Hello,

> >There could be a compiler bug, or something else that keeps it from
> >functioning properly.  Can you please run the OpenSSL crypto test suite?
> 
> 	if possible, could you run regression test under
> 	src/regress/lib/libcrpto (committed very recently)
> 	with very current libcrypto?

Ok, I just checked out basesrc/regress/lib/libcrypto and did run `make`
on my Amiga (same source which I tried with the sparc and hp300 - amiga
shows the same problem like hp300 so it must be m68k related) and got:

all ===> idea
cc -O2 -m68060 -Werror   -c /usr/src/regress/lib/libcrypto/idea/../../../../crypto/dist/openssl/crypto/idea/ideatest.c
cc   -o ideatest ideatest.o -lcrypto_idea -lcrypto
ld: -lcrypto_idea: no match
collect2: ld returned 1 exit status
*** Error code 1

So I commented out idea from the Makefile... and got:

all ===> rc5
cc -O2 -m68060 -Werror   -c /usr/src/regress/lib/libcrypto/rc5/../../../../crypto/dist/openssl/crypto/rc5/rc5test.c
cc   -o rc5test rc5test.o -lcrypto_rc5 -lcrypto
ld: -lcrypto_rc5: no match
collect2: ld returned 1 exit status
*** Error code 1

So I commented that one out too - maybe ist because the source from anoncvs
checked out at 01.10.2000 and the regression one I checked out from current
has differences.

A `make regress` in rsa gave me that:

==> running rsatest
PKCS #1 v1.5 encryption/decryption ok
OAEP encryption/decryption ok
PKCS #1 v1.5 encryption/decryption ok
OAEP encryption/decryption ok
PKCS #1 v1.5 encryption/decryption ok
OAEP encryption/decryption ok

Anything else I should try? The output of the complete regression test
(without idea and rc5) is too long to post here...

Greets
      Thorsten