Skip to contentSkip to conversation

xfs: testcases for shrinking free space in the last AG

RFC PATCH Superseded Gao Xiang 3 participants 12 messages Mar 15, 2021 – Mar 29, 2021

v3 of 6 Superseded Latest is v6 →

lore
Share & tools
lore

Gao Xiang

author
#0 lore raw

v3 cover xfs: testcases for shrinking free space in the last AG

Hi,

This version matches
kernel v8: https://lore.kernel.org/r/[email protected]
xfsprogs RFC: https://lore.kernel.org/r/[email protected]

and mainly addresses comments for the previous version.

Thanks,
Gao Xinag

changes since RFC v2:
- [1/3] add _require_scratch to _require_xfs_shrink() (Zorro);
- [1/3] drop unneeded _scratch_unmount (Zorro);
- [2/3] add _repair_scratch_fs for each test (Zorro);
- [2/3] check new size after shrinking (Zorro);
- [2/3] drop unneeded _scratch_unmount (Zorro);
- [3/3] update group as "auto growfs ioctl prealloc stress" to follow xfs/104 (Zorro);
- [3/3] use generic falloc instead of resvsp (Darrick).

Gao Xiang:
common/xfs: add a _require_xfs_shrink helper
xfs: basic functionality test for shrinking free space in the last AG
xfs: stress test for shrinking free space in the last AG

 common/xfs        |  11 +++++
 tests/xfs/990     |  70 ++++++++++++++++++++++++++
 tests/xfs/990.out |  12 +++++
 tests/xfs/991     | 122 ++++++++++++++++++++++++++++++++++++++++++++++
 tests/xfs/991.out |   8 +++
 tests/xfs/group   |   2 +
 6 files changed, 225 insertions(+)
 create mode 100755 tests/xfs/990
 create mode 100644 tests/xfs/990.out
 create mode 100755 tests/xfs/991
 create mode 100644 tests/xfs/991.out
Signature · 1 line
2.27.0
author
#1 lore raw

RFC PATCH v3 1/3 common/xfs: add a _require_xfs_shrink helper

In order to detect whether the current kernel supports XFS shrinking.

---
 common/xfs | 11 +++++++++++
 1 file changed, 11 insertions(+)
common/xfs +11 -0
index 2156749d..ea3b6cab 100644
@@ -432,6 +432,17 @@ _supports_xfs_scrub()
 	return 0
 }
 
+_require_xfs_shrink()
+{
+	_require_scratch
+
+	_scratch_mkfs_xfs > /dev/null
+	_scratch_mount
+	$XFS_GROWFS_PROG -D1 "$SCRATCH_MNT" 2>&1 | grep -q 'Invalid argument' || \
+		_notrun "kernel does not support shrinking"
+	_scratch_unmount
+}
+
 # run xfs_check and friends on a FS.
 _check_xfs_filesystem()
 {
Signature · 1 line
2.27.0
#2 lore raw

Re: [RFC PATCH v3 1/3] common/xfs: add a _require_xfs_shrink helper

Quoted 12 lines · Gao Xiang

In order to detect whether the current kernel supports XFS shrinking.

---
 common/xfs | 11 +++++++++++
 1 file changed, 11 insertions(+)
common/xfs +0 -0
index 2156749d..ea3b6cab 100644

common/xfs _supports_xfs_scrub() · line 437

Quoted 6 lines
common/xfs +3 -0
@@ -432,6 +432,17 @@ _supports_xfs_scrub()
 	return 0
 }
 
+_require_xfs_shrink()
+{
+	_require_scratch

_require_command "$XFS_GROWFS_PROG" xfs_growfs

common/xfs _supports_xfs_scrub() · line 442

Quoted 5 lines
common/xfs +5 -0
+
+	_scratch_mkfs_xfs > /dev/null
+	_scratch_mount
+	$XFS_GROWFS_PROG -D1 "$SCRATCH_MNT" 2>&1 | grep -q 'Invalid argument' || \
+		_notrun "kernel does not support shrinking"

Better to describe the behavior here to explain why EINVAL means kernel supports shrink.

Thanks,
Eryu

Quoted 8 lines
common/xfs +3 -0
+	_scratch_unmount
+}
+
 # run xfs_check and friends on a FS.
 _check_xfs_filesystem()
 {
Signature · 1 line
2.27.0
author
#3 lore raw

Re: [RFC PATCH v3 1/3] common/xfs: add a _require_xfs_shrink helper

Quoted 16 lines · Eryu Guan
Quoted 15 lines · Gao Xiang

In order to detect whether the current kernel supports XFS shrinking.

---
 common/xfs | 11 +++++++++++
 1 file changed, 11 insertions(+)
common/xfs +0 -0
index 2156749d..ea3b6cab 100644
@@ -432,6 +432,17 @@ _supports_xfs_scrub()
 	return 0
 }
 
Quoted 5 lines

common/xfs _supports_xfs_scrub() · line 437

Quoted 3 lines
common/xfs +3 -0
+_require_xfs_shrink()
+{
+	_require_scratch

_require_command "$XFS_GROWFS_PROG" xfs_growfs

will update.

Quoted 9 lines

common/xfs _supports_xfs_scrub() · line 442

Quoted 6 lines
common/xfs +5 -0
+
+	_scratch_mkfs_xfs > /dev/null
+	_scratch_mount
+	$XFS_GROWFS_PROG -D1 "$SCRATCH_MNT" 2>&1 | grep -q 'Invalid argument' || \
+		_notrun "kernel does not support shrinking"

Better to describe the behavior here to explain why EINVAL means kernel supports shrink.

hmmm.. I also don't think this check is correct now. Maybe try to shrink one fs-block would be better. Will update later.

Thanks,
Gao Xiang

Quoted 13 lines

Thanks,
Eryu

Quoted 8 lines
common/xfs +3 -0
+	_scratch_unmount
+}
+
 # run xfs_check and friends on a FS.
 _check_xfs_filesystem()
 {
Signature · 1 line
2.27.0
author
#4 lore raw

RFC PATCH v3 2/3 xfs: basic functionality test for shrinking free space in the last AG

In reply to Gao Xiang

Add basic test to make sure the functionality works as expected.

