Exercise the situation that cause ext4 to BUG_ON() when we use
zero range to zero a range which starts within the isize but ends
past the isize but still in the same block. This particular problem
has only been seen on systems with page_size > block_size.
This tests exercises the problem fixed in kernel with commit
a94b84e9ea3922220ae8bd654f6cb202904bbcac
ext4: Allocate entire range in zero range
Signed-off-by: Lukas Czerner <[email protected]>
---
tests/generic/084 | 60 +++++++++++++++++++++++++++++++++++++++++++++++++++
tests/generic/084.out | 1 +
tests/generic/group | 1 +
3 files changed, 62 insertions(+)
create mode 100644 tests/generic/084
create mode 100644 tests/generic/084.out
diff --git a/tests/generic/084 b/tests/generic/084
new file mode 100644
index 0000000..5066b14
--- /dev/null
+++ b/tests/generic/084
@@ -0,0 +1,60 @@
+#! /bin/bash
+# FS QA Test 084
+#
+# Exercise the situation that cause ext4 to BUG_ON() when we use
+# zero range to zero a range which starts within the isize but ends
+# past the isize but still in the same block. This particular problem
+# has only been seen on systems with page_size > block_size.
+#
+#-----------------------------------------------------------------------
+# Copyright (c) 2015 Red Hat, Inc., Lukas Czerner <[email protected]>
+#
+# This program is free software; you can redistribute it and/or
+# modify it under the terms of the GNU General Public License as
+# published by the Free Software Foundation.
+#
+# This program is distributed in the hope that it would be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write the Free Software Foundation,
+# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+#-----------------------------------------------------------------------
+#
+
+seq=`basename $0`
+seqres=$RESULT_DIR/$seq
+echo "QA output created by $seq"
+
+here=`pwd`
+tmp=/tmp/$$
+status=1 # failure is the default!
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+_cleanup()
+{
+ cd /
+ rm -f $tmp.*
+}
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
+# real QA test starts here
+
+# Modify as appropriate.
+_supported_fs generic
+_supported_os IRIX Linux
+_require_test
+_require_xfs_io_command "fzero"
+
+testfile=$TEST_DIR/$seq.$$
+
+$XFS_IO_PROG -f -c "pwrite 4096 512" -c "fzero 4351 512" $testfile > $seqres.full 2>&1
+
+# success, all done
+status=0
+exit
diff --git a/tests/generic/084.out b/tests/generic/084.out
new file mode 100644
index 0000000..d0aa380
--- /dev/null
+++ b/tests/generic/084.out
@@ -0,0 +1 @@
+QA output created by 084
diff --git a/tests/generic/group b/tests/generic/group
index c6ea3d8..6cb7dc3 100644
--- a/tests/generic/group
+++ b/tests/generic/group
@@ -86,6 +86,7 @@
081 auto quick
082 auto prealloc preallocrw quick
083 rw auto enospc stress
+084 auto prealloc quick
088 perms auto quick
089 metadata auto
091 rw auto quick
--
1.8.3.1
On Fri, Apr 24, 2015 at 04:38:29PM +0200, Lukas Czerner wrote:
> Exercise the situation that cause ext4 to BUG_ON() when we use
> zero range to zero a range which starts within the isize but ends
> past the isize but still in the same block. This particular problem
> has only been seen on systems with page_size > block_size.
>
> This tests exercises the problem fixed in kernel with commit
> a94b84e9ea3922220ae8bd654f6cb202904bbcac
> ext4: Allocate entire range in zero range
Seems the commit id is 0f2af21aae11972fa924374ddcf52e88347cf5a8 in Linus tree
Tested with 1k block size ext4 without the kernel patch, ext4 hit BUG_ON
as expected. xfs and btrfs passed this test.
