[{"id":1787410,"web_url":"http://patchwork.ozlabs.org/comment/1787410/","msgid":"<87y3objlrk.fsf@concordia.ellerman.id.au>","date":"2017-10-16T12:54:39","subject":"Re: [PATCH 2/2] powerpc/hotplug: Ensure nodes initialized for\n\thotplug","submitter":{"id":46580,"url":"http://patchwork.ozlabs.org/api/people/46580/","name":"Michael Ellerman","email":"mpe@ellerman.id.au"},"content":"Michael Bringmann writes:\n\n> powerpc/hotplug: On systems like PowerPC which allow 'hot-add' of CPU,\n> it may occur that the new resources are to be inserted into nodes\n> that were not used for memory resources at bootup. Many different\n> configurations of PowerPC resources may need to be supported depending\n> upon the environment.\n\nGive me some detail please?!\n\n> This patch fixes some problems encountered at\n\nWhat problems?\n\n> runtime with configurations that support memory-less nodes, but which\n> allow CPUs to be added at and after boot.\n\nHow does it fix those problems?\n\n> diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c\n> index b385cd0..e811dd1 100644\n> --- a/arch/powerpc/mm/numa.c\n> +++ b/arch/powerpc/mm/numa.c\n> @@ -1325,6 +1325,17 @@ static long vphn_get_associativity(unsigned long cpu,\n> \treturn rc;\n> }\n> \n> +static int verify_node_preparation(int nid)\n> +{\n\nI would not expect a function called \"verify\" ...\n\n> +\tif ((NODE_DATA(nid) == NULL) ||\n> +\t (NODE_DATA(nid)->node_spanned_pages == 0)) {\n> +\t\tif (try_online_node(nid))\n\n.. to do something like online a node.\n\n> +\t\t\treturn first_online_node;\n> +\t}\n> +\n> +\treturn nid;\n> +}\n> +\n> /*\n> * Update the CPU maps and sysfs entries for a single CPU when its NUMA\n> * characteristics change. This function doesn't perform any locking and is\n> @@ -1433,9 +1444,11 @@ int numa_update_cpu_topology(bool cpus_locked)\n> \t\t/* Use associativity from first thread for all siblings */\n> \t\tvphn_get_associativity(cpu, associativity);\n> \t\tnew_nid = associativity_to_nid(associativity);\n> -\t\tif (new_nid < 0 || !node_online(new_nid))\n> +\t\tif (new_nid < 0 || !node_possible(new_nid))\n> \t\t\tnew_nid = first_online_node;\n> \n> +\t\tnew_nid = verify_node_preparation(new_nid);\n\nYou're being called part-way through CPU hotplug here, are we sure it's\nsafe to go and do memory hotplug from there? What's the locking\nsituation?\n\ncheers","headers":{"Return-Path":"","X-Original-To":["patchwork-incoming@ozlabs.org","linuxppc-dev@lists.ozlabs.org"],"Delivered-To":["patchwork-incoming@ozlabs.org","linuxppc-dev@lists.ozlabs.org"],"Received":["from lists.ozlabs.org (lists.ozlabs.org [IPv6:2401:3900:2:1::3])\n\t(using TLSv1.2 with cipher ADH-AES256-GCM-SHA384 (256/256 bits))\n\t(No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yFyx52r52z9t3Z\n\tfor ;\n\tMon, 16 Oct 2017 23:55:53 +1100 (AEDT)","from lists.ozlabs.org (lists.ozlabs.org [IPv6:2401:3900:2:1::3])\n\tby lists.ozlabs.org (Postfix) with ESMTP id 3yFyx50hzBzDrLK\n\tfor ;\n\tMon, 16 Oct 2017 23:55:53 +1100 (AEDT)","from ozlabs.org (bilbo.ozlabs.org [103.22.144.67])\n\t(using TLSv1.2 with cipher ADH-AES256-GCM-SHA384 (256/256 bits))\n\t(No client certificate requested)\n\tby lists.ozlabs.org (Postfix) with ESMTPS id 3yFyvj11XszDrJ3\n\tfor ;\n\tMon, 16 Oct 2017 23:54:41 +1100 (AEDT)","from authenticated.ozlabs.org (localhost [127.0.0.1])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPSA id 3yFyvh5XVsz9t3R;\n\tMon, 16 Oct 2017 23:54:40 +1100 (AEDT)"],"From":"Michael Ellerman ","To":"Michael Bringmann , linuxppc-dev@lists.ozlabs.org,\n\tlinux-kernel@vger.kernel.org","Subject":"Re: [PATCH 2/2] powerpc/hotplug: Ensure nodes initialized for\n\thotplug","In-Reply-To":"","References":"","Date":"Mon, 16 Oct 2017 23:54:39 +1100","Message-ID":"<87y3objlrk.fsf@concordia.ellerman.id.au>","MIME-Version":"1.0","Content-Type":"text/plain","X-BeenThere":"linuxppc-dev@lists.ozlabs.org","X-Mailman-Version":"2.1.24","Precedence":"list","List-Id":"Linux on PowerPC Developers Mail List\n\t","List-Unsubscribe":",\n\t","List-Archive":"","List-Post":"","List-Help":"","List-Subscribe":",\n\t","Cc":"John Allen ,\n\tMichael Bringmann ,\n\tNathan Fontenot ","Errors-To":"linuxppc-dev-bounces+patchwork-incoming=ozlabs.org@lists.ozlabs.org","Sender":"\"Linuxppc-dev\"\n\t"}},{"id":1788525,"web_url":"http://patchwork.ozlabs.org/comment/1788525/","msgid":"<240499bb-081e-fb9e-1083-bd4c72ef404c@linux.vnet.ibm.com>","date":"2017-10-17T15:08:51","subject":"Re: [PATCH 2/2] powerpc/hotplug: Ensure nodes initialized for\n\thotplug","submitter":{"id":65104,"url":"http://patchwork.ozlabs.org/api/people/65104/","name":"Michael Bringmann","email":"mwb@linux.vnet.ibm.com"},"content":"On 10/16/2017 07:54 AM, Michael Ellerman wrote:\n> Michael Bringmann writes:\n> \n>> powerpc/hotplug: On systems like PowerPC which allow 'hot-add' of CPU,\n>> it may occur that the new resources are to be inserted into nodes\n>> that were not used for memory resources at bootup. Many different\n>> configurations of PowerPC resources may need to be supported depending\n>> upon the environment.