Re: [PATCH 1/5] generic/449: not run on tmpfs earlier

["Darrick J. Wong" <djwong@kernel.org>]

On Fri, Jun 14, 2024 at 06:17:24AM +0000, Daniel Gomez wrote:
> From: Hugh Dickins <hughd@google.com>
> 
> Do not waste 14 minutes to discover that tmpfs succeeds in
> setting acls despite running out of space for user attrs.
> 
> Signed-off-by: Hugh Dickins <hughd@google.com>
> ---
>  tests/generic/449 | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/tests/generic/449 b/tests/generic/449
> index 2b77a6a49..ffbad12e9 100755
> --- a/tests/generic/449
> +++ b/tests/generic/449
> @@ -23,6 +23,12 @@ _require_scratch
>  _require_test
>  _require_acls
>  _require_attrs trusted
> +_require_block_device $SCRATCH_DEV # hack to exclude tmpfs for now

If you're going to _notrun tmpfs below, why is this ^^^ hack needed?

> +if [ "$FSTYP" = "tmpfs" ]; then
> +	# Do not waste 14 minutes to discover this:
> +	_notrun "$FSTYP succeeds in setting acls despite running out of space for user attrs"

and this should explain /why/ this test should be skipped for tmpfs:

_notrun "$FSTYP does not allocate acls and data from the same space pools"

(assuming tmpfs does not in fact charge acl and data to the same account
like ondisk filesystems have to)

--D

> +fi
>  
>  _scratch_mkfs_sized $((256 * 1024 * 1024)) >> $seqres.full 2>&1
>  _scratch_mount || _fail "mount failed"
> -- 
> 2.43.0
>