Re: Can't set CRC result by inital value?
From: Oliver Betz (obetz_at_despammed.com)
Date: 03/15/05
- Next message: MrPepper11: "Crack in Computer Security Code Raises Red Flag"
- Previous message: spinoza1111_at_yahoo.com: "Re: A unique number for every "person" - can it be done?"
- In reply to: Andrew Swallow: "Re: Can't set CRC result by inital value?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Date: Tue, 15 Mar 2005 14:25:44 +0100
Andrew Swallow <am.swallow@btopenworld.com> wrote:
>> the reason for the question was that I could avoid storage for the
>> expected CRC result if I set the initial value so that it always
>> yields zero.
>
>Then you have to store the initial value so no storage is saved.
sorry, my explanation was too short. I meant RAM storage. If I check
for correct value at the end of a transmission, I must either save the
correct CRC at the beginning, or transmit it at the end.
In my example (microcontroller Flash loader as you might have
guessed), I send BASE64 encoded data in blocks, and I want to store
the last block only if the CRC of the whole transmission is correct
(authentication).
So when I have to decide whether to store or not, the block buffer is
still occupied, therefore "transmit CRC later" is not suitable.
But I can send an initial CRC value at the beginning of the whole
procedure, along with the unique IV necessary for RC4, without
additional effort.
Oliver
-- Oliver Betz, Munich
- Next message: MrPepper11: "Crack in Computer Security Code Raises Red Flag"
- Previous message: spinoza1111_at_yahoo.com: "Re: A unique number for every "person" - can it be done?"
- In reply to: Andrew Swallow: "Re: Can't set CRC result by inital value?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Relevant Pages
|