---
 tests/xfs/990     | 70 +++++++++++++++++++++++++++++++++++++++++++++++
 tests/xfs/990.out | 12 ++++++++
 tests/xfs/group   |  1 +
 3 files changed, 83 insertions(+)
 create mode 100755 tests/xfs/990
 create mode 100644 tests/xfs/990.out
tests/xfs/990 new file +70 -0
new file mode 100755
index 00000000..3c79186e
@@ -0,0 +1,70 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2021 Red Hat, Inc.  All Rights Reserved.
+#
+# FS QA Test 990
+#
+# XFS shrinkfs basic functionality test
+#
+# This test attempts to shrink with a small size (512K), half AG size and
+# an out-of-bound size (agsize + 1) to observe if it works as expected.
+#
+seq=`basename $0`
+seqres=$RESULT_DIR/$seq
+echo "QA output created by $seq"
+
+here=`pwd`
+tmp=/tmp/$$
+status=1    # failure is the default!
+trap "rm -f $tmp.*; exit \$status" 0 1 2 3 15
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
+test_shrink()
+{
+	$XFS_GROWFS_PROG -D"$1" $SCRATCH_MNT >> $seqres.full 2>&1
+	ret=$?
+
+	_scratch_unmount
+	_repair_scratch_fs >> $seqres.full
+	_scratch_mount >> $seqres.full
+
+	$XFS_INFO_PROG $SCRATCH_MNT 2>&1 | _filter_mkfs 2>$tmp.growfs >/dev/null
+	. $tmp.growfs
+	[ $ret -eq 0 -a $1 -eq $dblocks ]
+}
+
+# real QA test starts here
+_supported_fs xfs
+_require_xfs_shrink
+
+rm -f $seqres.full
+echo "Format and mount"
+_scratch_mkfs -dsize="$((512 * 1024 * 1024))" -dagcount=3 2>&1 | \
+	tee -a $seqres.full | _filter_mkfs 2>$tmp.mkfs
+. $tmp.mkfs
+t_dblocks=$dblocks
+_scratch_mount >> $seqres.full
+
+echo "Shrink fs (small size)"
+test_shrink $((t_dblocks-512*1024/dbsize)) || \
+	_fail "Shrink fs (small size) failure"
+
+echo "Shrink fs (half AG)"
+test_shrink $((t_dblocks-agsize/2)) || \
+	_fail "Shrink fs (half AG) failure"
+
+echo "Shrink fs (out-of-bound)"
+test_shrink $((t_dblocks-agsize-1)) && \
+	_fail "Shrink fs (out-of-bound) failure"
+[ $dblocks -ne $((t_dblocks-agsize/2)) ] && \
+	_fail "dblocks changed after shrinking failure"
+
+$XFS_INFO_PROG $SCRATCH_MNT >> $seqres.full
+echo "*** done"
+
+# success, all done
+status=0
+exit
tests/xfs/990.out new file +12 -0
new file mode 100644
index 00000000..812f89ef
@@ -0,0 +1,12 @@
+QA output created by 990
+Format and mount
+meta-data=DDEV isize=XXX agcount=N, agsize=XXX blks
+data     = bsize=XXX blocks=XXX, imaxpct=PCT
+         = sunit=XXX swidth=XXX, unwritten=X
+naming   =VERN bsize=XXX
+log      =LDEV bsize=XXX blocks=XXX
+realtime =RDEV extsz=XXX blocks=XXX, rtextents=XXX
+Shrink fs (small size)
+Shrink fs (half AG)
+Shrink fs (out-of-bound)
+*** done
tests/xfs/group +1 -0
index e861cec9..a7981b67 100644
@@ -525,3 +525,4 @@
 525 auto quick mkfs
 526 auto quick mkfs
 527 auto quick quota
+990 auto quick growfs
Signature · 1 line
2.27.0
#5 lore raw

Re: [RFC PATCH v3 2/3] xfs: basic functionality test for shrinking free space in the last AG

Quoted 35 lines · Gao Xiang

Add basic test to make sure the functionality works as expected.

---
 tests/xfs/990     | 70 +++++++++++++++++++++++++++++++++++++++++++++++
 tests/xfs/990.out | 12 ++++++++
 tests/xfs/group   |  1 +
 3 files changed, 83 insertions(+)
 create mode 100755 tests/xfs/990
 create mode 100644 tests/xfs/990.out
tests/xfs/990 new file +18 -0
new file mode 100755
index 00000000..3c79186e
@@ -0,0 +1,70 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2021 Red Hat, Inc.  All Rights Reserved.
+#
+# FS QA Test 990
+#
+# XFS shrinkfs basic functionality test
+#
+# This test attempts to shrink with a small size (512K), half AG size and
+# an out-of-bound size (agsize + 1) to observe if it works as expected.
+#
+seq=`basename $0`
+seqres=$RESULT_DIR/$seq
+echo "QA output created by $seq"
+
+here=`pwd`
+tmp=/tmp/$$
+status=1    # failure is the default!

tests/xfs/990 line 28

Quoted 10 lines
tests/xfs/990 new file +10 -0
+trap "rm -f $tmp.*; exit \$status" 0 1 2 3 15
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
+test_shrink()
+{
+	$XFS_GROWFS_PROG -D"$1" $SCRATCH_MNT >> $seqres.full 2>&1
+	ret=$?

I'm not sure what's the output of xfs_growfs when shrinking filesystem, if it's easy to do filter, but it'd be good if we could just dump the output and let .out file match & check the test result.

tests/xfs/990 line 31

Quoted 3 lines
tests/xfs/990 new file +3 -0
+
+	_scratch_unmount
+	_repair_scratch_fs >> $seqres.full

_repair_scratch_fs will fix corruption if there's any, and always return 0 if completes without problems. Is _check_scratch_fs() what you want?

tests/xfs/990 line 36

Quoted 5 lines
tests/xfs/990 new file +5 -0
+	_scratch_mount >> $seqres.full
+
+	$XFS_INFO_PROG $SCRATCH_MNT 2>&1 | _filter_mkfs 2>$tmp.growfs >/dev/null
+	. $tmp.growfs
+	[ $ret -eq 0 -a $1 -eq $dblocks ]

Just dump the expected size after shrink if possible.

tests/xfs/990 line 40

Quoted 4 lines
tests/xfs/990 new file +4 -0
+}
+
+# real QA test starts here
+_supported_fs xfs

Still missing _require_scratch

tests/xfs/990 line 46

Quoted 6 lines
tests/xfs/990 new file +6 -0
+_require_xfs_shrink
+
+rm -f $seqres.full
+echo "Format and mount"
+_scratch_mkfs -dsize="$((512 * 1024 * 1024))" -dagcount=3 2>&1 | \
+	tee -a $seqres.full | _filter_mkfs 2>$tmp.mkfs

Better to explain why mkfs with agcount=3

tests/xfs/990 line 53

Quoted 7 lines
tests/xfs/990 new file +7 -0
+. $tmp.mkfs
+t_dblocks=$dblocks
+_scratch_mount >> $seqres.full
+
+echo "Shrink fs (small size)"
+test_shrink $((t_dblocks-512*1024/dbsize)) || \
+	_fail "Shrink fs (small size) failure"

If it's possible to turn test_shrink to .out file matching way, _fail is not needed and below.

