Skip to contentSkip to conversation

f2fs/021: test quota mount option

PATCH Superseded Chao Yu 2 participants 7 messages Sep 1, 2025 – Sep 28, 2025

v1 of 3 Superseded Latest is v3 →

lore
Share & tools
lore

Chao Yu

author
#0 lore raw

PATCH 1/2 f2fs/021: test quota mount option

This testcase tries to check whether f2fs can handle "usrjquota=" during remount correctly, it expects kernel will encounter NULL pointer dereference bug w/o the fix ("f2fs: fix to avoid NULL pointer dereference in f2fs_check_quota_consistency()").

Cc Jaegeuk Kim <[email protected]>
---
 tests/f2fs/021     | 30 ++++++++++++++++++++++++++++++
 tests/f2fs/021.out |  2 ++
 2 files changed, 32 insertions(+)
 create mode 100755 tests/f2fs/021
 create mode 100644 tests/f2fs/021.out
tests/f2fs/021 new file +30 -0
new file mode 100755
index 00000000..c41760fc
@@ -0,0 +1,30 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2025 Chao Yu.  All Rights Reserved.
+#
+# FS QA Test No. f2fs/021
+#
+# This testcase tries to check whether f2fs can handle "usrjquota="
+# during remount correctly
+#
+. ./common/preamble
+_begin_fstest auto quick mount
+
+_fixed_by_kernel_commit xxxxxxxxxxxx \
+	"f2fs: fix to avoid NULL pointer dereference in f2fs_check_quota_consistency()"
+
+_require_scratch
+
+_scratch_mkfs >> $seqres.full
+_scratch_mount "-o usrquota"
+quotacheck -uc $SCRATCH_MNT
+_scratch_unmount
+
+_scratch_mount "-o usrjquota=aquota.user,jqfmt=vfsold"
+_scratch_mount "-o remount,usrjquota=,jqfmt=vfsold"
+_scratch_unmount
+
+echo "Silence is golden"
+
+status=0
+exit
tests/f2fs/021.out new file +2 -0
new file mode 100644
index 00000000..09f4062d
@@ -0,0 +1,2 @@
+QA output created by 021
+Silence is golden
Signature · 1 line
2.49.0
author
#1 lore raw

PATCH 2/2 f2fs/022: do sanity check on footer of non inode dnode

This is a regression test:
1. create foo & bar
2. write 8M data to foo
3. use inject.f2fs to inject i_nid[0] of foo w/ ino of bar
4. fpunch in foo w/ specified range

If we haven't applied kernel patch ("f2fs: fix to do sanity check on node footer for non inode dnode"), f2fs may missed to do sanity check on corrupted dnode, result in panic in step 4).

Cc Jaegeuk Kim <[email protected]>
---
 tests/f2fs/022     | 52 ++++++++++++++++++++++++++++++++++++++++++++++
 tests/f2fs/022.out |  2 ++
 2 files changed, 54 insertions(+)
 create mode 100755 tests/f2fs/022
 create mode 100644 tests/f2fs/022.out
tests/f2fs/022 new file +52 -0
new file mode 100755
index 00000000..a5e19a71
@@ -0,0 +1,52 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2025 Chao Yu.  All Rights Reserved.
+#
+# FS QA Test No. f2fs/022
+#
+# This is a regression test:
+# 1. create foo & bar
+# 2. write 8M data to foo
+# 3. use inject.f2fs to inject i_nid[0] of foo w/ ino of bar
+# 4. fpunch in foo w/ specified range
+#
+. ./common/preamble
+_begin_fstest auto quick rw
+
+. ./common/attr
+
+_fixed_by_kernel_commit xxxxxxxxxxxx \
+	"f2fs: fix to do sanity check on node footer for non inode dnode"
+
+_require_scratch_nocheck
+_require_command "$F2FS_INJECT_PROG" inject.f2fs
+_require_xfs_io_command "fpunch"
+_require_kernel_config CONFIG_F2FS_CHECK_FS
+
+# remove all mkfs options to avoid layout change of on-disk inode
+export MKFS_OPTIONS=""
+
+foo_path=$SCRATCH_MNT/foo
+bar_path=$SCRATCH_MNT/bar
+
+_scratch_mkfs >> $seqres.full
+_scratch_mount
+
+touch $foo_path
+touch $bar_path
+$XFS_IO_PROG $foo_path -c "pwrite 0 8M" >> $seqres.full
+sync
+foo_ino=`stat -c '%i' $foo_path`
+bar_ino=`stat -c '%i' $bar_path`
+_scratch_unmount
+
+# inject foo inode to replace i_nid[0] w/ to bar ino
+$F2FS_INJECT_PROG --node --mb i_nid --nid $foo_ino --idx 0 --val $bar_ino $SCRATCH_DEV >> $seqres.full || _fail "failed to inject"
+
+_scratch_mount
+# expect to trigger panic
+$XFS_IO_PROG $foo_path -c "fpunch 6984k 4k"
+_scratch_unmount
+
+status=0
+exit
tests/f2fs/022.out new file +2 -0
new file mode 100644
index 00000000..5307e3d7
@@ -0,0 +1,2 @@
+QA output created by 022
+fallocate: Structure needs cleaning
Signature · 1 line
2.49.0
#2 lore raw

