diff mbox series

[2/2] generic/059: also test that the file's mtime and ctime are updated

Message ID 20190619120624.9922-1-fdmanana@kernel.org (mailing list archive)
State New, archived
Headers show
Series [1/2] generic: test dir fsync after deleting dentry post eviction of its inode | expand

Commit Message

Filipe Manana June 19, 2019, 12:06 p.m. UTC 
From: Filipe Manana <fdmanana@suse.com>

Test as well that hole punch operations that affect a single file block
also update the file's mtime and ctime.

This is motivated by a bug a found in btrfs which is fixed by the
following patch for the linux kernel:

 "Btrfs: add missing inode version, ctime and mtime updates when
  punching hole"

Signed-off-by: Filipe Manana <fdmanana@suse.com>
---
 tests/generic/059 | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

Comments

Eryu Guan June 21, 2019, 10:36 a.m. UTC | #1 
On Wed, Jun 19, 2019 at 01:06:24PM +0100, fdmanana@kernel.org wrote:
> From: Filipe Manana <fdmanana@suse.com>
> 
> Test as well that hole punch operations that affect a single file block
> also update the file's mtime and ctime.
> 
> This is motivated by a bug a found in btrfs which is fixed by the
> following patch for the linux kernel:
> 
>  "Btrfs: add missing inode version, ctime and mtime updates when
>   punching hole"
> 
> Signed-off-by: Filipe Manana <fdmanana@suse.com>
> ---
>  tests/generic/059 | 18 ++++++++++++++++++
>  1 file changed, 18 insertions(+)
> 
> diff --git a/tests/generic/059 b/tests/generic/059
> index e8cb93d8..fd44b2ea 100755
> --- a/tests/generic/059
> +++ b/tests/generic/059
> @@ -18,6 +18,9 @@
>  #
>  #  Btrfs: add missing inode update when punching hole
>  #
> +# Also test the mtime and ctime properties of the file change after punching
> +# holes with ranges that operate only on a single block of the file.
> +#
>  seq=`basename $0`
>  seqres=$RESULT_DIR/$seq
>  echo "QA output created by $seq"
> @@ -68,6 +71,13 @@ $XFS_IO_PROG -c "fsync" $SCRATCH_MNT/foo
>  # fsync log.
>  sync
>  
> +# Sleep for 1 second, because we want to check that the next punch operations we
> +# do update the file's mtime and ctime.
> +sleep 1

Is this supposed to be after recording the initial c/mtime? i.e. moving
it after c/mtime_before?

Thanks,
Eryu

> +
> +mtime_before=$(stat -c %Y $SCRATCH_MNT/foo)
> +ctime_before=$(stat -c %Z $SCRATCH_MNT/foo)
> +
>  # Punch a hole in our file. This small range affects only 1 page.
>  # This made the btrfs hole punching implementation write only some zeroes in
>  # one page, but it did not update the btrfs inode fields used to determine if
> @@ -94,5 +104,13 @@ _flakey_drop_and_remount
>  echo "File content after:"
>  od -t x1 $SCRATCH_MNT/foo
>  
> +mtime_after=$(stat -c %Y $SCRATCH_MNT/foo)
> +ctime_after=$(stat -c %Z $SCRATCH_MNT/foo)
> +
> +[ $mtime_after -gt $mtime_before ] || \
> +	echo "mtime did not increase (before: $mtime_before after: $mtime_after"
> +[ $ctime_after -gt $ctime_before ] || \
> +	echo "ctime did not increase (before: $ctime_before after: $mtime_after"
> +
>  status=0
>  exit
> -- 
> 2.11.0
>
Filipe Manana June 21, 2019, 10:48 a.m. UTC | #2 
On Fri, Jun 21, 2019 at 11:36 AM Eryu Guan <guaneryu@gmail.com> wrote:
>
> On Wed, Jun 19, 2019 at 01:06:24PM +0100, fdmanana@kernel.org wrote:
> > From: Filipe Manana <fdmanana@suse.com>
> >
> > Test as well that hole punch operations that affect a single file block
> > also update the file's mtime and ctime.
> >
> > This is motivated by a bug a found in btrfs which is fixed by the
> > following patch for the linux kernel:
> >
> >  "Btrfs: add missing inode version, ctime and mtime updates when
> >   punching hole"
> >
> > Signed-off-by: Filipe Manana <fdmanana@suse.com>
> > ---
> >  tests/generic/059 | 18 ++++++++++++++++++
> >  1 file changed, 18 insertions(+)
> >
> > diff --git a/tests/generic/059 b/tests/generic/059
> > index e8cb93d8..fd44b2ea 100755
> > --- a/tests/generic/059
> > +++ b/tests/generic/059
> > @@ -18,6 +18,9 @@
> >  #
> >  #  Btrfs: add missing inode update when punching hole
> >  #
> > +# Also test the mtime and ctime properties of the file change after punching
> > +# holes with ranges that operate only on a single block of the file.
> > +#
> >  seq=`basename $0`
> >  seqres=$RESULT_DIR/$seq
> >  echo "QA output created by $seq"
> > @@ -68,6 +71,13 @@ $XFS_IO_PROG -c "fsync" $SCRATCH_MNT/foo
> >  # fsync log.
> >  sync
> >
> > +# Sleep for 1 second, because we want to check that the next punch operations we
> > +# do update the file's mtime and ctime.
> > +sleep 1
>
> Is this supposed to be after recording the initial c/mtime? i.e. moving
> it after c/mtime_before?

Either way is fine. Capturing the times right before or right after
the sleep, gives the same values as nothing changed the file.

