Skip to contentSkip to conversation

f2fs/006: add testcase to check out-of-space case

PATCH Superseded Chao Yu 2 participants 8 messages Oct 23, 2024 – Oct 28, 2024

v2 of 4 Superseded Latest is v4 →

lore
Share & tools
lore

Chao Yu

author
#0 lore raw

PATCH v2 1/2 f2fs/006: add testcase to check out-of-space case

This is a regression test to check whether f2fs handles dirty data correctly when checkpoint is disabled, if lfs mode is on, it will trigger OPU for all overwritten data, this will cost free segments, so f2fs must account overwritten data as OPU data when calculating free space, otherwise, it may run out of free segments in f2fs' allocation function, resulting in panic.

Cc Jaegeuk Kim <[email protected]>

---
v2:
- add _fixed_by_kernel_commit()
- use _scratch_mkfs_sized() rather than formating size-specified loop device
- code cleanup

 tests/f2fs/006     | 38 ++++++++++++++++++++++++++++++++++++++
 tests/f2fs/006.out |  6 ++++++
 2 files changed, 44 insertions(+)
 create mode 100755 tests/f2fs/006
 create mode 100644 tests/f2fs/006.out
tests/f2fs/006 new file +38 -0
new file mode 100755
index 00000000..63d00018
@@ -0,0 +1,38 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2024 Oppo.  All Rights Reserved.
+#
+# FS QA Test No. f2fs/006
+#
+# This is a regression test to check whether f2fs handles dirty
+# data correctly when checkpoint is disabled, if lfs mode is on,
+# it will trigger OPU for all overwritten data, this will cost
+# free segments, so f2fs must account overwritten data as OPU
+# data when calculating free space, otherwise, it may run out
+# of free segments in f2fs' allocation function, resulting in
+# panic.
+#
+. ./common/preamble
+_begin_fstest auto quick
+
+_fixed_by_kernel_commit xxxxxxxxxxxx \
+	"f2fs: fix to account dirty data in __get_secs_required()"
+
+testfile=$SCRATCH_MNT/testfile
+
+_require_scratch
+_scratch_mkfs_sized $((1024*1024*100)) >> $seqres.full
+
+# use mode=lfs to let f2fs always triggers OPU
+_scratch_mount -o mode=lfs,checkpoint=disable:10%,noinline_dentry >> $seqres.full
+
+dd if=/dev/zero of=$testfile bs=1M count=50 2>/dev/null
+
+# it may run out of free space of f2fs and hang kernel
+dd if=/dev/zero of=$testfile bs=1M count=50 conv=notrunc conv=fsync
+dd if=/dev/zero of=$testfile bs=1M count=50 conv=notrunc conv=fsync
+
+_scratch_remount checkpoint=enable
+
+status=0
+exit
tests/f2fs/006.out new file +6 -0
new file mode 100644
index 00000000..0d7b3910
@@ -0,0 +1,6 @@
+QA output created by 006
+50+0 records in
+50+0 records out
+dd: error writing '/mnt/scratch_f2fs/testfile': No space left on device
+3+0 records in
+2+0 records out
Signature · 1 line
2.40.1
author
#1 lore raw

PATCH v2 2/2 f2fs/007: add testcase to check consistency of compressed inode metadata

metadata of compressed inode should always be consistent after file compression, reservation, releasement and decompression, let's add a testcase to check it.

Cc Jaegeuk Kim <[email protected]>
Cc Qi Han <[email protected]>

---
v2:
- add _fixed_by_kernel_commit()
- code cleanup
- add error handling for _scratch_mkfs()

 tests/f2fs/007     | 116 +++++++++++++++++++++++++++++++++++++++++++++
 tests/f2fs/007.out |   4 ++
 2 files changed, 120 insertions(+)
 create mode 100755 tests/f2fs/007
 create mode 100644 tests/f2fs/007.out
tests/f2fs/007 new file +116 -0
new file mode 100755
index 00000000..8979c21c
@@ -0,0 +1,116 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2024 Oppo.  All Rights Reserved.
+#
+# FS QA Test No. f2fs/007
+#
+# This is a regression test to check whether compressed metadata
+# can become inconsistent after file compression, reservation
+# releasement, and decompression.
+#
+. ./common/preamble
+_begin_fstest auto quick rw compress
+
+_fixed_by_kernel_commit xxxxxxxxxxxx \
+        "f2fs: compress: fix inconsistent update of i_blocks in release_compress_blocks and reserve_compress_blocks"
+
+testfile_prefix=$SCRATCH_MNT/testfile
+fio_config=$tmp.fio
+
+cat >$fio_config <<EOF
+[write_compressed_data_30]
+name=mytest
+ioengine=psync
+rw=write
+direct=0
+bs=1M
+filesize=1M
+numjobs=1
+filename=${testfile_prefix}30
+buffer_compress_percentage=30
+
+[write_compressed_data_60]
+name=mytest
+ioengine=psync
+rw=write
+direct=0
+bs=1M
+filesize=1M
+numjobs=1
+filename=${testfile_prefix}60
+buffer_compress_percentage=60
+
+[write_compressed_data_90]
+name=mytest
+ioengine=psync
+rw=write
+direct=0
+bs=1M
+filesize=1M
+numjobs=1
+filename=${testfile_prefix}90
+buffer_compress_percentage=90
+EOF
+
+_require_scratch
+_require_fio $fio_config
+_scratch_mkfs "-f -O extra_attr,compression" >> $seqres.full || _fail "mkfs failed"
+_scratch_mount "-o compress_mode=user,compress_extension=*" >> $seqres.full
+
+echo -e "Run fio to initialize file w/ specified compress ratio" >> $seqres.full
+cat $fio_config >> $seqres.full
+$FIO_PROG $fio_config >> $seqres.full
+_scratch_unmount
+
+# force to repaire if filesystem is corrupted
+export FSCK_OPTIONS="-f $FSCK_OPTIONS"
+
+for i in 30 60 90; do
+	testfile=$testfile_prefix$i
+
+	_scratch_mount "-o compress_mode=user" >> $seqres.full
+	$F2FS_IO_PROG compress $testfile >> $seqres.full
+	cblocks=`$F2FS_IO_PROG get_cblocks $testfile`
+	echo "compression ratio is: "$cblocks" / 256"
+
+	_scratch_unmount
+
+	# 1. check after compression
+	_check_scratch_fs
+	if [ $? -ne 0 ]; then
+		_fail "filesystem becomes corrupted after compress"
+	fi
+
+	_scratch_mount >> $seqres.full
+	$F2FS_IO_PROG release_cblocks $testfile >> $seqres.full
+	_scratch_unmount
+
+	# 2. check after releasement
+	_check_scratch_fs
+	if [ $? -ne 0 ]; then
+		_fail "filesystem becomes corrupted after release_cblocks"
+	fi
+
+	_scratch_mount >> $seqres.full
+	$F2FS_IO_PROG reserve_cblocks $testfile >> $seqres.full
+	_scratch_unmount
+
+	# 3. check after rservation
+	_check_scratch_fs
+	if [ $? -ne 0 ]; then
+		_fail "filesystem becomes corrupted after reserve_cblocks"
+	fi
+
+	_scratch_mount "-o compress_mode=user" >> $seqres.full
+	$F2FS_IO_PROG decompress $testfile >> $seqres.full
+	_scratch_unmount
+
+	# 4. check after decompression
+	_check_scratch_fs
+	if [ $? -ne 0 ]; then
+		_fail "filesystem becomes corrupted after decompress"
+	fi
+done
+
+status=0
+exit
tests/f2fs/007.out new file +4 -0
new file mode 100644
index 00000000..2ea71c18
@@ -0,0 +1,4 @@
+QA output created by 007
+compression ratio is: 64 / 256
+compression ratio is: 128 / 256
+compression ratio is: 192 / 256
Signature · 1 line
2.40.1
#2 lore raw

Re: [PATCH v2 2/2] f2fs/007: add testcase to check consistency of compressed inode metadata

Quoted 69 lines · Chao Yu

metadata of compressed inode should always be consistent after file compression, reservation, releasement and decompression, let's add a testcase to check it.

