This discussion is archived
1 2 Previous Next 17 Replies Latest reply: Apr 30, 2006 10:04 PM by 843853 Go to original post RSS
  • 15. Re: CRC32C checksum in Java
    843853 Newbie
    Currently Being Moderated
    I just thought of something. I tried byte swapping the result
    from the first test and adding it to resulting test data like this:

         byte[] rawData = {
         0x01, (byte)0xC0, 0x00, 0x00,
         0x00, 0x00, 0x00, 0x00,
         0x00, 0x00, 0x00, 0x00,
         0x00, 0x00, 0x00, 0x00,
         0x01, (byte)0xFE, 0x60, (byte)0xAC,
         0x00, 0x00, 0x00, 0x08,
         0x00, 0x00, 0x00, 0x04,
         0x00, 0x00, 0x00, 0x09,
         0x25, 0x00, 0x00, 0x00,
         0x00, 0x00, 0x00, 0x00,
         0x00, 0x00, 0x00, 0x00,
         0x00, 0x00, 0x00, 0x00,
         (byte)0xeb, 0x75, 0x4f, 0x66
    };

    and got a result of 0. So, maybe I have to byte swap the result?

    If that's the case, is the test data in network (big endian) order? or does
    it need to be swapped also, perhaps? I'll try that and see what happens

    Thanks,
    Eric
  • 16. Re: CRC32C checksum in Java
    843853 Newbie
    Currently Being Moderated
    and get the results, in this case 0x664f75eb, that
    you should be able
    to append that result to the original data, run the
    test again and get a result of 0.
    No!

    <snip>
    Am I misunderstanding
    omething about how CRC32C works, or is there a
    problem with the code?
    You are 'misunderstanding somthing'!
  • 17. Re: CRC32C checksum in Java
    843853 Newbie
    Currently Being Moderated
    I suspect your test is invalid ...
    This implementation does ultimately work. I accidentally munged some of
    my test data for a while (which caused no end of grief). But then I went back
    to the iSCSI rfc and realized that I also need to take the results from the above code, do a 1s complement on it and byte swap it. After all that, the
    results match test data.

    Thanks for the help!!
    Eric
    data on one run and then

    Edited by: EJP on 9/02/2012 10:04: redacted
1 2 Previous Next