gfs2: fix timestamp handling on quota inodes
authorJeff Layton <jlayton@kernel.org>
Thu, 13 Jul 2023 13:52:48 +0000 (09:52 -0400)
committerChristian Brauner <brauner@kernel.org>
Mon, 24 Jul 2023 08:30:08 +0000 (10:30 +0200)
commitd85f1b5bad5752831b11bdeb5a1c403075673c0d
tree78452a9f36e7ce7b391135e383b1ffdc1092fc1d
parent13bc24457850583a2e7203ded05b7209ab4bc5ef
gfs2: fix timestamp handling on quota inodes

While these aren't generally visible from userland, it's best to be
consistent with timestamp handling. When adjusting the quota, update the
mtime and ctime like we would with a write operation on any other inode,
and avoid updating the atime which should only be done for reads.

Signed-off-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Andreas Gruenbacher <agruenba@redhat.com>
Message-Id: <20230713135249.153796-1-jlayton@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
fs/gfs2/quota.c