From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Hellwig Subject: Re: RFC: exporting per-superblock statistics to user space Date: Wed, 23 Mar 2005 11:06:57 +0000 Message-ID: <20050323110657.GA3745@infradead.org> References: <42360D49.7070902@citi.umich.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-fsdevel@vger.kernel.org Received: from pentafluge.infradead.org ([213.146.154.40]:17902 "EHLO pentafluge.infradead.org") by vger.kernel.org with ESMTP id S261534AbVCWLG5 (ORCPT ); Wed, 23 Mar 2005 06:06:57 -0500 To: Chuck Lever Content-Disposition: inline In-Reply-To: <42360D49.7070902@citi.umich.edu> Sender: linux-fsdevel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On Mon, Mar 14, 2005 at 05:16:41PM -0500, Chuck Lever wrote: > we still have a need to provide "iostat" like statistics for NFS > clients. attached are a couple of patches, against 2.6.11.3, which > prototype an approach for providing this kind of data to user programs. > i'd like some comment on the approach. > > 01-mountstats.patch adds a new file called /proc/self/mountstats and a > new file system hook called show_stats. this just replicates > /proc/mounts and the show_options hook. I don't think this makes a lot of sense. The stats aren't per namespace or mountpoint but per fs-instance, right?