diff mbox series

[v2] ext4: Regression test of ext4_lblk_t overflow

Message ID 20231120070327.1922631-1-libaokun1@huawei.com (mailing list archive)
State New, archived
Headers show
Series [v2] ext4: Regression test of ext4_lblk_t overflow | expand

Commit Message

Baokun Li Nov. 20, 2023, 7:03 a.m. UTC 
Append writes to a file approaching 16T and observe if a kernel crash is
caused by ext4_lblk_t overflow triggering BUG_ON at ext4_mb_new_inode_pa().
This is a regression test for commit bc056e7163ac ("ext4: fix BUG in
ext4_mb_new_inode_pa() due to overflow")

Signed-off-by: Baokun Li <libaokun1@huawei.com>
---
V1->V2:
	Changes to make the use case more generic, not just for testing
	ext4.(ext4 and xfs have been tested)

 tests/generic/737     | 55 +++++++++++++++++++++++++++++++++++++++++++
 tests/generic/737.out |  2 ++
 2 files changed, 57 insertions(+)
 create mode 100755 tests/generic/737
 create mode 100644 tests/generic/737.out

Comments

Zorro Lang Nov. 21, 2023, 9:28 a.m. UTC | #1 
On Mon, Nov 20, 2023 at 03:03:27PM +0800, Baokun Li wrote:
> Append writes to a file approaching 16T and observe if a kernel crash is
> caused by ext4_lblk_t overflow triggering BUG_ON at ext4_mb_new_inode_pa().
> This is a regression test for commit bc056e7163ac ("ext4: fix BUG in
> ext4_mb_new_inode_pa() due to overflow")
> 
> Signed-off-by: Baokun Li <libaokun1@huawei.com>
> ---
> V1->V2:
> 	Changes to make the use case more generic, not just for testing
> 	ext4.(ext4 and xfs have been tested)

Thanks for this change,

> 
>  tests/generic/737     | 55 +++++++++++++++++++++++++++++++++++++++++++
>  tests/generic/737.out |  2 ++
>  2 files changed, 57 insertions(+)
>  create mode 100755 tests/generic/737
>  create mode 100644 tests/generic/737.out
> 
> diff --git a/tests/generic/737 b/tests/generic/737
> new file mode 100755
> index 00000000..a65c448b
> --- /dev/null
> +++ b/tests/generic/737
> @@ -0,0 +1,55 @@
> +#! /bin/bash
> +# SPDX-License-Identifier: GPL-2.0
> +# Copyright (c) 2023 HUAWEI.  All Rights Reserved.
> +#
> +# FS QA Test No. 737
> +#
> +# Append writes to a file approaching 16T and observe if a kernel crash is
> +# caused by ext4_lblk_t overflow triggering BUG_ON at ext4_mb_new_inode_pa().
> +# This is a regression test for commit
> +# bc056e7163ac ("ext4: fix BUG in ext4_mb_new_inode_pa() due to overflow")
> +#
> +#set -x

Useless comment "set -x"

> +. ./common/preamble
> +. ./common/populate
> +_begin_fstest auto quick insert prealloc
> +
> +# real QA test starts here
> +_fixed_by_kernel_commit bc056e7163ac \
> +	"ext4: fix BUG in ext4_mb_new_inode_pa() due to overflow"

[[ "$FSTYP" =~ ext* ]] && ....

> +
> +_require_xfs_io_command "falloc"
> +_require_xfs_io_command "finsert"
> +
> +# Initalize a 4k 100M fs

It might not be 4k block size, so this comment is useless. Or you can say
better to use "4k" block size to tigger the bug:)

> +dev_size=$((100 * 1024 * 1024))
> +_scratch_mkfs_sized $dev_size >>$seqres.full 2>&1 || _fail "mkfs failed"
> +
> +_scratch_mount
> +blksz="$(_get_block_size ${SCRATCH_MNT})"
> +_require_odirect $blksz

I think the _require_odirect can be move up, before the mkfs calling.

> +
> +# Reserve 1M space
> +$XFS_IO_PROG -f -c "falloc 0 1M" "${SCRATCH_MNT}/tmp" >> $seqres.full
> +
> +# Create a file (~16T) with logical block numbers close to overflow
> +$XFS_IO_PROG -f -c "falloc 0 10M" "${SCRATCH_MNT}/file" >> $seqres.full
> +insert_size=$((blksz * 4096 - 10 - 3))
> +$XFS_IO_PROG -f -c "finsert 1M ${insert_size}M" "${SCRATCH_MNT}/file" >> $seqres.full
> +
> +# Filling up the free space ensures that the pre-allocated space is the reserved space.
> +nr_free=$(stat -f -c '%f' ${SCRATCH_MNT})
> +_fill_fs $((nr_free * blksz)) ${SCRATCH_MNT}/fill $blksz 0 >> $seqres.full 2>&1
> +sync
> +
> +# Remove reserved space to gain free space for allocation
> +rm -f ${SCRATCH_MNT}/tmp
> +
> +# Trying to allocate two blocks triggers BUG_ON.
> +$XFS_IO_PROG -c "open -ad ${SCRATCH_MNT}/file" -c "pwrite -S 0xff 0 8192" >> $seqres.full

Shouldn't the 8192 be "$((2 * blksz))", if the comment says "two blocks" ?

