From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from e23smtp08.au.ibm.com (e23smtp08.au.ibm.com [202.81.31.141]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ozlabs.org (Postfix) with ESMTPS id 7F8331400BE for ; Tue, 8 Apr 2014 18:25:27 +1000 (EST) Received: from /spool/local by e23smtp08.au.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Tue, 8 Apr 2014 18:25:19 +1000 Received: from d23relay03.au.ibm.com (d23relay03.au.ibm.com [9.190.235.21]) by d23dlp01.au.ibm.com (Postfix) with ESMTP id 8745E2CE803F for ; Tue, 8 Apr 2014 18:25:17 +1000 (EST) Received: from d23av01.au.ibm.com (d23av01.au.ibm.com [9.190.234.96]) by d23relay03.au.ibm.com (8.13.8/8.13.8/NCO v10.0) with ESMTP id s388P3SL10486156 for ; Tue, 8 Apr 2014 18:25:03 +1000 Received: from d23av01.au.ibm.com (localhost [127.0.0.1]) by d23av01.au.ibm.com (8.14.4/8.14.4/NCO v10.0 AVout) with ESMTP id s388P9YY018639 for ; Tue, 8 Apr 2014 18:25:09 +1000 Message-ID: <5343B246.1080909@linux.vnet.ibm.com> Date: Tue, 08 Apr 2014 13:54:38 +0530 From: "Srivatsa S. Bhat" MIME-Version: 1.0 To: Michael wang Subject: Re: [PATCH v2] power, sched: stop updating inside arch_update_cpu_topology() when nothing to be update References: <533B8431.8090507@linux.vnet.ibm.com> <53436AC8.5020705@linux.vnet.ibm.com> In-Reply-To: <53436AC8.5020705@linux.vnet.ibm.com> Content-Type: text/plain; charset=ISO-8859-1 Cc: sfr@canb.auug.org.au, LKML , paulus@samba.org, alistair@popple.id.au, nfont@linux.vnet.ibm.com, Andrew Morton , rcj@linux.vnet.ibm.com, linuxppc-dev@lists.ozlabs.org List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On 04/08/2014 08:49 AM, Michael wang wrote: > Since v1: > Edited the comment according to Srivatsa's suggestion. > > During the testing, we encounter below WARN followed by Oops: > > WARNING: at kernel/sched/core.c:6218 > ... > NIP [c000000000101660] .build_sched_domains+0x11d0/0x1200 > LR [c000000000101358] .build_sched_domains+0xec8/0x1200 > PACATMSCRATCH [800000000000f032] > Call Trace: > [c00000001b103850] [c000000000101358] .build_sched_domains+0xec8/0x1200 > [c00000001b1039a0] [c00000000010aad4] .partition_sched_domains+0x484/0x510 > [c00000001b103aa0] [c00000000016d0a8] .rebuild_sched_domains+0x68/0xa0 > [c00000001b103b30] [c00000000005cbf0] .topology_work_fn+0x10/0x30 > ... > Oops: Kernel access of bad area, sig: 11 [#1] > ... > NIP [c00000000045c000] .__bitmap_weight+0x60/0xf0 > LR [c00000000010132c] .build_sched_domains+0xe9c/0x1200 > PACATMSCRATCH [8000000000029032] > Call Trace: > [c00000001b1037a0] [c000000000288ff4] .kmem_cache_alloc_node_trace+0x184/0x3a0 > [c00000001b103850] [c00000000010132c] .build_sched_domains+0xe9c/0x1200 > [c00000001b1039a0] [c00000000010aad4] .partition_sched_domains+0x484/0x510 > [c00000001b103aa0] [c00000000016d0a8] .rebuild_sched_domains+0x68/0xa0 > [c00000001b103b30] [c00000000005cbf0] .topology_work_fn+0x10/0x30 > ... > > This was caused by that 'sd->groups == NULL' after building groups, which > was caused by the empty 'sd->span'. > > The cpu's domain contained nothing because the cpu was assigned to a wrong > node, due to the following unfortunate sequence of events: > > 1. The hypervisor sent a topology update to the guest OS, to notify changes > to the cpu-node mapping. However, the update was actually redundant - i.e., > the "new" mapping was exactly the same as the old one. > > 2. Due to this, the 'updated_cpus' mask turned out to be empty after exiting > the 'for-loop' in arch_update_cpu_topology(). > > 3. So we ended up calling stop-machine() with an empty cpumask list, which made > stop-machine internally elect cpumask_first(cpu_online_mask), i.e., CPU0 as > the cpu to run the payload (the update_cpu_topology() function). > > 4. This causes update_cpu_topology() to be run by CPU0. And since 'updates' > is kzalloc()'ed inside arch_update_cpu_topology(), update_cpu_topology() > finds update->cpu as well as update->new_nid to be 0. In other words, we > end up assigning CPU0 (and eventually its siblings) to node 0, incorrectly. > > Along with the following wrong updating, it causes the sched-domain rebuild > code to break and crash the system. > > Fix this by skipping the topology update in cases where we find that > the topology has not actually changed in reality (ie., spurious updates). > > CC: Benjamin Herrenschmidt > CC: Paul Mackerras > CC: Nathan Fontenot > CC: Stephen Rothwell > CC: Andrew Morton > CC: Robert Jennings > CC: Jesse Larrew > CC: "Srivatsa S. Bhat" > CC: Alistair Popple > Suggested-by: "Srivatsa S. Bhat" > Signed-off-by: Michael Wang > --- Reviewed-by: Srivatsa S. Bhat Regards, Srivatsa S. Bhat > arch/powerpc/mm/numa.c | 15 +++++++++++++++ > 1 file changed, 15 insertions(+) > > diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c > index 30a42e2..4ebbb9e 100644 > --- a/arch/powerpc/mm/numa.c > +++ b/arch/powerpc/mm/numa.c > @@ -1591,6 +1591,20 @@ int arch_update_cpu_topology(void) > cpu = cpu_last_thread_sibling(cpu); > } > > + /* > + * In cases where we have nothing to update (because the updates list > + * is too short or because the new topology is same as the old one), > + * skip invoking update_cpu_topology() via stop-machine(). This is > + * necessary (and not just a fast-path optimization) since stop-machine > + * can end up electing a random CPU to run update_cpu_topology(), and > + * thus trick us into setting up incorrect cpu-node mappings (since > + * 'updates' is kzalloc()'ed). > + * > + * And for the similar reason, we will skip all the following updating. > + */ > + if (!cpumask_weight(&updated_cpus)) > + goto out; > + > stop_machine(update_cpu_topology, &updates[0], &updated_cpus); > > /* > @@ -1612,6 +1626,7 @@ int arch_update_cpu_topology(void) > changed = 1; > } > > +out: > kfree(updates); > return changed; > } >