* [PATCH 0/4] rollup of fstests fixes
@ 2017-11-10 0:44 Darrick J. Wong
2017-11-10 0:44 ` [PATCH 1/4] xfs/31[3-7]: hide pwrite fsync errors that are supposed to fail Darrick J. Wong
` (3 more replies)
0 siblings, 4 replies; 6+ messages in thread
From: Darrick J. Wong @ 2017-11-10 0:44 UTC (permalink / raw)
To: eguan, darrick.wong; +Cc: linux-xfs, fstests
This is a rollup of all the patches I've been sending to the list.
The first patch fixes some problems seen with the new xfsprogs 4.14
release candidate; the second patch removes pad fields from the
fuzzing infrastructure; and the rest are all repeats from last week.
--D
^ permalink raw reply [flat|nested] 6+ messages in thread
* [PATCH 1/4] xfs/31[3-7]: hide pwrite fsync errors that are supposed to fail
2017-11-10 0:44 [PATCH 0/4] rollup of fstests fixes Darrick J. Wong
@ 2017-11-10 0:44 ` Darrick J. Wong
2017-11-10 0:44 ` [PATCH 2/4] common/fuzzy: ignore padding fields on xfs Darrick J. Wong
` (2 subsequent siblings)
3 siblings, 0 replies; 6+ messages in thread
From: Darrick J. Wong @ 2017-11-10 0:44 UTC (permalink / raw)
To: eguan, darrick.wong; +Cc: linux-xfs, fstests
From: Darrick J. Wong <darrick.wong@oracle.com>
xfs_io 4.14 will gain the ability to print error messages when
pwrite+fsync fail. Certain tests use the error injector to cause
failures, so the errors are expected. Since we test for a shut down
filesystem after the error injection, we can push the error messages to
the log.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
tests/xfs/312 | 2 +-
tests/xfs/313 | 2 +-
tests/xfs/314 | 2 +-
tests/xfs/315 | 2 +-
tests/xfs/317 | 2 +-
tests/xfs/326 | 2 +-
6 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/tests/xfs/312 b/tests/xfs/312
index 0c10441..0987527 100755
--- a/tests/xfs/312
+++ b/tests/xfs/312
@@ -79,7 +79,7 @@ echo "Inject error"
_scratch_inject_error "bmap_finish_one"
echo "CoW a few blocks"
-$XFS_IO_PROG -c "pwrite -W -S 0x67 -b $sz $((10 * blksz)) $((10 * blksz))" $SCRATCH_MNT/file2 >> $seqres.full
+$XFS_IO_PROG -c "pwrite -W -S 0x67 -b $sz $((10 * blksz)) $((10 * blksz))" $SCRATCH_MNT/file2 >> $seqres.full 2>&1
echo "FS should be shut down, touch will fail"
touch $SCRATCH_MNT/badfs 2>&1 | _filter_scratch
diff --git a/tests/xfs/313 b/tests/xfs/313
index 1a2cff2..0904b57 100755
--- a/tests/xfs/313
+++ b/tests/xfs/313
@@ -79,7 +79,7 @@ echo "Inject error"
_scratch_inject_error "refcount_finish_one"
echo "CoW a few blocks"
-$XFS_IO_PROG -c "pwrite -W -S 0x67 -b $sz $((10 * blksz)) $((10 * blksz))" $SCRATCH_MNT/file2 >> $seqres.full
+$XFS_IO_PROG -c "pwrite -W -S 0x67 -b $sz $((10 * blksz)) $((10 * blksz))" $SCRATCH_MNT/file2 >> $seqres.full 2>&1
echo "FS should be shut down, touch will fail"
touch $SCRATCH_MNT/badfs 2>&1 | _filter_scratch
diff --git a/tests/xfs/314 b/tests/xfs/314
index ec16c6f..e2e4d7c 100755
--- a/tests/xfs/314
+++ b/tests/xfs/314
@@ -81,7 +81,7 @@ echo "Inject error"
_scratch_inject_error "rmap_finish_one"
echo "CoW a few blocks"
-$XFS_IO_PROG -c "pwrite -W -S 0x67 -b $sz $((10 * blksz)) $((10 * blksz))" $SCRATCH_MNT/file2 >> $seqres.full