Quoted 40 lines
tests/xfs/990 new file +17 -0
+
+echo "Shrink fs (half AG)"
+test_shrink $((t_dblocks-agsize/2)) || \
+	_fail "Shrink fs (half AG) failure"
+
+echo "Shrink fs (out-of-bound)"
+test_shrink $((t_dblocks-agsize-1)) && \
+	_fail "Shrink fs (out-of-bound) failure"
+[ $dblocks -ne $((t_dblocks-agsize/2)) ] && \
+	_fail "dblocks changed after shrinking failure"
+
+$XFS_INFO_PROG $SCRATCH_MNT >> $seqres.full
+echo "*** done"
+
+# success, all done
+status=0
+exit
tests/xfs/990.out new file +12 -0
new file mode 100644
index 00000000..812f89ef
@@ -0,0 +1,12 @@
+QA output created by 990
+Format and mount
+meta-data=DDEV isize=XXX agcount=N, agsize=XXX blks
+data     = bsize=XXX blocks=XXX, imaxpct=PCT
+         = sunit=XXX swidth=XXX, unwritten=X
+naming   =VERN bsize=XXX
+log      =LDEV bsize=XXX blocks=XXX
+realtime =RDEV extsz=XXX blocks=XXX, rtextents=XXX
+Shrink fs (small size)
+Shrink fs (half AG)
+Shrink fs (out-of-bound)
+*** done
tests/xfs/group +0 -0
index e861cec9..a7981b67 100644

tests/xfs/group line 528

Quoted 4 lines
tests/xfs/group +1 -0
@@ -525,3 +525,4 @@
 525 auto quick mkfs
 526 auto quick mkfs
 527 auto quick quota
+990 auto quick growfs

Maybe it's time to add a new 'shrinkfs' group?

Thanks,
Eryu

author
#6 lore raw

Re: [RFC PATCH v3 2/3] xfs: basic functionality test for shrinking free space in the last AG

Quoted 42 lines · Eryu Guan
Quoted 41 lines · Gao Xiang

Add basic test to make sure the functionality works as expected.

---
 tests/xfs/990     | 70 +++++++++++++++++++++++++++++++++++++++++++++++
 tests/xfs/990.out | 12 ++++++++
 tests/xfs/group   |  1 +
 3 files changed, 83 insertions(+)
 create mode 100755 tests/xfs/990
 create mode 100644 tests/xfs/990.out
tests/xfs/990 new file +24 -0
new file mode 100755
index 00000000..3c79186e
@@ -0,0 +1,70 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2021 Red Hat, Inc.  All Rights Reserved.
+#
+# FS QA Test 990
+#
+# XFS shrinkfs basic functionality test
+#
+# This test attempts to shrink with a small size (512K), half AG size and
+# an out-of-bound size (agsize + 1) to observe if it works as expected.
+#
+seq=`basename $0`
+seqres=$RESULT_DIR/$seq
+echo "QA output created by $seq"
+
+here=`pwd`
+tmp=/tmp/$$
+status=1    # failure is the default!
+trap "rm -f $tmp.*; exit \$status" 0 1 2 3 15
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
Quoted 8 lines

tests/xfs/990 line 28

Quoted 4 lines
tests/xfs/990 new file +4 -0
+test_shrink()
+{
+	$XFS_GROWFS_PROG -D"$1" $SCRATCH_MNT >> $seqres.full 2>&1
+	ret=$?

I'm not sure what's the output of xfs_growfs when shrinking filesystem, if it's easy to do filter, but it'd be good if we could just dump the output and let .out file match & check the test result.

Not quite sure if it's of some use (e.g. also need to update expected output when output changed), since I think just make sure the shrinked size is as expected and xfs_repair won't argue anything on the new size, that would be enough.

Quoted 7 lines

tests/xfs/990 line 31

Quoted 4 lines
tests/xfs/990 new file +3 -0
+
+	_scratch_unmount
+	_repair_scratch_fs >> $seqres.full

_repair_scratch_fs will fix corruption if there's any, and always return 0 if completes without problems. Is _check_scratch_fs() what you want?

I didn't notice this, will update.

Quoted 8 lines

tests/xfs/990 line 36

Quoted 6 lines
tests/xfs/990 new file +5 -0
+	_scratch_mount >> $seqres.full
+
+	$XFS_INFO_PROG $SCRATCH_MNT 2>&1 | _filter_mkfs 2>$tmp.growfs >/dev/null
+	. $tmp.growfs
+	[ $ret -eq 0 -a $1 -eq $dblocks ]

Just dump the expected size after shrink if possible.

er... my idea is that I don't want to continue the test case if any fails... if dump the expected size to output, I need to fix blocksize as well, not sure if it's necessary.

Quoted 7 lines

tests/xfs/990 line 40

Quoted 5 lines
tests/xfs/990 new file +4 -0
+}
+
+# real QA test starts here
+_supported_fs xfs

Still missing _require_scratch

No, _require_xfs_shrink will include that, just as _require_xfs_scratch_rmapbt.

Quoted 9 lines

tests/xfs/990 line 46

Quoted 7 lines
tests/xfs/990 new file +6 -0
+_require_xfs_shrink
+
+rm -f $seqres.full
+echo "Format and mount"
+_scratch_mkfs -dsize="$((512 * 1024 * 1024))" -dagcount=3 2>&1 | \
+	tee -a $seqres.full | _filter_mkfs 2>$tmp.mkfs

Better to explain why mkfs with agcount=3

just because agcount = 1 is forbidden on purpose, and I need to make sure shrinking to 2 AG is not possible yet. So agcount = 3 is the minimum number.

It's just a functionality test, so agcount = 3 also makes the result determinatively.

Quoted 11 lines

tests/xfs/990 line 53

Quoted 8 lines
tests/xfs/990 new file +7 -0
+. $tmp.mkfs
+t_dblocks=$dblocks
+_scratch_mount >> $seqres.full
+
+echo "Shrink fs (small size)"
+test_shrink $((t_dblocks-512*1024/dbsize)) || \
+	_fail "Shrink fs (small size) failure"

If it's possible to turn test_shrink to .out file matching way, _fail is not needed and below.

The same as above. I could try to fix blocksize if the opinion is strong.

Quoted 39 lines
Quoted 39 lines
tests/xfs/990 new file +17 -0
+
+echo "Shrink fs (half AG)"
+test_shrink $((t_dblocks-agsize/2)) || \
+	_fail "Shrink fs (half AG) failure"
+
+echo "Shrink fs (out-of-bound)"
+test_shrink $((t_dblocks-agsize-1)) && \
+	_fail "Shrink fs (out-of-bound) failure"
+[ $dblocks -ne $((t_dblocks-agsize/2)) ] && \
+	_fail "dblocks changed after shrinking failure"
+
+$XFS_INFO_PROG $SCRATCH_MNT >> $seqres.full
+echo "*** done"
+
+# success, all done
+status=0
+exit
tests/xfs/990.out new file +12 -0
new file mode 100644
index 00000000..812f89ef
@@ -0,0 +1,12 @@
+QA output created by 990
+Format and mount
+meta-data=DDEV isize=XXX agcount=N, agsize=XXX blks
+data     = bsize=XXX blocks=XXX, imaxpct=PCT
+         = sunit=XXX swidth=XXX, unwritten=X
+naming   =VERN bsize=XXX
+log      =LDEV bsize=XXX blocks=XXX
+realtime =RDEV extsz=XXX blocks=XXX, rtextents=XXX
+Shrink fs (small size)
+Shrink fs (half AG)
+Shrink fs (out-of-bound)
+*** done
tests/xfs/group +0 -1
index e861cec9..a7981b67 100644
--- a/tests/xfs/group
Quoted 8 lines

tests/xfs/group line 528

Quoted 6 lines

+++ b/tests/xfs/group

tests/xfs/group +1 -0
@@ -525,3 +525,4 @@
 525 auto quick mkfs
 526 auto quick mkfs
 527 auto quick quota
+990 auto quick growfs

Maybe it's time to add a new 'shrinkfs' group?

not sure if it needs, since shrinkfs reuses growfs ioctl.

