From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752643Ab0IPJ06 (ORCPT ); Thu, 16 Sep 2010 05:26:58 -0400 Received: from mx1.redhat.com ([209.132.183.28]:7050 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751768Ab0IPJ05 (ORCPT ); Thu, 16 Sep 2010 05:26:57 -0400 Message-ID: <4C91E2CC.9040709@redhat.com> Date: Thu, 16 Sep 2010 11:26:36 +0200 From: Avi Kivity User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.9) Gecko/20100907 Fedora/3.1.3-1.fc13 Lightning/1.0b3pre Thunderbird/3.1.3 MIME-Version: 1.0 To: Christopher Yeoh CC: Bryan Donlan , linux-kernel@vger.kernel.org, Linux Memory Management List , Ingo Molnar Subject: Re: [RFC][PATCH] Cross Memory Attach References: <20100915104855.41de3ebf@lilo> <4C90A6C7.9050607@redhat.com> <20100916104819.36d10acb@lilo> In-Reply-To: <20100916104819.36d10acb@lilo> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 09/16/2010 03:18 AM, Christopher Yeoh wrote: > On Wed, 15 Sep 2010 23:46:09 +0900 > Bryan Donlan wrote: > > > On Wed, Sep 15, 2010 at 19:58, Avi Kivity wrote: > > > > > Instead of those two syscalls, how about a vmfd(pid_t pid, ulong > > > start, ulong len) system call which returns an file descriptor that > > > represents a portion of the process address space. You can then > > > use preadv() and pwritev() to copy memory, and > > > io_submit(IO_CMD_PREADV) and io_submit(IO_CMD_PWRITEV) for > > > asynchronous variants (especially useful with a dma engine, since > > > that adds latency). > > > > > > With some care (and use of mmu_notifiers) you can even mmap() your > > > vmfd and access remote process memory directly. > > > > Rather than introducing a new vmfd() API for this, why not just add > > implementations for these more efficient operations to the existing > > /proc/$pid/mem interface? > > Perhaps I'm misunderstanding something here, but > accessing /proc/$pid/mem requires ptracing the target process. > We can't really have all these MPI processes ptraceing each other > just to send/receive a message.... > You could have each process open /proc/self/mem and pass the fd using SCM_RIGHTS. That eliminates a race; with copy_to_process(), by the time the pid is looked up it might designate a different process. -- error compiling committee.c: too many arguments to function