[Cluster-devel] [xfstests PATCH] gfs2: allow testing with gfs2
Eryu Guan
eguan at redhat.com
Thu Jul 27 05:18:30 UTC 2017
On Wed, Jul 26, 2017 at 02:43:31PM -0400, Jeff Layton wrote:
> From: Jeff Layton <jlayton at redhat.com>
>
> For now, just have it avoid locking. With a fixed kernel,
> generic/441 now passes.
>
> Signed-off-by: Jeff Layton <jlayton at redhat.com>
> ---
> common/config | 1 +
> common/rc | 7 +++++++
> tests/generic/441 | 2 +-
> 3 files changed, 9 insertions(+), 1 deletion(-)
>
> diff --git a/common/config b/common/config
> index 5091db9d5099..eb32efc105db 100644
> --- a/common/config
> +++ b/common/config
> @@ -244,6 +244,7 @@ case "$HOSTOS" in
> export MKFS_REISER4_PROG="`set_prog_path mkfs.reiser4`"
> export E2FSCK_PROG="`set_prog_path e2fsck`"
> export TUNE2FS_PROG="`set_prog_path tune2fs`"
> + export MKFS_GFS2_PROG="`set_prog_path mkfs.gfs2`"
> ;;
> esac
>
> diff --git a/common/rc b/common/rc
> index bd989bb5f4dd..70bb39225368 100644
> --- a/common/rc
> +++ b/common/rc
> @@ -158,6 +158,9 @@ case "$FSTYP" in
> ext4)
> [ "$MKFS_EXT4_PROG" = "" ] && _fatal "mkfs.ext4 not found"
> ;;
> + gfs2)
> + [ "$MKFS_GFS2_PROG" = "" ] && _fatal "mkfs.gfs2 not found"
> + ;;
The above two changes seem fine.
> f2fs)
> [ "$MKFS_F2FS_PROG" = "" ] && _fatal "mkfs.f2fs not found"
> ;;
> @@ -861,6 +864,10 @@ _scratch_mkfs()
> mkfs_cmd="yes | $MKFS_PROG -t $FSTYP --"
> mkfs_filter="grep -v -e ^mkfs\.ocfs2"
> ;;
> + gfs2)
> + mkfs_cmd="$MKFS_GFS2_PROG -p lock_nolock -O"
> + mkfs_filter="cat"
> + ;;
This seems unnecessary, we already have gfs2 support and the
"-p lock_nolock -O" part is already added to MKFS_OPTIONS in
_mkfs_opts() in common/config (when MKFS_OPTIONS is empty).
generic/411 runs and fails (fails due to kernel bug, not test config
issue) now on gfs2 even without this change.
> *)
> mkfs_cmd="yes | $MKFS_PROG -t $FSTYP --"
> mkfs_filter="cat"
> diff --git a/tests/generic/441 b/tests/generic/441
> index 075d87723ca1..b98f709cb6eb 100755
> --- a/tests/generic/441
> +++ b/tests/generic/441
> @@ -55,7 +55,7 @@ case $FSTYP in
> btrfs)
> _notrun "btrfs has a specialized test for this"
> ;;
> - ext3|ext4|xfs)
> + ext3|ext4|gfs2|xfs)
Without external log device support in _scratch_mkfs for gfs2, this
doesn't seem right to me. Perhaps we should add external log support for
gfs2 then update this test, in different patches?
Thanks,
Eryu
> # Do the more thorough test if we have a logdev
> _has_logdev && sflag=''
> ;;
> --
> 2.13.3
>
More information about the Cluster-devel
mailing list