[PULL,48/48] iotests: Test qemu-img convert -S 0 behavior

Commit Message

Kevin Wolf March 29, 2016, 3:08 p.m. UTC

From: Max Reitz <[e-mail]>

Passing -S 0 to qemu-img convert should result in all source data being
copied to the output, even if that source data is known to be 0. The
output image should therefore have exactly the same size on disk as an
image which we explicitly filled with data.

Signed-off-by: Max Reitz <[e-mail]>
Reviewed-by: Fam Zheng <[e-mail]>
Signed-off-by: Kevin Wolf <[e-mail]>
---
 tests/qemu-iotests/150     | 105 +++++++++++++++++++++++++++++++++++++++++++++
 tests/qemu-iotests/150.out |  14 ++++++
 tests/qemu-iotests/group   |   1 +
 3 files changed, 120 insertions(+)
 create mode 100755 tests/qemu-iotests/150
 create mode 100644 tests/qemu-iotests/150.out

Comments

Paolo Bonzini April 7, 2016, 2:40 p.m. UTC | #1

On 29/03/2016 17:08, Kevin Wolf wrote:
> From: Max Reitz <[e-mail]>
> 
> Passing -S 0 to qemu-img convert should result in all source data being
> copied to the output, even if that source data is known to be 0. The
> output image should therefore have exactly the same size on disk as an
> image which we explicitly filled with data.
> 
> Signed-off-by: Max Reitz <[e-mail]>
> Reviewed-by: Fam Zheng <[e-mail]>
> Signed-off-by: Kevin Wolf <[e-mail]>

This fails with XFS and raw format, which preallocates an extra cluster
at the end of the file created with qemu-io (so that the on-disk size is
2056 512-byte blocks).

Paolo

Fam Zheng April 8, 2016, 1:18 a.m. UTC | #2

On Thu, 04/07 16:40, Paolo Bonzini wrote:
> 
> 
> On 29/03/2016 17:08, Kevin Wolf wrote:
> > From: Max Reitz <[e-mail]>
> > 
> > Passing -S 0 to qemu-img convert should result in all source data being
> > copied to the output, even if that source data is known to be 0. The
> > output image should therefore have exactly the same size on disk as an
> > image which we explicitly filled with data.
> > 
> > Signed-off-by: Max Reitz <[e-mail]>
> > Reviewed-by: Fam Zheng <[e-mail]>
> > Signed-off-by: Kevin Wolf <[e-mail]>
> 
> This fails with XFS and raw format, which preallocates an extra cluster
> at the end of the file created with qemu-io (so that the on-disk size is
> 2056 512-byte blocks).

Maybe squash in http://patchwork.ozlabs.org/patch/603047/ ?

Fam

Kevin Wolf April 8, 2016, 10:21 a.m. UTC | #3

Am 08.04.2016 um 03:18 hat Fam Zheng geschrieben:
> On Thu, 04/07 16:40, Paolo Bonzini wrote:
> > 
> > 
> > On 29/03/2016 17:08, Kevin Wolf wrote:
> > > From: Max Reitz <[e-mail]>
> > > 
> > > Passing -S 0 to qemu-img convert should result in all source data being
> > > copied to the output, even if that source data is known to be 0. The
> > > output image should therefore have exactly the same size on disk as an
> > > image which we explicitly filled with data.
> > > 
> > > Signed-off-by: Max Reitz <[e-mail]>
> > > Reviewed-by: Fam Zheng <[e-mail]>
> > > Signed-off-by: Kevin Wolf <[e-mail]>
> > 
> > This fails with XFS and raw format, which preallocates an extra cluster
> > at the end of the file created with qemu-io (so that the on-disk size is
> > 2056 512-byte blocks).
> 
> Maybe squash in http://patchwork.ozlabs.org/patch/603047/ ?

It's too late to squash it in, but yes, that should be the fix for it.