Thanks,
Gao Xiang

Quoted 4 lines

Thanks,
Eryu

#7 lore raw

Re: [RFC PATCH v3 2/3] xfs: basic functionality test for shrinking free space in the last AG

Quoted 81 lines · Gao Xiang
Quoted 42 lines · Eryu Guan
Quoted 41 lines · Gao Xiang

Add basic test to make sure the functionality works as expected.

---
 tests/xfs/990     | 70 +++++++++++++++++++++++++++++++++++++++++++++++
 tests/xfs/990.out | 12 ++++++++
 tests/xfs/group   |  1 +
 3 files changed, 83 insertions(+)
 create mode 100755 tests/xfs/990
 create mode 100644 tests/xfs/990.out
tests/xfs/990 new file +24 -0
new file mode 100755
index 00000000..3c79186e
@@ -0,0 +1,70 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2021 Red Hat, Inc.  All Rights Reserved.
+#
+# FS QA Test 990
+#
+# XFS shrinkfs basic functionality test
+#
+# This test attempts to shrink with a small size (512K), half AG size and
+# an out-of-bound size (agsize + 1) to observe if it works as expected.
+#
+seq=`basename $0`
+seqres=$RESULT_DIR/$seq
+echo "QA output created by $seq"
+
+here=`pwd`
+tmp=/tmp/$$
+status=1    # failure is the default!
+trap "rm -f $tmp.*; exit \$status" 0 1 2 3 15
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
Quoted 8 lines

tests/xfs/990 line 28

Quoted 4 lines
tests/xfs/990 new file +4 -0
+test_shrink()
+{
+	$XFS_GROWFS_PROG -D"$1" $SCRATCH_MNT >> $seqres.full 2>&1
+	ret=$?

I'm not sure what's the output of xfs_growfs when shrinking filesystem, if it's easy to do filter, but it'd be good if we could just dump the output and let .out file match & check the test result.

Not quite sure if it's of some use (e.g. also need to update expected output when output changed), since I think just make sure the shrinked size is as expected and xfs_repair won't argue anything on the new size, that would be enough.

Quoted 7 lines

tests/xfs/990 line 31

Quoted 4 lines
tests/xfs/990 new file +3 -0
+
+	_scratch_unmount
+	_repair_scratch_fs >> $seqres.full

_repair_scratch_fs will fix corruption if there's any, and always return 0 if completes without problems. Is _check_scratch_fs() what you want?

I didn't notice this, will update.

Quoted 8 lines

tests/xfs/990 line 36

Quoted 6 lines
tests/xfs/990 new file +5 -0
+	_scratch_mount >> $seqres.full
+
+	$XFS_INFO_PROG $SCRATCH_MNT 2>&1 | _filter_mkfs 2>$tmp.growfs >/dev/null
+	. $tmp.growfs
+	[ $ret -eq 0 -a $1 -eq $dblocks ]

Just dump the expected size after shrink if possible.

er... my idea is that I don't want to continue the test case if any fails... if dump the expected size to output, I need to fix blocksize as well, not sure if it's necessary.

Quoted 1 line
Quoted 8 lines
Quoted 6 lines

tests/xfs/990 line 40

Quoted 4 lines
tests/xfs/990 new file +4 -0
+}
+
+# real QA test starts here
+_supported_fs xfs

Still missing _require_scratch

No, _require_xfs_shrink will include that, just as _require_xfs_scratch_rmapbt.

Then include the "scratch" part in the helper name, so it's clear that it requires the SCRATCH_DEV, i.e. _require_xfs_scratch_shrink

Quoted 11 lines
Quoted 10 lines

tests/xfs/990 line 46

Quoted 8 lines
tests/xfs/990 new file +6 -0
+_require_xfs_shrink
+
+rm -f $seqres.full
+echo "Format and mount"
+_scratch_mkfs -dsize="$((512 * 1024 * 1024))" -dagcount=3 2>&1 | \
+	tee -a $seqres.full | _filter_mkfs 2>$tmp.mkfs

Better to explain why mkfs with agcount=3

Quoted 5 lines

just because agcount = 1 is forbidden on purpose, and I need to make sure shrinking to 2 AG is not possible yet. So agcount = 3 is the minimum number.

It's just a functionality test, so agcount = 3 also makes the result determinatively.

Yeah, I'd guess the reason, but it'd be great to make it clear in comments.

Quoted 63 lines
Quoted 12 lines

tests/xfs/990 line 53

Quoted 9 lines
tests/xfs/990 new file +7 -0
+. $tmp.mkfs
+t_dblocks=$dblocks
+_scratch_mount >> $seqres.full
+
+echo "Shrink fs (small size)"
+test_shrink $((t_dblocks-512*1024/dbsize)) || \
+	_fail "Shrink fs (small size) failure"

If it's possible to turn test_shrink to .out file matching way, _fail is not needed and below.

The same as above. I could try to fix blocksize if the opinion is strong.

Quoted 39 lines
Quoted 39 lines
tests/xfs/990 new file +17 -0
+
+echo "Shrink fs (half AG)"
+test_shrink $((t_dblocks-agsize/2)) || \
+	_fail "Shrink fs (half AG) failure"
+
+echo "Shrink fs (out-of-bound)"
+test_shrink $((t_dblocks-agsize-1)) && \
+	_fail "Shrink fs (out-of-bound) failure"
+[ $dblocks -ne $((t_dblocks-agsize/2)) ] && \
+	_fail "dblocks changed after shrinking failure"
+
+$XFS_INFO_PROG $SCRATCH_MNT >> $seqres.full
+echo "*** done"
+
+# success, all done
+status=0
+exit
tests/xfs/990.out new file +12 -0
new file mode 100644
index 00000000..812f89ef
@@ -0,0 +1,12 @@
+QA output created by 990
+Format and mount
+meta-data=DDEV isize=XXX agcount=N, agsize=XXX blks
+data     = bsize=XXX blocks=XXX, imaxpct=PCT
+         = sunit=XXX swidth=XXX, unwritten=X
+naming   =VERN bsize=XXX
+log      =LDEV bsize=XXX blocks=XXX
+realtime =RDEV extsz=XXX blocks=XXX, rtextents=XXX
+Shrink fs (small size)
+Shrink fs (half AG)
+Shrink fs (out-of-bound)
+*** done
tests/xfs/group +0 -1
index e861cec9..a7981b67 100644
--- a/tests/xfs/group
Quoted 8 lines

tests/xfs/group line 528

Quoted 6 lines

+++ b/tests/xfs/group

tests/xfs/group +1 -0
@@ -525,3 +525,4 @@
 525 auto quick mkfs
 526 auto quick mkfs
 527 auto quick quota
+990 auto quick growfs

Maybe it's time to add a new 'shrinkfs' group?