Cc Jaegeuk Kim <[email protected]>
Cc Qi Han <[email protected]>

---
v2:
- add _fixed_by_kernel_commit()
- code cleanup
- add error handling for _scratch_mkfs()

 tests/f2fs/007     | 116 +++++++++++++++++++++++++++++++++++++++++++++
 tests/f2fs/007.out |   4 ++
 2 files changed, 120 insertions(+)
 create mode 100755 tests/f2fs/007
 create mode 100644 tests/f2fs/007.out
tests/f2fs/007 new file +45 -0
new file mode 100755
index 00000000..8979c21c
@@ -0,0 +1,116 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2024 Oppo.  All Rights Reserved.
+#
+# FS QA Test No. f2fs/007
+#
+# This is a regression test to check whether compressed metadata
+# can become inconsistent after file compression, reservation
+# releasement, and decompression.
+#
+. ./common/preamble
+_begin_fstest auto quick rw compress
+
+_fixed_by_kernel_commit xxxxxxxxxxxx \
+        "f2fs: compress: fix inconsistent update of i_blocks in release_compress_blocks and reserve_compress_blocks"
+
+testfile_prefix=$SCRATCH_MNT/testfile
+fio_config=$tmp.fio
+
+cat >$fio_config <<EOF
+[write_compressed_data_30]
+name=mytest
+ioengine=psync
+rw=write
+direct=0
+bs=1M
+filesize=1M
+numjobs=1
+filename=${testfile_prefix}30
+buffer_compress_percentage=30
+
+[write_compressed_data_60]
+name=mytest
+ioengine=psync
+rw=write
+direct=0
+bs=1M
+filesize=1M
+numjobs=1
+filename=${testfile_prefix}60
+buffer_compress_percentage=60
+
+[write_compressed_data_90]
+name=mytest
+ioengine=psync

tests/f2fs/007 line 55

Quoted 10 lines
tests/f2fs/007 new file +10 -0
+rw=write
+direct=0
+bs=1M
+filesize=1M
+numjobs=1
+filename=${testfile_prefix}90
+buffer_compress_percentage=90
+EOF
+
+_require_scratch

Better to move _require_scratch before "testfile_prefix=$SCRATCH_MNT/testfile". Due to we'd better to make sure there's $SCRATCH_MNT, before using it.

Quoted 26 lines
tests/f2fs/007 new file +26 -0
+_require_fio $fio_config
+_scratch_mkfs "-f -O extra_attr,compression" >> $seqres.full || _fail "mkfs failed"
+_scratch_mount "-o compress_mode=user,compress_extension=*" >> $seqres.full
+
+echo -e "Run fio to initialize file w/ specified compress ratio" >> $seqres.full
+cat $fio_config >> $seqres.full
+$FIO_PROG $fio_config >> $seqres.full
+_scratch_unmount
+
+# force to repaire if filesystem is corrupted
+export FSCK_OPTIONS="-f $FSCK_OPTIONS"
+
+for i in 30 60 90; do
+	testfile=$testfile_prefix$i
+
+	_scratch_mount "-o compress_mode=user" >> $seqres.full
+	$F2FS_IO_PROG compress $testfile >> $seqres.full
+	cblocks=`$F2FS_IO_PROG get_cblocks $testfile`
+	echo "compression ratio is: "$cblocks" / 256"
+
+	_scratch_unmount
+
+	# 1. check after compression
+	_check_scratch_fs
+	if [ $? -ne 0 ]; then
+		_fail "filesystem becomes corrupted after compress"

tests/f2fs/007 line 91

Quoted 10 lines
tests/f2fs/007 new file +10 -0
+	fi
+
+	_scratch_mount >> $seqres.full
+	$F2FS_IO_PROG release_cblocks $testfile >> $seqres.full
+	_scratch_unmount
+
+	# 2. check after releasement
+	_check_scratch_fs
+	if [ $? -ne 0 ]; then
+		_fail "filesystem becomes corrupted after release_cblocks"

The _fail is not necessary, due to _check_scratch_fs will do "exit 1" if it find a corruption. If you need some message, maybe echo it before calling _check_scratch_fs?

tests/f2fs/007 line 102

Quoted 11 lines
tests/f2fs/007 new file +11 -0
+	fi
+
+	_scratch_mount >> $seqres.full
+	$F2FS_IO_PROG reserve_cblocks $testfile >> $seqres.full
+	_scratch_unmount
+
+	# 3. check after rservation
+	_check_scratch_fs
+	if [ $? -ne 0 ]; then
+		_fail "filesystem becomes corrupted after reserve_cblocks"
+	fi

Same as above

tests/f2fs/007 line 112

Quoted 10 lines
tests/f2fs/007 new file +10 -0
+
+	_scratch_mount "-o compress_mode=user" >> $seqres.full
+	$F2FS_IO_PROG decompress $testfile >> $seqres.full
+	_scratch_unmount
+
+	# 4. check after decompression
+	_check_scratch_fs
+	if [ $? -ne 0 ]; then
+		_fail "filesystem becomes corrupted after decompress"
+	fi

Same as above.

Thanks,
Zorro

Quoted 17 lines
tests/f2fs/007 new file +4 -0
+done
+
+status=0
+exit
tests/f2fs/007.out new file +4 -0
new file mode 100644
index 00000000..2ea71c18
@@ -0,0 +1,4 @@
+QA output created by 007
+compression ratio is: 64 / 256
+compression ratio is: 128 / 256
+compression ratio is: 192 / 256
Signature · 1 line
2.40.1
author
#3 lore raw

Re: [PATCH v2 2/2] f2fs/007: add testcase to check consistency of compressed inode metadata

Quoted 139 lines · Zorro Lang
Quoted 69 lines · Chao Yu

metadata of compressed inode should always be consistent after file compression, reservation, releasement and decompression, let's add a testcase to check it.

Cc Jaegeuk Kim <[email protected]>
Cc Qi Han <[email protected]>

---
v2:
- add _fixed_by_kernel_commit()
- code cleanup
- add error handling for _scratch_mkfs()

  tests/f2fs/007     | 116 +++++++++++++++++++++++++++++++++++++++++++++
  tests/f2fs/007.out |   4 ++
  2 files changed, 120 insertions(+)

create mode 100755 tests/f2fs/007
create mode 100644 tests/f2fs/007.out

tests/f2fs/007 new file +45 -0
new file mode 100755
index 00000000..8979c21c
@@ -0,0 +1,116 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2024 Oppo.  All Rights Reserved.
+#
+# FS QA Test No. f2fs/007
+#
+# This is a regression test to check whether compressed metadata
+# can become inconsistent after file compression, reservation
+# releasement, and decompression.
+#
+. ./common/preamble
+_begin_fstest auto quick rw compress
+
+_fixed_by_kernel_commit xxxxxxxxxxxx \
+        "f2fs: compress: fix inconsistent update of i_blocks in release_compress_blocks and reserve_compress_blocks"
+
+testfile_prefix=$SCRATCH_MNT/testfile
+fio_config=$tmp.fio
+
+cat >$fio_config <<EOF
+[write_compressed_data_30]
+name=mytest
+ioengine=psync
+rw=write
+direct=0
+bs=1M
+filesize=1M
+numjobs=1
+filename=${testfile_prefix}30
+buffer_compress_percentage=30
+
+[write_compressed_data_60]
+name=mytest
+ioengine=psync
+rw=write
+direct=0
+bs=1M
+filesize=1M
+numjobs=1
+filename=${testfile_prefix}60
+buffer_compress_percentage=60
+
+[write_compressed_data_90]
+name=mytest
+ioengine=psync

tests/f2fs/007 line 55

Quoted 10 lines
tests/f2fs/007 new file +10 -0
+rw=write
+direct=0
+bs=1M
+filesize=1M
+numjobs=1
+filename=${testfile_prefix}90
+buffer_compress_percentage=90
+EOF
+
+_require_scratch

Better to move _require_scratch before "testfile_prefix=$SCRATCH_MNT/testfile". Due to we'd better to make sure there's $SCRATCH_MNT, before using it.