I tried 1024 blocksize, really can't reproduce this bug, do you know why? Maybe
related with the falloc/finsert size?
Anyway, it covers this bug on default block size of ext4, and can be a generic
test on other conditions.

Thanks,
Zorro

> +
> +echo "Silence is golden"
> +
> +# success, all done
> +status=0
> +exit
> diff --git a/tests/generic/737.out b/tests/generic/737.out
> new file mode 100644
> index 00000000..67b83d78
> --- /dev/null
> +++ b/tests/generic/737.out
> @@ -0,0 +1,2 @@
> +QA output created by 737
> +Silence is golden
> -- 
> 2.31.1
>
Baokun Li Nov. 22, 2023, 11:28 a.m. UTC | #2 
On 2023/11/21 17:28, Zorro Lang wrote:
> On Mon, Nov 20, 2023 at 03:03:27PM +0800, Baokun Li wrote:
>> Append writes to a file approaching 16T and observe if a kernel crash is
>> caused by ext4_lblk_t overflow triggering BUG_ON at ext4_mb_new_inode_pa().
>> This is a regression test for commit bc056e7163ac ("ext4: fix BUG in
>> ext4_mb_new_inode_pa() due to overflow")
>>
>> Signed-off-by: Baokun Li <libaokun1@huawei.com>
>> ---
>> V1->V2:
>> 	Changes to make the use case more generic, not just for testing
>> 	ext4.(ext4 and xfs have been tested)
> Thanks for this change,
>
>>   tests/generic/737     | 55 +++++++++++++++++++++++++++++++++++++++++++
>>   tests/generic/737.out |  2 ++
>>   2 files changed, 57 insertions(+)
>>   create mode 100755 tests/generic/737
>>   create mode 100644 tests/generic/737.out
>>
>> diff --git a/tests/generic/737 b/tests/generic/737
>> new file mode 100755
>> index 00000000..a65c448b
>> --- /dev/null
>> +++ b/tests/generic/737
>> @@ -0,0 +1,55 @@
>> +#! /bin/bash
>> +# SPDX-License-Identifier: GPL-2.0
>> +# Copyright (c) 2023 HUAWEI.  All Rights Reserved.
>> +#
>> +# FS QA Test No. 737
>> +#
>> +# Append writes to a file approaching 16T and observe if a kernel crash is
>> +# caused by ext4_lblk_t overflow triggering BUG_ON at ext4_mb_new_inode_pa().
>> +# This is a regression test for commit
>> +# bc056e7163ac ("ext4: fix BUG in ext4_mb_new_inode_pa() due to overflow")
>> +#
>> +#set -x
> Useless comment "set -x"
Oh, I'm sorry, it was for debugging and I forgot to delete it.
>> +. ./common/preamble
>> +. ./common/populate
>> +_begin_fstest auto quick insert prealloc
>> +
>> +# real QA test starts here
>> +_fixed_by_kernel_commit bc056e7163ac \
>> +	"ext4: fix BUG in ext4_mb_new_inode_pa() due to overflow"
> [[ "$FSTYP" =~ ext* ]] && ....
This looks good.
>
>> +
>> +_require_xfs_io_command "falloc"
>> +_require_xfs_io_command "finsert"
>> +
>> +# Initalize a 4k 100M fs
> It might not be 4k block size, so this comment is useless. Or you can say
> better to use "4k" block size to tigger the bug:)
OK!
>> +dev_size=$((100 * 1024 * 1024))
>> +_scratch_mkfs_sized $dev_size >>$seqres.full 2>&1 || _fail "mkfs failed"
>> +
>> +_scratch_mount
>> +blksz="$(_get_block_size ${SCRATCH_MNT})"
>> +_require_odirect $blksz
> I think the _require_odirect can be move up, before the mkfs calling.
Totally agree!
>> +
>> +# Reserve 1M space
>> +$XFS_IO_PROG -f -c "falloc 0 1M" "${SCRATCH_MNT}/tmp" >> $seqres.full
>> +
>> +# Create a file (~16T) with logical block numbers close to overflow
>> +$XFS_IO_PROG -f -c "falloc 0 10M" "${SCRATCH_MNT}/file" >> $seqres.full
>> +insert_size=$((blksz * 4096 - 10 - 3))
>> +$XFS_IO_PROG -f -c "finsert 1M ${insert_size}M" "${SCRATCH_MNT}/file" >> $seqres.full
>> +
>> +# Filling up the free space ensures that the pre-allocated space is the reserved space.
>> +nr_free=$(stat -f -c '%f' ${SCRATCH_MNT})
>> +_fill_fs $((nr_free * blksz)) ${SCRATCH_MNT}/fill $blksz 0 >> $seqres.full 2>&1
>> +sync
>> +
>> +# Remove reserved space to gain free space for allocation
>> +rm -f ${SCRATCH_MNT}/tmp
>> +
>> +# Trying to allocate two blocks triggers BUG_ON.
>> +$XFS_IO_PROG -c "open -ad ${SCRATCH_MNT}/file" -c "pwrite -S 0xff 0 8192" >> $seqres.full
> Shouldn't the 8192 be "$((2 * blksz))", if the comment says "two blocks" ?
Okay, I'll change it to "$((2 * blksz))".
> I tried 1024 blocksize, really can't reproduce this bug, do you know why? Maybe
> related with the falloc/finsert size?
I'm not sure why this isn't triggered at a block size of 1024, I haven't 
looked into this in depth.
> Anyway, it covers this bug on default block size of ext4, and can be a generic
> test on other conditions.
>
> Thanks,
> Zorro

