From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mga17.intel.com (mga17.intel.com [192.55.52.151]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 543A717F for ; Tue, 10 Aug 2021 21:20:15 +0000 (UTC) X-IronPort-AV: E=McAfee;i="6200,9189,10072"; a="195263193" X-IronPort-AV: E=Sophos;i="5.84,310,1620716400"; d="scan'208";a="195263193" Received: from fmsmga003.fm.intel.com ([10.253.24.29]) by fmsmga107.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 10 Aug 2021 14:20:10 -0700 X-IronPort-AV: E=Sophos;i="5.84,310,1620716400"; d="scan'208";a="515987352" Received: from akleen-mobl1.amr.corp.intel.com (HELO [10.209.69.62]) ([10.209.69.62]) by fmsmga003-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 10 Aug 2021 14:20:09 -0700 Subject: Re: [PATCH 1/5] mm: Add support for unaccepted memory To: Dave Hansen , "Kirill A. Shutemov" , Borislav Petkov , Andy Lutomirski , Sean Christopherson , Andrew Morton , Joerg Roedel Cc: Kuppuswamy Sathyanarayanan , David Rientjes , Vlastimil Babka , Tom Lendacky , Thomas Gleixner , Peter Zijlstra , Paolo Bonzini , Ingo Molnar , Varad Gautam , Dario Faggioli , x86@kernel.org, linux-mm@kvack.org, linux-coco@lists.linux.dev, linux-kernel@vger.kernel.org, "Kirill A. Shutemov" References: <20210810062626.1012-1-kirill.shutemov@linux.intel.com> <20210810062626.1012-2-kirill.shutemov@linux.intel.com> <9748c07c-4e59-89d0-f425-c57f778d1b42@linux.intel.com> <17b6a3a3-bd7d-f57e-8762-96258b16247a@intel.com> <796a4b20-7fa3-3086-efa0-2f728f35ae06@linux.intel.com> From: Andi Kleen Message-ID: <3caf5e73-c104-0057-680c-7851476e67ac@linux.intel.com> Date: Tue, 10 Aug 2021 14:20:08 -0700 User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:78.0) Gecko/20100101 Thunderbird/78.12.0 Precedence: bulk X-Mailing-List: linux-coco@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Content-Language: en-US > These boot blips are not the biggest issue in the world. But, it is > fully under the guest's control and I think the guest has some > responsibility to provide *some* mitigation for it. It sounds more like an exercise in preliminary optimization at this point. If it's a real problem we can still worry about it later. > 1. Do background acceptance, as opposed to relying 100% on demand-driven > acceptance. Guarantees a limited window in which blips can occur. Like Kirill wrote this was abandoned because it always allocates all memory on the host even if the guest doesn't need it. > 2. Do acceptance based on user input, like from sysfs. You can easily do that by running "memhog" at boot. No need for anything in the kernel. BTW I believe this is also configurable at the guest BIOS level. > 3. Add a command-line argument to accept everything up front, or at > least before userspace runs. Same. > 4. Add some statistic for how much unaccepted memory remains. Yes that makes sense. We should have statistic counters for all of this. Also I agree with your suggestion that we should get the slow path out of the zone locks/interrupt disable region. That should be easy enough and is an obvious improvement. -Andi