\n> \n> Give me some detail please?!\n\nConfigurations that demonstrated problems included 'memoryless' nodes\nthat possessed only CPUs at boot, and nodes that contained neither CPUs\nnor memory at boot. The calculations in the kernel resulted in a different\nnode use layout on many SAP HANA configured systems.\n\n> \n>> This patch fixes some problems encountered at\n> \n> What problems?\n\nThe previous implementation collapsed all node assignments after affinity\ncalculations to use only those nodes that had memory at boot. This\nresulted in calculation and configuration differences between the FSP\ncode and the Linux kernel.\n\n> \n>> runtime with configurations that support memory-less nodes, but which\n>> allow CPUs to be added at and after boot.\n> \n> How does it fix those problems?\n\nThe change involves completing the initialization of nodes that were not\nused at boot, but which were selected by VPHN affinity calculations during\nsubsequent hotplug operations.\n\nMichael\n\n> \n>> diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c\n>> index b385cd0..e811dd1 100644\n>> --- a/arch/powerpc/mm/numa.c\n>> +++ b/arch/powerpc/mm/numa.c\n>> @@ -1325,6 +1325,17 @@ static long vphn_get_associativity(unsigned long cpu,\n>> \treturn rc;\n>> }\n>> \n>> +static int verify_node_preparation(int nid)\n>> +{\n> \n> I would not expect a function called \"verify\" ...\n> \n>> +\tif ((NODE_DATA(nid) == NULL) ||\n>> +\t (NODE_DATA(nid)->node_spanned_pages == 0)) {\n>> +\t\tif (try_online_node(nid))\n> \n> .. to do something like online a node.\n> \n>> +\t\t\treturn first_online_node;\n>> +\t}\n>> +\n>> +\treturn nid;\n>> +}\n>> +\n>> /*\n>> * Update the CPU maps and sysfs entries for a single CPU when its NUMA\n>> * characteristics change. This function doesn't perform any locking and is\n>> @@ -1433,9 +1444,11 @@ int numa_update_cpu_topology(bool cpus_locked)\n>> \t\t/* Use associativity from first thread for all siblings */\n>> \t\tvphn_get_associativity(cpu, associativity);\n>> \t\tnew_nid = associativity_to_nid(associativity);\n>> -\t\tif (new_nid < 0 || !node_online(new_nid))\n>> +\t\tif (new_nid < 0 || !node_possible(new_nid))\n>> \t\t\tnew_nid = first_online_node;\n>> \n>> +\t\tnew_nid = verify_node_preparation(new_nid);\n> \n> You're being called part-way through CPU hotplug here, are we sure it's\n> safe to go and do memory hotplug from there? What's the locking\n> situation?\n> \n> cheers\n> \n>","headers":{"Return-Path":"","X-Original-To":["patchwork-incoming@ozlabs.org","linuxppc-dev@lists.ozlabs.org"],"Delivered-To":["patchwork-incoming@ozlabs.org","linuxppc-dev@lists.ozlabs.org"],"Received":["from lists.ozlabs.org (lists.ozlabs.org [IPv6:2401:3900:2:1::3])\n\t(using TLSv1.2 with cipher ADH-AES256-GCM-SHA384 (256/256 bits))\n\t(No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yGdtL0rd7z9sP1\n\tfor ;\n\tWed, 18 Oct 2017 02:10:50 +1100 (AEDT)","from lists.ozlabs.org (lists.ozlabs.org [IPv6:2401:3900:2:1::3])\n\tby lists.ozlabs.org (Postfix) with ESMTP id 3yGdtK72lzzDrKJ\n\tfor ;\n\tWed, 18 Oct 2017 02:10:49 +1100 (AEDT)","from mx0a-001b2d01.pphosted.com (mx0a-001b2d01.pphosted.com\n\t[148.163.156.1])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby lists.ozlabs.org (Postfix) with ESMTPS id 3yGdrr2vnwzDqlv\n\tfor ;\n\tWed, 18 Oct 2017 02:09:32 +1100 (AEDT)","from pps.filterd (m0098393.ppops.net [127.0.0.1])\n\tby mx0a-001b2d01.pphosted.com (8.16.0.21/8.16.0.21) with SMTP id\n\tv9HF9D9X014079\n\tfor ; Tue, 17 Oct 2017 11:09:30 -0400","from e35.co.us.ibm.com (e35.co.us.ibm.com [32.97.110.153])\n\tby mx0a-001b2d01.pphosted.com with ESMTP id 2dnkpkt59r-1\n\t(version=TLSv1.2 cipher=AES256-SHA bits=256 verify=NOT)\n\tfor ; Tue, 17 Oct 2017 11:09:16 -0400","from localhost\n\tby e35.co.us.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use\n\tOnly! Violators will be prosecuted\n\tfor from ;\n\tTue, 17 Oct 2017 09:08:59 -0600","from b03cxnp08027.gho.boulder.ibm.com (9.17.130.19)\n\tby e35.co.us.ibm.com (192.168.1.135) with IBM ESMTP SMTP Gateway:\n\tAuthorized Use Only! Violators will be prosecuted; \n\tTue, 17 Oct 2017 09:08:58 -0600","from b03ledav002.gho.boulder.ibm.com\n\t(b03ledav002.gho.boulder.ibm.com [9.17.130.233])\n\tby b03cxnp08027.gho.boulder.ibm.com (8.14.9/8.14.9/NCO v10.0) with\n\tESMTP id v9HF8vGQ65470530; Tue, 17 Oct 2017 08:08:57 -0700","from b03ledav002.gho.boulder.ibm.com (unknown [127.0.0.1])\n\tby IMSVA (Postfix) with ESMTP id A8109136043;\n\tTue, 17 Oct 2017 09:08:57 -0600 (MDT)","from oc1554177480.ibm.com (unknown [9.53.92.241])\n\tby b03ledav002.gho.boulder.ibm.com (Postfix) with ESMTP id\n\t5C581136040; Tue, 