use the possibility in req, so we can start to create CSR's with
keys from the nForce box.
WARNING: I've made *no* tests yet, mostly because I didn't implement
this on the machine where I have an nForce box to play with. All I
know is that it compiles cleanly on Linux...
Atalla card, you should be able to compile with the "hw-atalla" switch
with "./config" or "perl Configure", and then you can use the command-
line switch "-engine atalla" inside speed, s_cient and s_server (after
checking out note (1)).
Notes:
(1) I've turned on native name translation when loading the shared-
library, but this means that the Unix shared library needs to be
libatasi.so rather than atasi.so. I got around this in my testing
by creating a symbollic link from /usr/lib/libatasi.so to the real
library, but something better will be needed. It also assumes in
win32 that the DLL will be called atasi.dll - but as I don't have
a win32/atalla environment to try I have no idea yet if this is
the case.
(2) Currently DSA verifies are not accelerated because I haven't yet
got a mod_exp-based variant of BN_mod_exp2_mont() that yields
correct results.
(3) Currently the "init()" doesn't fail if the shared library can
load successfully but the card is not operational. In this case,
the ENGINE_init() call will succeed, but all RSA, DSA, DH, and
the two BN_*** operations will fail until the ENGINE is switched
back to something that does work. I expect to correct this next.
(4) Although the API for the Atalla card just has the one crypto
function suggesting an RSA private key operation - this is in
fact just a straight mod_exp function that ignores all the RSA
key parameters except the (private) exponent and modulus. This is
why the only accelerator work is taking place inside the mod_exp
function and there's no optimisation of RSA private key operations
based on CRT etc.
patches that Geoff had in a patch file in his play directory.
NOTE for openssl-cvs: THIS IS A CVS BRANCH (BRANCH_engine). IT IS
NOT FOR THE FAINTHEARTED TO PLAY WITH. The code works as it is, but
it's not at all sure it ends up in the OpenSSL distributio in this
form, so do not get dependent on it!
Those rsyncing the repository are considered warned!
