if (bp == NULL)
                        for (j=0; j<10000; j++)
                                BN_add(&c,&a,&b);

seem to be pretty useless, and bp never is NULL anyway.

The bn_cmp_part_words bug was only caught in the BN_mod_mul() test,
not in the BN_mul() test, so apparently the choice of parameters in
some cases is bad.

casts) used in the lhash code are about as horrible and evil as they can
be. For starters, the callback prototypes contain empty parameter lists.
Yuck.

This first change defines clearer prototypes - including "typedef"'d
function pointer types to use as "hash" and "compare" callbacks, as well as
the callbacks passed to the lh_doall and lh_doall_arg iteration functions.
Now at least more explicit (and clear) casting is required in all of the
dependant code - and that should be included in this commit.

The next step will be to hunt down and obliterate some of the function
pointer casting being used when it's not necessary - a particularly evil
variant exists in the implementation of lh_doall.

Don't dump core in ssltest.

First tentative impementation of Kerberos 5 cryptos and keys for SSL/TLS.  Implemented by Vern Staats <staatsvr@asc.hpc.mil>, further hacked and distributed by Jeffrey Altman <jaltnab@columbia.edu>

violation in test_kron on a 32 bit system.

But even if this is avoided, there are still segmentation violations
(during one of the BN_free()s at the end of test_kron
in some cases, in other cases during BN_kronecker, or
later in BN_sqrt; choosing a different exponentiation
algorithm in bntest.c appears to influence when the SIGSEGV
takes place).

Submitted by: Lenka Fibikova

systems where such an operation is supported.

Windows.  Really, this should probably be done on Unix as well, but
that will be a later story...

Fix BN_mod_exp_simple for a==0 (mod m).
Skip useless round in BN_mod_sqrt (1 is always a square, no need
to test BN_kronecker for it).

exists in bn_sqrt.c now

test_bn still fails in the BN_sqrt test because
small primes appear to turn into zero for no
obvious reason, leading to "div by zero" errors.

the initialisation and cleanup of "ex_data" elements to before an init()
handler and after a finish() handler respectively.

(we need the two's complement of BN_lsw then).

so we have to reduce the random numbers used in test_mont.

Before this change, test_mont failed in [debug-]solaris-sparcv9-gcc
configurations ("Montgomery multiplication test failed!" because
the multiplication result obtained with Montgomery multiplication
differed from the result obtained by BN_mod_mul).
Substituing the old version of bn_gcd.c (BN_mod_inverse) did not avoid
the problem.

The strange thing is that it I did not observe any problems
when using debug-solaris-sparcv8-gcc and solaris-sparcv9-cc,
as well as when compiling OpenSSL 0.9.6 in the solaric-sparcv9-gcc
configuration on the same system.

Improve readability of bn_shift.c.

Add comment in bn_lib.c (why zero data between top and max?)

Change bntest.c output for BN_kronecker test

I'm still investigating what caused the segementation fault
(maybe "make clean; make" will cure it ...).
But BN_bin2bn should always reset ret->neg.

This caused a segmentation fault in calls to malloc, so I cleaned up
bn_lib.c a little so that it is easier to see what is going on.
The bug turned out to be an off-by-one error in BN_bin2bn.

Modify "CHANGES" entry for BN_mod_inverse (it's not just avoiding BN_div
that increases performance, avoiding BN_mul also helps)

Get the BN_mod_exp_mont bugfix (for handling negative inputs) correct
this time.