From mboxrd@z Thu Jan 1 00:00:00 1970 Subject: Re: PREEMPT on PPC From: Michel =?ISO-8859-1?Q?D=E4nzer?= To: Benjamin Herrenschmidt Cc: linuxppc-dev list , "debian-powerpc@lists.debian.org" In-Reply-To: <1080523601.1209.46.camel@gaston> References: <1080197240.1210.2.camel@gaston> <1080201697.1217.3.camel@gaston> <1080399924.2514.14.camel@thor.asgaard.local> <1080523601.1209.46.camel@gaston> Content-Type: text/plain; charset=UTF-8 Message-Id: <1080524303.2441.108.camel@thor.asgaard.local> Mime-Version: 1.0 Date: Mon, 29 Mar 2004 03:38:24 +0200 Sender: owner-linuxppc-dev@lists.linuxppc.org List-Id: On Mon, 2004-03-29 at 03:26, Benjamin Herrenschmidt wrote: > > I appreciate a lot that you're working on this, but unfortunately, this > > patch doesn't seem to make any difference here. I still get random > > segfaults, weird behaviour in galeon, ... as soon as I hit swap. How can > > I provide useful debugging information? > > In general, is there a relationship with highmem ? that is do you > have highmem enabled and if yes, does it get better without it ? It's enabled, but diasbling it didn't seem to make a difference either. FWIW, http://penguinppc.org/~daenzer/config-2.6.5-rc2-ben0-ck1 is my current, preempt-less config. Note that I'm not sure how tightly the problems are tied to swap; after they start, they seem to persist even when swap is no longer used. It's just the best criterion I've been able to identify so far. -- Earthling Michel Dänzer | Debian (powerpc), X and DRI developer Libre software enthusiast | http://svcs.affero.net/rm.php?r=daenzer ** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/