From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:40695) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1fYB87-0000Tw-3x for qemu-devel@nongnu.org; Wed, 27 Jun 2018 10:09:08 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1fYB83-0001rQ-SJ for qemu-devel@nongnu.org; Wed, 27 Jun 2018 10:09:07 -0400 Received: from mx3-rdu2.redhat.com ([66.187.233.73]:54214 helo=mx1.redhat.com) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1fYB83-0001q3-OP for qemu-devel@nongnu.org; Wed, 27 Jun 2018 10:09:03 -0400 Date: Wed, 27 Jun 2018 16:08:59 +0200 From: Igor Mammedov Message-ID: <20180627160859.3733036d@redhat.com> In-Reply-To: References: MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] Some issues of hotpluging memory and cpus List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: linzhecheng Cc: QEMU , "mst@redhat.com" , "wangxin (U)" , "robert.moore@intel.com" , lidonglin , "Gonglei (Arei)" , Paolo Bonzini , "lersek@redhat.com" On Mon, 25 Jun 2018 13:30:10 +0000 linzhecheng wrote: > Hi, all, > I had some problems when doing memory hot plugging. After a lot of tests, I was able to stably reproduce the following issues with Windows 2012 R2: > > 1. If I hot-plug the memory immediately after hot-plugging the CPUs, these memory devices will not be detected by windows OS because I cannot see them in the device manager. > 2. If I add a certain delay (for example, 2s or more) between the hot-plugging memory and the CPU, the memory devices will be detected but some of them may not be online. > And Memory Module Properties displaying the following contents: > > This device cannot start. (Code 10) > {Conflicting Address Range} > The specified address range conflicts with the address space. it could help if you can find in device manger what it conflicts with > 3. usb tablet mouse would not work probability after hot adding memory. > > Due to the closed nature of windows OS, I can't know what's going on inside. > Is it related to acpi tables exposure? > I hope you can give me some advice, thank you. you also can try running QEMU with cphp_* and mhp_* trace events enabled [*], it should show ACPI interaction os Windows with QEMU including address assignment. *) docs/devel/tracing.txt