Quoted 26 lines
tests/f2fs/007 new file +26 -0
+_require_fio $fio_config
+_scratch_mkfs "-f -O extra_attr,compression" >> $seqres.full || _fail "mkfs failed"
+_scratch_mount "-o compress_mode=user,compress_extension=*" >> $seqres.full
+
+echo -e "Run fio to initialize file w/ specified compress ratio" >> $seqres.full
+cat $fio_config >> $seqres.full
+$FIO_PROG $fio_config >> $seqres.full
+_scratch_unmount
+
+# force to repaire if filesystem is corrupted
+export FSCK_OPTIONS="-f $FSCK_OPTIONS"
+
+for i in 30 60 90; do
+	testfile=$testfile_prefix$i
+
+	_scratch_mount "-o compress_mode=user" >> $seqres.full
+	$F2FS_IO_PROG compress $testfile >> $seqres.full
+	cblocks=`$F2FS_IO_PROG get_cblocks $testfile`
+	echo "compression ratio is: "$cblocks" / 256"
+
+	_scratch_unmount
+
+	# 1. check after compression
+	_check_scratch_fs
+	if [ $? -ne 0 ]; then
+		_fail "filesystem becomes corrupted after compress"

tests/f2fs/007 line 91

Quoted 10 lines
tests/f2fs/007 new file +10 -0
+	fi
+
+	_scratch_mount >> $seqres.full
+	$F2FS_IO_PROG release_cblocks $testfile >> $seqres.full
+	_scratch_unmount
+
+	# 2. check after releasement
+	_check_scratch_fs
+	if [ $? -ne 0 ]; then
+		_fail "filesystem becomes corrupted after release_cblocks"

The _fail is not necessary, due to _check_scratch_fs will do "exit 1" if it find a corruption. If you need some message, maybe echo it before calling _check_scratch_fs?

tests/f2fs/007 line 102

Quoted 11 lines
tests/f2fs/007 new file +11 -0
+	fi
+
+	_scratch_mount >> $seqres.full
+	$F2FS_IO_PROG reserve_cblocks $testfile >> $seqres.full
+	_scratch_unmount
+
+	# 3. check after rservation
+	_check_scratch_fs
+	if [ $? -ne 0 ]; then
+		_fail "filesystem becomes corrupted after reserve_cblocks"
+	fi

Same as above

Quoted 12 lines

tests/f2fs/007 line 112

Quoted 10 lines
tests/f2fs/007 new file +10 -0
+
+	_scratch_mount "-o compress_mode=user" >> $seqres.full
+	$F2FS_IO_PROG decompress $testfile >> $seqres.full
+	_scratch_unmount
+
+	# 4. check after decompression
+	_check_scratch_fs
+	if [ $? -ne 0 ]; then
+		_fail "filesystem becomes corrupted after decompress"
+	fi

Same as above.

Okay, let me update the patch.

Thanks,

Quoted 22 lines

Thanks,
Zorro

Quoted 17 lines
tests/f2fs/007 new file +4 -0
+done
+
+status=0
+exit
tests/f2fs/007.out new file +4 -0
new file mode 100644
index 00000000..2ea71c18
@@ -0,0 +1,4 @@
+QA output created by 007
+compression ratio is: 64 / 256
+compression ratio is: 128 / 256
+compression ratio is: 192 / 256
Signature · 1 line
2.40.1
#4 lore raw
In reply to Chao Yu
Quoted 51 lines · Chao Yu

This is a regression test to check whether f2fs handles dirty data correctly when checkpoint is disabled, if lfs mode is on, it will trigger OPU for all overwritten data, this will cost free segments, so f2fs must account overwritten data as OPU data when calculating free space, otherwise, it may run out of free segments in f2fs' allocation function, resulting in panic.

Cc Jaegeuk Kim <[email protected]>

---
v2:
- add _fixed_by_kernel_commit()
- use _scratch_mkfs_sized() rather than formating size-specified loop device
- code cleanup

 tests/f2fs/006     | 38 ++++++++++++++++++++++++++++++++++++++
 tests/f2fs/006.out |  6 ++++++
 2 files changed, 44 insertions(+)
 create mode 100755 tests/f2fs/006
 create mode 100644 tests/f2fs/006.out
tests/f2fs/006 new file +23 -0
new file mode 100755
index 00000000..63d00018
@@ -0,0 +1,38 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2024 Oppo.  All Rights Reserved.
+#
+# FS QA Test No. f2fs/006
+#
+# This is a regression test to check whether f2fs handles dirty
+# data correctly when checkpoint is disabled, if lfs mode is on,
+# it will trigger OPU for all overwritten data, this will cost
+# free segments, so f2fs must account overwritten data as OPU
+# data when calculating free space, otherwise, it may run out
+# of free segments in f2fs' allocation function, resulting in
+# panic.
+#
+. ./common/preamble
+_begin_fstest auto quick
+
+_fixed_by_kernel_commit xxxxxxxxxxxx \
+	"f2fs: fix to account dirty data in __get_secs_required()"
+
+testfile=$SCRATCH_MNT/testfile
+
+_require_scratch

tests/f2fs/006 line 33

Quoted 10 lines
tests/f2fs/006 new file +10 -0
+_scratch_mkfs_sized $((1024*1024*100)) >> $seqres.full
+
+# use mode=lfs to let f2fs always triggers OPU
+_scratch_mount -o mode=lfs,checkpoint=disable:10%,noinline_dentry >> $seqres.full
+
+dd if=/dev/zero of=$testfile bs=1M count=50 2>/dev/null
+
+# it may run out of free space of f2fs and hang kernel
+dd if=/dev/zero of=$testfile bs=1M count=50 conv=notrunc conv=fsync
+dd if=/dev/zero of=$testfile bs=1M count=50 conv=notrunc conv=fsync

What kind of failure should be printed at here if test on unfixed kernel? I got:

# diff -u /root/git/xfstests/tests/f2fs/006.out /root/git/xfstests/results//default/f2fs/006.out.bad|less
--- /root/git/xfstests/tests/f2fs/006.out       2024-10-25 11:33:54.693883281 +0800
+++ /root/git/xfstests/results//default/f2fs/006.out.bad        2024-10-25 11:34:55.907252401 +0800
@@ -1,6 +1,6 @@
 QA output created by 006
 50+0 records in
 50+0 records out
-dd: error writing '/mnt/scratch_f2fs/testfile': No space left on device
-3+0 records in
-2+0 records out
+dd: fsync failed for '/mnt/scratch/testfile': Input/output error
+50+0 records in
+50+0 records out

Does that mean the bug is reproduced?

Quoted 11 lines
tests/f2fs/006 new file +5 -0
+
+_scratch_remount checkpoint=enable
+
+status=0
+exit
tests/f2fs/006.out new file +0 -0
new file mode 100644
index 00000000..0d7b3910

tests/f2fs/006.out line 4

Quoted 4 lines
tests/f2fs/006.out new file +4 -0
@@ -0,0 +1,6 @@
+QA output created by 006
+50+0 records in
+50+0 records out
+dd: error writing '/mnt/scratch_f2fs/testfile': No space left on device

The "/mnt/scratch_f2fs" should be SCRATCH_MNT, please use _filter_scratch() by importing common/filter.

Quoted 5 lines
tests/f2fs/006.out new file +2 -0
+3+0 records in
+2+0 records out
Signature · 1 line
2.40.1
author
#5 lore raw
Quoted 56 lines · Zorro Lang
Quoted 55 lines · Chao Yu

This is a regression test to check whether f2fs handles dirty data correctly when checkpoint is disabled, if lfs mode is on, it will trigger OPU for all overwritten data, this will cost free segments, so f2fs must account overwritten data as OPU data when calculating free space, otherwise, it may run out of free segments in f2fs' allocation function, resulting in panic.