Reviewed-by: Eryu Guan <[email protected]>
(with two minor comments inline)
>
> Signed-off-by: Lukas Czerner <[email protected]>
> ---
> tests/generic/084 | 60 +++++++++++++++++++++++++++++++++++++++++++++++++++
> tests/generic/084.out | 1 +
> tests/generic/group | 1 +
> 3 files changed, 62 insertions(+)
> create mode 100644 tests/generic/084
Change mode to 755? So git won't report mode chage after running the test
> create mode 100644 tests/generic/084.out
>
> diff --git a/tests/generic/084 b/tests/generic/084
> new file mode 100644
> index 0000000..5066b14
> --- /dev/null
> +++ b/tests/generic/084
> @@ -0,0 +1,60 @@
> +#! /bin/bash
> +# FS QA Test 084
> +#
> +# Exercise the situation that cause ext4 to BUG_ON() when we use
> +# zero range to zero a range which starts within the isize but ends
> +# past the isize but still in the same block. This particular problem
> +# has only been seen on systems with page_size > block_size.
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2015 Red Hat, Inc., Lukas Czerner <[email protected]>
> +#
> +# This program is free software; you can redistribute it and/or
> +# modify it under the terms of the GNU General Public License as
> +# published by the Free Software Foundation.
> +#
> +# This program is distributed in the hope that it would be useful,
> +# but WITHOUT ANY WARRANTY; without even the implied warranty of
> +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> +# GNU General Public License for more details.
> +#
> +# You should have received a copy of the GNU General Public License
> +# along with this program; if not, write the Free Software Foundation,
> +# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
> +#-----------------------------------------------------------------------
> +#
> +
> +seq=`basename $0`
> +seqres=$RESULT_DIR/$seq
> +echo "QA output created by $seq"
> +
> +here=`pwd`
> +tmp=/tmp/$$
> +status=1 # failure is the default!
> +trap "_cleanup; exit \$status" 0 1 2 3 15
> +
> +_cleanup()
> +{
> + cd /
> + rm -f $tmp.*
> +}
> +
> +# get standard environment, filters and checks
> +. ./common/rc
> +. ./common/filter
> +
> +# real QA test starts here
> +
> +# Modify as appropriate.
> +_supported_fs generic
> +_supported_os IRIX Linux
> +_require_test
> +_require_xfs_io_command "fzero"
> +
> +testfile=$TEST_DIR/$seq.$$
> +
> +$XFS_IO_PROG -f -c "pwrite 4096 512" -c "fzero 4351 512" $testfile > $seqres.full 2>&1
> +
> +# success, all done
> +status=0
> +exit
> diff --git a/tests/generic/084.out b/tests/generic/084.out
> new file mode 100644
> index 0000000..d0aa380
> --- /dev/null
> +++ b/tests/generic/084.out
> @@ -0,0 +1 @@
> +QA output created by 084
Add "Silence is golden" to indicate no output from test is expected?
Thanks,
Eryu
> diff --git a/tests/generic/group b/tests/generic/group
> index c6ea3d8..6cb7dc3 100644
> --- a/tests/generic/group
> +++ b/tests/generic/group
> @@ -86,6 +86,7 @@
> 081 auto quick
> 082 auto prealloc preallocrw quick
> 083 rw auto enospc stress
> +084 auto prealloc quick
> 088 perms auto quick
> 089 metadata auto
> 091 rw auto quick
> --
> 1.8.3.1
>
> --
> To unsubscribe from this list: send the line "unsubscribe fstests" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
On Mon, 27 Apr 2015, Eryu Guan wrote:
> Date: Mon, 27 Apr 2015 12:08:03 +0800
> From: Eryu Guan <[email protected]>
> To: Lukas Czerner <[email protected]>
> Cc: [email protected], [email protected]
> Subject: Re: [PATCH] generic: test bug when zero range is crossing isize
> within single block
>
> On Fri, Apr 24, 2015 at 04:38:29PM +0200, Lukas Czerner wrote:
> > Exercise the situation that cause ext4 to BUG_ON() when we use
> > zero range to zero a range which starts within the isize but ends
> > past the isize but still in the same block. This particular problem
> > has only been seen on systems with page_size > block_size.
> >
> > This tests exercises the problem fixed in kernel with commit
> > a94b84e9ea3922220ae8bd654f6cb202904bbcac
> > ext4: Allocate entire range in zero range
>
> Seems the commit id is 0f2af21aae11972fa924374ddcf52e88347cf5a8 in Linus tree
Hmm it's a94b84e9ea3922220ae8bd654f6cb202904bbcac on my
repository, but I might have something custom forgotten there.
