[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20120917073532.GM9428@port70.net>
Date: Mon, 17 Sep 2012 09:35:32 +0200
From: Szabolcs Nagy <nsz@...t70.net>
To: musl@...ts.openwall.com
Subject: Re: musl 0.9.5 release and new website
* Rich Felker <dalias@...ifal.cx> [2012-09-16 23:02:41 -0400]:
> On Sun, Sep 16, 2012 at 11:42:08PM +0200, Szabolcs Nagy wrote:
> > is the 30K key limit reasonable?
>
> I don't know; can you explain the motivation?
>
allowing 1G long key is clearly wrong because of dos
disallowing 20 byte key is wrong as well
so i picked a random number in between
we can use 256 to be consistent with sha crypt
but md5 crypt is less attackable this way
> > -#define FF(a,b,c,d,w,s,t) a += F(b,c,d) + w + t; a = rol(a,s) + b
> > -#define GG(a,b,c,d,w,s,t) a += G(b,c,d) + w + t; a = rol(a,s) + b
> > -#define HH(a,b,c,d,w,s,t) a += H(b,c,d) + w + t; a = rol(a,s) + b
> > -#define II(a,b,c,d,w,s,t) a += I(b,c,d) + w + t; a = rol(a,s) + b
> > +#define FF(a,b,c,d,w,r,t) a += F(b,c,d) + w + t; a = rol(a,r) + b
> > +#define GG(a,b,c,d,w,r,t) a += G(b,c,d) + w + t; a = rol(a,r) + b
> > +#define HH(a,b,c,d,w,r,t) a += H(b,c,d) + w + t; a = rol(a,r) + b
> > +#define II(a,b,c,d,w,r,t) a += I(b,c,d) + w + t; a = rol(a,r) + b
>
> Is this changing anything but the argument name? Why the change?
>
yes
(r is for rot, t is for tab, i think it helps when
there are so many arguments)
> > +static const uint8_t idx[64] = {
> > +0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,
> > +1,6,11,0,5,10,15,4,9,14,3,8,13,2,7,12,
> > +5,8,11,14,1,4,7,10,13,0,3,6,9,12,15,2,
> > +0,7,14,5,12,3,10,1,8,15,6,13,4,11,2,9
> > +};
> > +static const uint8_t rot[64] = {
> > +7,12,17,22,7,12,17,22,7,12,17,22,7,12,17,22,
> > +5,9,14,20,5,9,14,20,5,9,14,20,5,9,14,20,
> > +4,11,16,23,4,11,16,23,4,11,16,23,4,11,16,23,
> > +6,10,15,21,6,10,15,21,6,10,15,21,6,10,15,21
>
> It would be nice if these could be done without tables. As-is, I'm not
> really sure the the de-unrolled code is all that much cleaner than the
> original, but at least it's slightly smaller...
>
if they are calculated inline then the code is even more slow
but not really smaller (the two tables are 128 bytes)
and not really cleaner:
...
for (; i < 48; i++) {
static const uint8_t rot[] = {4,11,16,23};
HH(a,b,c,d,W[(3*i+5)%16],rot[i%4],tab[i]);
x = d; d = c; c = b; b = a; a = x;
}
...
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