Cc Jaegeuk Kim <[email protected]>

---
v2:
- add _fixed_by_kernel_commit()
- use _scratch_mkfs_sized() rather than formating size-specified loop device
- code cleanup

  tests/f2fs/006     | 38 ++++++++++++++++++++++++++++++++++++++
  tests/f2fs/006.out |  6 ++++++
  2 files changed, 44 insertions(+)

create mode 100755 tests/f2fs/006
create mode 100644 tests/f2fs/006.out

tests/f2fs/006 new file +27 -0
new file mode 100755
index 00000000..63d00018
@@ -0,0 +1,38 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2024 Oppo.  All Rights Reserved.
+#
+# FS QA Test No. f2fs/006
+#
+# This is a regression test to check whether f2fs handles dirty
+# data correctly when checkpoint is disabled, if lfs mode is on,
+# it will trigger OPU for all overwritten data, this will cost
+# free segments, so f2fs must account overwritten data as OPU
+# data when calculating free space, otherwise, it may run out
+# of free segments in f2fs' allocation function, resulting in
+# panic.
+#
+. ./common/preamble
+_begin_fstest auto quick
+
+_fixed_by_kernel_commit xxxxxxxxxxxx \
+	"f2fs: fix to account dirty data in __get_secs_required()"
+
+testfile=$SCRATCH_MNT/testfile
+
+_require_scratch
+_scratch_mkfs_sized $((1024*1024*100)) >> $seqres.full
+
+# use mode=lfs to let f2fs always triggers OPU
+_scratch_mount -o mode=lfs,checkpoint=disable:10%,noinline_dentry >> $seqres.full
Quoted 8 lines

tests/f2fs/006 line 33

Quoted 6 lines
tests/f2fs/006 new file +6 -0
+
+dd if=/dev/zero of=$testfile bs=1M count=50 2>/dev/null
+
+# it may run out of free space of f2fs and hang kernel
+dd if=/dev/zero of=$testfile bs=1M count=50 conv=notrunc conv=fsync
+dd if=/dev/zero of=$testfile bs=1M count=50 conv=notrunc conv=fsync

What kind of failure should be printed at here if test on unfixed kernel?

It will panic kernel w/o fix, can you please check dmesg?

Quoted 27 lines

I got:

# diff -u /root/git/xfstests/tests/f2fs/006.out /root/git/xfstests/results//default/f2fs/006.out.bad|less
root/git/xfstests/results//default/f2fs/006.out.bad 2024-10-25 11:34:55.907252401 +0800 renamed from root/git/xfstests/tests/f2fs/006.out 2024-10-25 11:33:54.693883281 +0800 +3 -3
@@ -1,6 +1,6 @@
  QA output created by 006
  50+0 records in
  50+0 records out
-dd: error writing '/mnt/scratch_f2fs/testfile': No space left on device
-3+0 records in
-2+0 records out
+dd: fsync failed for '/mnt/scratch/testfile': Input/output error
+50+0 records in
+50+0 records out

Does that mean the bug is reproduced?

Quoted 10 lines
Quoted 1 line

+

tests/f2fs/006 new file +4 -0
+_scratch_remount checkpoint=enable
+
+status=0
+exit
tests/f2fs/006.out new file +0 -0
new file mode 100644
index 00000000..0d7b3910
Quoted 8 lines

tests/f2fs/006.out line 4

Quoted 5 lines
tests/f2fs/006.out new file +4 -0
@@ -0,0 +1,6 @@
+QA output created by 006
+50+0 records in
+50+0 records out
+dd: error writing '/mnt/scratch_f2fs/testfile': No space left on device

The "/mnt/scratch_f2fs" should be SCRATCH_MNT, please use _filter_scratch() by importing common/filter.

Correct, let me fix this.

Thanks,

Quoted 7 lines
Quoted 6 lines
tests/f2fs/006.out new file +2 -0
+3+0 records in
+2+0 records out
Signature · 1 line
2.40.1
#6 lore raw
Quoted 66 lines · Chao Yu
Quoted 56 lines · Zorro Lang
Quoted 55 lines · Chao Yu

This is a regression test to check whether f2fs handles dirty data correctly when checkpoint is disabled, if lfs mode is on, it will trigger OPU for all overwritten data, this will cost free segments, so f2fs must account overwritten data as OPU data when calculating free space, otherwise, it may run out of free segments in f2fs' allocation function, resulting in panic.

Cc Jaegeuk Kim <[email protected]>

---
v2:
- add _fixed_by_kernel_commit()
- use _scratch_mkfs_sized() rather than formating size-specified loop device
- code cleanup

  tests/f2fs/006     | 38 ++++++++++++++++++++++++++++++++++++++
  tests/f2fs/006.out |  6 ++++++
  2 files changed, 44 insertions(+)

create mode 100755 tests/f2fs/006
create mode 100644 tests/f2fs/006.out

tests/f2fs/006 new file +27 -0
new file mode 100755
index 00000000..63d00018
@@ -0,0 +1,38 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2024 Oppo.  All Rights Reserved.
+#
+# FS QA Test No. f2fs/006
+#
+# This is a regression test to check whether f2fs handles dirty
+# data correctly when checkpoint is disabled, if lfs mode is on,
+# it will trigger OPU for all overwritten data, this will cost
+# free segments, so f2fs must account overwritten data as OPU
+# data when calculating free space, otherwise, it may run out
+# of free segments in f2fs' allocation function, resulting in
+# panic.
+#
+. ./common/preamble
+_begin_fstest auto quick
+
+_fixed_by_kernel_commit xxxxxxxxxxxx \
+	"f2fs: fix to account dirty data in __get_secs_required()"
+
+testfile=$SCRATCH_MNT/testfile
+
+_require_scratch
+_scratch_mkfs_sized $((1024*1024*100)) >> $seqres.full
+
+# use mode=lfs to let f2fs always triggers OPU
+_scratch_mount -o mode=lfs,checkpoint=disable:10%,noinline_dentry >> $seqres.full
Quoted 8 lines

tests/f2fs/006 line 33

Quoted 6 lines
tests/f2fs/006 new file +6 -0
+
+dd if=/dev/zero of=$testfile bs=1M count=50 2>/dev/null
+
+# it may run out of free space of f2fs and hang kernel
+dd if=/dev/zero of=$testfile bs=1M count=50 conv=notrunc conv=fsync
+dd if=/dev/zero of=$testfile bs=1M count=50 conv=notrunc conv=fsync

What kind of failure should be printed at here if test on unfixed kernel?

Quoted 1 line

It will panic kernel w/o fix, can you please check dmesg?

The dmesg as below [2]. But it hit EIO [1], rather than ENOSPC, and didn't hang. If it's the same issue, or another issue, I think we can explain it in above comment.

Thanks,
Zorro

[1]

# diff -u /root/git/xfstests/tests/f2fs/006.out /root/git/xfstests/results//default/f2fs/006.out.bad
--- /root/git/xfstests/tests/f2fs/006.out       2024-10-28 20:51:08.381020424 +0800
+++ /root/git/xfstests/results//default/f2fs/006.out.bad        2024-10-28 20:54:33.252246497 +0800
@@ -1,6 +1,6 @@
 QA output created by 006
 50+0 records in
 50+0 records out
-dd: error writing '/mnt/scratch_f2fs/testfile': No space left on device
-3+0 records in
-2+0 records out
+dd: fsync failed for '/mnt/scratch/testfile': Input/output error
+50+0 records in
+50+0 records out

[2]

