Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 14 Mar 2002 03:17:03 -0800
From:      Michael Smith <msmith@freebsd.org>
To:        Garance A Drosihn <drosih@rpi.edu>
Cc:        Dag-Erling Smorgrav <des@ofug.org>, Josef Karthauser <joe@tao.org.uk>, arch@FreeBSD.ORG, peter@FreeBSD.ORG, jake@FreeBSD.ORG
Subject:   Re: dumpsys() rewrite 
Message-ID:  <200203141117.g2EBH4D02622@mass.dis.org>
In-Reply-To: Your message of "Wed, 13 Mar 2002 10:42:41 EST." <p05101560b8b522a7f8a1@[128.113.24.47]> 

next in thread | previous in thread | raw e-mail | index | archive | help
> >dumpsys() controls what is written and when.  It is perfectly
> >conceivable for dumpsys() to do the same kind of hole compression
> >that savecore(1) does, but it would be *much* slower and ...
> 
> Would it make any sense to compress it in the 'gzip' sense of
> the word (or some simpler algorithm)?  Or does it already do
> some of that?

The problem here is that when dumpsys is called, the system is in an 
unknown but probably hosed state.  Compression algorithms typically 
require considerable amounts of memory, which would have to be tied down 
at startup to avoid depending on a possibly broken, deadlocked or corrupt 
allocator at dump time.

-- 
To announce that there must be no criticism of the president,
or that we are to stand by the president, right or wrong, is not
only unpatriotic and servile, but is morally treasonable to 
the American public.  - Theodore Roosevelt



To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-arch" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200203141117.g2EBH4D02622>