Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 20 Mar 2003 11:49:06 -0300
From:      Mario Sergio Fujikawa Ferreira <lioux@FreeBSD.org>
To:        Phil Regnauld <regnauld@catpipe.net>
Cc:        Anton Berezin <tobez@tobez.org>, Dag-Erling Sm?rgrav <des@ofug.org>, Wes Peters <wes@softweyr.com>, Juli Mallett <jmallett@FreeBSD.org>, Eivind Eklund <eivind@FreeBSD.org>, Mike Silbersack <silby@silby.com>, David Schultz <das@FreeBSD.org>, src-committers@FreeBSD.org, cvs-src@FreeBSD.org, cvs-all@FreeBSD.org
Subject:   Re: cvs commit: src/sys/vm vm_map.c vm_map.h vm_pageout.c
Message-ID:  <20030320144928.46460.qmail@exxodus.fedaykin.here>
In-Reply-To: <20030320085707.GD47243@catpipe.net>
References:  <200303122313.h2CNDHMU046431@repoman.freebsd.org> <20030314012954.A42430@FreeBSD.org> <20030314101857.A98861@FreeBSD.org> <200303171156.40901.wes@softweyr.com> <20030319143619.GA47243@catpipe.net> <xzpu1dzyohn.fsf@flood.ping.uio.no> <20030320081820.GA81566@heechee.tobez.org> <20030320085707.GD47243@catpipe.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, Mar 20, 2003 at 09:56:45AM +0100, Phil Regnauld wrote:
> Anton Berezin (tobez) writes:
> > > 
> > > man setrlimit
> > 
> > setrlimit is a contract signed by a process, while what Phil suggests is
> > a contract signed by a process _and_ the kernel.
> 
> 	Actually, I was thinking about the possibility to extend setrlimit
> 	semantics to support this.  i.e. an extra parameter which would
> 	indicate that we want to be notified (via the return value)
> 	that our contract request could not be fulfilled.   Using this
> 	extra parameter would modify's the kernel's behaviour in that
> 	the kernel will "agree" to preserver this process as much as possible
> 	in case of resource starvation (that's the main difference with
> 	setrlimit).

	One thing though. This would only be safe within
trust boundaries. Suppose a process requests the kernel for a
contract of all available memory. No other processes will be able
to request contracts anymore.
	I am not saying this shouldn't be implemented. I am just
voicing a possible drawback of the solution.

-- 
Mario S F Ferreira - DF - Brazil - "I guess this is a signature."
FreeBSD Committer | CS Developer
flames to beloved devnull@someotherworldbeloworabove.org
feature, n: a documented bug | bug, n: an undocumented feature

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




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030320144928.46460.qmail>