Date: Mon, 08 Nov 2004 19:03:07 -0500 From: Stephan Uphoff <ups@freebsd.org> To: Julian Elischer <julian@elischer.org> Cc: julian@freebsd.org Subject: Re: preemption stable under 5.3? Message-ID: <1099958587.24619.453.camel@palm.tree.com> In-Reply-To: <418FE0E8.1020702@elischer.org> References: <Pine.NEB.3.96L.1041108182215.73102M-100000@fledge.watson.org> <418FE0E8.1020702@elischer.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 2004-11-08 at 16:11, Julian Elischer wrote: > Robert Watson wrote: > > >On Mon, 8 Nov 2004, Mipam wrote: > > > > > > > >>Thanks for your reply, okay, then i'd like to enable preemption. I > >>noticed it's not in the GENERIC kernel config file. So: options > >>PREEMPTION would suffice to enable it i guess? Any experience with > >>preemption. noticable changes? So the problem: "PREEMPTION triggers > >>frequent hangs" is resolved? Btw, is RELENG_5 also stable or only for > >>early adopters? I really would like to see ule working stable in > >>combination with preemption, but in 5.3 it won't happen. Maybe ule will > >>be enabled later in the 5 series? > >> > >> > > > >There was a series of bugs in the scheduler which got tickled by > >preemption; I'm unclear as to whether they were all resolved before 5.3 or > >whether they require fixes in HEAD that haven't yet been merged. It may > >well be safe, but I make no promises. Hopefully we can trick Julian or > >John into responding to this thread. :-) Having it off by default on 5.3 > >is certainly the more conservative (and reasonable) position, but if it > >helps your environment and appears stable, there should be no reason not > >to turn it on. It should substantially improve latency in interrupt > >processing as well as packet processing. > > > > I think that PREEMPTION with SCHED_4BSD might be ok.. > It's hard to say because it's always harder to prove something correct > than to prove it broken :-) I agree - PREEMPTION should be ok (On head and on 5.3) as far as the SCHED_4BSD is concerned. ( I ran most tests with PREEMPTION and FULL_PREEMPTION) However running with PREEMPTION may expose locking problems elsewhere. I think at least PREEMPTION (and FULL_PREEMPTION?) should be made standard in head for the archs that support them. ( Otherwise locking will never really get exercised on UP systems) > Hopefully with the rush off, we can sit down and try "prove it ok" and > take some cleanup passes over it. > I still owe my wife a significant "chunk-o-time" (TM) however so count > me out for a while . > Hopefully however ups@ is coming online again this week. I recovered enough to at least debug problems and will resume coding tomorrow. Have fun - I will try to hold the ford ;-) Stephan
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1099958587.24619.453.camel>