+$XFS_IO_PROG -c "pwrite -W -S 0x67 -b $sz $((10 * blksz)) $((10 * blksz))" $SCRATCH_MNT/file2 >> $seqres.full 2>&1
echo "FS should be shut down, touch will fail"
touch $SCRATCH_MNT/badfs 2>&1 | _filter_scratch
diff --git a/tests/xfs/315 b/tests/xfs/315
index 6170540..db26555 100755
--- a/tests/xfs/315
+++ b/tests/xfs/315
@@ -77,7 +77,7 @@ echo "Inject error"
_scratch_inject_error "free_extent"
echo "CoW a few blocks"
-$XFS_IO_PROG -c "pwrite -W -S 0x67 -b $sz 0 $sz" $SCRATCH_MNT/file1 >> $seqres.full
+$XFS_IO_PROG -c "pwrite -W -S 0x67 -b $sz 0 $sz" $SCRATCH_MNT/file1 >> $seqres.full 2>&1
sync
echo "FS should be shut down, touch will fail"
diff --git a/tests/xfs/317 b/tests/xfs/317
index 507c2a9..651a067 100755
--- a/tests/xfs/317
+++ b/tests/xfs/317
@@ -71,7 +71,7 @@ echo "Inject error"
_scratch_inject_error "rmap_finish_one"
echo "Write files"
-$XFS_IO_PROG -c "pwrite -W -S 0x67 -b $sz 0 $sz" $SCRATCH_MNT/file1 >> $seqres.full
+$XFS_IO_PROG -c "pwrite -W -S 0x67 -b $sz 0 $sz" $SCRATCH_MNT/file1 >> $seqres.full 2>&1
echo "FS should be shut down, touch will fail"
touch $SCRATCH_MNT/badfs 2>&1 | _filter_scratch
diff --git a/tests/xfs/326 b/tests/xfs/326
index 36bdcf4..929d3e6 100755
--- a/tests/xfs/326
+++ b/tests/xfs/326
@@ -87,7 +87,7 @@ echo "Inject error"
_scratch_inject_error "refcount_finish_one"
echo "CoW a few blocks"
-$XFS_IO_PROG -c "pwrite -W -S 0x67 -b $sz $((10 * blksz)) $((10 * blksz))" $SCRATCH_MNT/file2 >> $seqres.full
+$XFS_IO_PROG -c "pwrite -W -S 0x67 -b $sz $((10 * blksz)) $((10 * blksz))" $SCRATCH_MNT/file2 >> $seqres.full 2>&1
echo "FS should be shut down, touch will fail"
touch $SCRATCH_MNT/badfs 2>&1 | _filter_scratch
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH 2/4] common/fuzzy: ignore padding fields on xfs
2017-11-10 0:44 [PATCH 0/4] rollup of fstests fixes Darrick J. Wong
2017-11-10 0:44 ` [PATCH 1/4] xfs/31[3-7]: hide pwrite fsync errors that are supposed to fail Darrick J. Wong
@ 2017-11-10 0:44 ` Darrick J. Wong
2017-11-10 0:44 ` [PATCH 3/4] xfs: test that we don't leak inodes and dquots during failed cow recovery Darrick J. Wong
2017-11-10 0:44 ` [PATCH 4/4] generic/204: use available blocks to determine the number of files to create Darrick J. Wong
3 siblings, 0 replies; 6+ messages in thread
From: Darrick J. Wong @ 2017-11-10 0:44 UTC (permalink / raw)
To: eguan, darrick.wong; +Cc: linux-xfs, fstests
From: Darrick J. Wong <darrick.wong@oracle.com>
Padding fields are never checked and can have arbitrary values (if we
ever put them to use there'll be a feature flag) so there's no point
in fuzz-testing them.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
common/fuzzy | 14 ++++++++++----
1 file changed, 10 insertions(+), 4 deletions(-)
diff --git a/common/fuzzy b/common/fuzzy
index 87f7acc..b9c7399 100644
--- a/common/fuzzy
+++ b/common/fuzzy
@@ -79,6 +79,14 @@ _scratch_scrub() {
esac
}
+# Filter out any keys with an array index >= 10, collapse any array range
+# ("[1-195]") to the first item, and ignore padding fields.
