[Cluster-devel] [PATCH gfs2/for-next] gfs2: Fix deadlock dumping resource group glocks

[Cluster-devel] [PATCH gfs2/for-next] gfs2: Fix deadlock dumping resource group glocks

[Alexander Aring]

Commit 0e539ca1bbbe ("gfs2: Fix NULL pointer dereference in gfs2_rgrp_dump")
introduced additional locking in gfs2_rgrp_go_dump, which is also used for
dumping resource group glocks via debugfs.  However, on that code path, the
glock spin lock is already taken in dump_glock, and taking it again in
gfs2_glock2rgrp leads to deadlock.  This can be reproduced with:

  $ mkfs.gfs2 -O -p lock_nolock /dev/FOO
  $ mount /dev/FOO /mnt/foo
  $ touch /mnt/foo/bar
  $ cat /sys/kernel/debug/gfs2/FOO/glocks

Fix that by not taking the glock spin lock inside the go_dump callback.

Fixes: 0e539ca1bbbe ("gfs2: Fix NULL pointer dereference in gfs2_rgrp_dump")
Signed-off-by: Alexander Aring <aahringo@redhat.com>
---
 fs/gfs2/glops.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c
index 67f2921ae8d4..6cedeefb7b3f 100644
--- a/fs/gfs2/glops.c
+++ b/fs/gfs2/glops.c
@@ -245,7 +245,7 @@ static void rgrp_go_inval(struct gfs2_glock *gl, int flags)
 static void gfs2_rgrp_go_dump(struct seq_file *seq, struct gfs2_glock *gl,
 			      const char *fs_id_buf)
 {
-	struct gfs2_rgrpd *rgd = gfs2_glock2rgrp(gl);
+	struct gfs2_rgrpd *rgd = gl->gl_object;
 
 	if (rgd)
 		gfs2_rgrp_dump(seq, rgd, fs_id_buf);
-- 
2.26.2

[Cluster-devel] [PATCH gfs2/for-next] gfs2: Fix deadlock dumping resource group glocks

[Andreas Gruenbacher]

On Mon, Nov 23, 2020 at 12:10 AM Alexander Aring <aahringo@redhat.com> wrote:
> Commit 0e539ca1bbbe ("gfs2: Fix NULL pointer dereference in gfs2_rgrp_dump")
> introduced additional locking in gfs2_rgrp_go_dump, which is also used for
> dumping resource group glocks via debugfs.  However, on that code path, the
> glock spin lock is already taken in dump_glock, and taking it again in
> gfs2_glock2rgrp leads to deadlock.  This can be reproduced with:
>
>   $ mkfs.gfs2 -O -p lock_nolock /dev/FOO
>   $ mount /dev/FOO /mnt/foo
>   $ touch /mnt/foo/bar
>   $ cat /sys/kernel/debug/gfs2/FOO/glocks
>
> Fix that by not taking the glock spin lock inside the go_dump callback.
>
> Fixes: 0e539ca1bbbe ("gfs2: Fix NULL pointer dereference in gfs2_rgrp_dump")
> Signed-off-by: Alexander Aring <aahringo@redhat.com>
> ---
>  fs/gfs2/glops.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c
> index 67f2921ae8d4..6cedeefb7b3f 100644
> --- a/fs/gfs2/glops.c
> +++ b/fs/gfs2/glops.c
> @@ -245,7 +245,7 @@ static void rgrp_go_inval(struct gfs2_glock *gl, int flags)
>  static void gfs2_rgrp_go_dump(struct seq_file *seq, struct gfs2_glock *gl,
>                               const char *fs_id_buf)
>  {
> -       struct gfs2_rgrpd *rgd = gfs2_glock2rgrp(gl);
> +       struct gfs2_rgrpd *rgd = gl->gl_object;
>
>         if (rgd)
>                 gfs2_rgrp_dump(seq, rgd, fs_id_buf);
> --
> 2.26.2
>

Thanks. I've added this to for-next for now, but I'll try to push this
to Linus before the 5.10 release.

Andreas