phc-discussions - Re: [PHC] A review per day

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 linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date: Sat, 30 Aug 2014 07:35:29 -0400
From: Bill Cox <waywardgeek@...il.com>
To: "discussions@...sword-hashing.net" <discussions@...sword-hashing.net>
Subject: Re: [PHC] A review per day - TwoCats

On Sat, Aug 30, 2014 at 12:50 AM, Solar Designer <solar@...nwall.com> wrote:

> > How fast does bcrypt do random reads?
>
> According to that posting I referenced above, it's:
>
> 2176000*3072 = 6.7 billion/s per FX-8120 chip
> 2176000*493 = 1.1 billion/s with 1 instance (with other cores idle)
>
> That's for defensive-use bcrypt code, which is what's relevant here.
>
> So your "2.8B 16-byte random reads per second on Ivy Bridge" is very
> good if it's for 1 thread, but not good enough if it's for entire chip.
> And you also need to compare the available parallelism vs. bcrypt's.
>

2.8B reads/s was for 4 threads on my quad-core Ivy Bridge processor.  I
tweaked the inner loop just a bit, and benchmarked it against the simplest
loop I could write for doing unpredictable reads.  Here's my one thread
numbers vs what is possible:

Max possible w/ no parallelism: 0.56B/s
TwoCats w/ no parallelism: 0.435B/s
Max possible w/ 4-way parallelism: 1.28B/s

TwoCats is very close to maxed-out speed-wise for no parallelism.  However,
bcrypt has 4-way parallelism in the reads, meaning that we can do all four
reads in parallel.  When I unroll the inner loop 4 times and read 4
different locations in parallel in my test code, I get 1.29B reads/second.
This should match bcrypt's upper speed fairly well, assuming that the
rounds computations dominate.  I seem to be doing about 1/3 as well as
Bcrypt on one thread.  Not as good as Yescrypt, but not too bad :-)

Bill

Content of type "text/html" skipped