+__filter_xfs_db_keys() {
+ sed -e '/\([a-z]*\)\[\([0-9][0-9]\+\)\].*/d' \
+ -e 's/\([a-zA-Z0-9_]*\)\[\([0-9]*\)-[0-9]*\]/\1[\2]/g' \
+ -e '/pad/d'
+}
+
# Filter the xfs_db print command's field debug information
# into field name and type.
__filter_xfs_db_print_fields() {
@@ -87,15 +95,13 @@ __filter_xfs_db_print_fields() {
filter='^'
fi
grep ' = ' | while read key equals value; do
- # Filter out any keys with an array index >= 10, and
- # collapse any array range ("[1-195]") to the first item.
- fuzzkey="$(echo "${key}" | sed -e '/\([a-z]*\)\[\([0-9][0-9]\+\)\].*/d' -e 's/\([a-zA-Z0-9_]*\)\[\([0-9]*\)-[0-9]*\]/\1[\2]/g')"
+ fuzzkey="$(echo "${key}" | __filter_xfs_db_keys)"
if [ -z "${fuzzkey}" ]; then
continue
elif [[ "${value}" == "["* ]]; then
echo "${value}" | sed -e 's/^.//g' -e 's/.$//g' -e 's/,/\n/g' | while read subfield; do
echo "${fuzzkey}.${subfield}"
- done
+ done | __filter_xfs_db_keys
else
echo "${fuzzkey}"
fi
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH 3/4] xfs: test that we don't leak inodes and dquots during failed cow recovery
2017-11-10 0:44 [PATCH 0/4] rollup of fstests fixes Darrick J. Wong
2017-11-10 0:44 ` [PATCH 1/4] xfs/31[3-7]: hide pwrite fsync errors that are supposed to fail Darrick J. Wong
2017-11-10 0:44 ` [PATCH 2/4] common/fuzzy: ignore padding fields on xfs Darrick J. Wong
@ 2017-11-10 0:44 ` Darrick J. Wong
2017-11-10 3:32 ` Eryu Guan
2017-11-10 0:44 ` [PATCH 4/4] generic/204: use available blocks to determine the number of files to create Darrick J. Wong
3 siblings, 1 reply; 6+ messages in thread
From: Darrick J. Wong @ 2017-11-10 0:44 UTC (permalink / raw)
To: eguan, darrick.wong; +Cc: linux-xfs, fstests
From: Darrick J. Wong <darrick.wong@oracle.com>
Add a couple of tests to check that we don't leak inodes or dquots
if CoW recovery fails and therefore the mount fails.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
common/rc | 1
tests/xfs/703 | 111 +++++++++++++++++++++++++++++++++++++++++++++++++++++
tests/xfs/703.out | 9 ++++
tests/xfs/704 | 89 ++++++++++++++++++++++++++++++++++++++++++
tests/xfs/704.out | 5 ++
tests/xfs/705 | 105 ++++++++++++++++++++++++++++++++++++++++++++++++++
tests/xfs/705.out | 9 ++++
tests/xfs/group | 3 +
8 files changed, 332 insertions(+)
create mode 100755 tests/xfs/703
create mode 100644 tests/xfs/703.out
create mode 100755 tests/xfs/704
create mode 100644 tests/xfs/704.out
create mode 100755 tests/xfs/705
create mode 100644 tests/xfs/705.out
diff --git a/common/rc b/common/rc
index ca420b0..c700102 100644
--- a/common/rc
+++ b/common/rc
@@ -3327,6 +3327,7 @@ _check_dmesg()
-e "(INFO|ERR): suspicious RCU usage" \
-e "INFO: possible circular locking dependency detected" \
-e "general protection fault:" \
+ -e "BUG .* remaining" \
$seqres.dmesg
if [ $? -eq 0 ]; then
_dump_err "_check_dmesg: something found in dmesg (see $seqres.dmesg)"
diff --git a/tests/xfs/703 b/tests/xfs/703
new file mode 100755
index 0000000..b2e37e6
--- /dev/null
+++ b/tests/xfs/703
@@ -0,0 +1,111 @@
+#! /bin/bash
+# FS QA Test No. 703
+#
+# Ensure that we don't leak quota inodes when CoW recovery fails.
