From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ryan Harper Subject: Re: x86_64 DOM0 SMP oops Date: Mon, 22 Aug 2005 14:32:51 -0500 Message-ID: <20050822193251.GJ5523@us.ibm.com> References: <20050822185220.GH5523@us.ibm.com> <20050822191254.GM7901@shell0.pdx.osdl.net> <20050822191500.GI5523@us.ibm.com> <20050822192029.GA7762@shell0.pdx.osdl.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20050822192029.GA7762@shell0.pdx.osdl.net> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Chris Wright Cc: xen-devel@lists.xensource.com List-Id: xen-devel@lists.xenproject.org * Chris Wright [2005-08-22 14:26]: > * Ryan Harper (ryanh@us.ibm.com) wrote: > > > > Unable to handle kernel NULL pointer dereference at 000000000000001c RIP: > > > > {retint_kernel+0} > > > > PGD aae067 PUD aaf067 PMD 0 > > > > Oops: 0000 [1] PREEMPT SMP > > > > > > Not surprising, w/ PREEMPT. Is it stable w/out PREEMPT? > > > > Really? I thought CONFIG_PREEMPT was tested since it was on for x86_32 > > SMP. > > No, I expect it to be broken (and you seem to be able to easily prove > that ;-) > > > I'll give that a shot and let you know. That seems to have done it. Built several kernels back-to-back, no oops. Now if I could just get this to go away: Timer ISR/1: Time went backwards: delta=-169458624 cpu_delta=1110558976 shadow=57067465250 off=133093726 processed=57370017600 cpu_processed=56090000000 0: 57370017600 1: 56090000000 Anyone else with a dual-opteron see this? It shows up on everything (x86_64 UP, x86_32 UP/SMP). -- Ryan Harper Software Engineer; Linux Technology Center IBM Corp., Austin, Tx (512) 838-9253 T/L: 678-9253 ryanh@us.ibm.com