Re: [PATCH 2/2] f2fs/022: do sanity check on footer of non inode dnode

Quoted 39 lines · Chao Yu

This is a regression test:
1. create foo & bar
2. write 8M data to foo
3. use inject.f2fs to inject i_nid[0] of foo w/ ino of bar
4. fpunch in foo w/ specified range

If we haven't applied kernel patch ("f2fs: fix to do sanity check on node footer for non inode dnode"), f2fs may missed to do sanity check on corrupted dnode, result in panic in step 4).

Cc Jaegeuk Kim <[email protected]>
---
 tests/f2fs/022     | 52 ++++++++++++++++++++++++++++++++++++++++++++++
 tests/f2fs/022.out |  2 ++
 2 files changed, 54 insertions(+)
 create mode 100755 tests/f2fs/022
 create mode 100644 tests/f2fs/022.out
tests/f2fs/022 new file +14 -0
new file mode 100755
index 00000000..a5e19a71
@@ -0,0 +1,52 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2025 Chao Yu.  All Rights Reserved.
+#
+# FS QA Test No. f2fs/022
+#
+# This is a regression test:
+# 1. create foo & bar
+# 2. write 8M data to foo
+# 3. use inject.f2fs to inject i_nid[0] of foo w/ ino of bar
+# 4. fpunch in foo w/ specified range
+#
+. ./common/preamble
+_begin_fstest auto quick rw

tests/f2fs/022 line 24

Quoted 10 lines
tests/f2fs/022 new file +10 -0
+
+. ./common/attr
+
+_fixed_by_kernel_commit xxxxxxxxxxxx \
+	"f2fs: fix to do sanity check on node footer for non inode dnode"
+
+_require_scratch_nocheck
+_require_command "$F2FS_INJECT_PROG" inject.f2fs
+_require_xfs_io_command "fpunch"
+_require_kernel_config CONFIG_F2FS_CHECK_FS

Looks like the CONFIG_F2FS_CHECK_FS is a necessary condition to reproduce that bug, right? I'm wondering is there a better way to check if CONFIG_F2FS_CHECK_FS is enabled, likes reading someone file in /sys or what ever else? Checking kernel config isn't always a recommended way, except no better idea :)

Quoted 39 lines
tests/f2fs/022 new file +28 -0
+
+# remove all mkfs options to avoid layout change of on-disk inode
+export MKFS_OPTIONS=""
+
+foo_path=$SCRATCH_MNT/foo
+bar_path=$SCRATCH_MNT/bar
+
+_scratch_mkfs >> $seqres.full
+_scratch_mount
+
+touch $foo_path
+touch $bar_path
+$XFS_IO_PROG $foo_path -c "pwrite 0 8M" >> $seqres.full
+sync
+foo_ino=`stat -c '%i' $foo_path`
+bar_ino=`stat -c '%i' $bar_path`
+_scratch_unmount
+
+# inject foo inode to replace i_nid[0] w/ to bar ino
+$F2FS_INJECT_PROG --node --mb i_nid --nid $foo_ino --idx 0 --val $bar_ino $SCRATCH_DEV >> $seqres.full || _fail "failed to inject"
+
+_scratch_mount
+# expect to trigger panic
+$XFS_IO_PROG $foo_path -c "fpunch 6984k 4k"
+_scratch_unmount
+
+status=0
+exit
tests/f2fs/022.out new file +2 -0
new file mode 100644
index 00000000..5307e3d7
@@ -0,0 +1,2 @@
+QA output created by 022
+fallocate: Structure needs cleaning
Signature · 1 line
2.49.0
author
#3 lore raw

Re: [PATCH 2/2] f2fs/022: do sanity check on footer of non inode dnode

Quoted 46 lines · Zorro Lang
Quoted 45 lines · Chao Yu

This is a regression test:
1. create foo & bar
2. write 8M data to foo
3. use inject.f2fs to inject i_nid[0] of foo w/ ino of bar
4. fpunch in foo w/ specified range

If we haven't applied kernel patch ("f2fs: fix to do sanity check on node footer for non inode dnode"), f2fs may missed to do sanity check on corrupted dnode, result in panic in step 4).