17 Oct 2017 09:08:57 -0600 (MDT)"],"Authentication-Results":"ozlabs.org;\n\tspf=none (mailfrom) smtp.mailfrom=linux.vnet.ibm.com\n\t(client-ip=148.163.156.1; helo=mx0a-001b2d01.pphosted.com;\n\tenvelope-from=mwb@linux.vnet.ibm.com; receiver=)","Subject":"Re: [PATCH 2/2] powerpc/hotplug: Ensure nodes initialized for\n\thotplug","To":"Michael Ellerman , linuxppc-dev@lists.ozlabs.org,\n\tlinux-kernel@vger.kernel.org","References":"\n\t<87y3objlrk.fsf@concordia.ellerman.id.au>","From":"Michael Bringmann ","Organization":"IBM Linux Technology Center","Date":"Tue, 17 Oct 2017 10:08:51 -0500","User-Agent":"Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101\n\tThunderbird/52.2.0","MIME-Version":"1.0","In-Reply-To":"<87y3objlrk.fsf@concordia.ellerman.id.au>","Content-Type":"text/plain; charset=utf-8","Content-Language":"en-US","Content-Transfer-Encoding":"8bit","X-TM-AS-GCONF":"00","x-cbid":"17101715-0012-0000-0000-000015289DE3","X-IBM-SpamModules-Scores":"","X-IBM-SpamModules-Versions":"BY=3.00007906; HX=3.00000241; KW=3.00000007;\n\tPH=3.00000004; SC=3.00000237; SDB=6.00932465; UDB=6.00469574;\n\tIPR=6.00712755; \n\tBA=6.00005642; NDR=6.00000001; ZLA=6.00000005; ZF=6.00000009;\n\tZB=6.00000000; \n\tZP=6.00000000; ZH=6.00000000; ZU=6.00000002; MB=3.00017576;\n\tXFM=3.00000015; UTC=2017-10-17 15:08:59","X-IBM-AV-DETECTION":"SAVI=unused REMOTE=unused XFE=unused","x-cbparentid":"17101715-0013-0000-0000-00004FE92240","Message-Id":"<240499bb-081e-fb9e-1083-bd4c72ef404c@linux.vnet.ibm.com>","X-Proofpoint-Virus-Version":"vendor=fsecure engine=2.50.10432:, ,\n\tdefinitions=2017-10-17_11:, , signatures=0","X-Proofpoint-Spam-Details":"rule=outbound_notspam policy=outbound score=0\n\tspamscore=0 suspectscore=0\n\tmalwarescore=0 phishscore=0 adultscore=0 bulkscore=0 classifier=spam\n\tadjust=0 reason=mlx scancount=1 engine=8.0.1-1707230000\n\tdefinitions=main-1710170213","X-BeenThere":"linuxppc-dev@lists.ozlabs.org","X-Mailman-Version":"2.1.24","Precedence":"list","List-Id":"Linux on PowerPC Developers Mail List\n\t","List-Unsubscribe":",\n\t","List-Archive":"","List-Post":"","List-Help":"","List-Subscribe":",\n\t","Cc":"John Allen ,\n\tNathan Fontenot ","Errors-To":"linuxppc-dev-bounces+patchwork-incoming=ozlabs.org@lists.ozlabs.org","Sender":"\"Linuxppc-dev\"\n\t"}},{"id":1805229,"web_url":"http://patchwork.ozlabs.org/comment/1805229/","msgid":"<3be6fd7b-36ec-3beb-21ce-1def39ef0938@linux.vnet.ibm.com>","date":"2017-11-15T18:28:19","subject":"Re: [PATCH 2/2] powerpc/hotplug: Ensure nodes initialized for\n\thotplug","submitter":{"id":65104,"url":"http://patchwork.ozlabs.org/api/people/65104/","name":"Michael Bringmann","email":"mwb@linux.vnet.ibm.com"},"content":"Hello:\n See below.\n\nOn 10/16/2017 07:54 AM, Michael Ellerman wrote:\n> Michael Bringmann writes:\n> \n>> powerpc/hotplug: On systems like PowerPC which allow 'hot-add' of CPU,\n>> it may occur that the new resources are to be inserted into nodes\n>> that were not used for memory resources at bootup. Many different\n>> configurations of PowerPC resources may need to be supported depending\n>> upon the environment.\n> \n> Give me some detail please?!\n\nThe most important characteristics that I have observed are:\n\n* Dedicated vs. shared resources. Shared resources require information\n such as the VPHN hcall for CPU assignment to nodes.\n* memoryless nodes at boot. Nodes need to be defined as 'possible' at\n boot for operation with other code modules. Previously, the powerpc\n code would limit the set of possible/online nodes to those which have\n memory assigned at boot. Subsequent add/remove of CPUs or memory would\n only work with this subset of possible nodes.\n* memoryless nodes with CPUs at boot. Due to the previous restriction on\n nodes, nodes that had CPUs but no memory were being collapsed into other\n nodes that did have memory at boot. In practice this meant that the\n node assignment presented by the runtime kernel differed from the affinity\n and associativity attirbutes presented by the device tree or VPHN hcalls.\n Nodes that might be known to the pHyp were not 'possible' in the runtime\n kernel because they did not have memory at boot.\n\n> \n>> This patch fixes some problems encountered at\n> \n> What problems?\n\nThis patch set fixes a couple of problems.\n\n* Nodes known to powerpc to be memoryless at boot, but to have CPUs in them\n are allowed to be 'possible' and 'online'. Memory allocations for those\n nodes are taken from another node that does have memory until and if memory\n is hot-added to the node.\n* Nodes which have no resources assigned at boot, but which may still be\n referenced subsequently by affinity or associativity attributes, are kept\n in the list of 'possible' nodes for powerpc. Hot-add of memory or CPUs\n to the system can reference these nodes and bring them online instead of\n redirecting the resources to the set of nodes known to have memory at boot.