On Sun, Nov 03, 2019 at 07:49:39PM -0600, Eric Sandeen wrote: > On 11/3/19 9:24 AM, Eryu Guan wrote: > > On Tue, Oct 29, 2019 at 12:53:48PM -0500, Eric Sandeen wrote: > >> The ability to use a mounted device node as the primary argument > >> to xfs_growfs will be added back in, because it was an undocumented > >> behavior that some userspace depended on. This test exercises that > >> functionality. > >> > >> Signed-off-by: Eric Sandeen <sandeen@xxxxxxxxxx> > >> --- > >> > >> diff --git a/tests/xfs/148 b/tests/xfs/148 > >> new file mode 100755 > >> index 00000000..357ae01c > >> --- /dev/null > >> +++ b/tests/xfs/148 > >> @@ -0,0 +1,100 @@ > >> +#! /bin/bash > >> +# SPDX-License-Identifier: GPL-2.0 > >> +# Copyright (c) 2019 Red Hat, Inc. All Rights Reserved. > >> +# > >> +# FS QA Test 148 > >> +# > >> +# Test to ensure xfs_growfs command accepts device nodes if & only > >> +# if they are mounted. > >> +# This functionality, though undocumented, worked until xfsprogs v4.12 > >> +# It was added back and documented after xfsprogs v5.2 > > > > I'm testing with xfsprogs from for-next branch, which is v5.3.0-rc1 > > based xfs_growfs, but I still see failures like > > > > === xfs_growfs - check device node === > > +xfs_growfs: /dev/loop0 is not a mounted XFS filesystem > > === xfs_growfs - check device symlink === > > +xfs_growfs: /mnt/test/loop_symlink.21781 is not a mounted XFS filesystem > > === unmount === > > > > If it's already fixed, would you please list the related commits in > > commit log as well? > > I haven't merged the fix yet. Ah, that explains. I saw "It was added back and documented after xfsprogs v5.2", so I expected it to be PASS when testing with v5.3-rc1. > > If you like I can resend the test when it's merged. > Either way is fine, as long as the fix is referenced somewhere (either in commit log or in test description, and refer to the patch summary if it's not merged yet). Thanks, Eryu