Cc Jaegeuk Kim <[email protected]>
---
  tests/f2fs/022     | 52 ++++++++++++++++++++++++++++++++++++++++++++++
  tests/f2fs/022.out |  2 ++
  2 files changed, 54 insertions(+)

create mode 100755 tests/f2fs/022
create mode 100644 tests/f2fs/022.out

tests/f2fs/022 new file +20 -0
new file mode 100755
index 00000000..a5e19a71
@@ -0,0 +1,52 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2025 Chao Yu.  All Rights Reserved.
+#
+# FS QA Test No. f2fs/022
+#
+# This is a regression test:
+# 1. create foo & bar
+# 2. write 8M data to foo
+# 3. use inject.f2fs to inject i_nid[0] of foo w/ ino of bar
+# 4. fpunch in foo w/ specified range
+#
+. ./common/preamble
+_begin_fstest auto quick rw
+
+. ./common/attr
+
+_fixed_by_kernel_commit xxxxxxxxxxxx \
+	"f2fs: fix to do sanity check on node footer for non inode dnode"
+
Quoted 8 lines

tests/f2fs/022 line 24

Quoted 4 lines
tests/f2fs/022 new file +4 -0
+_require_scratch_nocheck
+_require_command "$F2FS_INJECT_PROG" inject.f2fs
+_require_xfs_io_command "fpunch"
+_require_kernel_config CONFIG_F2FS_CHECK_FS

Looks like the CONFIG_F2FS_CHECK_FS is a necessary condition to reproduce that bug, right? I'm wondering is there a better way to check if CONFIG_F2FS_CHECK_FS is enabled, likes reading someone file in /sys or what ever else? Checking kernel config

No, I don't see any sysfs entry for checing if CONFIG_F2FS_CHECK_FS is enabled.

I checked the testcase w/ image which has CONFIG_F2FS_CHECK_FS=off, the testcase falls into a deadloop inside kernel, so it's fine to get rid of the CONFIG_F2FS_CHECK_FS requirement? Thoughts?

Thanks,

Quoted 42 lines

isn't always a recommended way, except no better idea :)

Quoted 39 lines
tests/f2fs/022 new file +28 -0
+
+# remove all mkfs options to avoid layout change of on-disk inode
+export MKFS_OPTIONS=""
+
+foo_path=$SCRATCH_MNT/foo
+bar_path=$SCRATCH_MNT/bar
+
+_scratch_mkfs >> $seqres.full
+_scratch_mount
+
+touch $foo_path
+touch $bar_path
+$XFS_IO_PROG $foo_path -c "pwrite 0 8M" >> $seqres.full
+sync
+foo_ino=`stat -c '%i' $foo_path`
+bar_ino=`stat -c '%i' $bar_path`
+_scratch_unmount
+
+# inject foo inode to replace i_nid[0] w/ to bar ino
+$F2FS_INJECT_PROG --node --mb i_nid --nid $foo_ino --idx 0 --val $bar_ino $SCRATCH_DEV >> $seqres.full || _fail "failed to inject"
+
+_scratch_mount
+# expect to trigger panic
+$XFS_IO_PROG $foo_path -c "fpunch 6984k 4k"
+_scratch_unmount
+
+status=0
+exit
tests/f2fs/022.out new file +2 -0
new file mode 100644
index 00000000..5307e3d7
@@ -0,0 +1,2 @@
+QA output created by 022
+fallocate: Structure needs cleaning
Signature · 1 line
2.49.0
#4 lore raw

Re: [PATCH 2/2] f2fs/022: do sanity check on footer of non inode dnode

Quoted 56 lines · Chao Yu
Quoted 46 lines · Zorro Lang
Quoted 45 lines · Chao Yu

This is a regression test:
1. create foo & bar
2. write 8M data to foo
3. use inject.f2fs to inject i_nid[0] of foo w/ ino of bar
4. fpunch in foo w/ specified range

If we haven't applied kernel patch ("f2fs: fix to do sanity check on node footer for non inode dnode"), f2fs may missed to do sanity check on corrupted dnode, result in panic in step 4).

