Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Mon, 17 Aug 2015 19:09:53 +0300
From: Solar Designer <>
Subject: Re: BENCHMARK_LENGTH bugs

On Mon, Aug 17, 2015 at 06:06:54PM +0200, Frank Dittrich wrote:
> On 08/17/2015 05:22 PM, Solar Designer wrote:
> > The "cq" format, presumably like many others, wrongly set
> > BENCHMARK_LENGTH to -1, thereby inhibiting the Many salts vs. Only one
> > salt benchmarks.  Yet it's a fast, salted format, for which separate
> > benchmarks are needed.  The attached patch corrects this.
> If we fix that for several formats, can/should relbench somehow be
> changed to compare against "Raw:" speed instead of "Many salts:" speed,
> and vice versa? May be depending on a new option?

Maybe.  Feel free to propose a patch.

> Otherwise, we'll have several formats where we can't compare speeds...

Only between versions before/after this change.


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.