\n\n> \n>> runtime with configurations that support memory-less nodes, but which\n>> allow CPUs to be added at and after boot.\n> \n> How does it fix those problems?\n\nThis problem was fixed in a couple of ways. First, the code now checks\nwhether the node to which a CPU is mapped by 'numa_update_cpu_topology' /\n'arch_update_cpu_topology' has been initialized and has memory available.\nIf either test is false, a call is made to 'try_online_node()' to finish\nthe data structure initialization. Only if we are unable to initialize\nthe node at this point will the CPU node assignment be collapsed into an\nexisting node. After initialization by 'try_online_node()', calls to\n'local_memory_node' no longer crash for these memoryless nodes.\n\n> \n>> diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c\n>> index b385cd0..e811dd1 100644\n>> --- a/arch/powerpc/mm/numa.c\n>> +++ b/arch/powerpc/mm/numa.c\n>> @@ -1325,6 +1325,17 @@ static long vphn_get_associativity(unsigned long cpu,\n>> \treturn rc;\n>> }\n>> \n>> +static int verify_node_preparation(int nid)\n>> +{\n> \n> I would not expect a function called \"verify\" ...\n> \n>> +\tif ((NODE_DATA(nid) == NULL) ||\n>> +\t (NODE_DATA(nid)->node_spanned_pages == 0)) {\n>> +\t\tif (try_online_node(nid))\n> \n> .. to do something like online a node.\n\nWe have changed the function name to 'find_cpu_nid'.\n\n> \n>> +\t\t\treturn first_online_node;\n>> +\t}\n>> +\n>> +\treturn nid;\n>> +}\n>> +\n>> /*\n>> * Update the CPU maps and sysfs entries for a single CPU when its NUMA\n>> * characteristics change. This function doesn't perform any locking and is\n>> @@ -1433,9 +1444,11 @@ int numa_update_cpu_topology(bool cpus_locked)\n>> \t\t/* Use associativity from first thread for all siblings */\n>> \t\tvphn_get_associativity(cpu, associativity);\n>> \t\tnew_nid = associativity_to_nid(associativity);\n>> -\t\tif (new_nid < 0 || !node_online(new_nid))\n>> +\t\tif (new_nid < 0 || !node_possible(new_nid))\n>> \t\t\tnew_nid = first_online_node;\n>> \n>> +\t\tnew_nid = verify_node_preparation(new_nid);\n> \n> You're being called part-way through CPU hotplug here, are we sure it's\n> safe to go and do memory hotplug from there? What's the locking\n> situation?\n\nWe are not doing memory hotplug. We are initializing a node that may be used\nby CPUs or memory before it can be referenced as invalid by a CPU hotplug\noperation. CPU hotplug operations are protected by a range of APIs including\ncpu_maps_update_begin/cpu_maps_update_done, cpus_read/write_lock / cpus_read/write_unlock,\ndevice locks, and more. Memory hotplug operations, including try_online_node,\nare protected by mem_hotplug_begin/mem_hotplug_done, device locks, and more.\nIn the case of CPUs being hot-added to a previously memoryless node, the\ntry_online_node operation occurs wholly within the CPU locks with no overlap.\nUsing HMC hot-add/hot-remove operations, I have been able to add and remove\nCPUs to any possible node without failures. HMC operations involve a degree\nself-serialization, though.\n\n> \n> cheers\n> \n>","headers":{"Return-Path":"","X-Original-To":["patchwork-incoming@ozlabs.org","linuxppc-dev@lists.ozlabs.org"],"Delivered-To":["patchwork-incoming@ozlabs.org","linuxppc-dev@lists.ozlabs.org"],"Received":["from lists.ozlabs.org (lists.ozlabs.org [103.22.144.68])\n\t(using TLSv1.2 with cipher ADH-AES256-GCM-SHA384 (256/256 bits))\n\t(No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3ycY8C1T41z9s7G\n\tfor ;\n\tThu, 16 Nov 2017 05:39:55 +1100 (AEDT)","from lists.ozlabs.org (lists.ozlabs.org [IPv6:2401:3900:2:1::3])\n\tby lists.ozlabs.org (Postfix) with ESMTP id 3ycY8B70zYzDqr8\n\tfor ;\n\tThu, 16 Nov 2017 05:39:54 +1100 (AEDT)","from mx0a-001b2d01.pphosted.com (mx0a-001b2d01.pphosted.com\n\t[148.163.156.1])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby lists.ozlabs.org (Postfix) with ESMTPS id 3ycXty3b4QzDqpF\n\tfor ;\n\tThu, 16 Nov 2017 05:28:26 +1100 (AEDT)","from pps.filterd (m0098409.ppops.net [127.0.0.1])\n\tby mx0a-001b2d01.pphosted.com (8.16.0.21/8.16.0.21) with SMTP id\n\tvAFIQvBU071907\n\tfor ; Wed, 15 Nov 2017 13:28:24 -0500","from e35.co.us.ibm.com (e35.co.us.ibm.com [32.97.110.153])\n\tby mx0a-001b2d01.pphosted.com with ESMTP id 2e8qjdk0gw-1\n\t(version=TLSv1.2 cipher=AES256-SHA bits=256 verify=NOT)\n\tfor ; Wed, 15 Nov 2017 13:28:23 -0500","from localhost\n\tby e35.co.us.