+#
+# Use xfs_fsr to inject bmap redo items in the log for a linked file and
+# an unlinked file; enable quota so that we always mount with the quota
+# inodes; and then corrupt the refcount btree to ensure that the CoW
+# garbage collection (and therefore the mount) fail.
+#
+# On a subsequent mount attempt, we should be able to replay the bmap
+# items for the linked and unlinked files without prematurely truncating
+# the unlinked inode and without leaking the linked inode, and we should
+# be able to release the quota inodes when we're aborting the mount. We
+# also should not leak dquots.
+#
+#-----------------------------------------------------------------------
+# Copyright (c) 2017, Oracle and/or its affiliates. All Rights Reserved.
+#
+# 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 -rf "$tmp".*
+}
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+. ./common/attr
+. ./common/reflink
+. ./common/inject
+. ./common/quota
+. ./common/module
+
+# real QA test starts here
+_supported_os Linux
+_supported_fs xfs
+_require_loadable_fs_module "xfs"
+_require_quota
+_require_scratch_reflink
+_require_cp_reflink
+_require_command "$XFS_FSR_PROG" "xfs_fsr"
+_require_xfs_io_error_injection "bmap_finish_one"
+_require_xfs_scratch_rmapbt
+
+rm -f "$seqres.full"
+
+echo "Format and mount"
+_scratch_mkfs > "$seqres.full" 2>&1
+_scratch_mount -o noquota >> "$seqres.full" 2>&1
+
+testdir="$SCRATCH_MNT/test-$seq"
+blksz=65536
+blks=3
+mkdir "$testdir"
+
+echo "Create a many-block file"
+_pwrite_byte 0x62 0 $((blksz * blks)) $testdir/file1 >> $seqres.full
+_pwrite_byte 0x63 0 $blksz $testdir/file2 >> $seqres.full
+_reflink_range $testdir/file2 0 $testdir/file1 $blksz $blksz >> $seqres.full
+_scratch_cycle_mount noquota
+
+echo "Inject error"
+_scratch_inject_error "bmap_finish_one"
+
+echo "Defrag the file"
+$XFS_FSR_PROG -v -d $testdir/file1 >> $seqres.full 2>&1
+
+echo "FS should be shut down, touch will fail"
+touch $SCRATCH_MNT/badfs 2>&1 | _filter_scratch
+
+echo "Remount to replay log" | tee /dev/ttyprintk
+_scratch_unmount
+_scratch_dump_log >> $seqres.full
+_scratch_xfs_db -x -c 'agf 0' -c 'addr refcntroot' -c 'fuzz -d recs[1].startblock ones' >> $seqres.full
+_scratch_xfs_db -x -c 'agf 0' -c 'addr refcntroot' -c p >> $seqres.full
+
+# Suddenly enable quota to test if we can leak the quotacheck dquots!
