Date: Thu, 22 Nov 2001 23:01:53 -0800 (PST) From: Julian Elischer <julian@elischer.org> To: Terry Lambert <tlambert2@mindspring.com> Cc: Marcel Moolenaar <marcel@xcllnt.net>, John Baldwin <jhb@FreeBSD.ORG>, arch@FreeBSD.ORG Subject: Re: Kernel Thread scheduler Message-ID: <Pine.BSF.4.21.0111222257010.46415-100000@InterJet.elischer.org> In-Reply-To: <3BFDD227.8F9A9540@mindspring.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 22 Nov 2001, Terry Lambert wrote: > Julian Elischer wrote: > > > > Interstingly only about 4 people downloaded all 7 pictures. > > (according to the logs).. This is a bit disappointing. > > Quite a few more looked at the first one, and then didn;t bother with the > > rest.. I wonder if they had problems with it? was it impossible to look at > > them for some people? Should I have done something different with them? > > Or maybe the topic of shecduling thasks with multiple threads is just > > boring :-) > > I am incredible disagreement about the idea of virtual processors > (KSEG's), but then you know that from two years ago. I believe things have gone forward since then. KSEGs are not virtual processors. They are simply an entity which collects statistics for, and acts as a contact point for, a group of KSEs which are sharing an allotment of CPU. The KSEs in turn are co-operating to execute asyncronous syscalls on behalf of the user. Each active syscall is a 'thread' in the kernel and may individually block or run. > > It won't scale. There's no need to repeat everything ad infinitum, > just because it's easier to implement. The resulting code is actually > _less_ complex, though its conceptual complexity is higher. "Repeat what?" I ask.. I see no reason that it will not scale. > > If you go don the virtual processor route, expect me to not ride in > the bandwagon. > > -- Terry > 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?Pine.BSF.4.21.0111222257010.46415-100000>