Quoted 1 line

not sure if it needs, since shrinkfs reuses growfs ioctl.

Ok, I have no strong preference on this, just think it's a bit easier to run only shrink tests by ./check -g shrinkfs.

Thanks,
Eryu

author
#8 lore raw

Re: [RFC PATCH v3 2/3] xfs: basic functionality test for shrinking free space in the last AG

Hi Eryu,

Quoted 89 lines · Eryu Guan
Quoted 88 lines · Gao Xiang
Quoted 42 lines · Eryu Guan
Quoted 41 lines · Gao Xiang

Add basic test to make sure the functionality works as expected.

---
 tests/xfs/990     | 70 +++++++++++++++++++++++++++++++++++++++++++++++
 tests/xfs/990.out | 12 ++++++++
 tests/xfs/group   |  1 +
 3 files changed, 83 insertions(+)
 create mode 100755 tests/xfs/990
 create mode 100644 tests/xfs/990.out
tests/xfs/990 new file +24 -0
new file mode 100755
index 00000000..3c79186e
@@ -0,0 +1,70 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2021 Red Hat, Inc.  All Rights Reserved.
+#
+# FS QA Test 990
+#
+# XFS shrinkfs basic functionality test
+#
+# This test attempts to shrink with a small size (512K), half AG size and
+# an out-of-bound size (agsize + 1) to observe if it works as expected.
+#
+seq=`basename $0`
+seqres=$RESULT_DIR/$seq
+echo "QA output created by $seq"
+
+here=`pwd`
+tmp=/tmp/$$
+status=1    # failure is the default!
+trap "rm -f $tmp.*; exit \$status" 0 1 2 3 15
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
Quoted 8 lines

tests/xfs/990 line 28

Quoted 4 lines
tests/xfs/990 new file +4 -0
+test_shrink()
+{
+	$XFS_GROWFS_PROG -D"$1" $SCRATCH_MNT >> $seqres.full 2>&1
+	ret=$?

I'm not sure what's the output of xfs_growfs when shrinking filesystem, if it's easy to do filter, but it'd be good if we could just dump the output and let .out file match & check the test result.

Not quite sure if it's of some use (e.g. also need to update expected output when output changed), since I think just make sure the shrinked size is as expected and xfs_repair won't argue anything on the new size, that would be enough.

Quoted 7 lines

tests/xfs/990 line 31

Quoted 4 lines
tests/xfs/990 new file +3 -0
+
+	_scratch_unmount
+	_repair_scratch_fs >> $seqres.full

_repair_scratch_fs will fix corruption if there's any, and always return 0 if completes without problems. Is _check_scratch_fs() what you want?

I didn't notice this, will update.

Quoted 8 lines

tests/xfs/990 line 36

Quoted 6 lines
tests/xfs/990 new file +5 -0
+	_scratch_mount >> $seqres.full
+
+	$XFS_INFO_PROG $SCRATCH_MNT 2>&1 | _filter_mkfs 2>$tmp.growfs >/dev/null
+	. $tmp.growfs
+	[ $ret -eq 0 -a $1 -eq $dblocks ]

Just dump the expected size after shrink if possible.

er... my idea is that I don't want to continue the test case if any fails... if dump the expected size to output, I need to fix blocksize as well, not sure if it's necessary.

Quoted 7 lines

tests/xfs/990 line 40

Quoted 5 lines
tests/xfs/990 new file +4 -0
+}
+
+# real QA test starts here
+_supported_fs xfs

Still missing _require_scratch

Quoted 4 lines
Quoted 1 line

No, _require_xfs_shrink will include that, just as _require_xfs_scratch_rmapbt.

Then include the "scratch" part in the helper name, so it's clear that it requires the SCRATCH_DEV, i.e. _require_xfs_scratch_shrink

ok, will rename this later. I'm working on the other stuff now.

Quoted 12 lines
Quoted 12 lines
Quoted 11 lines

tests/xfs/990 line 46

Quoted 9 lines
tests/xfs/990 new file +6 -0
+_require_xfs_shrink
+
+rm -f $seqres.full
+echo "Format and mount"
+_scratch_mkfs -dsize="$((512 * 1024 * 1024))" -dagcount=3 2>&1 | \
+	tee -a $seqres.full | _filter_mkfs 2>$tmp.mkfs

Better to explain why mkfs with agcount=3

Quoted 8 lines
Quoted 5 lines

just because agcount = 1 is forbidden on purpose, and I need to make sure shrinking to 2 AG is not possible yet. So agcount = 3 is the minimum number.

It's just a functionality test, so agcount = 3 also makes the result determinatively.

Yeah, I'd guess the reason, but it'd be great to make it clear in comments.

ok, will try to add some description.

Quoted 64 lines
Quoted 64 lines
Quoted 13 lines

tests/xfs/990 line 53

Quoted 10 lines
tests/xfs/990 new file +7 -0
+. $tmp.mkfs
+t_dblocks=$dblocks
+_scratch_mount >> $seqres.full
+
+echo "Shrink fs (small size)"
+test_shrink $((t_dblocks-512*1024/dbsize)) || \
+	_fail "Shrink fs (small size) failure"

If it's possible to turn test_shrink to .out file matching way, _fail is not needed and below.

The same as above. I could try to fix blocksize if the opinion is strong.

Quoted 47 lines
Quoted 41 lines
tests/xfs/990 new file +17 -0
+
+echo "Shrink fs (half AG)"
+test_shrink $((t_dblocks-agsize/2)) || \
+	_fail "Shrink fs (half AG) failure"
+
+echo "Shrink fs (out-of-bound)"
+test_shrink $((t_dblocks-agsize-1)) && \
+	_fail "Shrink fs (out-of-bound) failure"
+[ $dblocks -ne $((t_dblocks-agsize/2)) ] && \
+	_fail "dblocks changed after shrinking failure"
+
+$XFS_INFO_PROG $SCRATCH_MNT >> $seqres.full
+echo "*** done"
+
+# success, all done
+status=0
+exit
tests/xfs/990.out new file +12 -0
new file mode 100644
index 00000000..812f89ef
@@ -0,0 +1,12 @@
+QA output created by 990
+Format and mount
+meta-data=DDEV isize=XXX agcount=N, agsize=XXX blks
+data     = bsize=XXX blocks=XXX, imaxpct=PCT
+         = sunit=XXX swidth=XXX, unwritten=X
+naming   =VERN bsize=XXX
+log      =LDEV bsize=XXX blocks=XXX
+realtime =RDEV extsz=XXX blocks=XXX, rtextents=XXX
+Shrink fs (small size)
+Shrink fs (half AG)
+Shrink fs (out-of-bound)
+*** done
tests/xfs/group +0 -0
index e861cec9..a7981b67 100644

tests/xfs/group line 528

Quoted 4 lines
tests/xfs/group +1 -0
@@ -525,3 +525,4 @@
 525 auto quick mkfs
 526 auto quick mkfs
 527 auto quick quota
+990 auto quick growfs

Maybe it's time to add a new 'shrinkfs' group?

