On Wed, 2022-06-15 at 12:15 +0800, Zorro Lang wrote:
> On Tue, Jun 14, 2022 at 01:08:43PM +0800, An Long wrote:
> > _scratch_mkfs_sized only receive integer number of bytes as a valid
> > input. But if the MKFS_OPTIONS variable exists, it will use the
> > value of
> > block size in MKFS_OPTIONS to override input. In case of
> > MKFS_OPTIONS="-b 4k", would result in blocksize=4 but not 4096.
> > This
>
> Hi An,
>
> Good catch. Can you provide an example that a case fails on this
> issue, and
> then it tests passed after having this patch? I'm wondering why no
> one notice
> that before.
>
For example, I set MKFS_OPTIONS="-b 4k" and then run generic/416 on
ext4 filesystem.
# ./check tests/generic/416
FSTYP -- ext4
PLATFORM -- Linux/x86_64 bogon 5.3.18-57-default #1 SMP Wed Apr 28
10:54:41 UTC 2021 (ba3c2e9)
MKFS_OPTIONS -- -b 4k /dev/loop1
MOUNT_OPTIONS -- -o acl,user_xattr /dev/loop1 /mnt/scratch
generic/416 90s ... [failed, exit status 1]- output mismatch (see
/home/lan/work/xfstests/results//generic/416.out.bad)
--- tests/generic/416.out 2021-06-25 23:18:09.595254118 +0800
+++ /home/lan/work/xfstests/results//generic/416.out.bad 2022-
06-15 16:33:19.584206316 +0800
@@ -1,3 +1,4 @@
QA output created by 416
-wrote 16777216/16777216 bytes at offset 0
-XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+mount: /mnt/scratch: wrong fs type, bad option, bad superblock on
/dev/loop1, missing codepage or helper program, or other error.
+mount -o acl,user_xattr /dev/loop1 /mnt/scratch failed
+(see /home/lan/work/xfstests/results//generic/416.full for
details)
...
(Run 'diff -u /home/lan/work/xfstests/tests/generic/416.out
/home/lan/work/xfstests/results//generic/416.out.bad' to see the
entire diff)
Ran: generic/416
Failures: generic/416
Failed 1 of 1 tests
But the more common result as below:
# ./check tests/generic/416
FSTYP -- ext4
PLATFORM -- Linux/x86_64 bogon 5.3.18-57-default #1 SMP Wed Apr 28
10:54:41 UTC 2021 (ba3c2e9)
MKFS_OPTIONS -- -b 4k /dev/loop1
MOUNT_OPTIONS -- -o acl,user_xattr /dev/loop1 /mnt/scratch
umount: /mnt/scratch: target is busy.
generic/416 77s ... 90s
Ran: generic/416
Passed all 1 tests
Test passed due to _require_scratch_nocheck() ignore the umount error.
However, "mkfs.ext4: invalid block size - 4" was found in the 416.full
Apart from this, the error only occurs when MKFS_OPTIONS is set, and
doesn't affect xfs or btrfs. Since ext4 only supports 4k blocksize on
most platforms, we usually don't set blocksize in MKFS_OPTIONS.
I think that's why this problem hasn't been found before. In fact, I
found this issue when I tested on ppc64 with 64k kernel and 4k
subblocksize fs.
> > will give errors to ext2/3/4 etc, and brings potential bugs to xfs
> > or
> > btrfs.
> >
> > In addition, since we can receive various strings, so remove
> > integer
> > number check.
> >
> > Signed-off-by: An Long <lan@xxxxxxxx>
> > ---
> > common/rc | 18 ++++++------------
> > 1 file changed, 6 insertions(+), 12 deletions(-)
> >
> > diff --git a/common/rc b/common/rc
> > index 22050bc2..026007d3 100644
> > --- a/common/rc
> > +++ b/common/rc
> > @@ -1077,7 +1077,7 @@ _parse_size_from_string()
> > }
> >
> > # Create fs of certain size on scratch device
> > -# _scratch_mkfs_sized <size in bytes> [optional blocksize]
> > +# _scratch_mkfs_sized <size> [optional blocksize]
> > _scratch_mkfs_sized()
> > {
> > local fssize=$1
> > @@ -1086,13 +1086,13 @@ _scratch_mkfs_sized()
> >
> > case $FSTYP in
> > xfs)
> > - def_blksz=`echo $MKFS_OPTIONS | sed -rn 's/.*-b ?size=
> > ?+([0-9]+).*/\1/p'`
> > + def_blksz=`echo $MKFS_OPTIONS | sed -rn 's/.*-b ?size=
> > ?+([0-9]+[a-zA-Z]?).*/\1/p'`
> > ;;
> > btrfs)
> > - def_blksz=`echo $MKFS_OPTIONS | sed -rn 's/.*-s ?+([0-
> > 9]+).*/\1/p'`
> > + def_blksz=`echo $MKFS_OPTIONS | sed -rn 's/.*-s ?+([0-
> > 9]+[a-zA-Z]?).*/\1/p'`
> > ;;
> > ext2|ext3|ext4|ext4dev|udf|reiser4|ocfs2|reiserfs)
> > - def_blksz=`echo $MKFS_OPTIONS | sed -rn 's/.*-b ?+([0-
> > 9]+).*/\1/p'`
> > + def_blksz=`echo $MKFS_OPTIONS | sed -rn 's/.*-b ?+([0-
> > 9]+[a-zA-Z]?).*/\1/p'`
> > ;;
> > jfs)
> > def_blksz=4096
> > @@ -1101,14 +1101,8 @@ _scratch_mkfs_sized()
> >
> > [ -n "$def_blksz" ] && blocksize=$def_blksz
> > [ -z "$blocksize" ] && blocksize=4096
> > -
> > - local re='^[0-9]+$'
> > - if ! [[ $fssize =~ $re ]] ; then
> > - _notrun "error: _scratch_mkfs_sized: fs size
> > \"$fssize\" not an integer."
> > - fi
> > - if ! [[ $blocksize =~ $re ]] ; then
> > - _notrun "error: _scratch_mkfs_sized: block size
> > \"$blocksize\" not an integer."
> > - fi
> > + blocksize=$(_parse_size_from_string $blocksize)
> > + fssize=$(_parse_size_from_string $fssize)
>
> For now, the _parse_size_from_string is only used for this patch, I
> think these
> two patches can be merged into one patch, as it trys to fix one
> single problem.
>
There are a total of two logical changes. In particular it also include
a Function update. So it's better to keep two separate patch. But I'll
add patch depends on info to commit message.
Thank you for your time!
> Thanks,
> Zorro
>
> >
> > local blocks=`expr $fssize / $blocksize`
> >
> > --
> > 2.35.3
> >
--
An Long <lan@xxxxxxxx>
SUSE QE LSG, QE 2, Beijing