Cc Jaegeuk Kim <[email protected]>
---
  tests/f2fs/022     | 52 ++++++++++++++++++++++++++++++++++++++++++++++
  tests/f2fs/022.out |  2 ++
  2 files changed, 54 insertions(+)

create mode 100755 tests/f2fs/022
create mode 100644 tests/f2fs/022.out

tests/f2fs/022 new file +20 -0
new file mode 100755
index 00000000..a5e19a71
@@ -0,0 +1,52 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2025 Chao Yu.  All Rights Reserved.
+#
+# FS QA Test No. f2fs/022
+#
+# This is a regression test:
+# 1. create foo & bar
+# 2. write 8M data to foo
+# 3. use inject.f2fs to inject i_nid[0] of foo w/ ino of bar
+# 4. fpunch in foo w/ specified range
+#
+. ./common/preamble
+_begin_fstest auto quick rw
+
+. ./common/attr
+
+_fixed_by_kernel_commit xxxxxxxxxxxx \
+	"f2fs: fix to do sanity check on node footer for non inode dnode"
+
Quoted 8 lines

tests/f2fs/022 line 24

Quoted 4 lines
tests/f2fs/022 new file +4 -0
+_require_scratch_nocheck
+_require_command "$F2FS_INJECT_PROG" inject.f2fs
+_require_xfs_io_command "fpunch"
+_require_kernel_config CONFIG_F2FS_CHECK_FS

Looks like the CONFIG_F2FS_CHECK_FS is a necessary condition to reproduce that bug, right? I'm wondering is there a better way to check if CONFIG_F2FS_CHECK_FS is enabled, likes reading someone file in /sys or what ever else? Checking kernel config

Quoted 6 lines

No, I don't see any sysfs entry for checing if CONFIG_F2FS_CHECK_FS is enabled.

I checked the testcase w/ image which has CONFIG_F2FS_CHECK_FS=off, the testcase falls into a deadloop inside kernel, so it's fine to get rid of the CONFIG_F2FS_CHECK_FS requirement? Thoughts?

Sure, let's remove the unnecessary requirement :) And ...

Quoted 21 lines

Thanks,

Quoted 18 lines

isn't always a recommended way, except no better idea :)

Quoted 16 lines
tests/f2fs/022 new file +16 -0
+
+# remove all mkfs options to avoid layout change of on-disk inode
+export MKFS_OPTIONS=""
+
+foo_path=$SCRATCH_MNT/foo
+bar_path=$SCRATCH_MNT/bar
+
+_scratch_mkfs >> $seqres.full
+_scratch_mount
+
+touch $foo_path
+touch $bar_path
+$XFS_IO_PROG $foo_path -c "pwrite 0 8M" >> $seqres.full
+sync
+foo_ino=`stat -c '%i' $foo_path`
+bar_ino=`stat -c '%i' $bar_path`
Quoted 8 lines
Quoted 8 lines

tests/f2fs/022 line 48

Quoted 8 lines
tests/f2fs/022 new file +8 -0
+_scratch_unmount
+
+# inject foo inode to replace i_nid[0] w/ to bar ino
+$F2FS_INJECT_PROG --node --mb i_nid --nid $foo_ino --idx 0 --val $bar_ino $SCRATCH_DEV >> $seqres.full || _fail "failed to inject"
+
+_scratch_mount
+# expect to trigger panic
+$XFS_IO_PROG $foo_path -c "fpunch 6984k 4k"

... if you might be better to change this comment a bit :)

If you'd like to resend this patchset, please change the patch 1/2 by the way. then I can merge your next version directly :)

Thanks,
Zorro

Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
tests/f2fs/022 new file +4 -0
+_scratch_unmount
+
+status=0
+exit
tests/f2fs/022.out new file +0 -1
new file mode 100644
index 00000000..5307e3d7
--- /dev/null
Quoted 9 lines
Quoted 8 lines
Quoted 7 lines

+++ b/tests/f2fs/022.out

tests/f2fs/022.out new file +2 -0
@@ -0,0 +1,2 @@
+QA output created by 022
+fallocate: Structure needs cleaning
Signature · 1 line
2.49.0

_______________________________________________
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

Reviewed-by on v1 1/2
#5 lore raw
In reply to Chao Yu
Quoted 21 lines · Chao Yu

This testcase tries to check whether f2fs can handle "usrjquota=" during remount correctly, it expects kernel will encounter NULL pointer dereference bug w/o the fix ("f2fs: fix to avoid NULL pointer dereference in f2fs_check_quota_consistency()").

