Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Sun, 20 Nov 2011 05:07:18 +0400
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: Was: RE: [john-users] comparing two benchmarks

Jim -

On Fri, Nov 11, 2011 at 12:48:52PM -0600, jfoug wrote:
> I have patched the script.  The parsing only worked for systems which listed
> real: "Cnt[K/M] c/s  virtual: Cnt[K/M] c/s"    Cygwin/mingw (and likely
> others), do not have this format.  They simply have " Cnt[K/M] c/s"

Thank you for noticing this issue.  I completely forgot about that other
output format.  In fact, I think we can change the Cygwin build to
output virtual times as well by now.  The reason why those were disabled
was that they'd be erroneous on Windows 9x.  Even Windows NT 4 already
displayed them correctly, if enabled.

> Thus, I put in a quick 'fix'.  If the parsing cannot find the first form, it
> tries for the form that does not have virtual.  If it finds that, then after
> building the 'real' time, it sets virtual to be the same as real.  Thus all
> the percentages are the same for real/virt in the output, but so what, the
> script works, and the change was trivial.
> 
> It's up on the wiki, as patch 0008.  The script 'fixed' was the original
> jumbo-8

I ended up patching this slightly differently for the revision of the
script that I plan to include in 1.7.9.  Attached in case you want to
test this one as well.

Thanks again,

Alexander

View attachment "relbench.pl" of type "text/plain" (4725 bytes)

Powered by blists - more mailing lists

Your e-mail address:

Powered by Openwall GNU/*/Linux - Powered by OpenVZ