2008-12-05 14:54:14

by Jarod Wilson

[permalink] [raw]
Subject: Re: [PATCH] crypto: des: fix inverted weak key flag check

Jarod Wilson wrote:
> Based on the fact the testmgr code sets the weak key flag for
> vectors with wk = 1, the weak key flag check in des_setkey()
> is backwards. Also, add a warning when a weak key is rejected,
> otherwise you silently get a bogus result back.
>
> Signed-off-by: Jarod Wilson <[email protected]>

Self-NAK. Due to mail server issues, this might show up before the
original...

My understanding of the flag's meaning appears to be backwards, Herbert
pointed out that the use of this flag is self-consistent. Also, the
printk probably shouldn't be there, the caller should check retval and
complain if need be.


--
Jarod Wilson
[email protected]