Cc Jaegeuk Kim <[email protected]>
---
 tests/f2fs/021     | 30 ++++++++++++++++++++++++++++++
 tests/f2fs/021.out |  2 ++
 2 files changed, 32 insertions(+)
 create mode 100755 tests/f2fs/021
 create mode 100644 tests/f2fs/021.out
tests/f2fs/021 new file +1 -0
new file mode 100755
index 00000000..c41760fc
@@ -0,0 +1,30 @@
+#! /bin/bash

tests/f2fs/021 line 11

Quoted 10 lines
tests/f2fs/021 new file +10 -0
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2025 Chao Yu.  All Rights Reserved.
+#
+# FS QA Test No. f2fs/021
+#
+# This testcase tries to check whether f2fs can handle "usrjquota="
+# during remount correctly
+#
+. ./common/preamble
+_begin_fstest auto quick mount
                                  ^^^^^^^^^^^^^
                                  quota remount

I'll help to add these two tags when I merge it. Others look good to me, if you've maken sure it can trigger the bug you want to uncover.

Quoted 30 lines
tests/f2fs/021 new file +19 -0
+
+_fixed_by_kernel_commit xxxxxxxxxxxx \
+	"f2fs: fix to avoid NULL pointer dereference in f2fs_check_quota_consistency()"
+
+_require_scratch
+
+_scratch_mkfs >> $seqres.full
+_scratch_mount "-o usrquota"
+quotacheck -uc $SCRATCH_MNT
+_scratch_unmount
+
+_scratch_mount "-o usrjquota=aquota.user,jqfmt=vfsold"
+_scratch_mount "-o remount,usrjquota=,jqfmt=vfsold"
+_scratch_unmount
+
+echo "Silence is golden"
+
+status=0
+exit
tests/f2fs/021.out new file +2 -0
new file mode 100644
index 00000000..09f4062d
@@ -0,0 +1,2 @@
+QA output created by 021
+Silence is golden
Signature · 1 line
2.49.0
author
#6 lore raw
Quoted 35 lines · Zorro Lang
Quoted 21 lines · Chao Yu

This testcase tries to check whether f2fs can handle "usrjquota=" during remount correctly, it expects kernel will encounter NULL pointer dereference bug w/o the fix ("f2fs: fix to avoid NULL pointer dereference in f2fs_check_quota_consistency()").

Cc Jaegeuk Kim <[email protected]>
---
  tests/f2fs/021     | 30 ++++++++++++++++++++++++++++++
  tests/f2fs/021.out |  2 ++
  2 files changed, 32 insertions(+)

create mode 100755 tests/f2fs/021
create mode 100644 tests/f2fs/021.out

tests/f2fs/021 new file +1 -0
new file mode 100755
index 00000000..c41760fc
@@ -0,0 +1,30 @@
+#! /bin/bash

tests/f2fs/021 line 11

Quoted 10 lines
tests/f2fs/021 new file +10 -0
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2025 Chao Yu.  All Rights Reserved.
+#
+# FS QA Test No. f2fs/021
+#
+# This testcase tries to check whether f2fs can handle "usrjquota="
+# during remount correctly
+#
+. ./common/preamble
+_begin_fstest auto quick mount
                                   ^^^^^^^^^^^^^
                                   quota remount
Quoted 4 lines

I'll help to add these two tags when I merge it. Others look good to me, if you've maken sure it can trigger the bug you want to uncover.

Thanks for the fix.

Thanks,

Quoted 32 lines
Quoted 31 lines
tests/f2fs/021 new file +19 -0
+
+_fixed_by_kernel_commit xxxxxxxxxxxx \
+	"f2fs: fix to avoid NULL pointer dereference in f2fs_check_quota_consistency()"
+
+_require_scratch
+
+_scratch_mkfs >> $seqres.full
+_scratch_mount "-o usrquota"
+quotacheck -uc $SCRATCH_MNT
+_scratch_unmount
+
+_scratch_mount "-o usrjquota=aquota.user,jqfmt=vfsold"
+_scratch_mount "-o remount,usrjquota=,jqfmt=vfsold"
+_scratch_unmount
+
+echo "Silence is golden"
+
+status=0
+exit
tests/f2fs/021.out new file +2 -0
new file mode 100644
index 00000000..09f4062d
@@ -0,0 +1,2 @@
+QA output created by 021
+Silence is golden
Signature · 1 line
2.49.0