+_scratch_mount -o quota >> $seqres.full 2>&1
+_scratch_unmount 2> /dev/null
+rm -f ${RESULT_DIR}/require_scratch
+
+echo "See if we leak"
+_reload_fs_module "xfs"
+
+# success, all done
+status=0
+exit
diff --git a/tests/xfs/703.out b/tests/xfs/703.out
new file mode 100644
index 0000000..d811880
--- /dev/null
+++ b/tests/xfs/703.out
@@ -0,0 +1,9 @@
+QA output created by 703
+Format and mount
+Create a many-block file
+Inject error
+Defrag the file
+FS should be shut down, touch will fail
+touch: cannot touch 'SCRATCH_MNT/badfs': Input/output error
+Remount to replay log
+See if we leak
diff --git a/tests/xfs/704 b/tests/xfs/704
new file mode 100755
index 0000000..3f4d704
--- /dev/null
+++ b/tests/xfs/704
@@ -0,0 +1,89 @@
+#! /bin/bash
+# FS QA Test No. 704
+#
+# Ensure that we don't leak dquots when CoW recovery fails.
+#
+# Corrupt the refcount btree to ensure that the CoW garbage collection
+# (and therefore the mount) fail.
+#
+# On a subsequent mount attempt, we should be able to release the quota
+# inodes when we're aborting the mount. We also should not leak dquots.
+#
+#-----------------------------------------------------------------------
+# Copyright (c) 2017, Oracle and/or its affiliates. All Rights Reserved.
+#
+# 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 -rf "$tmp".*
+}
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+. ./common/attr
+. ./common/reflink
+. ./common/quota
+. ./common/module
+
+# real QA test starts here
+_supported_os Linux
+_supported_fs xfs
+_require_loadable_fs_module "xfs"
+_require_quota
+_require_scratch_reflink
+_require_cp_reflink
+
+rm -f "$seqres.full"
+
+echo "Format and mount"
+_scratch_mkfs > "$seqres.full" 2>&1
+_scratch_mount -o quota >> "$seqres.full" 2>&1
+
+testdir="$SCRATCH_MNT/test-$seq"
+blksz=65536
+blks=3
+mkdir "$testdir"
+
+echo "Create a many-block file"
+_pwrite_byte 0x62 0 $((blksz * blks)) $testdir/file1 >> $seqres.full
+_pwrite_byte 0x63 0 $blksz $testdir/file2 >> $seqres.full
+_reflink_range $testdir/file2 0 $testdir/file1 $blksz $blksz >> $seqres.full
+
+echo "Remount to check recovery" | tee /dev/ttyprintk
+_scratch_unmount
+_scratch_xfs_db -x -c 'agf 0' -c 'addr refcntroot' -c 'fuzz -d recs[1].startblock ones' >> $seqres.full
+_scratch_xfs_db -x -c 'agf 0' -c 'addr refcntroot' -c p >> $seqres.full
+_scratch_mount -o quota >> $seqres.full 2>&1
+_scratch_unmount 2> /dev/null
+rm -f ${RESULT_DIR}/require_scratch
+
+echo "See if we leak"
+_reload_fs_module "xfs"
+
+# success, all done
+status=0
+exit
diff --git a/tests/xfs/704.out b/tests/xfs/704.out
new file mode 100644
index 0000000..b237b2d
--- /dev/null
+++ b/tests/xfs/704.out
@@ -0,0 +1,5 @@
+QA output created by 704
+Format and mount
+Create a many-block file
+Remount to check recovery
+See if we leak
diff --git a/tests/xfs/705 b/tests/xfs/705
new file mode 100755
index 0000000..6019a4e
--- /dev/null
+++ b/tests/xfs/705
@@ -0,0 +1,105 @@
+#! /bin/bash
+# FS QA Test No. 705
+#
+# Ensure that we don't leak inodes when CoW recovery fails.
