From mboxrd@z Thu Jan 1 00:00:00 1970 From: Gleb Natapov Subject: Re: [PATCH 0/4] KVM: Clean up and optimize __kvm_set_memory_region() - part1 Date: Thu, 17 Jan 2013 18:35:40 +0200 Message-ID: <20130117163540.GA10107@redhat.com> References: <20130111182518.6c5975d9.yoshikawa_takuya_b1@lab.ntt.co.jp> <20130117124202.GB31120@redhat.com> <20130117232653.c7b8c5c723441e01453f14aa@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Takuya Yoshikawa , mtosatti@redhat.com, kvm@vger.kernel.org To: Takuya Yoshikawa Return-path: Received: from mx1.redhat.com ([209.132.183.28]:63989 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756987Ab3AQQfq (ORCPT ); Thu, 17 Jan 2013 11:35:46 -0500 Content-Disposition: inline In-Reply-To: <20130117232653.c7b8c5c723441e01453f14aa@gmail.com> Sender: kvm-owner@vger.kernel.org List-ID: On Thu, Jan 17, 2013 at 11:26:53PM +0900, Takuya Yoshikawa wrote: > On Thu, 17 Jan 2013 14:42:02 +0200 > Gleb Natapov wrote: > > > Applied 1-3. It is not clear whether kvm_iommu_map_pages() should be called > > when flags change, so not applying 4 for now. > > Thank you. > > Although I confirmed that kvm_iommu_map_pages() does nothing with > the flags, it needs to be confirmed by the author. > Yes, it may be a bug. I asked Alex in other email. > Another thing I'm not 100% sure about the API is whether we can > change the flags when we move the base: flags + move. The current > implementation allows it, but not sure if it's a good thing. > > Takuya -- Gleb.