On 16.05.19 15:06, Nir Soffer wrote:
> On Thu, May 16, 2019 at 2:03 PM Max Reitz <mre...@redhat.com
> <mailto:mre...@redhat.com>> wrote:
> 
>     On 13.05.19 17:52, Max Reitz wrote:
>     > It is possible for an empty file to take up blocks on a filesystem.
>     > Make iotest 175 take this into account.
>     >
>     > Reported-by: Thomas Huth <th...@redhat.com <mailto:th...@redhat.com>>
>     > Signed-off-by: Max Reitz <mre...@redhat.com
>     <mailto:mre...@redhat.com>>
>     > ---
>     > v2: [Nir]
>     > - Use a function for filtering
>     > - s/empty_blocks/extra_blocks/
>     > ---
>     >  tests/qemu-iotests/175     | 26 ++++++++++++++++++++++----
>     >  tests/qemu-iotests/175.out |  8 ++++----
>     >  2 files changed, 26 insertions(+), 8 deletions(-)
>     >
>     > diff --git a/tests/qemu-iotests/175 b/tests/qemu-iotests/175
>     > index d0ffc495c2..b5eb0aa856 100755
>     > --- a/tests/qemu-iotests/175
>     > +++ b/tests/qemu-iotests/175
>     > @@ -28,10 +28,25 @@ status=1  # failure is the default!
>     > 
>     >  _cleanup()
>     >  {
>     > -     _cleanup_test_img
>     > +    _cleanup_test_img
>     > +    rm -f "$TEST_DIR/empty"
>     >  }
>     >  trap "_cleanup; exit \$status" 0 1 2 3 15
>     > 
>     > +# Some file systems sometimes allocate extra blocks independently of
>     > +# the file size.  This function hides the resulting difference in the
>     > +# stat -c '%b' output.
>     > +# Parameter 1: Number of blocks an empty file occupies
>     > +# Parameter 2: Image size in bytes
>     > +_filter_blocks()
>     > +{
>     > +    extra_blocks=$1
>     > +    img_size=$2
>     > +
>     > +    sed -e "s/blocks=$extra_blocks/nothing allocated/" \
>     > +        -e "s/blocks=$((extra_blocks + img_size /
>     512))/everything allocated/"
> 
>     Only now got around to creating an FS with similar characteristics to
>     Thomas’s.  Turns out this test still fails there because it prints
>     “blocks=nothing allocated050”...
> 
>     Unless objections arise, I’ll just add a \$ to each pattern to fix that.
>      (I could also just swap the lines, but that isn’t really right.)
> 
> 
> Sounds right.
> 
> Can you share how to create file system which reproduces this?
> Maybe adding the details in the test would be useful.

Sure:

$ mkfs.ext4 -b 1024 -I 128 /dev/loop0

-I 128 is the important one.  The block size just changes the number of
extra blocks (2 with 1024, 8 with 4096).

...I guess I’ll just a v3 now.

Max

Attachment: signature.asc
Description: OpenPGP digital signature

Reply via email to