[3370744.465936] run fstests f2fs/006 at 2024-10-28 20:54:27
[3370746.308401] F2FS-fs (sda6): Adjust unusable cap for checkpoint=disable = 1530 / 10%
[3370746.318664] F2FS-fs (sda6): Found nat_bits in checkpoint
[3370746.341354] F2FS-fs (sda6): Start checkpoint disabled!
[3370746.347782] F2FS-fs (sda6): Mounted with checkpoint version = 355eea66
[3370747.846817] F2FS-fs (sda6): Stopped filesystem due to reason: 7
[3370747.853002] ------------[ cut here ]------------
[3370747.857826] WARNING: CPU: 1 PID: 791405 at fs/f2fs/segment.c:2748 new_curseg+0xc7e/0x1ef0 [f2fs]
[3370747.866938] Modules linked in: f2fs crc32_generic lz4hc_compress lz4_compress dm_thin_pool dm_persistent_data dm_bio_prison dm_log_writes xfs mlx5_ib ib_uverbs macsec ib_core nft_fib_inet nft_fib_ipv4 nft_fib_ipv6 nft_fib intel_rapl_msr intel_rapl_common nft_reject_inet nf_reject_ipv4 nf_reject_ipv6 intel_uncore_frequency nft_reject intel_uncore_frequency_common i10nm_edac skx_edac_common nft_ct nfit x86_pkg_temp_thermal intel_powerclamp nft_chain_nat nf_nat coretemp rfkill nf_conntrack nf_defrag_ipv6 kvm_intel nf_defrag_ipv4 snd_pcsp mlx5_core dax_hmem snd_pcm kvm ip_set cxl_acpi snd_timer spi_nor dell_pc iTCO_wdt rapl cxl_core intel_pmc_bxt snd dell_smbios acpi_power_meter iTCO_vendor_support mtd mlxfw ipmi_ssif intel_cstate platform_profile nf_tables dcdbas nd_pmem isst_if_mmio isst_if_mbox_pci psample dax_pmem nd_btt intel_uncore dell_wmi_descriptor wmi_bmof einj soundcore tls mei_me tg3 intel_th_gth spi_intel_pci i2c_i801 isst_if_common pci_hyperv_intf mei spi_intel intel_th_pci i2c_smbus intel_pch_thermal
[3370747.867283]  ipmi_si intel_th intel_vsec acpi_ipmi ipmi_devintf ipmi_msghandler fuse loop nfnetlink zram crct10dif_pclmul crc32_pclmul crc32c_intel polyval_clmulni polyval_generic ghash_clmulni_intel sha512_ssse3 nd_e820 sha256_ssse3 libnvdimm sha1_ssse3 megaraid_sas mgag200 i2c_algo_bit wmi [last unloaded: scsi_debug]
[3370747.985397] CPU: 1 UID: 0 PID: 791405 Comm: dd Tainted: G        W         -------  ---  6.11.0-0.rc6.49.fc42.x86_64+debug #1
[3370747.996895] Tainted: [W]=WARN
[3370748.000067] Hardware name: Dell Inc. PowerEdge R750/0PJ80M, BIOS 1.5.4 12/17/2021
[3370748.007747] RIP: 0010:new_curseg+0xc7e/0x1ef0 [f2fs]
[3370748.012949] Code: 8e 38 0b 00 00 3b ab d8 11 00 00 0f 82 b3 f8 ff ff 48 8b 7c 24 28 e8 f1 26 5b f0 ba 07 00 00 00 31 f6 48 89 df e8 42 6b f7 ff <0f> 0b be 08 00 00 00 48 8d bb 20 01 00 00 e8 3f db d1 ed f0 80 8b
[3370748.031898] RSP: 0018:ffa00000365cea90 EFLAGS: 00010296
[3370748.037323] RAX: 0000000000000000 RBX: ff1100110a61c000 RCX: 0000000000000000
[3370748.044656] RDX: 0000000000000033 RSI: ffffffffb2c56fe0 RDI: fff3fc0006cb9d23
[3370748.051989] RBP: 000000000000002a R08: 0000000000000001 R09: fff3fc0006cb9ce8
[3370748.059319] R10: ffa00000365ce747 R11: 0000000000000000 R12: 0000000000000001
[3370748.066652] R13: ff110009482c4178 R14: ff11001096338e00 R15: ffe21c02214c3a3b
[3370748.073984] FS:  00007f9d7a3d4740(0000) GS:ff11002031000000(0000) knlGS:0000000000000000
[3370748.082269] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[3370748.088211] CR2: 000056272e350c08 CR3: 0000001100088002 CR4: 0000000000771ef0
[3370748.095546] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[3370748.102906] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[3370748.110236] PKRU: 55555554
[3370748.113149] Call Trace:
[3370748.115800]  <TASK>
[3370748.118106]  ? __warn.cold+0x5b/0x1af
[3370748.121975]  ? new_curseg+0xc7e/0x1ef0 [f2fs]
[3370748.126575]  ? report_bug+0x1fc/0x3d0
[3370748.130449]  ? handle_bug+0x3c/0x80
[3370748.134141]  ? exc_invalid_op+0x17/0x40
[3370748.138181]  ? asm_exc_invalid_op+0x1a/0x20
[3370748.142591]  ? new_curseg+0xc7e/0x1ef0 [f2fs]
[3370748.147200]  ? update_segment_mtime+0x144/0x4c0 [f2fs]
[3370748.152591]  f2fs_allocate_data_block+0xbc9/0x4070 [f2fs]
[3370748.158244]  ? mark_lock+0xf5/0x16d0
[3370748.162021]  ? __pfx___lock_acquire+0x10/0x10
[3370748.166603]  ? __pfx_f2fs_allocate_data_block+0x10/0x10 [f2fs]
[3370748.172676]  ? __pfx___get_segment_type+0x10/0x10 [f2fs]
[3370748.178231]  ? folio_memcg_unlock+0x61/0x120
[3370748.182707]  ? local_clock_noinstr+0xd/0x100
[3370748.187191]  do_write_page+0x156/0xdf0 [f2fs]
[3370748.191789]  ? __pfx___update_extent_cache+0x10/0x10 [f2fs]
[3370748.197596]  ? mark_held_locks+0x94/0xe0
[3370748.201727]  ? __pfx_do_write_page+0x10/0x10 [f2fs]
[3370748.206865]  ? __folio_start_writeback+0x266/0x870
[3370748.211873]  f2fs_outplace_write_data+0x198/0x310 [f2fs]
[3370748.217419]  ? __pfx_f2fs_outplace_write_data+0x10/0x10 [f2fs]
[3370748.223503]  f2fs_do_write_data_page+0xaf9/0x1120 [f2fs]
[3370748.229175]  ? __pfx_f2fs_do_write_data_page+0x10/0x10 [f2fs]
[3370748.235176]  ? __pfx___lock_acquire+0x10/0x10
[3370748.239748]  f2fs_write_single_data_page+0xf9f/0x16e0 [f2fs]
[3370748.245674]  ? __pfx_f2fs_write_single_data_page+0x10/0x10 [f2fs]
[3370748.252007]  ? local_clock_noinstr+0xd/0x100
[3370748.256482]  ? folio_clear_dirty_for_io+0x20b/0x5b0
[3370748.261567]  ? local_clock_noinstr+0xd/0x100
[3370748.266081]  f2fs_write_cache_pages+0xa46/0x1ec0 [f2fs]
[3370748.271573]  ? __pfx_f2fs_write_cache_pages+0x10/0x10 [f2fs]
[3370748.277486]  ? __pfx___lock_acquire+0x10/0x10
[3370748.282051]  ? mark_lock+0xf5/0x16d0
[3370748.285869]  ? f2fs_write_data_pages+0x844/0xc00 [f2fs]
[3370748.291352]  ? rcu_is_watching+0x12/0xc0
[3370748.295482]  ? trace_contention_end+0xd4/0x110
[3370748.300190]  f2fs_write_data_pages+0x85d/0xc00 [f2fs]
[3370748.305491]  ? __pfx_f2fs_write_data_pages+0x10/0x10 [f2fs]
[3370748.311293]  ? __pfx___lock_acquire+0x10/0x10
[3370748.315889]  do_writepages+0x176/0x780
[3370748.319860]  ? __pfx_do_writepages+0x10/0x10
[3370748.324329]  ? filemap_fdatawrite_wbc+0xd3/0x180
[3370748.329162]  ? do_raw_spin_unlock+0x58/0x1f0
[3370748.333641]  ? _raw_spin_unlock+0x2d/0x50
[3370748.337859]  ? wbc_attach_and_unlock_inode+0x3da/0x7d0
[3370748.343204]  filemap_fdatawrite_wbc+0x113/0x180
[3370748.347939]  __filemap_fdatawrite_range+0xaf/0xf0
[3370748.352855]  ? __pfx___filemap_fdatawrite_range+0x10/0x10
[3370748.358515]  ? __pfx_lock_release+0x10/0x10
[3370748.362909]  file_write_and_wait_range+0x9b/0x110
[3370748.367819]  f2fs_do_sync_file+0x27b/0x1ab0 [f2fs]
[3370748.372882]  ? __pfx_f2fs_do_sync_file+0x10/0x10 [f2fs]
[3370748.378348]  ? __pfx_vfs_read+0x10/0x10
[3370748.382452]  ? __mark_inode_dirty+0x6ef/0x9d0
[3370748.387018]  ? ksys_read+0xfb/0x1d0
[3370748.390715]  ? vfs_fsync_range+0x11b/0x220
[3370748.395024]  __x64_sys_fsync+0x59/0xa0
[3370748.398980]  do_syscall_64+0x97/0x190
[3370748.402861]  ? __pfx_ksys_read+0x10/0x10
[3370748.406995]  ? lockdep_hardirqs_on_prepare+0x171/0x400
[3370748.412336]  ? do_syscall_64+0xa3/0x190
[3370748.416380]  ? lockdep_hardirqs_on+0x7c/0x100
[3370748.420941]  ? do_syscall_64+0xa3/0x190
[3370748.424981]  ? lockdep_hardirqs_on_prepare+0x171/0x400
[3370748.430324]  ? do_syscall_64+0xa3/0x190
[3370748.434364]  ? lockdep_hardirqs_on+0x7c/0x100
[3370748.438924]  ? do_syscall_64+0xa3/0x190
[3370748.442969]  ? do_syscall_64+0xa3/0x190
[3370748.447009]  ? lockdep_hardirqs_on_prepare+0x171/0x400
[3370748.452349]  ? do_syscall_64+0xa3/0x190
[3370748.456394]  ? lockdep_hardirqs_on+0x7c/0x100
[3370748.460953]  ? do_syscall_64+0xa3/0x190
[3370748.464994]  ? do_syscall_64+0xa3/0x190
[3370748.469039]  ? lockdep_hardirqs_on_prepare+0x171/0x400
[3370748.474380]  ? do_syscall_64+0xa3/0x190
[3370748.478427]  ? lockdep_hardirqs_on+0x7c/0x100
[3370748.483043]  ? clear_bhb_loop+0x25/0x80
[3370748.487082]  ? clear_bhb_loop+0x25/0x80
[3370748.491127]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
[3370748.496383] RIP: 0033:0x7f9d7a4c11a4
[3370748.500187] Code: 00 00 0f 1f 40 00 f3 0f 1e fa 48 8b 3d 9d 5c 10 00 e9 10 73 f1 ff f3 0f 1e fa 80 3d c5 5e 10 00 00 74 13 b8 4a 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 3c c3 0f 1f 00 55 4
8 89 e5 48 83 ec 10 89 7d
[3370748.519133] RSP: 002b:00007fff6c1e62c8 EFLAGS: 00000202 ORIG_RAX: 000000000000004a
[3370748.526905] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f9d7a4c11a4
[3370748.534239] RDX: 0000000000000032 RSI: 0000000000000000 RDI: 0000000000000001
[3370748.541568] RBP: 00007fff6c1e6320 R08: 0000000000000000 R09: 00007f9d7a60c380
[3370748.548932] R10: 0000000000000022 R11: 0000000000000202 R12: 00007f9d7a3d46c8
[3370748.556268] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000032
[3370748.563625]  </TASK>
[3370748.566016] irq event stamp: 342211
[3370748.569711] hardirqs last  enabled at (342221): [<ffffffffaf41b9f6>] console_unlock+0x236/0x2c0
[3370748.578611] hardirqs last disabled at (342230): [<ffffffffaf41b9db>] console_unlock+0x21b/0x2c0
[3370748.587504] softirqs last  enabled at (342178): [<ffffffffaf247c3b>] __irq_exit_rcu+0xbb/0x1c0
[3370748.596307] softirqs last disabled at (342157): [<ffffffffaf247c3b>] __irq_exit_rcu+0xbb/0x1c0
[3370748.605115] ---[ end trace 0000000000000000 ]---
[3370748.792483] F2FS-fs (sda6): Adjust unusable cap for checkpoint=disable = 1530 / 10%
[3370750.194429] F2FS-fs (sda6): Disable nat_bits due to incorrect cp_ver (4137321793606052454, 18446744073709551615)
[3370750.259391] F2FS-fs (sda6): Mounted with checkpoint version = 355eea67
Quoted 49 lines
Quoted 28 lines