+#
+# Use xfs_fsr to inject bmap redo items in the log for a linked file and
+# an unlinked file; and then corrupt the refcount btree to ensure that
+# the CoW garbage collection (and therefore the mount) fail.
+#
+# On a subsequent mount attempt, we should be able to replay the bmap
+# items for the linked and unlinked files without prematurely truncating
+# the unlinked inode and without leaking the linked inode, and we should
+# be able to release all the inodes when we're aborting the mount.
+#
+#-----------------------------------------------------------------------
+# Copyright (c) 2017, Oracle and/or its affiliates. All Rights Reserved.
+#
+# 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 -rf "$tmp".*
+}
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+. ./common/attr
+. ./common/reflink
+. ./common/inject
+. ./common/module
+
+# real QA test starts here
+_supported_os Linux
+_supported_fs xfs
+_require_loadable_fs_module "xfs"
+_require_scratch_reflink
+_require_cp_reflink
+_require_command "$XFS_FSR_PROG" "xfs_fsr"
+_require_xfs_io_error_injection "bmap_finish_one"
+_require_xfs_scratch_rmapbt
+
+rm -f "$seqres.full"
+
+echo "Format and mount"
+_scratch_mkfs > "$seqres.full" 2>&1
+_scratch_mount -o noquota >> "$seqres.full" 2>&1
+
+testdir="$SCRATCH_MNT/test-$seq"
+blksz=65536
+blks=3
+mkdir "$testdir"
+
+echo "Create a many-block file"
+_pwrite_byte 0x62 0 $((blksz * blks)) $testdir/file1 >> $seqres.full
+_pwrite_byte 0x63 0 $blksz $testdir/file2 >> $seqres.full
+_reflink_range $testdir/file2 0 $testdir/file1 $blksz $blksz >> $seqres.full
+_scratch_cycle_mount noquota
+
+echo "Inject error"
+_scratch_inject_error "bmap_finish_one"
+
+echo "Defrag the file"
+$XFS_FSR_PROG -v -d $testdir/file1 >> $seqres.full 2>&1
+
+echo "FS should be shut down, touch will fail"
+touch $SCRATCH_MNT/badfs 2>&1 | _filter_scratch
+
+echo "Remount to replay log" | tee /dev/ttyprintk
+_scratch_unmount
+_scratch_dump_log >> $seqres.full
+_scratch_xfs_db -x -c 'agf 0' -c 'addr refcntroot' -c 'fuzz -d recs[1].startblock ones' >> $seqres.full
+_scratch_xfs_db -x -c 'agf 0' -c 'addr refcntroot' -c p >> $seqres.full
+_scratch_mount -o noquota >> $seqres.full 2>&1
+_scratch_unmount 2> /dev/null
+rm -f ${RESULT_DIR}/require_scratch
+
+echo "See if we leak"
+_reload_fs_module "xfs"
+
+# success, all done
+status=0
+exit
diff --git a/tests/xfs/705.out b/tests/xfs/705.out
new file mode 100644
index 0000000..0bcd9ab
--- /dev/null
+++ b/tests/xfs/705.out
@@ -0,0 +1,9 @@
+QA output created by 705
+Format and mount
+Create a many-block file
+Inject error
+Defrag the file
+FS should be shut down, touch will fail
+touch: cannot touch 'SCRATCH_MNT/badfs': Input/output error
+Remount to replay log
+See if we leak
diff --git a/tests/xfs/group b/tests/xfs/group
index b439842..e56666b 100644
--- a/tests/xfs/group
+++ b/tests/xfs/group
@@ -431,3 +431,6 @@
431 auto quick dangerous
432 auto quick dir metadata
433 auto quick attr
+703 auto quick clone fsr
+704 auto quick clone
+705 auto quick clone fsr
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH 4/4] generic/204: use available blocks to determine the number of files to create
2017-11-10 0:44 [PATCH 0/4] rollup of fstests fixes Darrick J. Wong
` (2 preceding siblings ...)
