Date: Wed, 7 Jul 2004 13:23:41 -0400 From: Kenneth Culver <culverk@sweetdreamsracing.biz> To: Niki Denev <nike_d@cytexbg.com> Cc: Robert Watson <rwatson@freebsd.org> Subject: Re: cvs commit: src/sys/kern sched_ule.c Message-ID: <20040707132341.8nkpl1ngg4cw0wwc@www.sweetdreamsracing.biz> In-Reply-To: <cone.1089217959.659982.586.1001@phobos.totalterror.net> References: <20040707155348.GC75364@cell.sick.ru> <Pine.NEB.3.96L.1040707120618.37929B-100000@fledge.watson.org> <20040707161049.GA75603@cell.sick.ru> <cone.1089217959.659982.586.1001@phobos.totalterror.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Quoting Niki Denev <nike_d@cytexbg.com>: > Gleb Smirnoff writes: > >> On Wed, Jul 07, 2004 at 12:06:34PM -0400, Robert Watson wrote: >> R> > J> > Modified files: >> R> > J> > sys/kern sched_ule.c >> R> > J> > Log: >> R> > J> > Temporarily disable preemption in SCHED_ULE due to >> reported panics and >> R> > J> > hangs due to recent preemption changes. This change >> appears to remove >> R> > J> > the panic that I was running into, but at the cost of increasing >> R> > J> > ithread scheduling latency, and as such is a temporary >> band-aid until >> R> > J> > jhb has a chance to resolve the ule<->preemption >> interaction that is >> R> > J> > the source of the problem. If it doesn't fix the >> problem for others-- >> R> > J> > sorry! >> R> > J> R> > J> It it just broken on SMP? I actually did the commit >> from a UP preemption R> > J> kernel using SCHED_ULE from within X11 >> with KDE, etc. running. >> R> > R> > it is broken on UP, too. See thread in -current. R> R> An >> SMP kernel on a UP system, or a UP kernel on a UP system? >> >> UP kernel on a UP system >> >> -- Totus tuus, Glebius. >> GLEBIUS-RIPN GLEB-RIPE > > Well, i'm running UP kernel (and without APIC) on UP machine (Laptop IBM TP > X31), cvsupped on 5-th, 2 days ago. With SCHED_ULE, and i have not > experienced any signs of instability... not a single panic :) > > -- > niki Same here, except I'm running with apic on my x86 install, and I also tried to reproduce this on my amd64 install. Neither one misbehaved at all. Ken
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040707132341.8nkpl1ngg4cw0wwc>