From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tom Parker Subject: Re: Using device-mapper with many targets Date: Sat, 11 Aug 2007 01:53:49 +0200 Message-ID: <46BCFA8D.4070607@tevp.net> References: <46AFA7D0.8040900@tevp.net> Reply-To: device-mapper development Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <46AFA7D0.8040900@tevp.net> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: dm-devel-bounces@redhat.com Errors-To: dm-devel-bounces@redhat.com To: device-mapper development List-Id: dm-devel.ids Tom Parker wrote: > With RAID 1 targets, I can get up to about 600 targets before the whole > thing seems to lock hard (dmsetup itself gets OOM'ed, and I can't get > back to the shell or to any other of the virtual terminals). While > scaling up to that point, I find that for every RAID 1 target, > device-mapper seems to eat ~.5MB of RAM. The targets themselves are > currently very small (1MB I think) as this was more a test in many > targets, rather than in large targets. I've been trying to investigate this on my own, and ran into the issue that there's both a dm-raid1.c file in the 2.6.22 kernel, *and* in the device-mapper CVS. Additionally, the dm-raid1.c in the device-mapper CVS appears to be out of sync with current kernels. This does raise the issue of which one I should be investigating regarding the memory issues.... Any good reason why this is? Should I just be looking at the current kernel versions of that file? Thanks, Tom P.S. If anyone knows any good tools for getting per-module memory usage (preferably with alloc/free tracing) out of the kernel, I'd love to know about them. I'm tempted at the moment to go with manual printk adding just because it's simple and it'll work.