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use\n\tOnly! Violators will be prosecuted\n\tfor from ;\n\tWed, 15 Nov 2017 11:28:23 -0700","from b03cxnp07029.gho.boulder.ibm.com (9.17.130.16)\n\tby e35.co.us.ibm.com (192.168.1.135) with IBM ESMTP SMTP Gateway:\n\tAuthorized Use Only! Violators will be prosecuted; \n\tWed, 15 Nov 2017 11:28:21 -0700","from b03ledav003.gho.boulder.ibm.com\n\t(b03ledav003.gho.boulder.ibm.com [9.17.130.234])\n\tby b03cxnp07029.gho.boulder.ibm.com (8.14.9/8.14.9/NCO v10.0) with\n\tESMTP id vAFISK5G10092914\n\tfor ; Wed, 15 Nov 2017 11:28:20 -0700","from b03ledav003.gho.boulder.ibm.com (unknown [127.0.0.1])\n\tby IMSVA (Postfix) with ESMTP id 9A8E56A03D\n\tfor ;\n\tWed, 15 Nov 2017 11:28:20 -0700 (MST)","from oc2832402873.ibm.com (unknown [9.53.92.161])\n\tby b03ledav003.gho.boulder.ibm.com (Postfix) with ESMTP id 6CB586A03F\n\tfor ;\n\tWed, 15 Nov 2017 11:28:20 -0700 (MST)"],"Authentication-Results":"ozlabs.org;\n\tspf=none (mailfrom) smtp.mailfrom=linux.vnet.ibm.com\n\t(client-ip=148.163.156.1; helo=mx0a-001b2d01.pphosted.com;\n\tenvelope-from=mwb@linux.vnet.ibm.com; receiver=)","Subject":"Re: [PATCH 2/2] powerpc/hotplug: Ensure nodes initialized for\n\thotplug","To":"linuxppc-dev@lists.ozlabs.org","References":"\n\t<87y3objlrk.fsf@concordia.ellerman.id.au>","From":"Michael Bringmann ","Organization":"IBM Linux Technology Center","Date":"Wed, 15 Nov 2017 12:28:19 -0600","User-Agent":"Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101\n\tThunderbird/52.4.0","MIME-Version":"1.0","In-Reply-To":"<87y3objlrk.fsf@concordia.ellerman.id.au>","Content-Type":"text/plain; charset=utf-8","Content-Language":"en-US","Content-Transfer-Encoding":"8bit","X-TM-AS-GCONF":"00","x-cbid":"17111518-0012-0000-0000-0000154CACF6","X-IBM-SpamModules-Scores":"","X-IBM-SpamModules-Versions":"BY=3.00008068; HX=3.00000241; KW=3.00000007;\n\tPH=3.00000004; SC=3.00000240; SDB=6.00946326; UDB=6.00477671;\n\tIPR=6.00726623; \n\tBA=6.00005692; NDR=6.00000001; ZLA=6.00000005; ZF=6.00000009;\n\tZB=6.00000000; \n\tZP=6.00000000; ZH=6.00000000; ZU=6.00000002; MB=3.00018033;\n\tXFM=3.00000015; UTC=2017-11-15 18:28:21","X-IBM-AV-DETECTION":"SAVI=unused REMOTE=unused XFE=unused","x-cbparentid":"17111518-0013-0000-0000-000050474C3E","Message-Id":"<3be6fd7b-36ec-3beb-21ce-1def39ef0938@linux.vnet.ibm.com>","X-Proofpoint-Virus-Version":"vendor=fsecure engine=2.50.10432:, ,\n\tdefinitions=2017-11-15_09:, , signatures=0","X-Proofpoint-Spam-Details":"rule=outbound_notspam policy=outbound score=0\n\tpriorityscore=1501\n\tmalwarescore=0 suspectscore=1 phishscore=0 bulkscore=0 spamscore=0\n\tclxscore=1015 lowpriorityscore=0 impostorscore=0 adultscore=0\n\tclassifier=spam adjust=0 reason=mlx scancount=1\n\tengine=8.0.1-1709140000\n\tdefinitions=main-1711150244","X-BeenThere":"linuxppc-dev@lists.ozlabs.org","X-Mailman-Version":"2.1.24","Precedence":"list","List-Id":"Linux on PowerPC Developers Mail List\n\t","List-Unsubscribe":",\n\t","List-Archive":"","List-Post":"","List-Help":"","List-Subscribe":",\n\t","Errors-To":"linuxppc-dev-bounces+patchwork-incoming=ozlabs.org@lists.ozlabs.org","Sender":"\"Linuxppc-dev\"\n\t"}},{"id":1805828,"web_url":"http://patchwork.ozlabs.org/comment/1805828/","msgid":"<2c095594-7040-2bf9-6444-4583854c957a@linux.vnet.ibm.com>","date":"2017-11-16T16:57:54","subject":"Re: [PATCH 2/2] powerpc/hotplug: Ensure nodes initialized for\n\thotplug","submitter":{"id":17146,"url":"http://patchwork.ozlabs.org/api/people/17146/","name":"Nathan Fontenot","email":"nfont@linux.vnet.ibm.com"},"content":"On 11/15/2017 12:28 PM, Michael Bringmann wrote:\n> Hello:\n> See below.\n> \n> On 10/16/2017 07:54 AM, Michael Ellerman wrote:\n>> Michael Bringmann writes:\n>>\n>>> powerpc/hotplug: On systems like PowerPC which allow 'hot-add' of CPU,\n>>> it may occur that the new resources are to be inserted into nodes\n>>> that were not used for memory resources at bootup. Many different\n>>> configurations of PowerPC resources may need to be supported depending\n>>> upon the environment.\n>>\n>> Give me some detail please?!\n> \n> The most important characteristics that I have observed are:\n> \n> * Dedicated vs. shared resources. Shared resources require information\n> such as the VPHN hcall for CPU assignment to nodes.\n> * memoryless nodes at boot. Nodes need to be defined as 'possible' at\n> boot for operation with other code modules. Previously, the powerpc\n> code would limit the set of possible/online nodes to those which have\n> memory assigned at boot. Subsequent add/remove of CPUs or memory would\n> only work with this subset of possible nodes.\n> * memoryless nodes with CPUs at boot. Due to the previous restriction on\n> nodes, nodes that had CPUs but no memory were being collapsed into other\n> nodes that did have memory at boot. In practice this meant that the\n> node assignment presented by the runtime kernel differed from the affinity\n> and associativity attirbutes presented by the device tree or VPHN hcalls.\n> Nodes that might be known to the pHyp were not 'possible' in the runtime\n> kernel because they did not have memory at boot.\n> \n>>\n>>> This patch fixes some problems encountered at\n>>\n>> What problems?