I got:

# diff -u /root/git/xfstests/tests/f2fs/006.out /root/git/xfstests/results//default/f2fs/006.out.bad|less
root/git/xfstests/results//default/f2fs/006.out.bad 2024-10-25 11:34:55.907252401 +0800 renamed from root/git/xfstests/tests/f2fs/006.out 2024-10-25 11:33:54.693883281 +0800 +3 -3
@@ -1,6 +1,6 @@
  QA output created by 006
  50+0 records in
  50+0 records out
-dd: error writing '/mnt/scratch_f2fs/testfile': No space left on device
-3+0 records in
-2+0 records out
+dd: fsync failed for '/mnt/scratch/testfile': Input/output error
+50+0 records in
+50+0 records out

Does that mean the bug is reproduced?

Quoted 10 lines
Quoted 1 line

+

tests/f2fs/006 new file +4 -0
+_scratch_remount checkpoint=enable
+
+status=0
+exit
tests/f2fs/006.out new file +0 -0
new file mode 100644
index 00000000..0d7b3910
Quoted 8 lines

tests/f2fs/006.out line 4

Quoted 5 lines
tests/f2fs/006.out new file +4 -0
@@ -0,0 +1,6 @@
+QA output created by 006
+50+0 records in
+50+0 records out
+dd: error writing '/mnt/scratch_f2fs/testfile': No space left on device

The "/mnt/scratch_f2fs" should be SCRATCH_MNT, please use _filter_scratch() by importing common/filter.

Correct, let me fix this.

Thanks,

Quoted 7 lines
Quoted 6 lines
tests/f2fs/006.out new file +2 -0
+3+0 records in
+2+0 records out
Signature · 1 line
2.40.1
author
#7 lore raw
Quoted 67 lines · Zorro Lang
Quoted 66 lines · Chao Yu
Quoted 64 lines · Zorro Lang
Quoted 51 lines · Chao Yu

This is a regression test to check whether f2fs handles dirty data correctly when checkpoint is disabled, if lfs mode is on, it will trigger OPU for all overwritten data, this will cost free segments, so f2fs must account overwritten data as OPU data when calculating free space, otherwise, it may run out of free segments in f2fs' allocation function, resulting in panic.

Cc Jaegeuk Kim <[email protected]>

---
v2:
- add _fixed_by_kernel_commit()
- use _scratch_mkfs_sized() rather than formating size-specified loop device
- code cleanup

   tests/f2fs/006     | 38 ++++++++++++++++++++++++++++++++++++++
   tests/f2fs/006.out |  6 ++++++
   2 files changed, 44 insertions(+)

create mode 100755 tests/f2fs/006
create mode 100644 tests/f2fs/006.out

tests/f2fs/006 new file +23 -0
new file mode 100755
index 00000000..63d00018
@@ -0,0 +1,38 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2024 Oppo.  All Rights Reserved.
+#
+# FS QA Test No. f2fs/006
+#
+# This is a regression test to check whether f2fs handles dirty
+# data correctly when checkpoint is disabled, if lfs mode is on,
+# it will trigger OPU for all overwritten data, this will cost
+# free segments, so f2fs must account overwritten data as OPU
+# data when calculating free space, otherwise, it may run out
+# of free segments in f2fs' allocation function, resulting in
+# panic.
+#
+. ./common/preamble
+_begin_fstest auto quick
+
+_fixed_by_kernel_commit xxxxxxxxxxxx \
+	"f2fs: fix to account dirty data in __get_secs_required()"
+
+testfile=$SCRATCH_MNT/testfile
+
+_require_scratch

tests/f2fs/006 line 33

Quoted 10 lines
tests/f2fs/006 new file +10 -0
+_scratch_mkfs_sized $((1024*1024*100)) >> $seqres.full
+
+# use mode=lfs to let f2fs always triggers OPU
+_scratch_mount -o mode=lfs,checkpoint=disable:10%,noinline_dentry >> $seqres.full
+
+dd if=/dev/zero of=$testfile bs=1M count=50 2>/dev/null
+
+# it may run out of free space of f2fs and hang kernel
+dd if=/dev/zero of=$testfile bs=1M count=50 conv=notrunc conv=fsync
+dd if=/dev/zero of=$testfile bs=1M count=50 conv=notrunc conv=fsync

What kind of failure should be printed at here if test on unfixed kernel?

