lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  PHC 
Open Source and information security mailing list archives
 
Hash Suite for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Thu, 3 Apr 2014 00:42:42 +0000
From: Brandon Enright <bmenrigh@...ndonenright.net>
To: discussions@...sword-hashing.net
Cc: waywardgeek@...il.com, bmenrigh@...ndonenright.net
Subject: Re: [PHC] data-dependent branching (Re: [PHC] A little nit which
 bothers me...)

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Wed, 2 Apr 2014 20:29:07 -0400
Bill Cox <waywardgeek@...il.com> wrote:

> Data dependent branching doesn't seem to buy as much ASIC resistance
> as some PHC submitters seem to think, IMO.  In an ASIC or FPGA, the
> if-then-else statements synthesize to muxes in hardware, and then the
> optimizer gets to reuse blocks that are in mutually exclusive block
> statements.

You're right.  The way I see it:

* Data dependant branching defeats extremely wide SIMD / Stream
  processing (GPUs)

* Large memory usage and random memory access defeats FPGAs and ASICs

One could probably argue that large memory and random memory accesses
also defeat GPUs but that hasn't actually panned out very well.  From
what I've read, the 128 KiB scrypt usage for litecoin is a few orders of
magnitude faster on GPUs than CPUs.  Yeah 128 KiB isn't much memory but
still, I think that's enough cause for concern for memory usage being
the only protection against GPUs.

Brandon

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (GNU/Linux)

iEYEARECAAYFAlM8rogACgkQqaGPzAsl94IgcwCfYRUl/1J17oGdnI6UvdSnDaJY
tNsAn0TBA2sVeuMzkOg0SQKcgpXEFMTu
=j3+S
-----END PGP SIGNATURE-----

Powered by blists - more mailing lists