From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx0a-001b2d01.pphosted.com (mx0b-001b2d01.pphosted.com [148.163.158.5]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by lists.ozlabs.org (Postfix) with ESMTPS id 3y7mxv4HwVzDqhg for ; Fri, 6 Oct 2017 22:05:07 +1100 (AEDT) Received: from pps.filterd (m0098419.ppops.net [127.0.0.1]) by mx0b-001b2d01.pphosted.com (8.16.0.21/8.16.0.21) with SMTP id v96B4x3O034961 for ; Fri, 6 Oct 2017 07:05:04 -0400 Received: from e36.co.us.ibm.com (e36.co.us.ibm.com [32.97.110.154]) by mx0b-001b2d01.pphosted.com with ESMTP id 2de6q45s2y-1 (version=TLSv1.2 cipher=AES256-SHA bits=256 verify=NOT) for ; Fri, 06 Oct 2017 07:05:03 -0400 Received: from localhost by e36.co.us.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Fri, 6 Oct 2017 05:04:58 -0600 Subject: Re: Possible LMB hot unplug bug in 4.13+ kernels To: Michael Ellerman , Nathan Fontenot , linuxppc-dev@lists.ozlabs.org References: <94618f16-b47f-b714-9cb5-4bbbf7fdccdf@linux.vnet.ibm.com> <8760bszxww.fsf@concordia.ellerman.id.au> From: Daniel Henrique Barboza Date: Fri, 6 Oct 2017 08:04:52 -0300 MIME-Version: 1.0 In-Reply-To: <8760bszxww.fsf@concordia.ellerman.id.au> Content-Type: text/plain; charset=utf-8; format=flowed Message-Id: <3bdf3e9d-a2f4-4820-6ecd-a9b14b3c45c4@linux.vnet.ibm.com> List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , > Unless you (Daniel) think there's some reason lmb_is_removable() is > incorrectly returning false. But most likely it's correct and there's > just an unmovable allocation in that range. I am not educated enough to say that the current behavior is wrong. What I can say is that in 4.11 and older kernels that supports LMB hot plug/unplug I didn't see this kernel "refusal" to remove a LMB that was just hotplugged. Assuming that the kernel is behaving as intended, a QEMU guest started with 4Gb of RAM that receives an extra 1Gb of RAM will not unplug this same 1Gb. It seems off from the user perspective that a recently added memory is being considered not removable, thus QEMU will need to keep this limitation in mind when dealing with future LMB bugs in 4.13+ kernels. Thanks, Daniel