Quoted 4 lines
Quoted 1 line

It will panic kernel w/o fix, can you please check dmesg?

The dmesg as below [2]. But it hit EIO [1], rather than ENOSPC, and didn't hang. If it's the same issue, or another issue, I think we can explain it in above comment.

It's the same issue, let me explain more about this.

You will encounter kernel hang after you enable CONFIG_F2FS_CHECK_FS config.

Thanks,

Quoted 201 lines

Thanks,
Zorro

[1]

# diff -u /root/git/xfstests/tests/f2fs/006.out /root/git/xfstests/results//default/f2fs/006.out.bad
root/git/xfstests/results//default/f2fs/006.out.bad 2024-10-28 20:54:33.252246497 +0800 renamed from root/git/xfstests/tests/f2fs/006.out 2024-10-28 20:51:08.381020424 +0800 +3 -3
@@ -1,6 +1,6 @@
  QA output created by 006
  50+0 records in
  50+0 records out
-dd: error writing '/mnt/scratch_f2fs/testfile': No space left on device
-3+0 records in
-2+0 records out
+dd: fsync failed for '/mnt/scratch/testfile': Input/output error
+50+0 records in
+50+0 records out

[2]

[3370744.465936] run fstests f2fs/006 at 2024-10-28 20:54:27
[3370746.308401] F2FS-fs (sda6): Adjust unusable cap for checkpoint=disable = 1530 / 10%
[3370746.318664] F2FS-fs (sda6): Found nat_bits in checkpoint
[3370746.341354] F2FS-fs (sda6): Start checkpoint disabled!
[3370746.347782] F2FS-fs (sda6): Mounted with checkpoint version = 355eea66
[3370747.846817] F2FS-fs (sda6): Stopped filesystem due to reason: 7
[3370747.853002] ------------[ cut here ]------------
[3370747.857826] WARNING: CPU: 1 PID: 791405 at fs/f2fs/segment.c:2748 new_curseg+0xc7e/0x1ef0 [f2fs]
[3370747.866938] Modules linked in: f2fs crc32_generic lz4hc_compress lz4_compress dm_thin_pool dm_persistent_data dm_bio_prison dm_log_writes xfs mlx5_ib ib_uverbs macsec ib_core nft_fib_inet nft_fib_ipv4 nft_fib_ipv6 nft_fib intel_rapl_msr intel_rapl_common nft_reject_inet nf_reject_ipv4 nf_reject_ipv6 intel_uncore_frequency nft_reject intel_uncore_frequency_common i10nm_edac skx_edac_common nft_ct nfit x86_pkg_temp_thermal intel_powerclamp nft_chain_nat nf_nat coretemp rfkill nf_conntrack nf_defrag_ipv6 kvm_intel nf_defrag_ipv4 snd_pcsp mlx5_core dax_hmem snd_pcm kvm ip_set cxl_acpi snd_timer spi_nor dell_pc iTCO_wdt rapl cxl_core intel_pmc_bxt snd dell_smbios acpi_power_meter iTCO_vendor_support mtd mlxfw ipmi_ssif intel_cstate platform_profile nf_tables dcdbas nd_pmem isst_if_mmio isst_if_mbox_pci psample dax_pmem nd_btt intel_uncore dell_wmi_descriptor wmi_bmof einj soundcore tls mei_me tg3 intel_th_gth spi_intel_pci i2c_i801 isst_if_common pci_hyperv_intf mei spi_intel intel_th_pci i2c_smbus intel_pch_thermal
[3370747.867283]  ipmi_si intel_th intel_vsec acpi_ipmi ipmi_devintf ipmi_msghandler fuse loop nfnetlink zram crct10dif_pclmul crc32_pclmul crc32c_intel polyval_clmulni polyval_generic ghash_clmulni_intel sha512_ssse3 nd_e820 sha256_ssse3 libnvdimm sha1_ssse3 megaraid_sas mgag200 i2c_algo_bit wmi [last unloaded: scsi_debug]
[3370747.985397] CPU: 1 UID: 0 PID: 791405 Comm: dd Tainted: G        W         -------  ---  6.11.0-0.rc6.49.fc42.x86_64+debug #1
[3370747.996895] Tainted: [W]=WARN
[3370748.000067] Hardware name: Dell Inc. PowerEdge R750/0PJ80M, BIOS 1.5.4 12/17/2021
[3370748.007747] RIP: 0010:new_curseg+0xc7e/0x1ef0 [f2fs]
[3370748.012949] Code: 8e 38 0b 00 00 3b ab d8 11 00 00 0f 82 b3 f8 ff ff 48 8b 7c 24 28 e8 f1 26 5b f0 ba 07 00 00 00 31 f6 48 89 df e8 42 6b f7 ff <0f> 0b be 08 00 00 00 48 8d bb 20 01 00 00 e8 3f db d1 ed f0 80 8b
[3370748.031898] RSP: 0018:ffa00000365cea90 EFLAGS: 00010296
[3370748.037323] RAX: 0000000000000000 RBX: ff1100110a61c000 RCX: 0000000000000000
[3370748.044656] RDX: 0000000000000033 RSI: ffffffffb2c56fe0 RDI: fff3fc0006cb9d23
[3370748.051989] RBP: 000000000000002a R08: 0000000000000001 R09: fff3fc0006cb9ce8
[3370748.059319] R10: ffa00000365ce747 R11: 0000000000000000 R12: 0000000000000001
[3370748.066652] R13: ff110009482c4178 R14: ff11001096338e00 R15: ffe21c02214c3a3b
[3370748.073984] FS:  00007f9d7a3d4740(0000) GS:ff11002031000000(0000) knlGS:0000000000000000
[3370748.082269] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[3370748.088211] CR2: 000056272e350c08 CR3: 0000001100088002 CR4: 0000000000771ef0
[3370748.095546] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[3370748.102906] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[3370748.110236] PKRU: 55555554
[3370748.113149] Call Trace:
[3370748.115800]  <TASK>
[3370748.118106]  ? __warn.cold+0x5b/0x1af
[3370748.121975]  ? new_curseg+0xc7e/0x1ef0 [f2fs]
[3370748.126575]  ? report_bug+0x1fc/0x3d0
[3370748.130449]  ? handle_bug+0x3c/0x80
[3370748.134141]  ? exc_invalid_op+0x17/0x40
[3370748.138181]  ? asm_exc_invalid_op+0x1a/0x20
[3370748.142591]  ? new_curseg+0xc7e/0x1ef0 [f2fs]
[3370748.147200]  ? update_segment_mtime+0x144/0x4c0 [f2fs]
[3370748.152591]  f2fs_allocate_data_block+0xbc9/0x4070 [f2fs]
[3370748.158244]  ? mark_lock+0xf5/0x16d0
[3370748.162021]  ? __pfx___lock_acquire+0x10/0x10
[3370748.166603]  ? __pfx_f2fs_allocate_data_block+0x10/0x10 [f2fs]
[3370748.172676]  ? __pfx___get_segment_type+0x10/0x10 [f2fs]
[3370748.178231]  ? folio_memcg_unlock+0x61/0x120
[3370748.182707]  ? local_clock_noinstr+0xd/0x100
[3370748.187191]  do_write_page+0x156/0xdf0 [f2fs]
[3370748.191789]  ? __pfx___update_extent_cache+0x10/0x10 [f2fs]
[3370748.197596]  ? mark_held_locks+0x94/0xe0
[3370748.201727]  ? __pfx_do_write_page+0x10/0x10 [f2fs]
[3370748.206865]  ? __folio_start_writeback+0x266/0x870
[3370748.211873]  f2fs_outplace_write_data+0x198/0x310 [f2fs]
[3370748.217419]  ? __pfx_f2fs_outplace_write_data+0x10/0x10 [f2fs]
[3370748.223503]  f2fs_do_write_data_page+0xaf9/0x1120 [f2fs]
[3370748.229175]  ? __pfx_f2fs_do_write_data_page+0x10/0x10 [f2fs]
[3370748.235176]  ? __pfx___lock_acquire+0x10/0x10
[3370748.239748]  f2fs_write_single_data_page+0xf9f/0x16e0 [f2fs]
[3370748.245674]  ? __pfx_f2fs_write_single_data_page+0x10/0x10 [f2fs]
[3370748.252007]  ? local_clock_noinstr+0xd/0x100
[3370748.256482]  ? folio_clear_dirty_for_io+0x20b/0x5b0
[3370748.261567]  ? local_clock_noinstr+0xd/0x100
[3370748.266081]  f2fs_write_cache_pages+0xa46/0x1ec0 [f2fs]
[3370748.271573]  ? __pfx_f2fs_write_cache_pages+0x10/0x10 [f2fs]
[3370748.277486]  ? __pfx___lock_acquire+0x10/0x10
[3370748.282051]  ? mark_lock+0xf5/0x16d0
[3370748.285869]  ? f2fs_write_data_pages+0x844/0xc00 [f2fs]
[3370748.291352]  ? rcu_is_watching+0x12/0xc0
[3370748.295482]  ? trace_contention_end+0xd4/0x110
[3370748.300190]  f2fs_write_data_pages+0x85d/0xc00 [f2fs]
[3370748.305491]  ? __pfx_f2fs_write_data_pages+0x10/0x10 [f2fs]
[3370748.311293]  ? __pfx___lock_acquire+0x10/0x10
[3370748.315889]  do_writepages+0x176/0x780
[3370748.319860]  ? __pfx_do_writepages+0x10/0x10
[3370748.324329]  ? filemap_fdatawrite_wbc+0xd3/0x180
[3370748.329162]  ? do_raw_spin_unlock+0x58/0x1f0
[3370748.333641]  ? _raw_spin_unlock+0x2d/0x50
[3370748.337859]  ? wbc_attach_and_unlock_inode+0x3da/0x7d0
[3370748.343204]  filemap_fdatawrite_wbc+0x113/0x180
[3370748.347939]  __filemap_fdatawrite_range+0xaf/0xf0
[3370748.352855]  ? __pfx___filemap_fdatawrite_range+0x10/0x10
[3370748.358515]  ? __pfx_lock_release+0x10/0x10
[3370748.362909]  file_write_and_wait_range+0x9b/0x110
[3370748.367819]  f2fs_do_sync_file+0x27b/0x1ab0 [f2fs]
[3370748.372882]  ? __pfx_f2fs_do_sync_file+0x10/0x10 [f2fs]
[3370748.378348]  ? __pfx_vfs_read+0x10/0x10
[3370748.382452]  ? __mark_inode_dirty+0x6ef/0x9d0
[3370748.387018]  ? ksys_read+0xfb/0x1d0
[3370748.390715]  ? vfs_fsync_range+0x11b/0x220
[3370748.395024]  __x64_sys_fsync+0x59/0xa0
[3370748.398980]  do_syscall_64+0x97/0x190
[3370748.402861]  ? __pfx_ksys_read+0x10/0x10
[3370748.406995]  ? lockdep_hardirqs_on_prepare+0x171/0x400
[3370748.412336]  ? do_syscall_64+0xa3/0x190
[3370748.416380]  ? lockdep_hardirqs_on+0x7c/0x100
[3370748.420941]  ? do_syscall_64+0xa3/0x190
[3370748.424981]  ? lockdep_hardirqs_on_prepare+0x171/0x400
[3370748.430324]  ? do_syscall_64+0xa3/0x190
[3370748.434364]  ? lockdep_hardirqs_on+0x7c/0x100
[3370748.438924]  ? do_syscall_64+0xa3/0x190
[3370748.442969]  ? do_syscall_64+0xa3/0x190
[3370748.447009]  ? lockdep_hardirqs_on_prepare+0x171/0x400
[3370748.452349]  ? do_syscall_64+0xa3/0x190
[3370748.456394]  ? lockdep_hardirqs_on+0x7c/0x100
[3370748.460953]  ? do_syscall_64+0xa3/0x190
[3370748.464994]  ? do_syscall_64+0xa3/0x190
[3370748.469039]  ? lockdep_hardirqs_on_prepare+0x171/0x400
[3370748.474380]  ? do_syscall_64+0xa3/0x190
[3370748.478427]  ? lockdep_hardirqs_on+0x7c/0x100
[3370748.483043]  ? clear_bhb_loop+0x25/0x80
[3370748.487082]  ? clear_bhb_loop+0x25/0x80
[3370748.491127]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
[3370748.496383] RIP: 0033:0x7f9d7a4c11a4
[3370748.500187] Code: 00 00 0f 1f 40 00 f3 0f 1e fa 48 8b 3d 9d 5c 10 00 e9 10 73 f1 ff f3 0f 1e fa 80 3d c5 5e 10 00 00 74 13 b8 4a 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 3c c3 0f 1f 00 55 4
8 89 e5 48 83 ec 10 89 7d
[3370748.519133] RSP: 002b:00007fff6c1e62c8 EFLAGS: 00000202 ORIG_RAX: 000000000000004a
[3370748.526905] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f9d7a4c11a4
[3370748.534239] RDX: 0000000000000032 RSI: 0000000000000000 RDI: 0000000000000001
[3370748.541568] RBP: 00007fff6c1e6320 R08: 0000000000000000 R09: 00007f9d7a60c380
[3370748.548932] R10: 0000000000000022 R11: 0000000000000202 R12: 00007f9d7a3d46c8
[3370748.556268] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000032
[3370748.563625]  </TASK>
[3370748.566016] irq event stamp: 342211
[3370748.569711] hardirqs last  enabled at (342221): [<ffffffffaf41b9f6>] console_unlock+0x236/0x2c0
[3370748.578611] hardirqs last disabled at (342230): [<ffffffffaf41b9db>] console_unlock+0x21b/0x2c0
[3370748.587504] softirqs last  enabled at (342178): [<ffffffffaf247c3b>] __irq_exit_rcu+0xbb/0x1c0
[3370748.596307] softirqs last disabled at (342157): [<ffffffffaf247c3b>] __irq_exit_rcu+0xbb/0x1c0
[3370748.605115] ---[ end trace 0000000000000000 ]---
[3370748.792483] F2FS-fs (sda6): Adjust unusable cap for checkpoint=disable = 1530 / 10%
[3370750.194429] F2FS-fs (sda6): Disable nat_bits due to incorrect cp_ver (4137321793606052454, 18446744073709551615)
[3370750.259391] F2FS-fs (sda6): Mounted with checkpoint version = 355eea67
Quoted 49 lines
Quoted 28 lines