\n> \n> This patch set fixes a couple of problems.\n> \n> * Nodes known to powerpc to be memoryless at boot, but to have CPUs in them\n> are allowed to be 'possible' and 'online'. Memory allocations for those\n> nodes are taken from another node that does have memory until and if memory\n> is hot-added to the node.\n> * Nodes which have no resources assigned at boot, but which may still be\n> referenced subsequently by affinity or associativity attributes, are kept\n> in the list of 'possible' nodes for powerpc. Hot-add of memory or CPUs\n> to the system can reference these nodes and bring them online instead of\n> redirecting the resources to the set of nodes known to have memory at boot.\n> \n>>\n>>> runtime with configurations that support memory-less nodes, but which\n>>> allow CPUs to be added at and after boot.\n>>\n>> How does it fix those problems?\n> \n> This problem was fixed in a couple of ways. First, the code now checks\n> whether the node to which a CPU is mapped by 'numa_update_cpu_topology' /\n> 'arch_update_cpu_topology' has been initialized and has memory available.\n> If either test is false, a call is made to 'try_online_node()' to finish\n> the data structure initialization. Only if we are unable to initialize\n> the node at this point will the CPU node assignment be collapsed into an\n> existing node. After initialization by 'try_online_node()', calls to\n> 'local_memory_node' no longer crash for these memoryless nodes.\n> \n>>\n>>> diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c\n>>> index b385cd0..e811dd1 100644\n>>> --- a/arch/powerpc/mm/numa.c\n>>> +++ b/arch/powerpc/mm/numa.c\n>>> @@ -1325,6 +1325,17 @@ static long vphn_get_associativity(unsigned long cpu,\n>>> \treturn rc;\n>>> }\n>>> \n>>> +static int verify_node_preparation(int nid)\n>>> +{\n>>\n>> I would not expect a function called \"verify\" ...\n>>\n>>> +\tif ((NODE_DATA(nid) == NULL) ||\n>>> +\t (NODE_DATA(nid)->node_spanned_pages == 0)) {\n>>> +\t\tif (try_online_node(nid))\n>>\n>> .. to do something like online a node.\n> \n> We have changed the function name to 'find_cpu_nid'.\n\nOk, but I would still not expect 'find_cpu_nid' to online the node.\n\n> \n>>\n>>> +\t\t\treturn first_online_node;\n>>> +\t}\n>>> +\n>>> +\treturn nid;\n>>> +}\n>>> +\n>>> /*\n>>> * Update the CPU maps and sysfs entries for a single CPU when its NUMA\n>>> * characteristics change. This function doesn't perform any locking and is\n>>> @@ -1433,9 +1444,11 @@ int numa_update_cpu_topology(bool cpus_locked)\n>>> \t\t/* Use associativity from first thread for all siblings */\n>>> \t\tvphn_get_associativity(cpu, associativity);\n>>> \t\tnew_nid = associativity_to_nid(associativity);\n>>> -\t\tif (new_nid < 0 || !node_online(new_nid))\n>>> +\t\tif (new_nid < 0 || !node_possible(new_nid))\n>>> \t\t\tnew_nid = first_online_node;\n>>> \n>>> +\t\tnew_nid = verify_node_preparation(new_nid);\n>>\n>> You're being called part-way through CPU hotplug here, are we sure it's\n>> safe to go and do memory hotplug from there? What's the locking\n>> situation?\n> \n> We are not doing memory hotplug. We are initializing a node that may be used\n> by CPUs or memory before it can be referenced as invalid by a CPU hotplug\n> operation. CPU hotplug operations are protected by a range of APIs including\n> cpu_maps_update_begin/cpu_maps_update_done, cpus_read/write_lock / cpus_read/write_unlock,\n> device locks, and more. Memory hotplug operations, including try_online_node,\n> are protected by mem_hotplug_begin/mem_hotplug_done, device locks, and more.\n> In the case of CPUs being hot-added to a previously memoryless node, the\n> try_online_node operation occurs wholly within the CPU locks with no overlap.\n> Using HMC hot-add/hot-remove operations, I have been able to add and remove\n> CPUs to any possible node without failures. HMC operations involve a degree\n> self-serialization, though.\n\nFor both memory and cpu DLPAR operations we hold the device hotplug lock.\n\n-Nathan\n \n> \n>>\n>> cheers\n>>\n>>\n>","headers":{"Return-Path":"","X-Original-To":["patchwork-incoming@ozlabs.org","linuxppc-dev@lists.ozlabs.org"],"Delivered-To":["patchwork-incoming@ozlabs.org","linuxppc-dev@lists.ozlabs.org"],"Received":["from lists.ozlabs.org (lists.ozlabs.org [103.22.144.68])\n\t(using TLSv1.2 with cipher ADH-AES256-GCM-SHA384 (256/256 bits))\n\t(No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yd6t05Jj4z9ryv\n\tfor ;\n\tFri, 17 Nov 2017 03:59:36 +1100 (AEDT)","from lists.ozlabs.org (lists.ozlabs.org [IPv6:2401:3900:2:1::3])\n\tby lists.ozlabs.org (Postfix) with ESMTP id 3yd6t04LzFzDr62\n\tfor ;\n\tFri, 17 Nov 2017 03:59:36 +1100 (AEDT)","from mx0a-001b2d01.pphosted.com (mx0b-001b2d01.pphosted.com\n\t[148.163.158.5])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby lists.ozlabs.org (Postfix) with ESMTPS id 3yd6r94ZSTzDqtL\n\tfor ;\n\tFri, 17 Nov 2017 03:58:01 +1100 (AEDT)","from pps.filterd (m0098414.ppops.net [127.0.0.1])\n\tby mx0b-001b2d01.pphosted.com (8.16.0.21/8.16.0.21) with SMTP id\n\tvAGGsh0s075969\n\tfor ; Thu, 16 Nov 2017 11:57:58 -0500","from e15.ny.us.ibm.com (e15.ny.us.ibm.com [129.33.205.205])\n\tby mx0b-001b2d01.pphosted.com with ESMTP id 2e9de9487w-1\n\t(version=TLSv1.2 cipher=AES256-SHA bits=256 verify=NOT)\n\tfor ; Thu, 16 Nov 2017 11:57:57 -0500","from localhost\n\tby e15.ny.us.