Btw, I had noticed the other day that the second "echo" has
$mtime_after instead of $ctime_after (copy-paste mistake).
Do you want me to send a v2 fixing that typo, or you can do it
yourself when you pick the patch?

Thanks.
>
> Thanks,
> Eryu
>
> > +
> > +mtime_before=$(stat -c %Y $SCRATCH_MNT/foo)
> > +ctime_before=$(stat -c %Z $SCRATCH_MNT/foo)
> > +
> >  # Punch a hole in our file. This small range affects only 1 page.
> >  # This made the btrfs hole punching implementation write only some zeroes in
> >  # one page, but it did not update the btrfs inode fields used to determine if
> > @@ -94,5 +104,13 @@ _flakey_drop_and_remount
> >  echo "File content after:"
> >  od -t x1 $SCRATCH_MNT/foo
> >
> > +mtime_after=$(stat -c %Y $SCRATCH_MNT/foo)
> > +ctime_after=$(stat -c %Z $SCRATCH_MNT/foo)
> > +
> > +[ $mtime_after -gt $mtime_before ] || \
> > +     echo "mtime did not increase (before: $mtime_before after: $mtime_after"
> > +[ $ctime_after -gt $ctime_before ] || \
> > +     echo "ctime did not increase (before: $ctime_before after: $mtime_after"
> > +
> >  status=0
> >  exit
> > --
> > 2.11.0
> >
Eryu Guan June 21, 2019, 1:52 p.m. UTC | #3 
On Fri, Jun 21, 2019 at 11:48:57AM +0100, Filipe Manana wrote:
> On Fri, Jun 21, 2019 at 11:36 AM Eryu Guan <guaneryu@gmail.com> wrote:
> >
> > On Wed, Jun 19, 2019 at 01:06:24PM +0100, fdmanana@kernel.org wrote:
> > > From: Filipe Manana <fdmanana@suse.com>
> > >
> > > Test as well that hole punch operations that affect a single file block
> > > also update the file's mtime and ctime.
> > >
> > > This is motivated by a bug a found in btrfs which is fixed by the
> > > following patch for the linux kernel:
> > >
> > >  "Btrfs: add missing inode version, ctime and mtime updates when
> > >   punching hole"
> > >
> > > Signed-off-by: Filipe Manana <fdmanana@suse.com>
> > > ---
> > >  tests/generic/059 | 18 ++++++++++++++++++
> > >  1 file changed, 18 insertions(+)
> > >
> > > diff --git a/tests/generic/059 b/tests/generic/059
> > > index e8cb93d8..fd44b2ea 100755
> > > --- a/tests/generic/059
> > > +++ b/tests/generic/059
> > > @@ -18,6 +18,9 @@
> > >  #
> > >  #  Btrfs: add missing inode update when punching hole
> > >  #
> > > +# Also test the mtime and ctime properties of the file change after punching
> > > +# holes with ranges that operate only on a single block of the file.
> > > +#
> > >  seq=`basename $0`
> > >  seqres=$RESULT_DIR/$seq
> > >  echo "QA output created by $seq"
> > > @@ -68,6 +71,13 @@ $XFS_IO_PROG -c "fsync" $SCRATCH_MNT/foo
> > >  # fsync log.
> > >  sync
> > >
> > > +# Sleep for 1 second, because we want to check that the next punch operations we
> > > +# do update the file's mtime and ctime.
> > > +sleep 1
> >
> > Is this supposed to be after recording the initial c/mtime? i.e. moving
> > it after c/mtime_before?
> 
> Either way is fine. Capturing the times right before or right after
> the sleep, gives the same values as nothing changed the file.

Ah, you're right.

> 
> Btw, I had noticed the other day that the second "echo" has
> $mtime_after instead of $ctime_after (copy-paste mistake).
> Do you want me to send a v2 fixing that typo, or you can do it
> yourself when you pick the patch?

I can fix it on commit, thanks for pointing it out!

Thanks,
Eryu
diff mbox series

Patch

diff --git a/tests/generic/059 b/tests/generic/059
index e8cb93d8..fd44b2ea 100755
--- a/tests/generic/059
+++ b/tests/generic/059
@@ -18,6 +18,9 @@ 
 #
 #  Btrfs: add missing inode update when punching hole
 #
+# Also test the mtime and ctime properties of the file change after punching
+# holes with ranges that operate only on a single block of the file.
+#
 seq=`basename $0`
 seqres=$RESULT_DIR/$seq
 echo "QA output created by $seq"
@@ -68,6 +71,13 @@  $XFS_IO_PROG -c "fsync" $SCRATCH_MNT/foo
 # fsync log.
 sync
 
+# Sleep for 1 second, because we want to check that the next punch operations we
+# do update the file's mtime and ctime.
+sleep 1
+
+mtime_before=$(stat -c %Y $SCRATCH_MNT/foo)
+ctime_before=$(stat -c %Z $SCRATCH_MNT/foo)
+
 # Punch a hole in our file. This small range affects only 1 page.
 # This made the btrfs hole punching implementation write only some zeroes in
 # one page, but it did not update the btrfs inode fields used to determine if
@@ -94,5 +104,13 @@  _flakey_drop_and_remount
 echo "File content after:"
 od -t x1 $SCRATCH_MNT/foo
 
+mtime_after=$(stat -c %Y $SCRATCH_MNT/foo)
+ctime_after=$(stat -c %Z $SCRATCH_MNT/foo)
+
+[ $mtime_after -gt $mtime_before ] || \
+	echo "mtime did not increase (before: $mtime_before after: $mtime_after"
+[ $ctime_after -gt $ctime_before ] || \
+	echo "ctime did not increase (before: $ctime_before after: $mtime_after"
+
 status=0
 exit