From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Michael S. Tsirkin" Subject: Re: [PATCH] rdma: don't make pages writeable if not requiested Date: Thu, 21 Mar 2013 20:03:00 +0200 Message-ID: <20130321180300.GB4366@redhat.com> References: <20130321070357.GD28328@redhat.com> <20130321085107.GE28328@redhat.com> <20130321093946.GG28328@redhat.com> <20130321171115.GA653@obsidianresearch.com> <20130321171525.GE2994@redhat.com> <20130321172150.GA3118@obsidianresearch.com> <20130321174237.GA4060@redhat.com> <20130321175732.GA3263@obsidianresearch.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20130321175732.GA3263-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org> Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Jason Gunthorpe Cc: Roland Dreier , "Michael R. Hines" , Sean Hefty , Hal Rosenstock , Yishai Hadas , Christoph Lameter , "linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , LKML , qemu-devel-qX2TKyscuCcdnm+yROfE0A@public.gmane.org List-Id: linux-rdma@vger.kernel.org On Thu, Mar 21, 2013 at 11:57:32AM -0600, Jason Gunthorpe wrote: > On Thu, Mar 21, 2013 at 07:42:37PM +0200, Michael S. Tsirkin wrote: > > > It doesn't actually, and our app would sometimes write to these pages. > > It simply does not care which version does the remote get in this case > > since we track writes and resend later. > > Heh, somehow I thought you might say that :) > > A new flag seems like the only way then - maybe: > IBV_ACCESS_NON_COHERENT - The adaptor and the CPU do not share > a coherent view of registered memory. Memory writes from the CPU > after ibv_reg_mr completes may not be reflected in the memory > viewed by the adaptor. > > Can only be combined with read only access permissions. > > Jason I kind of like _GIFT for name, gifts are nice :) But yes that's exactly the semantics we need. -- MST -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752175Ab3CUSCw (ORCPT ); Thu, 21 Mar 2013 14:02:52 -0400 Received: from mx1.redhat.com ([209.132.183.28]:42492 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752016Ab3CUSCq (ORCPT ); Thu, 21 Mar 2013 14:02:46 -0400 Date: Thu, 21 Mar 2013 20:03:00 +0200 From: "Michael S. Tsirkin" To: Jason Gunthorpe Cc: Roland Dreier , "Michael R. Hines" , Sean Hefty , Hal Rosenstock , Yishai Hadas , Christoph Lameter , "linux-rdma@vger.kernel.org" , LKML , qemu-devel@nongnu.org Subject: Re: [PATCH] rdma: don't make pages writeable if not requiested Message-ID: <20130321180300.GB4366@redhat.com> References: <20130321070357.GD28328@redhat.com> <20130321085107.GE28328@redhat.com> <20130321093946.GG28328@redhat.com> <20130321171115.GA653@obsidianresearch.com> <20130321171525.GE2994@redhat.com> <20130321172150.GA3118@obsidianresearch.com> <20130321174237.GA4060@redhat.com> <20130321175732.GA3263@obsidianresearch.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20130321175732.GA3263@obsidianresearch.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Mar 21, 2013 at 11:57:32AM -0600, Jason Gunthorpe wrote: > On Thu, Mar 21, 2013 at 07:42:37PM +0200, Michael S. Tsirkin wrote: > > > It doesn't actually, and our app would sometimes write to these pages. > > It simply does not care which version does the remote get in this case > > since we track writes and resend later. > > Heh, somehow I thought you might say that :) > > A new flag seems like the only way then - maybe: > IBV_ACCESS_NON_COHERENT - The adaptor and the CPU do not share > a coherent view of registered memory. Memory writes from the CPU > after ibv_reg_mr completes may not be reflected in the memory > viewed by the adaptor. > > Can only be combined with read only access permissions. > > Jason I kind of like _GIFT for name, gifts are nice :) But yes that's exactly the semantics we need. -- MST From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:53816) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UIjpF-0000L2-B1 for qemu-devel@nongnu.org; Thu, 21 Mar 2013 14:02:57 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1UIjp5-0000Xi-3q for qemu-devel@nongnu.org; Thu, 21 Mar 2013 14:02:53 -0400 Received: from mx1.redhat.com ([209.132.183.28]:4302) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UIjp4-0000XO-Qv for qemu-devel@nongnu.org; Thu, 21 Mar 2013 14:02:43 -0400 Date: Thu, 21 Mar 2013 20:03:00 +0200 From: "Michael S. Tsirkin" Message-ID: <20130321180300.GB4366@redhat.com> References: <20130321070357.GD28328@redhat.com> <20130321085107.GE28328@redhat.com> <20130321093946.GG28328@redhat.com> <20130321171115.GA653@obsidianresearch.com> <20130321171525.GE2994@redhat.com> <20130321172150.GA3118@obsidianresearch.com> <20130321174237.GA4060@redhat.com> <20130321175732.GA3263@obsidianresearch.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20130321175732.GA3263@obsidianresearch.com> Subject: Re: [Qemu-devel] [PATCH] rdma: don't make pages writeable if not requiested List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Jason Gunthorpe Cc: Roland Dreier , qemu-devel@nongnu.org, "linux-rdma@vger.kernel.org" , Yishai Hadas , LKML , "Michael R. Hines" , Hal Rosenstock , Sean Hefty , Christoph Lameter On Thu, Mar 21, 2013 at 11:57:32AM -0600, Jason Gunthorpe wrote: > On Thu, Mar 21, 2013 at 07:42:37PM +0200, Michael S. Tsirkin wrote: > > > It doesn't actually, and our app would sometimes write to these pages. > > It simply does not care which version does the remote get in this case > > since we track writes and resend later. > > Heh, somehow I thought you might say that :) > > A new flag seems like the only way then - maybe: > IBV_ACCESS_NON_COHERENT - The adaptor and the CPU do not share > a coherent view of registered memory. Memory writes from the CPU > after ibv_reg_mr completes may not be reflected in the memory > viewed by the adaptor. > > Can only be combined with read only access permissions. > > Jason I kind of like _GIFT for name, gifts are nice :) But yes that's exactly the semantics we need. -- MST