Kevin

Fam Zheng April 8, 2016, 10:42 a.m. UTC | #4

On Fri, 04/08 12:21, Kevin Wolf wrote:
> Am 08.04.2016 um 03:18 hat Fam Zheng geschrieben:
> > On Thu, 04/07 16:40, Paolo Bonzini wrote:
> > > 
> > > 
> > > On 29/03/2016 17:08, Kevin Wolf wrote:
> > > > From: Max Reitz <[e-mail]>
> > > > 
> > > > Passing -S 0 to qemu-img convert should result in all source data being
> > > > copied to the output, even if that source data is known to be 0. The
> > > > output image should therefore have exactly the same size on disk as an
> > > > image which we explicitly filled with data.
> > > > 
> > > > Signed-off-by: Max Reitz <[e-mail]>
> > > > Reviewed-by: Fam Zheng <[e-mail]>
> > > > Signed-off-by: Kevin Wolf <[e-mail]>
> > > 
> > > This fails with XFS and raw format, which preallocates an extra cluster
> > > at the end of the file created with qemu-io (so that the on-disk size is
> > > 2056 512-byte blocks).
> > 
> > Maybe squash in http://patchwork.ozlabs.org/patch/603047/ ?
> 
> It's too late to squash it in, but yes, that should be the fix for it.
> 

Yes, didn't know it's already merged.

Fam

Patch

diff --git a/tests/qemu-iotests/150 b/tests/qemu-iotests/150
new file mode 100755
index 0000000..97d2a35
--- /dev/null
+++ b/tests/qemu-iotests/150
@@ -0,0 +1,105 @@ 
+#!/bin/bash
+#
+# Test that qemu-img convert -S 0 fully allocates the target image
+#
+# Copyright (C) 2016 Red Hat, Inc.
+#
+# 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; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will 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, see <http://www.gnu.org/licenses/>.
+#
+
+# creator
+owner=[e-mail]
+
+seq="$(basename $0)"
+echo "QA output created by $seq"
+
+here="$PWD"
+tmp=/tmp/$$
+status=1	# failure is the default!
+
+_cleanup()
+{
+    _cleanup_test_img
+}
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+# get standard environment, filters and checks
+. ./common.rc
+. ./common.filter
+
+_supported_fmt generic
+_supported_proto file
+_supported_os Linux
+
+
+on_disk_size()
+{
+    du "$@" | sed -e 's/\t\+.*//'
+}
+
+
+img_size=1048576
+
+
+echo
+echo '=== Comparing empty image against sparse conversion ==='
+echo
+
+_make_test_img $img_size
+
+empty_size=$(on_disk_size "$TEST_IMG")
+
+
+$QEMU_IMG_PROG convert -O "$IMGFMT" -S 512 \
+    "json:{ 'driver': 'null-co', 'size': $img_size, 'read-zeroes': true }" \
+    "$TEST_IMG"
+
+sparse_convert_size=$(on_disk_size "$TEST_IMG")
+
+
+if [ "$empty_size" -eq "$sparse_convert_size" ]; then
+    echo 'Equal image size'
+else
+    echo 'Different image size'
+fi
+
+
+echo
+echo '=== Comparing full image against non-sparse conversion ==='
+echo
+
+_make_test_img $img_size
+$QEMU_IO -c "write 0 $img_size" "$TEST_IMG" | _filter_qemu_io
+
+full_size=$(on_disk_size "$TEST_IMG")
+
+
+$QEMU_IMG convert -O "$IMGFMT" -S 0 \
+    "json:{ 'driver': 'null-co', 'size': $img_size, 'read-zeroes': true }" \
+    "$TEST_IMG"
+
+non_sparse_convert_size=$(on_disk_size "$TEST_IMG")
+
+
+if [ "$full_size" -eq "$non_sparse_convert_size" ]; then
+    echo 'Equal image size'
+else
+    echo 'Different image size'
+fi
+
+
+# success, all done
+echo "*** done"
+rm -f $seq.full
+status=0
diff --git a/tests/qemu-iotests/150.out b/tests/qemu-iotests/150.out
new file mode 100644
index 0000000..2d29da1
--- /dev/null
+++ b/tests/qemu-iotests/150.out
@@ -0,0 +1,14 @@ 
+QA output created by 150
+
+=== Comparing empty image against sparse conversion ===
+
+Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=1048576
+Equal image size
+
+=== Comparing full image against non-sparse conversion ===
+
+Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=1048576
+wrote 1048576/1048576 bytes at offset 0
+1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+Equal image size
+*** done
diff --git a/tests/qemu-iotests/group b/tests/qemu-iotests/group
index 92ae8ec..2952b9d 100644
--- a/tests/qemu-iotests/group
+++ b/tests/qemu-iotests/group
@@ -151,3 +151,4 @@ 
 146 auto quick
 148 rw auto quick
 149 rw auto sudo
+150 rw auto quick