From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sasha Levin Date: Wed, 8 Jul 2020 11:42:07 -0400 Subject: [Cluster-devel] [PATCH AUTOSEL 4.14 4/5] gfs2: read-only mounts should grab the sd_freeze_gl glock In-Reply-To: <20200708154208.3200232-1-sashal@kernel.org> References: <20200708154208.3200232-1-sashal@kernel.org> Message-ID: <20200708154208.3200232-4-sashal@kernel.org> List-Id: To: cluster-devel.redhat.com MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit From: Bob Peterson [ Upstream commit b780cc615ba4795a7ef0e93b19424828a5ad456a ] Before this patch, only read-write mounts would grab the freeze glock in read-only mode, as part of gfs2_make_fs_rw. So the freeze glock was never initialized. That meant requests to freeze, which request the glock in EX, were granted without any state transition. That meant you could mount a gfs2 file system, which is currently frozen on a different cluster node, in read-only mode. This patch makes read-only mounts lock the freeze glock in SH mode, which will block for file systems that are frozen on another node. Signed-off-by: Bob Peterson Signed-off-by: Sasha Levin --- fs/gfs2/ops_fstype.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index 7ed0359ebac61..2de67588ac2d8 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -1179,7 +1179,17 @@ static int fill_super(struct super_block *sb, struct gfs2_args *args, int silent goto fail_per_node; } - if (!sb_rdonly(sb)) { + if (sb_rdonly(sb)) { + struct gfs2_holder freeze_gh; + + error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_SHARED, + GL_EXACT, &freeze_gh); + if (error) { + fs_err(sdp, "can't make FS RO: %d\n", error); + goto fail_per_node; + } + gfs2_glock_dq_uninit(&freeze_gh); + } else { error = gfs2_make_fs_rw(sdp); if (error) { fs_err(sdp, "can't make FS RW: %d\n", error); -- 2.25.1