Quoted 4 lines
Quoted 1 line

not sure if it needs, since shrinkfs reuses growfs ioctl.

Ok, I have no strong preference on this, just think it's a bit easier to run only shrink tests by ./check -g shrinkfs.

Currently only 2 test cases are available. ok, if that is needed, I could add a new shrinkfs group.

(Also noted that xfsprogs patch haven't been merged yet. So I think the
testcase needs to be formally merged after xfsprogs side is settled.)

Thanks,
Gao Xiang

Quoted 4 lines

Thanks,
Eryu

author
#9 lore raw

RFC PATCH v3 3/3 xfs: stress test for shrinking free space in the last AG

In reply to Gao Xiang

This adds a stress testcase to shrink free space as much as possible in the last AG with background fsstress workload.

The expectation is that no crash happens with expected output.

---
 tests/xfs/991     | 122 ++++++++++++++++++++++++++++++++++++++++++++++
 tests/xfs/991.out |   8 +++
 tests/xfs/group   |   1 +
 3 files changed, 131 insertions(+)
 create mode 100755 tests/xfs/991
 create mode 100644 tests/xfs/991.out
tests/xfs/991 new file +122 -0
new file mode 100755
index 00000000..7e7d318e
@@ -0,0 +1,122 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2020-2021 Red Hat, Inc.  All Rights Reserved.
+#
+# FS QA Test 991
+#
+# XFS online shrinkfs stress test
+#
+# This test attempts to shrink unused space as much as possible with
+# background fsstress workload. It will decrease the shrink size if
+# larger size fails. And totally repeat 2 * TIME_FACTOR times.
+#
+seq=`basename $0`
+seqres=$RESULT_DIR/$seq
+echo "QA output created by $seq"
+
+here=`pwd`
+tmp=/tmp/$$
+status=1	# failure is the default!
+trap "rm -f $tmp.*; exit \$status" 0 1 2 3 15
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
+create_scratch()
+{
+	_scratch_mkfs_xfs $@ | tee -a $seqres.full | \
+		_filter_mkfs 2>$tmp.mkfs >/dev/null
+	. $tmp.mkfs
+
+	if ! _try_scratch_mount 2>/dev/null; then
+		echo "failed to mount $SCRATCH_DEV"
+		exit 1
+	fi
+
+	# fix the reserve block pool to a known size so that the enospc
+	# calculations work out correctly.
+	_scratch_resvblks 1024 > /dev/null 2>&1
+}
+
+fill_scratch()
+{
+	$XFS_IO_PROG -f -c "falloc -k 0 $1" $SCRATCH_MNT/resvfile
+}
+
+stress_scratch()
+{
+	procs=3
+	nops=$((1000 * LOAD_FACTOR))
+	# -w ensures that the only ops are ones which cause write I/O
+	FSSTRESS_ARGS=`_scale_fsstress_args -d $SCRATCH_MNT -w -p $procs \
+	    -n $nops $FSSTRESS_AVOID`
+	$FSSTRESS_PROG $FSSTRESS_ARGS >> $seqres.full 2>&1 &
+}
+
+# real QA test starts here
+_supported_fs xfs
+_require_xfs_shrink
+_require_xfs_io_command "falloc"
+
+rm -f $seqres.full
+_scratch_mkfs_xfs | tee -a $seqres.full | _filter_mkfs 2>$tmp.mkfs
+. $tmp.mkfs	# extract blocksize and data size for scratch device
+
+decsize=`expr  42 \* 1048576`	# shrink in chunks of this size at most
+endsize=`expr 125 \* 1048576`	# stop after shrinking this big
+[ `expr $endsize / $dbsize` -lt $dblocks ] || _notrun "Scratch device too small"
+
+nags=2
+totalcount=$((2 * TIME_FACTOR))
+
+while [ $totalcount -gt 0 ]; do
+	size=`expr 1010 \* 1048576`	# 1010 megabytes initially
+	logblks=$(_scratch_find_xfs_min_logblocks -dsize=${size} -dagcount=${nags})
+
+	create_scratch -lsize=${logblks}b -dsize=${size} -dagcount=${nags}
+
+	for i in `seq 125 -1 90`; do
+		fillsize=`expr $i \* 1048576`
+		out="$(fill_scratch $fillsize 2>&1)"
+		echo "$out" | grep -q 'No space left on device' && continue
+		test -n "${out}" && echo "$out"
+		break
+	done
+
+	while [ $size -gt $endsize ]; do
+		stress_scratch
+		sleep 1
+
+		decb=`expr $decsize / $dbsize`    # in data blocks
+		while [ $decb -gt 0 ]; do
+			sizeb=`expr $size / $dbsize - $decb`
+
+			$XFS_GROWFS_PROG -D ${sizeb} $SCRATCH_MNT \
+				>> $seqres.full 2>&1 && break
+
+			[ $decb -gt 100 ] && decb=`expr $decb + $RANDOM % 10`
+			decb=`expr $decb / 2`
+		done
+
+		wait
+		[ $decb -eq 0 ] && break
+
+		# get latest dblocks
+		$XFS_INFO_PROG $SCRATCH_MNT 2>&1 | _filter_mkfs 2>$tmp.growfs >/dev/null
+		. $tmp.growfs
+
+		size=`expr $dblocks \* $dbsize`
+		_scratch_unmount
+		_repair_scratch_fs >> $seqres.full
+		_scratch_mount
+	done
+
+	_scratch_unmount
+	_repair_scratch_fs >> $seqres.full
+	totalcount=`expr $totalcount - 1`
+done
+
+echo "*** done"
+status=0
+exit
tests/xfs/991.out new file +8 -0
new file mode 100644
index 00000000..e8209672
@@ -0,0 +1,8 @@
+QA output created by 991
+meta-data=DDEV isize=XXX agcount=N, agsize=XXX blks
+data     = bsize=XXX blocks=XXX, imaxpct=PCT
+         = sunit=XXX swidth=XXX, unwritten=X
+naming   =VERN bsize=XXX
+log      =LDEV bsize=XXX blocks=XXX
+realtime =RDEV extsz=XXX blocks=XXX, rtextents=XXX
+*** done
tests/xfs/group +1 -0
index a7981b67..cf190b59 100644
@@ -526,3 +526,4 @@
 526 auto quick mkfs
 527 auto quick quota
 990 auto quick growfs
+991 auto growfs ioctl prealloc stress
Signature · 1 line
2.27.0
#10 lore raw

Re: [RFC PATCH v3 3/3] xfs: stress test for shrinking free space in the last AG

Quoted 45 lines · Gao Xiang

This adds a stress testcase to shrink free space as much as possible in the last AG with background fsstress workload.

The expectation is that no crash happens with expected output.

