Re: blowfish in ecb mode false ciphertext

2007-10-23 Thread Victor Duchovni
On Tue, Oct 23, 2007 at 05:55:37PM +0200, Erno Kovacs wrote: > > The 0.9.8e version did not use the key length correctly. > > http://cvs.openssl.org/chngview?cn=15978 > > The 0.9.8d and 0.9.8g code should be correct. > > 0.9.8g on windows (provided by Shining Light) is still incorrect. > Shou

Re[2]: blowfish in ecb mode false ciphertext

2007-10-23 Thread Erno Kovacs
> The 0.9.8e version did not use the key length correctly. > http://cvs.openssl.org/chngview?cn=15978 > The 0.9.8d and 0.9.8g code should be correct. 0.9.8g on windows (provided by Shining Light) is still incorrect. Should i report this bug directly to them? Erno ___

Re: blowfish in ecb mode false ciphertext

2007-10-23 Thread Victor Duchovni
On Tue, Oct 23, 2007 at 04:50:08PM +0200, Erno Kovacs wrote: > One of my applications broke just after i upgraded Openssl from 0.9.8d. > After several hours of debugging I realized, something is fishy around > the blowfish cipher: it just doesnt produces the same ciphertext as it > did using the 0

blowfish in ecb mode false ciphertext

2007-10-23 Thread Erno Kovacs
One of my applications broke just after i upgraded Openssl from 0.9.8d. After several hours of debugging I realized, something is fishy around the blowfish cipher: it just doesnt produces the same ciphertext as it did using the 0.9.8d. You can find the really small and stupid source code which pro