Re: counter mode and data integrity
From: Andrew Swallow (am.swallow_at_eatspam.btinternet.com)
Date: 10/31/03
- Next message: Marcel Martin: "Re: NSA chooses ECC"
- Previous message: John Hairell: "Re: NSA chooses ECC"
- In reply to: Joerg Platte: "Re: counter mode and data integrity"
- Next in thread: flip: "Re: counter mode and data integrity"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Date: Fri, 31 Oct 2003 16:46:30 +0000 (UTC)
"Joerg Platte" <Joerg.Platte@uni-dortmund.de> wrote in message
news:bnu0dc$lcc$1@nx6.HRZ.Uni-Dortmund.DE...
> Michael Amling wrote:
>
> > If you're storing a hash-derived counter value, you may as well store
> > a sequence-derived counter value.
> > Your goal was "But I don't want to store the counter value for every
> > block to save some bytes."
>
> Yes. But to ensure data integrity I need a hash value. And if this value
can
> also be the counter it's possible to omit an additional counter.
>
Are you storing data like the number of times the record has
been updated or the time of the last update?
Store this information in plain text and it can be used to form a field
within the counter. AES counters are 128 bits long, probably longer
than a numeric index field.
Andrew Swallow.
- Next message: Marcel Martin: "Re: NSA chooses ECC"
- Previous message: John Hairell: "Re: NSA chooses ECC"
- In reply to: Joerg Platte: "Re: counter mode and data integrity"
- Next in thread: flip: "Re: counter mode and data integrity"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Relevant Pages
|
