Return-Path: Received: from mx0b-00082601.pphosted.com ([67.231.153.30]:5005 "EHLO mx0a-00082601.pphosted.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S933734AbcHJSeV (ORCPT ); Wed, 10 Aug 2016 14:34:21 -0400 Subject: Re: [lkp] [nfsd] b44061d0b9: BUG: Dentry ffff880027d7c540{i=1846f,n=0a} still in use (1) [unmount of btrfs vda] To: Linus Torvalds References: <20160810053947.GG21941@yexl-desktop> CC: kernel test robot , Al Viro , Chris Mason , David Sterba , "J. Bruce Fields" , LKML , LKP , linux-btrfs , Linux NFS Mailing List From: Josef Bacik Message-ID: Date: Wed, 10 Aug 2016 14:32:47 -0400 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset="utf-8"; format=flowed Sender: linux-nfs-owner@vger.kernel.org List-ID: On 08/10/2016 02:25 PM, Linus Torvalds wrote: > On Wed, Aug 10, 2016 at 11:22 AM, Josef Bacik wrote: >> On 08/10/2016 02:06 PM, Linus Torvalds wrote: >>> >>> More information in the original email on lkml. >> >> I'm not subscribed to lkml and for some reason I can't find the original >> email in any of the lkml/linux-nfs archives. Could you forward more of the >> details? > > Done. > Looks like an NFS problem. Before in the !resfhp->fh_dentry case we would do a lookup_one_len() and carry on. Now we do the lookup_one_len(), call into nfsd_create_locked() and do a dget on resfhp->f_dentry, which from looks of it is the dentry we just did a lookup_one_len() on, so we have one more ref on the dentry than we did previously. Thanks, Josef