From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mx1.redhat.com ([209.132.183.28]:36484 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932953AbeAXKkZ (ORCPT ); Wed, 24 Jan 2018 05:40:25 -0500 Date: Wed, 24 Jan 2018 18:40:23 +0800 From: Eryu Guan Subject: Re: [PATCH v2 0/7] Overlayfs exportfs tests Message-ID: <20180124104023.GB18267@eguan.usersys.redhat.com> References: <1516741152-20345-1-git-send-email-amir73il@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1516741152-20345-1-git-send-email-amir73il@gmail.com> Sender: fstests-owner@vger.kernel.org To: Amir Goldstein Cc: Miklos Szeredi , linux-unionfs@vger.kernel.org, fstests@vger.kernel.org List-ID: On Tue, Jan 23, 2018 at 10:59:05PM +0200, Amir Goldstein wrote: > Eryu, > > This is the 2nd revision of test for overlayfs NFS export. > > Changes since v1: > - Added new generic/exportfs test instead of changing existing test > - Addressed review comments on open_by_handle [-i|-o] patch > - Added the test that uses open_by_handle [-i|-o] in the same patch > - Split new overlay/exportfs tests to four tests, a pair of tests > for samefs and a pair of tests for non-samefs configuration. > - For each configuration, one test requires and enables redirect_dir > for directory rename test cases and both tests require and enable > nfs_export feature. > > The new overlay tests don't run on upstream kernel and pass with > my overlayfs NFS export V3 branch [1]. You may want to wait with > merging those tests until overlayfs NFS support is merged. Yeah, I probably will wait and let the kernel patches go first. (And I'll go through the overlay specific patches later.) > > Merging patches 1-3 seems like a good idea to me regardless of > overlayfs NFS export support. Patches 1-3 look good to me. One minor issue is that my compiler complained about "may be used uninitialized" on 'infile' and 'outfile' open_by_handle.c:350:12: warning: 'outfile' may be used uninitialized in this function [-Wmaybe-uninitialized] though I don't think we'll use them uninitialized in practice, I'd like to initialize them with NULL on commit. Thanks for the revision! Eryu