Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757963AbYJNM3b (ORCPT ); Tue, 14 Oct 2008 08:29:31 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1756143AbYJNM3W (ORCPT ); Tue, 14 Oct 2008 08:29:22 -0400 Received: from e3.ny.us.ibm.com ([32.97.182.143]:40404 "EHLO e3.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755621AbYJNM3V (ORCPT ); Tue, 14 Oct 2008 08:29:21 -0400 Subject: Re: [PATCH]IB/ehca:reject dynamic memory add/remove From: Dave Hansen To: Stefan Roscher Cc: Stefan Roscher , Roland Dreier , general@lists.openfabrics.org, LinuxPPC-Dev , LKML , hnguyen@de.ibm.com In-Reply-To: <200810141423.48111.stefan.roscher@vnet.de.ibm.com> References: <200810131310.34413.ossrosch@linux.vnet.ibm.com> <1223917766.29877.17.camel@nimitz> <200810141423.48111.stefan.roscher@vnet.de.ibm.com> Content-Type: text/plain Date: Tue, 14 Oct 2008 05:29:13 -0700 Message-Id: <1223987353.29877.40.camel@nimitz> Mime-Version: 1.0 X-Mailer: Evolution 2.22.3.1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1555 Lines: 35 On Tue, 2008-10-14 at 14:23 +0200, Stefan Roscher wrote: > On Monday 13 October 2008 07:09:26 pm Dave Hansen wrote: > > On Mon, 2008-10-13 at 13:10 +0200, Stefan Roscher wrote: > > > Since the ehca device driver does not support dynamic memory add and remove > > > operations, the driver must explicitly reject such requests in order to prevent > > > unpredictable behaviors related to memory regions already occupied and being > > > used by InfiniBand applications. > > > The solution is to add a memory notifier to the ehca device driver and if a request > > > for dynamic memory add or remove comes in, ehca will always reject it. > > > > Why doesn't the driver support it? > > > > This seems like an awfully extreme action to take. Do you have plans to > > support this in the driver soon? > > > There is currently a slight incompatibility how openfabrics uses MRs > and how System p does DMEM add/remove, which basically disables this > support. > If you want to talk to the firmware developpers, I can give you the right contacts. I wish I knew what an 'MR' is. :( Could you be a bit more specific so we can get a better changelog? Perhaps if we understand the situation better, we can come up with a better solution. Does this have anything in common with the problems with 16GB pages? -- Dave -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/