From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:32833) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1atHOM-0005P9-N0 for qemu-devel@nongnu.org; Thu, 21 Apr 2016 12:23:47 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1atHOJ-0004kB-FP for qemu-devel@nongnu.org; Thu, 21 Apr 2016 12:23:46 -0400 Received: from mail-wm0-x235.google.com ([2a00:1450:400c:c09::235]:32969) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1atHOI-0004jS-VO for qemu-devel@nongnu.org; Thu, 21 Apr 2016 12:23:43 -0400 Received: by mail-wm0-x235.google.com with SMTP id 127so4725984wmz.0 for ; Thu, 21 Apr 2016 09:23:42 -0700 (PDT) Date: Thu, 21 Apr 2016 18:23:48 +0200 From: Christoffer Dall Message-ID: <20160421162348.GA24178@cbox> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Subject: [Qemu-devel] Performance regression using KVM/ARM List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Peter Maydell , "Michael S. Tsirkin" , qemu-devel@nongnu.org Cc: Marc Zyngier , Alexander Graf Hi, Commit 9fac18f (oslib: allocate PROT_NONE pages on top of RAM, 2015-09-10) had the unfortunate side effect that memory slots registered with KVM no longer contain a userspace address that is aligned to a 2M boundary, causing the use of THP to fail in the kernel. I fail to see where in the QEMU code we should be asking for a 2M alignment of our memory region. Can someone help pointing me to the right place to fix this or suggest a patch? This causes a performance regssion of hackbench on KVM/ARM of about 62% compared to the workload running with THP. We have verified that this is indeed the cause of the failure by adding various prints to QEMU and the kernel, but unfortunatley my QEMU knowledge is not sufficient for me to fix it myself. Any help would be much appreciated! Thanks, -Christoffer