---
 tests/xfs/991     | 122 ++++++++++++++++++++++++++++++++++++++++++++++
 tests/xfs/991.out |   8 +++
 tests/xfs/group   |   1 +
 3 files changed, 131 insertions(+)
 create mode 100755 tests/xfs/991
 create mode 100644 tests/xfs/991.out
tests/xfs/991 new file +25 -0
new file mode 100755
index 00000000..7e7d318e
@@ -0,0 +1,122 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2020-2021 Red Hat, Inc.  All Rights Reserved.
+#
+# FS QA Test 991
+#
+# XFS online shrinkfs stress test
+#
+# This test attempts to shrink unused space as much as possible with
+# background fsstress workload. It will decrease the shrink size if
+# larger size fails. And totally repeat 2 * TIME_FACTOR times.
+#
+seq=`basename $0`
+seqres=$RESULT_DIR/$seq
+echo "QA output created by $seq"
+
+here=`pwd`
+tmp=/tmp/$$
+status=1	# failure is the default!
+trap "rm -f $tmp.*; exit \$status" 0 1 2 3 15
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+

tests/xfs/991 line 35

Quoted 10 lines
tests/xfs/991 new file +10 -0
+create_scratch()
+{
+	_scratch_mkfs_xfs $@ | tee -a $seqres.full | \
+		_filter_mkfs 2>$tmp.mkfs >/dev/null
+	. $tmp.mkfs
+
+	if ! _try_scratch_mount 2>/dev/null; then
+		echo "failed to mount $SCRATCH_DEV"
+		exit 1
+	fi

_scratch_mount will do the check and _fail the test on mount failure.

Quoted 5 lines
tests/xfs/991 new file +5 -0
+
+	# fix the reserve block pool to a known size so that the enospc
+	# calculations work out correctly.
+	_scratch_resvblks 1024 > /dev/null 2>&1
+}

tests/xfs/991 line 50

Quoted 10 lines
tests/xfs/991 new file +10 -0
+
+fill_scratch()
+{
+	$XFS_IO_PROG -f -c "falloc -k 0 $1" $SCRATCH_MNT/resvfile
+}
+
+stress_scratch()
+{
+	procs=3
+	nops=$((1000 * LOAD_FACTOR))

Declare procs and nops as local.

tests/xfs/991 line 54

Quoted 4 lines
tests/xfs/991 new file +4 -0
+	# -w ensures that the only ops are ones which cause write I/O
+	FSSTRESS_ARGS=`_scale_fsstress_args -d $SCRATCH_MNT -w -p $procs \
+	    -n $nops $FSSTRESS_AVOID`
+	$FSSTRESS_PROG $FSSTRESS_ARGS >> $seqres.full 2>&1 &

I think it's more explicit to run run stress_scratch in background, instead run fsstress in background implicit.

tests/xfs/991 line 58

Quoted 4 lines
tests/xfs/991 new file +4 -0
+}
+
+# real QA test starts here
+_supported_fs xfs

_require_scratch

Quoted 20 lines
tests/xfs/991 new file +20 -0
+_require_xfs_shrink
+_require_xfs_io_command "falloc"
+
+rm -f $seqres.full
+_scratch_mkfs_xfs | tee -a $seqres.full | _filter_mkfs 2>$tmp.mkfs
+. $tmp.mkfs	# extract blocksize and data size for scratch device
+
+decsize=`expr  42 \* 1048576`	# shrink in chunks of this size at most
+endsize=`expr 125 \* 1048576`	# stop after shrinking this big
+[ `expr $endsize / $dbsize` -lt $dblocks ] || _notrun "Scratch device too small"
+
+nags=2
+totalcount=$((2 * TIME_FACTOR))
+
+while [ $totalcount -gt 0 ]; do
+	size=`expr 1010 \* 1048576`	# 1010 megabytes initially
+	logblks=$(_scratch_find_xfs_min_logblocks -dsize=${size} -dagcount=${nags})
+
+	create_scratch -lsize=${logblks}b -dsize=${size} -dagcount=${nags}
+

tests/xfs/991 line 88

Quoted 10 lines
tests/xfs/991 new file +10 -0
+	for i in `seq 125 -1 90`; do
+		fillsize=`expr $i \* 1048576`
+		out="$(fill_scratch $fillsize 2>&1)"
+		echo "$out" | grep -q 'No space left on device' && continue
+		test -n "${out}" && echo "$out"
+		break
+	done
+
+	while [ $size -gt $endsize ]; do
+		stress_scratch

So just call

stress_scratch &

here? So it's clear that we put stress_scratch in background, and the 'wait' below is waiting for it.

Thanks,
Eryu

Quoted 59 lines
tests/xfs/991 new file +34 -0
+		sleep 1
+
+		decb=`expr $decsize / $dbsize`    # in data blocks
+		while [ $decb -gt 0 ]; do
+			sizeb=`expr $size / $dbsize - $decb`
+
+			$XFS_GROWFS_PROG -D ${sizeb} $SCRATCH_MNT \
+				>> $seqres.full 2>&1 && break
+
+			[ $decb -gt 100 ] && decb=`expr $decb + $RANDOM % 10`
+			decb=`expr $decb / 2`
+		done
+
+		wait
+		[ $decb -eq 0 ] && break
+
+		# get latest dblocks
+		$XFS_INFO_PROG $SCRATCH_MNT 2>&1 | _filter_mkfs 2>$tmp.growfs >/dev/null
+		. $tmp.growfs
+
+		size=`expr $dblocks \* $dbsize`
+		_scratch_unmount
+		_repair_scratch_fs >> $seqres.full
+		_scratch_mount
+	done
+
+	_scratch_unmount
+	_repair_scratch_fs >> $seqres.full
+	totalcount=`expr $totalcount - 1`
+done
+
+echo "*** done"
+status=0
+exit
tests/xfs/991.out new file +8 -0
new file mode 100644
index 00000000..e8209672
@@ -0,0 +1,8 @@
+QA output created by 991
+meta-data=DDEV isize=XXX agcount=N, agsize=XXX blks
+data     = bsize=XXX blocks=XXX, imaxpct=PCT
+         = sunit=XXX swidth=XXX, unwritten=X
+naming   =VERN bsize=XXX
+log      =LDEV bsize=XXX blocks=XXX
+realtime =RDEV extsz=XXX blocks=XXX, rtextents=XXX
+*** done
tests/xfs/group +1 -0
index a7981b67..cf190b59 100644
@@ -526,3 +526,4 @@
 526 auto quick mkfs
 527 auto quick quota
 990 auto quick growfs
+991 auto growfs ioctl prealloc stress
Signature · 1 line
2.27.0
author
#11 lore raw

Re: [RFC PATCH v3 3/3] xfs: stress test for shrinking free space in the last AG

Quoted 50 lines · Eryu Guan
Quoted 49 lines · Gao Xiang

This adds a stress testcase to shrink free space as much as possible in the last AG with background fsstress workload.

The expectation is that no crash happens with expected output.