Will resend, Thanks.
>
> Tested with 1k block size ext4 without the kernel patch, ext4 hit BUG_ON
> as expected. xfs and btrfs passed this test.
>
> Reviewed-by: Eryu Guan <[email protected]>
>
> (with two minor comments inline)
>
> >
> > Signed-off-by: Lukas Czerner <[email protected]>
> > ---
> > tests/generic/084 | 60 +++++++++++++++++++++++++++++++++++++++++++++++++++
> > tests/generic/084.out | 1 +
> > tests/generic/group | 1 +
> > 3 files changed, 62 insertions(+)
> > create mode 100644 tests/generic/084
>
> Change mode to 755? So git won't report mode chage after running the test
>
> > create mode 100644 tests/generic/084.out
> >
> > diff --git a/tests/generic/084 b/tests/generic/084
> > new file mode 100644
> > index 0000000..5066b14
> > --- /dev/null
> > +++ b/tests/generic/084
> > @@ -0,0 +1,60 @@
> > +#! /bin/bash
> > +# FS QA Test 084
> > +#
> > +# Exercise the situation that cause ext4 to BUG_ON() when we use
> > +# zero range to zero a range which starts within the isize but ends
> > +# past the isize but still in the same block. This particular problem
> > +# has only been seen on systems with page_size > block_size.
> > +#
> > +#-----------------------------------------------------------------------
> > +# Copyright (c) 2015 Red Hat, Inc., Lukas Czerner <[email protected]>
> > +#
> > +# This program is free software; you can redistribute it and/or
> > +# modify it under the terms of the GNU General Public License as
> > +# published by the Free Software Foundation.
> > +#
> > +# This program is distributed in the hope that it would be useful,
> > +# but WITHOUT ANY WARRANTY; without even the implied warranty of
> > +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> > +# GNU General Public License for more details.
> > +#
> > +# You should have received a copy of the GNU General Public License
> > +# along with this program; if not, write the Free Software Foundation,
> > +# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
> > +#-----------------------------------------------------------------------
> > +#
> > +
> > +seq=`basename $0`
> > +seqres=$RESULT_DIR/$seq
> > +echo "QA output created by $seq"
> > +
> > +here=`pwd`
> > +tmp=/tmp/$$
> > +status=1 # failure is the default!
> > +trap "_cleanup; exit \$status" 0 1 2 3 15
> > +
> > +_cleanup()
> > +{
> > + cd /
> > + rm -f $tmp.*
> > +}
> > +
> > +# get standard environment, filters and checks
> > +. ./common/rc
> > +. ./common/filter
> > +
> > +# real QA test starts here
> > +
> > +# Modify as appropriate.
> > +_supported_fs generic
> > +_supported_os IRIX Linux
> > +_require_test
> > +_require_xfs_io_command "fzero"
> > +
> > +testfile=$TEST_DIR/$seq.$$
> > +
> > +$XFS_IO_PROG -f -c "pwrite 4096 512" -c "fzero 4351 512" $testfile > $seqres.full 2>&1
> > +
> > +# success, all done
> > +status=0
> > +exit
> > diff --git a/tests/generic/084.out b/tests/generic/084.out
> > new file mode 100644
> > index 0000000..d0aa380
> > --- /dev/null
> > +++ b/tests/generic/084.out
> > @@ -0,0 +1 @@
> > +QA output created by 084
>
> Add "Silence is golden" to indicate no output from test is expected?
>
> Thanks,
> Eryu
> > diff --git a/tests/generic/group b/tests/generic/group
> > index c6ea3d8..6cb7dc3 100644
> > --- a/tests/generic/group
> > +++ b/tests/generic/group
> > @@ -86,6 +86,7 @@
> > 081 auto quick
> > 082 auto prealloc preallocrw quick
> > 083 rw auto enospc stress
> > +084 auto prealloc quick
> > 088 perms auto quick
> > 089 metadata auto
> > 091 rw auto quick
> > --
> > 1.8.3.1
> >
> > --
> > To unsubscribe from this list: send the line "unsubscribe fstests" in
> > the body of a message to [email protected]
> > More majordomo info at http://vger.kernel.org/majordomo-info.html
>