Return-Path: Received: from mail-yk0-f179.google.com ([209.85.160.179]:36345 "EHLO mail-yk0-f179.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756790AbbJHMyx (ORCPT ); Thu, 8 Oct 2015 08:54:53 -0400 Received: by ykba192 with SMTP id a192so43391214ykb.3 for ; Thu, 08 Oct 2015 05:54:53 -0700 (PDT) Message-ID: <1444308880.43040.1.camel@primarydata.com> Subject: Re: RCU caching regression in kernel v4.1+ From: Trond Myklebust To: Alexander Viro Cc: Linux NFS Mailing List , lawa@nvidia.com, Linux FS-devel Mailing List Date: Thu, 08 Oct 2015 08:54:40 -0400 In-Reply-To: References: Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Sender: linux-nfs-owner@vger.kernel.org List-ID: On Wed, 2015-10-07 at 14:57 -0400, Trond Myklebust wrote: > Hi Al, > > Please could you take a look at the bugzilla entry in > https://bugzilla.kernel.org/show_bug.cgi?id=104911 ? > > It describes a NFS caching regression that appears to be caused by > commit 766c4cbfacd8634d7580bac6a1b8456e63de3e84 ("namei: > d_is_negative() should be checked before ->d_seq validation"). > > Shouldn't that test for 'if (negative) return -ENOENT;' happen after > the call to d_revalidate() in lookup_fast()? If not, we can end up > caching negative dentries forever, AFAICS... > > Cheers > Trond Leandro, can you please test if the following patch helps in any way? Cheers Trond 8<-----------------------------------------------------------------