tests/basic fails sometimes...

Werner Koch wk at gnupg.org
Sat Dec 14 21:17:59 CET 2013

On Sat, 14 Dec 2013 00:58, wk at gnupg.org said:

> Thanks.  Meanwhile I was abale to trigger that bug myself.  I'll look at
> it.

The reason for the problem is that we check whether the decryption
returns a specific error code if a OAEP encrypted data is decrypted as
PKCS#1 encrypted data.  Now if the decrypted frame starts with 0x02
followed by a few random bytes, a 0x00 and more bytes, this looks like
correct PKCS#1 encrytpted data and Libgcrypt return with success and not
with the expected error code.  Later the value is checked and that
triggers the sceond error - as expected.

I'll add a fix to the test.



Die Gedanken sind frei.  Ausnahmen regelt ein Bundesgesetz.

More information about the Gcrypt-devel mailing list