Re: Crypto problems in Vista

"Laszlo Hars" <Laszlo Hars@xxxxxxxxxxxxxxxxxxxxxxxxx> wrote in message
news:CEC21216-2439-4BC7-94B0-ED22DB1F9A07@xxxxxxxxxxxxxxxx
My Crypto questions remained unanswered in the MSDN Forum for over a week:
http://forums.microsoft.com/MSDN/ShowPost.aspx?PostID=2457667&SiteID=1
Is there a way to get some more information about the cryptographic CSP of
Windows Vista?

I managed to call AES and SHA-1 functions from the "Microsoft Enhanced RSA
and AES Cryptographic Provider" CSP in Vista. It works, I can encrypt and
hash messages, export and import keys. However, something is funny with
AES
encryption:

1. Messages are always padded. I need to encrypt 16-byte blocks, and the
padding algorithm makes them all 32-bytes. Does anyone know a way to
switch
off padding? I tried calling from the advapi32.dll CryptSetKeyParam() with
dwParam = KP_PADDING and pbData pointing to a DWORD containing one of the
numbers -1,0,1,2,3., without any effects.

AES is a block cipher with a 16 byte block-length. However, as noted on the
MSDN documentation:

"As a rule, if a stream cipher is used, the ciphertext is the same size as
the plaintext. If a block cipher is used, the ciphertext is up to a block
length larger than the plaintext." (From
http://msdn2.microsoft.com/en-us/library/aa379924.aspx, documentation for
CryptEncrypt)

Okay, so the ciphertext may be as much as a block length larger than the
plaintext. That appears to be what you are seeing, though I'm perhaps not
the expert you wanted.

2. Selecting CRYPT_MODE_CTS (CipherText Stealing) mode should return a
ciphertext of the same length as the plaintext, but it always rounds up
the
length to an integer multiple of 16 bytes, the block length of AES. I
could
not find any difference between it and CRYPT_MODE_CBC (Cipher Block
Chaining)
mode. Is CTS supported? If yes, how to invoke it?

If CryptSetKeyParam appears to accept CRYPT_MODE_CTS, but you see no sign
that it's working, you might call CryptGetKeyParam to see whether the mode
was set. Maybe CryptSetKeyParam is ignoring an option it can't accept?

3. CRYPT_MODE_CFB seems to work, but CRYPT_MODE_OFB does not. When this
encryption mode is selected, the buffer of both CryptEncrypt() and
CryptDecrypt() remains unchanged. Is OFB supported?

http://msdn2.microsoft.com/en-us/library/aa379949.aspx - "CRYPT_MODE_OFB -
The cipher mode is Output Feedback (OFB). Microsoft CSPs currently do not
support Output Feedback Mode."

Doesn't look supported to me.

Alun.
~~~~


.

Relevant Pages

  • Re: Only people who originally frequent sci.crypt reply to this
    ... The mode of a cipher is one of the many, ... you need to get right in order to turn a secure algorithm into a secure ... there are no known attacks against AES. ... attack of any kind against a cipher, ...
    (sci.crypt)
  • Re: Is a cryptographic monoculture hurting us all?
    ... AES may well remain secure for a while but it could be broken in the ... Or worse you get protocol attacks like the GSM. ... centeralisation means that the cipher has gotta be damn tough to break. ... safe that are roughly equal in strength. ...
    (sci.crypt)
  • Quadruple Algorithms
    ... occurring" (a fatal flaw being found in AES, ... the most likely attack on your entire system, ... Threat one: Your implementation of AES has an undiscovered ... with the output of one cipher feeding ...
    (sci.crypt)
  • Re: Cascading/Layering Group Ciphers
    ... >with AES using one key and then encrypt that ciphertext again with AES ... cipher, ... We'd need a precise unambiguous bit-level specification of what mode ...
    (sci.crypt)
  • Re: Encryption > access denied after importing key with Certificate
    ... group on Sept 11 2002 "Does Enhanced CSP support AES" ... I'd like to know if the MS Enhanced Provider does support ...
    (microsoft.public.windowsxp.security_admin)