Subject: Re: Additional features for veriexecgen(8)
To: None <tech-userlevel@NetBSD.org>
From: Elad Efrat <elad@NetBSD.org>
Date: 10/09/2006 23:34:33
David Young wrote:
> Elad, my proposal does not entail any regression. The script that would
> invoke veriexecgen could understand all of the same flags. I thought
> that much would be understood. Sorry I was not clear.
> The performance that M J mentions is certainly important. I believe
> performance of find | veriexecgen would be alike to veriexecgen itself
> on most of my servers, but they have lots of memory dedicated to inode
> caching, and not every server does.
That is nearly 14 minute run-time compared with less than a second,
generated on my 2ghz/1gb ram amd64 machine. There is no way we are going
back to maintain a shell script.