---
 tests/xfs/991     | 122 ++++++++++++++++++++++++++++++++++++++++++++++
 tests/xfs/991.out |   8 +++
 tests/xfs/group   |   1 +
 3 files changed, 131 insertions(+)
 create mode 100755 tests/xfs/991
 create mode 100644 tests/xfs/991.out
tests/xfs/991 new file +29 -0
new file mode 100755
index 00000000..7e7d318e
@@ -0,0 +1,122 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2020-2021 Red Hat, Inc.  All Rights Reserved.
+#
+# FS QA Test 991
+#
+# XFS online shrinkfs stress test
+#
+# This test attempts to shrink unused space as much as possible with
+# background fsstress workload. It will decrease the shrink size if
+# larger size fails. And totally repeat 2 * TIME_FACTOR times.
+#
+seq=`basename $0`
+seqres=$RESULT_DIR/$seq
+echo "QA output created by $seq"
+
+here=`pwd`
+tmp=/tmp/$$
+status=1	# failure is the default!
+trap "rm -f $tmp.*; exit \$status" 0 1 2 3 15
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
+create_scratch()
+{
+	_scratch_mkfs_xfs $@ | tee -a $seqres.full | \
+		_filter_mkfs 2>$tmp.mkfs >/dev/null
Quoted 8 lines

tests/xfs/991 line 35

Quoted 6 lines
tests/xfs/991 new file +6 -0
+	. $tmp.mkfs
+
+	if ! _try_scratch_mount 2>/dev/null; then
+		echo "failed to mount $SCRATCH_DEV"
+		exit 1
+	fi

_scratch_mount will do the check and _fail the test on mount failure.

It was just copied & pasted from somewhere else rather than wrote from scratch.

Quoted 10 lines
Quoted 10 lines
tests/xfs/991 new file +9 -0
+
+	# fix the reserve block pool to a known size so that the enospc
+	# calculations work out correctly.
+	_scratch_resvblks 1024 > /dev/null 2>&1
+}
+
+fill_scratch()
+{
+	$XFS_IO_PROG -f -c "falloc -k 0 $1" $SCRATCH_MNT/resvfile
Quoted 8 lines

tests/xfs/991 line 50

Quoted 6 lines
tests/xfs/991 new file +6 -0
+}
+
+stress_scratch()
+{
+	procs=3
+	nops=$((1000 * LOAD_FACTOR))

Declare procs and nops as local.

Will update, thanks for the suggestion.

Quoted 8 lines

tests/xfs/991 line 54

Quoted 5 lines
tests/xfs/991 new file +4 -0
+	# -w ensures that the only ops are ones which cause write I/O
+	FSSTRESS_ARGS=`_scale_fsstress_args -d $SCRATCH_MNT -w -p $procs \
+	    -n $nops $FSSTRESS_AVOID`
+	$FSSTRESS_PROG $FSSTRESS_ARGS >> $seqres.full 2>&1 &

I think it's more explicit to run run stress_scratch in background, instead run fsstress in background implicit.

Will update as well.

Quoted 7 lines

tests/xfs/991 line 58

Quoted 5 lines
tests/xfs/991 new file +4 -0
+}
+
+# real QA test starts here
+_supported_fs xfs

_require_scratch

No, _require_xfs_shrink also checks this.

Quoted 34 lines
Quoted 21 lines
tests/xfs/991 new file +20 -0
+_require_xfs_shrink
+_require_xfs_io_command "falloc"
+
+rm -f $seqres.full
+_scratch_mkfs_xfs | tee -a $seqres.full | _filter_mkfs 2>$tmp.mkfs
+. $tmp.mkfs	# extract blocksize and data size for scratch device
+
+decsize=`expr  42 \* 1048576`	# shrink in chunks of this size at most
+endsize=`expr 125 \* 1048576`	# stop after shrinking this big
+[ `expr $endsize / $dbsize` -lt $dblocks ] || _notrun "Scratch device too small"
+
+nags=2
+totalcount=$((2 * TIME_FACTOR))
+
+while [ $totalcount -gt 0 ]; do
+	size=`expr 1010 \* 1048576`	# 1010 megabytes initially
+	logblks=$(_scratch_find_xfs_min_logblocks -dsize=${size} -dagcount=${nags})
+
+	create_scratch -lsize=${logblks}b -dsize=${size} -dagcount=${nags}
+

tests/xfs/991 line 88

Quoted 10 lines
tests/xfs/991 new file +10 -0
+	for i in `seq 125 -1 90`; do
+		fillsize=`expr $i \* 1048576`
+		out="$(fill_scratch $fillsize 2>&1)"
+		echo "$out" | grep -q 'No space left on device' && continue
+		test -n "${out}" && echo "$out"
+		break
+	done
+
+	while [ $size -gt $endsize ]; do
+		stress_scratch

So just call

Quoted 4 lines

stress_scratch &

here? So it's clear that we put stress_scratch in background, and the 'wait' below is waiting for it.

Will update.

Thanks,
Gao Xiang

Quoted 64 lines

Thanks,
Eryu

Quoted 59 lines
tests/xfs/991 new file +34 -0
+		sleep 1
+
+		decb=`expr $decsize / $dbsize`    # in data blocks
+		while [ $decb -gt 0 ]; do
+			sizeb=`expr $size / $dbsize - $decb`
+
+			$XFS_GROWFS_PROG -D ${sizeb} $SCRATCH_MNT \
+				>> $seqres.full 2>&1 && break
+
+			[ $decb -gt 100 ] && decb=`expr $decb + $RANDOM % 10`
+			decb=`expr $decb / 2`
+		done
+
+		wait
+		[ $decb -eq 0 ] && break
+
+		# get latest dblocks
+		$XFS_INFO_PROG $SCRATCH_MNT 2>&1 | _filter_mkfs 2>$tmp.growfs >/dev/null
+		. $tmp.growfs
+
+		size=`expr $dblocks \* $dbsize`
+		_scratch_unmount
+		_repair_scratch_fs >> $seqres.full
+		_scratch_mount
+	done
+
+	_scratch_unmount
+	_repair_scratch_fs >> $seqres.full
+	totalcount=`expr $totalcount - 1`
+done
+
+echo "*** done"
+status=0
+exit
tests/xfs/991.out new file +8 -0
new file mode 100644
index 00000000..e8209672
@@ -0,0 +1,8 @@
+QA output created by 991
+meta-data=DDEV isize=XXX agcount=N, agsize=XXX blks
+data     = bsize=XXX blocks=XXX, imaxpct=PCT
+         = sunit=XXX swidth=XXX, unwritten=X
+naming   =VERN bsize=XXX
+log      =LDEV bsize=XXX blocks=XXX
+realtime =RDEV extsz=XXX blocks=XXX, rtextents=XXX
+*** done
tests/xfs/group +1 -0
index a7981b67..cf190b59 100644
@@ -526,3 +526,4 @@
 526 auto quick mkfs
 527 auto quick quota
 990 auto quick growfs
+991 auto growfs ioctl prealloc stress
Signature · 1 line
2.27.0