I got:

# diff -u /root/git/xfstests/tests/f2fs/006.out /root/git/xfstests/results//default/f2fs/006.out.bad|less
root/git/xfstests/results//default/f2fs/006.out.bad 2024-10-25 11:34:55.907252401 +0800 renamed from root/git/xfstests/tests/f2fs/006.out 2024-10-25 11:33:54.693883281 +0800 +3 -3
@@ -1,6 +1,6 @@
   QA output created by 006
   50+0 records in
   50+0 records out
-dd: error writing '/mnt/scratch_f2fs/testfile': No space left on device
-3+0 records in
-2+0 records out
+dd: fsync failed for '/mnt/scratch/testfile': Input/output error
+50+0 records in
+50+0 records out

Does that mean the bug is reproduced?

Quoted 10 lines
Quoted 1 line

+

tests/f2fs/006 new file +4 -0
+_scratch_remount checkpoint=enable
+
+status=0
+exit
tests/f2fs/006.out new file +0 -0
new file mode 100644
index 00000000..0d7b3910
Quoted 8 lines

tests/f2fs/006.out line 4

Quoted 5 lines
tests/f2fs/006.out new file +4 -0
@@ -0,0 +1,6 @@
+QA output created by 006
+50+0 records in
+50+0 records out
+dd: error writing '/mnt/scratch_f2fs/testfile': No space left on device

The "/mnt/scratch_f2fs" should be SCRATCH_MNT, please use _filter_scratch() by importing common/filter.

Correct, let me fix this.

Thanks,

Quoted 7 lines
Quoted 6 lines
tests/f2fs/006.out new file +2 -0
+3+0 records in
+2+0 records out
Signature · 1 line
2.40.1