Re: [Qemu-devel] [PATCH 11/21] iotests: Prepare for refcount_width option

[Eric Blake <eblake@redhat.com>]

[-- Attachment #1: Type: text/plain, Size: 3726 bytes --]

On 11/10/2014 06:45 AM, Max Reitz wrote:
> Some tests do not work well with certain refcount widths (i.e. you
> cannot create internal snapshots with refcount_width=1), so make those
> widths unsupported.
> 
> Furthermore, add another filter to _filter_img_create in common.filter
> which filters out the refcount_width value.
> 
> Signed-off-by: Max Reitz <mreitz@redhat.com>
> ---
>  tests/qemu-iotests/007           |  4 ++++
>  tests/qemu-iotests/015           |  1 +
>  tests/qemu-iotests/026           | 11 +++++++++++
>  tests/qemu-iotests/029           |  1 +
>  tests/qemu-iotests/051           |  1 +
>  tests/qemu-iotests/058           |  1 +
>  tests/qemu-iotests/067           |  7 +++++++
>  tests/qemu-iotests/079           |  1 +
>  tests/qemu-iotests/080           |  1 +
>  tests/qemu-iotests/089           |  7 +++++++
>  tests/qemu-iotests/090           |  1 +
>  tests/qemu-iotests/108           |  6 ++++++
>  tests/qemu-iotests/common.filter |  3 ++-
>  13 files changed, 44 insertions(+), 1 deletion(-)
> 
> diff --git a/tests/qemu-iotests/007 b/tests/qemu-iotests/007
> index fe1a743..de39d1b 100755
> --- a/tests/qemu-iotests/007
> +++ b/tests/qemu-iotests/007
> @@ -43,6 +43,10 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
>  _supported_fmt qcow2
>  _supported_proto generic
>  _supported_os Linux
> +# refcount_width must be at least 4 bits so we can create ten internal snapshots
> +# (1 bit supports none, 2 bits support three, 4 bits support 15)

Feels like an off-by-one comment.  A width of 1 bit support a max
refcount of 1 (therefore no snapshots), a width of 2 bits supports a max
refcount of 3 (therefore 2 snapshots in addition to the original), a
width of 4 bits supports a max refcount of 15 (therefore only 14 snapshots).

> +++ b/tests/qemu-iotests/067
> @@ -35,6 +35,13 @@ status=1	# failure is the default!
>  _supported_fmt qcow2
>  _supported_proto file
>  _supported_os Linux
> +# Because this would change the output of query-block
> +_unsupported_imgopts 'refcount_width=1[^0-9]' \
> +                     'refcount_width=2[^0-9]' \
> +                     'refcount_width=4[^0-9]' \
> +                     'refcount_width=8[^0-9]' \
> +                     'refcount_width=32[^0-9]' \
> +                     'refcount_width=64[^0-9]'

It might be more compact to exploit globbing and just say:

_unsupported_imgopts 'refcount_width=?[^6]'

which leaves refcount_width=16 as the only pattern that doesn't match
the glob.  But that feels more fragile, so I can live with your longer list.

> +++ b/tests/qemu-iotests/089
> @@ -41,6 +41,13 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
>  _supported_fmt qcow2
>  _supported_proto file
>  _supported_os Linux
> +# Because this would change the output of qemu_io -c info
> +_unsupported_imgopts 'refcount_width=1[^0-9]' \

I like how you give reasons for some tests...

> +++ b/tests/qemu-iotests/090
> @@ -41,6 +41,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
>  _supported_fmt qcow2
>  _supported_proto file nfs
>  _supported_os Linux
> +_unsupported_imgopts 'refcount_width=1[^0-9]'

...so why not do it for all tests?

At any rate, the patch makes sense, so whether or not you tweak comments,

Reviewed-by: Eric Blake <eblake@redhat.com>

I'm assuming that later in the series you add a test that explicitly
covers the error message given when a refcount_order=0 (width=1) image
is attempted to be used with snapshots, since that will fail (internal
snapshots are simply not possible without a refcount that can't exceed 1).

-- 
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 539 bytes --]