28 messages in com.xensource.lists.xen-develRe: [Xen-devel] LVM Snapshot Troubles
FromSent OnAttachments
Paul Dorman27 Sep 2004 12:23 
Paul Dorman27 Sep 2004 13:25 
Mark A. Williamson27 Sep 2004 14:45 
Paul Dorman27 Sep 2004 14:49 
Mark A. Williamson27 Sep 2004 14:52 
Ian Pratt27 Sep 2004 15:07 
Michael Vrable27 Sep 2004 15:18 
Paul Dorman27 Sep 2004 15:43 
Peri Hankey28 Sep 2004 01:23 
Ian Pratt28 Sep 2004 05:26 
Brian Wolfe28 Sep 2004 07:20 
Peri Hankey28 Sep 2004 07:54.Other, .Other
Ian Pratt28 Sep 2004 08:43 
Michael Vrable28 Sep 2004 11:08.Other
Michael Vrable28 Sep 2004 11:41 
Ian Pratt28 Sep 2004 13:06 
Michael Vrable28 Sep 2004 13:51 
Ian Pratt28 Sep 2004 14:23 
Mark A. Williamson28 Sep 2004 16:59.txt
Keir Fraser29 Sep 2004 01:39 
Christian Limpach29 Sep 2004 03:36 
Paul Dorman29 Sep 2004 12:46 
Ian Pratt29 Sep 2004 13:23 
Mark A. Williamson30 Sep 2004 13:58 
Mark A. Williamson30 Sep 2004 14:00 
Michael Vrable30 Sep 2004 14:02 
James Harper30 Sep 2004 16:04 
Ian Pratt01 Oct 2004 03:23 
Subject:Re: [Xen-devel] LVM Snapshot Troubles
From:Michael Vrable (mvra@cs.ucsd.edu)
Date:09/30/2004 02:02:01 PM
List:com.xensource.lists.xen-devel

I've been looking into the LVM snapshot/memory allocation troubles and will try to come up with a fix.

FYI: In doing more searching for information about the problem, I did come across this: http://www.redhat.com/archives/dm-devel/2004-January/msg00068.html

"The problem seems to be that dm-ioctl-v4.c sets the PF_MEMALLOC flag for the current process.

Lookaing at the memory allocator (__alloc_page) this means that the VM will think the memory allocation is already running (and this is a recursion) so it will not try to free pages / rebalance page or whatever." ...

I'm looking into sharing memory between the snapshots instead of giving each snapshot its own private allocation of pages for I/O. (As I'd like to scale to a large number of snapshots, and don't want to need >1 MB of kernel memory per snapshot.)