2017-11-10 0:44 ` [PATCH 3/4] xfs: test that we don't leak inodes and dquots during failed cow recovery Darrick J. Wong
@ 2017-11-10 0:44 ` Darrick J. Wong
3 siblings, 0 replies; 6+ messages in thread
From: Darrick J. Wong @ 2017-11-10 0:44 UTC (permalink / raw)
To: eguan, darrick.wong; +Cc: linux-xfs, fstests
From: Darrick J. Wong <darrick.wong@oracle.com>
Use the available block count to compute the number of files we think
we can create, rather than hardcoding a particular size. This fixes
the ENOSPC failures for xfs filesystems with rmap/reflink support.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
tests/generic/204 | 14 ++++++++++++--
1 file changed, 12 insertions(+), 2 deletions(-)
diff --git a/tests/generic/204 b/tests/generic/204
index 4c203a2..71bee83 100755
--- a/tests/generic/204
+++ b/tests/generic/204
@@ -69,7 +69,7 @@ _scratch_mount
# work out correctly. Space usages is based 22500 files and 1024 reserved blocks
# on a 4k block size 256 byte inode size filesystem.
resv_blks=1024
-space=97920000
+space=$(stat -f -c '%f * %S' $SCRATCH_MNT | $BC_PROG)
# decrease files for inode size.
# 22500 * (256 + 4k) = ~97MB
@@ -78,10 +78,20 @@ space=97920000
files=$((space / (isize + dbsize)))
+# Now do it again, but factor in the filename sizes too.
+# We naïvely assume 8 bytes for inode number, 1 byte for ftype,
+# and 1 more byte for namelen, then round up to the nearest 8
+# bytes.
+
+namelen="$(echo -n "$files" | wc -c)"
+direntlen="$(echo "(10 + $namelen + 7) / 8 * 8" | $BC_PROG)"
+
+files=$((space / (direntlen + isize + dbsize)))
+
echo files $files, resvblks $resv_blks >> $seqres.full
_scratch_resvblks $resv_blks >> $seqres.full 2>&1
-for i in `seq 1 $files`; do
+for i in `seq -w 1 $files`; do
echo -n > $SCRATCH_MNT/$i
echo XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX > $SCRATCH_MNT/$i
done
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH 3/4] xfs: test that we don't leak inodes and dquots during failed cow recovery
2017-11-10 0:44 ` [PATCH 3/4] xfs: test that we don't leak inodes and dquots during failed cow recovery Darrick J. Wong
@ 2017-11-10 3:32 ` Eryu Guan
0 siblings, 0 replies; 6+ messages in thread
From: Eryu Guan @ 2017-11-10 3:32 UTC (permalink / raw)
To: Darrick J. Wong; +Cc: linux-xfs, fstests
On Thu, Nov 09, 2017 at 04:44:40PM -0800, Darrick J. Wong wrote:
> From: Darrick J. Wong <darrick.wong@oracle.com>
>
> Add a couple of tests to check that we don't leak inodes or dquots
> if CoW recovery fails and therefore the mount fails.
>
> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
I thought I've picked them up in last update.. Will queue them in now.
Thanks,
Eryu
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2017-11-10 3:32 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2017-11-10 0:44 [PATCH 0/4] rollup of fstests fixes Darrick J. Wong
2017-11-10 0:44 ` [PATCH 1/4] xfs/31[3-7]: hide pwrite fsync errors that are supposed to fail Darrick J. Wong
2017-11-10 0:44 ` [PATCH 2/4] common/fuzzy: ignore padding fields on xfs Darrick J. Wong
2017-11-10 0:44 ` [PATCH 3/4] xfs: test that we don't leak inodes and dquots during failed cow recovery Darrick J. Wong
2017-11-10 3:32 ` Eryu Guan
2017-11-10 0:44 ` [PATCH 4/4] generic/204: use available blocks to determine the number of files to create Darrick J. Wong
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).