NOTES, THOUGHTS, and EVERYTHING
-------------------------------
(1) Maybe ENGINE_get_struct_size() isn't such a good idea. All ENGINEs
should be allocated from within OpenSSL (rather than, for example,
a loaded DSO). Two reasons, (i) DSOs authors are likely to stash
the return value as an assumed constant and so everything will
break down horribly when OpenSSL is changed/expanded, (ii) with
the structure allocated within OpenSSL, we could handle the case
where a DSO *really* wants to close down and lick its wounds even
if there are still references because we could simply NULL out the
pointers in the structure. If I change this, I should also
remember to get rid of the parameter in ENGINE_new() as it would
serve no purpose and is likely to confuse.
(2) Concurrency and locking ... I made a change to the ENGINE_free code
because I spotted a potential hold-up in proceedings (doing too
much inside a lock including calling a callback), there may be
other bits like this. What do the speed/optimisation freaks think
of this aspect of the code and design? There's lots of locking for
manipulation functions and I need that to keep things nice and
solid, but this manipulation is mostly (de)initialisation, I would
think that most run-time locking is purely in the ENGINE_init and
ENGINE_finish calls that might be made when getting handles for
RSA (and friends') structures, and these would be mostly reference
count operations as the functional references should always be 1
or greater at run-time to prevent init/deinit thrashing.
(3) Atalla isn't finished quite yet.
(4) The DH stuff was added to the CryptoSwift code without testing
because it should work trivially and didn't involve adding more of
the cropped bits from Rainbow's headers back into the vendor_defns
stuff. (Also, randomness should be easy to add soon when I sort
the headers out a bit more which would give hw_cswift a full
suite).
(5) Another make update is probably due ...
05d909c549