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use\n\tOnly! Violators will be prosecuted\n\tfor from ; \n\tThu, 16 Nov 2017 11:57:57 -0500","from b01cxnp22036.gho.pok.ibm.com (9.57.198.26)\n\tby e15.ny.us.ibm.com (146.89.104.202) with IBM ESMTP SMTP Gateway:\n\tAuthorized Use Only! Violators will be prosecuted; \n\tThu, 16 Nov 2017 11:57:55 -0500","from b01ledav006.gho.pok.ibm.com (b01ledav006.gho.pok.ibm.com\n\t[9.57.199.111])\n\tby b01cxnp22036.gho.pok.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP\n\tid vAGGvthd49545288\n\tfor ; Thu, 16 Nov 2017 16:57:55 GMT","from b01ledav006.gho.pok.ibm.com (unknown [127.0.0.1])\n\tby IMSVA (Postfix) with ESMTP id EAB5BAC043\n\tfor ;\n\tThu, 16 Nov 2017 11:58:47 -0500 (EST)","from [9.41.92.186] (unknown [9.41.92.186])\n\tby b01ledav006.gho.pok.ibm.com (Postfix) with ESMTP id BCCB9AC040\n\tfor ;\n\tThu, 16 Nov 2017 11:58:47 -0500 (EST)"],"Authentication-Results":"ozlabs.org;\n\tspf=none (mailfrom) smtp.mailfrom=linux.vnet.ibm.com\n\t(client-ip=148.163.158.5; helo=mx0a-001b2d01.pphosted.com;\n\tenvelope-from=nfont@linux.vnet.ibm.com; receiver=)","Subject":"Re: [PATCH 2/2] powerpc/hotplug: Ensure nodes initialized for\n\thotplug","To":"linuxppc-dev@lists.ozlabs.org","References":"\n\t<87y3objlrk.fsf@concordia.ellerman.id.au>\n\t<3be6fd7b-36ec-3beb-21ce-1def39ef0938@linux.vnet.ibm.com>","From":"Nathan Fontenot ","Date":"Thu, 16 Nov 2017 10:57:54 -0600","User-Agent":"Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101\n\tThunderbird/52.4.0","MIME-Version":"1.0","In-Reply-To":"<3be6fd7b-36ec-3beb-21ce-1def39ef0938@linux.vnet.ibm.com>","Content-Type":"text/plain; charset=utf-8","Content-Language":"en-US","Content-Transfer-Encoding":"7bit","X-TM-AS-GCONF":"00","x-cbid":"17111616-0036-0000-0000-0000028DAB7F","X-IBM-SpamModules-Scores":"","X-IBM-SpamModules-Versions":"BY=3.00008074; HX=3.00000241; KW=3.00000007;\n\tPH=3.00000004; SC=3.00000240; SDB=6.00946774; UDB=6.00477940;\n\tIPR=6.00727067; \n\tBA=6.00005694; NDR=6.00000001; ZLA=6.00000005; ZF=6.00000009;\n\tZB=6.00000000; \n\tZP=6.00000000; ZH=6.00000000; ZU=6.00000002; MB=3.00018047;\n\tXFM=3.00000015; UTC=2017-11-16 16:57:56","X-IBM-AV-DETECTION":"SAVI=unused REMOTE=unused XFE=unused","x-cbparentid":"17111616-0037-0000-0000-00004266E16B","Message-Id":"<2c095594-7040-2bf9-6444-4583854c957a@linux.vnet.ibm.com>","X-Proofpoint-Virus-Version":"vendor=fsecure engine=2.50.10432:, ,\n\tdefinitions=2017-11-16_06:, , signatures=0","X-Proofpoint-Spam-Details":"rule=outbound_notspam policy=outbound score=0\n\tpriorityscore=1501\n\tmalwarescore=0 suspectscore=1 phishscore=0 bulkscore=0 spamscore=0\n\tclxscore=1011 lowpriorityscore=0 impostorscore=0 adultscore=0\n\tclassifier=spam adjust=0 reason=mlx scancount=1\n\tengine=8.0.1-1709140000\n\tdefinitions=main-1711160227","X-BeenThere":"linuxppc-dev@lists.ozlabs.org","X-Mailman-Version":"2.1.24","Precedence":"list","List-Id":"Linux on PowerPC Developers Mail List\n\t","List-Unsubscribe":",\n\t","List-Archive":"","List-Post":"","List-Help":"","List-Subscribe":",\n\t","Errors-To":"linuxppc-dev-bounces+patchwork-incoming=ozlabs.org@lists.ozlabs.org","Sender":"\"Linuxppc-dev\"\n\t"}},{"id":1805864,"web_url":"http://patchwork.ozlabs.org/comment/1805864/","msgid":"<83153b17-3fcb-7c0b-f8ac-c728437748d6@linux.vnet.ibm.com>","date":"2017-11-16T17:36:01","subject":"Re: [PATCH 2/2] powerpc/hotplug: Ensure nodes initialized for\n\thotplug","submitter":{"id":65104,"url":"http://patchwork.ozlabs.org/api/people/65104/","name":"Michael Bringmann","email":"mwb@linux.vnet.ibm.com"},"content":">>>\n>>>> +\tif ((NODE_DATA(nid) == NULL) ||\n>>>> +\t (NODE_DATA(nid)->node_spanned_pages == 0)) {\n>>>> +\t\tif (try_online_node(nid))\n>>>\n>>> .. to do something like online a node.\n>>\n>> We have changed the function name to 'find_cpu_nid'.\n> \n> Ok, but I would still not expect 'find_cpu_nid' to online the node.\n> \n\nWe would have to talk to the developer that created try_online_node()\nwhich fully initializes the node and all of the related data structures.\nA few of the APIs are external, and 'numa.c' knows how to allocate the\nbase 'pgdat' structure, but everything else that the kernel depends\nupon for a node is handled in mm/page_alloc.c and mm/hotplug_memory.c.\nI was trying to avoid piecemeal changes to that code -- avoid any changes\nif it comes to it.