Date: Tue, 26 Mar 2002 12:03:10 -0800 From: Terry Lambert <tlambert2@mindspring.com> To: Dan Nelson <dnelson@allantgroup.com> Cc: Andrew <andrew@ugh.net.au>, Volker Stolz <stolz@hyperion.informatik.rwth-aachen.de>, Ian <freebsd@damnhippie.dyndns.org>, hackers@FreeBSD.ORG Subject: Re: idprio Message-ID: <3CA0D3FE.8113515C@mindspring.com> References: <20020326121046.A3952@margaux.informatik.rwth-aachen.de> <20020326232510.A21338-100000@starbug.ugh.net.au> <20020326153541.GF17825@dan.emsphone.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Dan Nelson wrote: > In the last episode (Mar 26), Andrew said: > > On Tue, 26 Mar 2002, Volker Stolz wrote: > > > Under FreeBSD system calls are currently never preempted, therefore > > > non- realtime processes can starve realtime processes, or idletime > > > processes can starve normal priority processes. > > > > Even so an idprio process can't be worse than a normal process. > > Sure it can, if the idprio process has locked a vnode trying to update > the contents of a file, and another non-idprio process starts consuming > 100% CPU. The idprio process never gets a chance to run again, and if > that vnode happened to be an important one (say for /), you may not be > able to kill the other process without rebooting. You meant that when you use priorities, you risk priority inversion? 8-p -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3CA0D3FE.8113515C>