[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Sun, 1 Apr 2012 23:02:10 +0300
From: Milen Rangelov <gat3way@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: fast hashes on GPU
Well, those are profiler stats from my sha256 and sha512 kernels:
sha256 (using uint4 vectors):
Used GPRs = 36
ALU ops = 1654
sha512 (using uint2 vectors):
Used GPRs = 51
ALU ops = 5283
So in fact SHA512 is doing ~3.2x more ALU operations to process 2x less
hashes. GPR utilization of 51 badly hampers occupancy as well.
On Sun, Apr 1, 2012 at 7:16 AM, Solar Designer <solar@...nwall.com> wrote:
>
> Yes, I expected that SHA-512 would be less efficient on AMD/ATI GPUs
> compared to many other hash types on the same hardware. 10x slower than
> SHA-256 is extreme, though. I thought/think that the difference should
> be less.
>
> > On NVidia, situation is somewhat better though.
>
> Yes, that's my expectation, although it's largely due to other
> ("competing") hash types being less efficient there (no rotate and
> bitselect anyway). I think the only major reason why SHA-512 may be
> more than twice slower than SHA-256 on current Nvidia GPUs is the
> increased register pressure. Right?
>
> As you probably recall, we picked raw SHA-512 (and its salted variants)
> for one of the first "fast" hashes to experiment with precisely because
> it is on the slower side of the fast hashes and thus should fit in the
> current formats interface better. For faster fast hashes, we'll need to
> proceed with more invasive changes.
>
> Alexander
>
Content of type "text/html" skipped
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.
