From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from userp2120.oracle.com ([156.151.31.85]:34644 "EHLO userp2120.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750745AbdLMGDS (ORCPT ); Wed, 13 Dec 2017 01:03:18 -0500 Subject: [PATCH 0/8] weekly fstests changes From: "Darrick J. Wong" Date: Tue, 12 Dec 2017 22:03:10 -0800 Message-ID: <151314499003.18893.8687182548758898133.stgit@magnolia> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Sender: fstests-owner@vger.kernel.org To: eguan@redhat.com, darrick.wong@oracle.com Cc: linux-xfs@vger.kernel.org, fstests@vger.kernel.org List-ID: Hi all, Here's the usual weekly pile of fstests changes. :) We start with the same patch as last time that scans kmemleak for leaks after each test. Then we move on to a fix in the scrub probe code because upstream xfsprogs changed xfs_io syntax again. Third is a patch that adds Unicode linedraw character tests to the two tests that check that we can store arbitrary byte patterns without screwing things up. The fourth patch amends various xfs tests to deal with our slow removal of zero-alloc transactions, which is causing unexpected fs shutdowns when free space nears zero. The fifth patch adds a new test to invoke dm-error on a fsstress run; this test simulates (for the filesystem) an internal "yank all the disk cables out" test. The sixth patch adds FICLONERANGE/FIDEDUPERANGE support to fsstress. This was helpful in finding a bunch of unhandled corner cases in the xfs reflink implementation. Patch seven test write-only fsstress to maximize testing of write paths where coding mistakes cannot be backed out of so easily. The final patch in the series fixes the dir/file counts that are hardcoded in the xfs/068 golden output, since with fsstress doing reflink now we might get a lot more "files" than we used to. --D