From dgregor@XXXXXXXXXX Thu Jan 10 10:41:31 2008 Delivered-To: mpifrm-mpi-21-outgoing@XXXXXXXXXXXXXXXXXXXXXXX X-Original-To: mpifrm-mpi-21@XXXXXXXXXXXXXXXXXXXXXXX Delivered-To: mpifrm-mpi-21@XXXXXXXXXXXXXXXXXXXXXXX From: Doug Gregor To: mpi-21@XXXXXXXXXXXXX In-Reply-To: Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit Mime-Version: 1.0 (Apple Message framework v915) Subject: Re: [mpi-21] Proposal EH2: add const keyword to the C bindings Date: Thu, 10 Jan 2008 11:41:20 -0500 References: X-Mailer: Apple Mail (2.915) X-Virus-Scanned: by amavisd-new-20030616-p10 (Debian) at mailbouncer.mcs.anl.gov Sender: owner-mpi-21@XXXXXXXXXXXXX Precedence: bulk Reply-To: mpi-21@XXXXXXXXXXXXX X-Virus-Scanned: by amavisd-new-20030616-p10 (Debian) at mailbouncer.mcs.anl.gov On Jan 10, 2008, at 8:53 AM, Richard Treumann wrote: > I have mentioned that I think the proposal for removing the send > buffer access restriction is worth examining. I have even said I > think it likely (not obvious) that the rule should be relaxed. First > we should conclude that keeping it has trivial potential for being > important to some MPI implementation in comparison the burden it > puts on application writers. Performance matters a lot and always > will. > Can anyone point to an MPI implementation *in use* that takes advantage of this rule? The only example I have heard about is a DMA engine that is not cache- coherent with the main processor, but we know that particular example is no longer relevant. - Doug