Welcome to OGeek Q&A Community for programmer and developer-Open, Learning and Share
Welcome To Ask or Share your Answers For Others

Categories

0 votes
126 views
in Technique[技术] by (71.8m points)

c# - Why are RijndaelManaged and AesCryptoServiceProvider returning different results?

Here is the example that I have run. It has the same Mode, Padding, BlockSize, KeySize. I am using the same init vector, key and data.

Using the RijndaelManaged produces an encrypted value of: 0x8d,0x81,0x27,0xc6,0x3c,0xe2,0x53,0x2f,0x35,0x78,0x90,0xc2,0x2e,0x3b,0x8a,0x61, 0x41,0x47,0xd6,0xd0,0xff,0x92,0x72,0x3d,0xc6,0x16,0x2b,0xd8,0xb5,0xd9,0x12,0x85

Using the AesCryptoServiceProvider produces an encrypted value of: 0x8d,0x9f,0x6e,0x99,0xe9,0x54,0x8b,0x12,0xa9,0x88,0x1a,0x3d,0x65,0x23,0x9c,0x4e, 0x18,0x5a,0x89,0x31,0xf5,0x75,0xc5,0x9e,0x0d,0x43,0xe9,0x86,0xd4,0xf3,0x64,0x3a

Here is the code I used to generate these results


   public partial class AesTest
   {
      private SymmetricAlgorithm mEncryptionType;
      private byte[] mPrivateKey;
      private byte[] mInitializationVector;
      private byte[] mData;

      public AesTest()
      {
         mPrivateKey = new byte[32] 
         { 
            0x22, 0x22, 0x22, 0x22, 
            0x22, 0x22, 0x22, 0x22, 
            0x22, 0x22, 0x22, 0x22, 
            0x22, 0x22, 0x22, 0x22,
            0x22, 0x22, 0x22, 0x22, 
            0x22, 0x22, 0x22, 0x22, 
            0x22, 0x22, 0x22, 0x22, 
            0x22, 0x22, 0x22, 0x22
         };

         mInitializationVector = new byte[16]
         { 
            0x33, 0x33, 0x33, 0x33,
            0x33, 0x33, 0x33, 0x33,
            0x33, 0x33, 0x33, 0x33,
            0x33, 0x33, 0x33, 0x33
         };

         mData = new byte[16]
         {
            0x44, 0x44, 0x44, 0x44,
            0x44, 0x44, 0x44, 0x44,
            0x44, 0x44, 0x44, 0x44,
            0x44, 0x44, 0x44, 0x44
         };

         mEncryptionType = new RijndaelManaged();
         mEncryptionType.Mode = CipherMode.CFB;
         mEncryptionType.Padding = PaddingMode.PKCS7;
         mEncryptionType.BlockSize = 128;
         mEncryptionType.KeySize = 256;

         byte[] rij_encrypted_data = Encrypt(mData);

         mEncryptionType = new AesCryptoServiceProvider();
         mEncryptionType.Mode = CipherMode.CFB;
         mEncryptionType.Padding = PaddingMode.PKCS7;
         mEncryptionType.BlockSize = 128;
         mEncryptionType.KeySize = 256;

         byte[] aes_encrypted_data = Encrypt(mData);
      }

      public virtual byte[] Encrypt(byte[] unencryptedData)
      {
         return TransformData(unencryptedData, mEncryptionType.CreateEncryptor(mPrivateKey, mInitializationVector));
      }

      private byte[] TransformData(byte[] dataToTransform, ICryptoTransform cryptoTransform)
      {
         byte[] result = new byte[0];
         if (dataToTransform != null && cryptoTransform != null && dataToTransform.Length > 0)
         {
            // Create the memory stream to store the results
            MemoryStream mem_stream = new MemoryStream();
            // Create the crypto stream to do the transformation
            CryptoStream crypto_stream = new CryptoStream(mem_stream, cryptoTransform, CryptoStreamMode.Write);
            // bytes are transformed on a write
            crypto_stream.Write(dataToTransform, 0, dataToTransform.Length);
            // Flush the final block
            crypto_stream.FlushFinalBlock();
            // Convert the transformed memory stream back to a byte array
            result = mem_stream.ToArray();
            // Close the streams
            mem_stream.Close();
            crypto_stream.Close();
         }
         return result;
      }
   }

I guess I'm just wondering if I missed something.

Update: Turns out that AesManaged will throw a CryptographicException ("The specified cipher mode is not valid for this algorithm") if you try and set the CipherMode to CFB. I feel that the AesCryptoServiceProvider should do that same, but it doesnt. Seems funny that the FIPS Certified class allows invalid cipher modes.

See Question&Answers more detail:os

与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…
Welcome To Ask or Share your Answers For Others

1 Reply

0 votes
by (71.8m points)

Response from Microsoft:

RijndaelManaged class and AesCryptoServiceProvider class are two different implementations. RijndaelManaged class is a kind of implementation of Rijndael algorithm in .net framework, which was not validated under NIST (National Institute of Standards and Technology) Cryptographic Module Validation Program (CMVP).

However, AesCryptoServiceProvider class calls the Windows Crypto API, which uses RSAENH.DLL, and has been validated by NIST in CMVP. Although Rijndael algorithm was the winner of the NIST competition to select the algorithm that would become AES, there are some differences between Rijndael and official AES. Therefore, RijndaelManaged class and AesCryptoServiceProvider class have subtle differences on implementation.

In addition, RijndaelManaged class cannot provide an equivalent implementation with AES. There is another class implemented in .net framework, AesManaged class. This class just wrapped RijndaelManaged class with a fixed block size and iteration count to achieve the AES standard. However, it does not support the feedback size, especially, when the mode is set as CFB or OFB, the CryptographicException will be thrown.

For more information, please refer to the following MSDN documents.

AesManaged Class and AesManaged.Mode Property

If you want to pick up standard AES as security algorithm in your application, we recommend using the AesCryptoServiceProvider class. If you want to mix the RijndaelManged class and AesCryptoServiceProvider class in your application, we suggest using CBC mode instead of CFB mode in your program, since the implementation of the CBC mode in both classes is the same.


与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…
OGeek|极客中国-欢迎来到极客的世界,一个免费开放的程序员编程交流平台!开放,进步,分享!让技术改变生活,让极客改变未来! Welcome to OGeek Q&A Community for programmer and developer-Open, Learning and Share
Click Here to Ask a Question

1.4m articles

1.4m replys

5 comments

57.0k users

...