From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753301AbbCFIuq (ORCPT ); Fri, 6 Mar 2015 03:50:46 -0500 Received: from casper.infradead.org ([85.118.1.10]:43927 "EHLO casper.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751126AbbCFIul (ORCPT ); Fri, 6 Mar 2015 03:50:41 -0500 Date: Fri, 6 Mar 2015 09:50:17 +0100 From: Peter Zijlstra To: He Kuang Cc: paulus@samba.org, mingo@redhat.com, acme@kernel.org, ericvh@gmail.com, rminnich@sandia.gov, lucho@ionkov.net, wangnan0@huawei.com, linux-kernel@vger.kernel.org Subject: Re: [BUG] uprobe: failed to work on 9pfs Message-ID: <20150306085017.GW21418@twins.programming.kicks-ass.net> References: <54F938B7.20700@huawei.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <54F938B7.20700@huawei.com> User-Agent: Mutt/1.5.21 (2012-12-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, Mar 06, 2015 at 01:18:47PM +0800, He Kuang wrote: > Uprobe uses inode address to index all registered uprobes in a > rb_tree, this works well in most filesystems but failed on 9pfs. > > 9pfs allocate more than one vfs inode to the same file, the inode > address when we create uprobe is not same as the inode address we > run later. As a result, neither perf record nor events/uprobe can > capture the predefined uprobe events. How does 9pfs work with the vfs/mm in this regard? The file cache very much assumes the same thing uprobes does. File caching is per inode, so if you have multiple inodes on the same file you get multiple copies of your data with all sorts of braindamaged corruption as a result. I would very much chalk this up to 9pfs being insane.