From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Sat, 30 Jun 2007 05:59:07 -0400 Message-ID: <87d4zdix04.wl%peter@chubb.wattle.id.au> From: Peter Chubb In-Reply-To: <4685D9C9.20504@de.ibm.com> References: <87myynt1m6.wl%peter@chubb.wattle.id.au> <468517E1.4050803@goop.org> <4685D9C9.20504@de.ibm.com> MIME-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka") Content-Type: text/plain; charset=US-ASCII Subject: Re: RFC: multiple address spaces for one process Sender: owner-linux-mm@kvack.org Return-Path: To: carsteno@de.ibm.com Cc: Jeremy Fitzhardinge , linux-mm@kvack.org, Peter Chubb , virtualization@lists.linux-foundation.org List-ID: In LinuxOnLinux, the guest is just another linux process (or rather, one per virtual CPU). So what this patch does is allow such a garden-variety process to have more than one address space and switch between them.2 Essentially what I'm doing is treating the standard linux VM as a huge software-loaded TLB. When the guest gets a TLB miss, the VMM calls mmap() to insert the translation. Without the patch the address space has to be remapped on every guest context switch. The patch adds address-space IDs to make that unnecessary. What I'm wondering is if there are any *other* uses for this (not just virtualisation). I know qemu uses mmap() to create its address spaces if CONFIG_SOFTMMU is not defined, and could thus benefit. -- Dr Peter Chubb http://www.gelato.unsw.edu.au peterc AT gelato.unsw.edu.au http://www.ertos.nicta.com.au ERTOS within National ICT Australia -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org