\n\nEven if it was not expected to put the node online, it is convenient,\nas otherwise the patches to 'numa.c' would have to put the node online\n-- that is expected for a CPU that is online.\n\nRegards,","headers":{"Return-Path":"","X-Original-To":["patchwork-incoming@ozlabs.org","linuxppc-dev@lists.ozlabs.org"],"Delivered-To":["patchwork-incoming@ozlabs.org","linuxppc-dev@lists.ozlabs.org"],"Received":["from lists.ozlabs.org (lists.ozlabs.org [IPv6:2401:3900:2:1::3])\n\t(using TLSv1.2 with cipher ADH-AES256-GCM-SHA384 (256/256 bits))\n\t(No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yd81w3Dm1z9s71\n\tfor ;\n\tFri, 17 Nov 2017 04:51:32 +1100 (AEDT)","from lists.ozlabs.org (lists.ozlabs.org [IPv6:2401:3900:2:1::3])\n\tby lists.ozlabs.org (Postfix) with ESMTP id 3yd81w0WL5zDqvV\n\tfor ;\n\tFri, 17 Nov 2017 04:51:32 +1100 (AEDT)","from mx0a-001b2d01.pphosted.com (mx0a-001b2d01.pphosted.com\n\t[148.163.156.1])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby lists.ozlabs.org (Postfix) with ESMTPS id 3yd7jP19BczDrDj\n\tfor ;\n\tFri, 17 Nov 2017 04:37:12 +1100 (AEDT)","from pps.filterd (m0098394.ppops.net [127.0.0.1])\n\tby mx0a-001b2d01.pphosted.com (8.16.0.21/8.16.0.21) with SMTP id\n\tvAGHb3C7061706\n\tfor ; Thu, 16 Nov 2017 12:37:11 -0500","from e36.co.us.ibm.com (e36.co.us.ibm.com [32.97.110.154])\n\tby mx0a-001b2d01.pphosted.com with ESMTP id 2e9f4fgcah-1\n\t(version=TLSv1.2 cipher=AES256-SHA bits=256 verify=NOT)\n\tfor ; Thu, 16 Nov 2017 12:37:04 -0500","from localhost\n\tby e36.co.us.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use\n\tOnly! Violators will be prosecuted\n\tfor from ;\n\tThu, 16 Nov 2017 10:36:06 -0700","from b03cxnp08025.gho.boulder.ibm.com (9.17.130.17)\n\tby e36.co.us.ibm.com (192.168.1.136) with IBM ESMTP SMTP Gateway:\n\tAuthorized Use Only! Violators will be prosecuted; \n\tThu, 16 Nov 2017 10:36:05 -0700","from b03ledav001.gho.boulder.ibm.com\n\t(b03ledav001.gho.boulder.ibm.com [9.17.130.232])\n\tby b03cxnp08025.gho.boulder.ibm.com (8.14.9/8.14.9/NCO v10.0) with\n\tESMTP id vAGHa28I36045004\n\tfor ; Thu, 16 Nov 2017 10:36:02 -0700","from b03ledav001.gho.boulder.ibm.com (unknown [127.0.0.1])\n\tby IMSVA (Postfix) with ESMTP id 2B9A66E044\n\tfor ;\n\tThu, 16 Nov 2017 10:36:02 -0700 (MST)","from oc2832402873.ibm.com (unknown [9.53.92.243])\n\tby b03ledav001.gho.boulder.ibm.com (Postfix) with ESMTP id 0ACC36E048\n\tfor ;\n\tThu, 16 Nov 2017 10:36:01 -0700 (MST)"],"Authentication-Results":"ozlabs.org;\n\tspf=none (mailfrom) smtp.mailfrom=linux.vnet.ibm.com\n\t(client-ip=148.163.156.1; helo=mx0a-001b2d01.pphosted.com;\n\tenvelope-from=mwb@linux.vnet.ibm.com; receiver=)","Subject":"Re: [PATCH 2/2] powerpc/hotplug: Ensure nodes initialized for\n\thotplug","To":"linuxppc-dev@lists.ozlabs.org","References":"\n\t<87y3objlrk.fsf@concordia.ellerman.id.au>\n\t<3be6fd7b-36ec-3beb-21ce-1def39ef0938@linux.vnet.ibm.com>\n\t<2c095594-7040-2bf9-6444-4583854c957a@linux.vnet.ibm.com>","From":"Michael Bringmann ","Organization":"IBM Linux Technology Center","Date":"Thu, 16 Nov 2017 11:36:01 -0600","User-Agent":"Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101\n\tThunderbird/52.4.0","MIME-Version":"1.0","In-Reply-To":"<2c095594-7040-2bf9-6444-4583854c957a@linux.vnet.ibm.com>","Content-Type":"text/plain; charset=utf-8","Content-Language":"en-US","Content-Transfer-Encoding":"7bit","X-TM-AS-GCONF":"00","x-cbid":"17111617-0020-0000-0000-00000D026728","X-IBM-SpamModules-Scores":"","X-IBM-SpamModules-Versions":"BY=3.00008074; HX=3.00000241; KW=3.00000007;\n\tPH=3.00000004; SC=3.00000240; SDB=6.00946787; UDB=6.00477948;\n\tIPR=6.00727079; \n\tBA=6.00005695; NDR=6.00000001; ZLA=6.00000005; ZF=6.00000009;\n\tZB=6.00000000; \n\tZP=6.00000000; ZH=6.00000000; ZU=6.00000002; MB=3.00018048;\n\tXFM=3.00000015; UTC=2017-11-16 17:36:06","X-IBM-AV-DETECTION":"SAVI=unused REMOTE=unused XFE=unused","x-cbparentid":"17111617-0021-0000-0000-00005EEDDD0C","Message-Id":"<83153b17-3fcb-7c0b-f8ac-c728437748d6@linux.vnet.ibm.com>","X-Proofpoint-Virus-Version":"vendor=fsecure engine=2.50.10432:, ,\n\tdefinitions=2017-11-16_06:, , signatures=0","X-Proofpoint-Spam-Details":"rule=outbound_notspam policy=outbound score=0\n\tpriorityscore=1501\n\tmalwarescore=0 suspectscore=1 phishscore=0 bulkscore=0 spamscore=0\n\tclxscore=1015 lowpriorityscore=0 impostorscore=0 adultscore=0\n\tclassifier=spam adjust=0 reason=mlx scancount=1\n\tengine=8.0.1-1709140000\n\tdefinitions=main-1711160235","X-BeenThere":"linuxppc-dev@lists.ozlabs.org","X-Mailman-Version":"2.1.24","Precedence":"list","List-Id":"Linux on PowerPC Developers Mail List\n\t","List-Unsubscribe":",\n\t","List-Archive":"","List-Post":"","List-Help":"","List-Subscribe":",\n\t","Errors-To":"linuxppc-dev-bounces+patchwork-incoming=ozlabs.org@lists.ozlabs.org","Sender":"\"Linuxppc-dev\"\n\t"}}]