Yes! Thank you very much for your review!

>> +
>> +echo "Silence is golden"
>> +
>> +# success, all done
>> +status=0
>> +exit
>> diff --git a/tests/generic/737.out b/tests/generic/737.out
>> new file mode 100644
>> index 00000000..67b83d78
>> --- /dev/null
>> +++ b/tests/generic/737.out
>> @@ -0,0 +1,2 @@
>> +QA output created by 737
>> +Silence is golden
>> -- 
>> 2.31.1
>>
Thanks!
diff mbox series

Patch

diff --git a/tests/generic/737 b/tests/generic/737
new file mode 100755
index 00000000..a65c448b
--- /dev/null
+++ b/tests/generic/737
@@ -0,0 +1,55 @@ 
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2023 HUAWEI.  All Rights Reserved.
+#
+# FS QA Test No. 737
+#
+# Append writes to a file approaching 16T and observe if a kernel crash is
+# caused by ext4_lblk_t overflow triggering BUG_ON at ext4_mb_new_inode_pa().
+# This is a regression test for commit
+# bc056e7163ac ("ext4: fix BUG in ext4_mb_new_inode_pa() due to overflow")
+#
+#set -x
+. ./common/preamble
+. ./common/populate
+_begin_fstest auto quick insert prealloc
+
+# real QA test starts here
+_fixed_by_kernel_commit bc056e7163ac \
+	"ext4: fix BUG in ext4_mb_new_inode_pa() due to overflow"
+
+_require_xfs_io_command "falloc"
+_require_xfs_io_command "finsert"
+
+# Initalize a 4k 100M fs
+dev_size=$((100 * 1024 * 1024))
+_scratch_mkfs_sized $dev_size >>$seqres.full 2>&1 || _fail "mkfs failed"
+
+_scratch_mount
+blksz="$(_get_block_size ${SCRATCH_MNT})"
+_require_odirect $blksz
+
+# Reserve 1M space
+$XFS_IO_PROG -f -c "falloc 0 1M" "${SCRATCH_MNT}/tmp" >> $seqres.full
+
+# Create a file (~16T) with logical block numbers close to overflow
+$XFS_IO_PROG -f -c "falloc 0 10M" "${SCRATCH_MNT}/file" >> $seqres.full
+insert_size=$((blksz * 4096 - 10 - 3))
+$XFS_IO_PROG -f -c "finsert 1M ${insert_size}M" "${SCRATCH_MNT}/file" >> $seqres.full
+
+# Filling up the free space ensures that the pre-allocated space is the reserved space.
+nr_free=$(stat -f -c '%f' ${SCRATCH_MNT})
+_fill_fs $((nr_free * blksz)) ${SCRATCH_MNT}/fill $blksz 0 >> $seqres.full 2>&1
+sync
+
+# Remove reserved space to gain free space for allocation
+rm -f ${SCRATCH_MNT}/tmp
+
+# Trying to allocate two blocks triggers BUG_ON.
+$XFS_IO_PROG -c "open -ad ${SCRATCH_MNT}/file" -c "pwrite -S 0xff 0 8192" >> $seqres.full
+
+echo "Silence is golden"
+
+# success, all done
+status=0
+exit
diff --git a/tests/generic/737.out b/tests/generic/737.out
new file mode 100644
index 00000000..67b83d78
--- /dev/null
+++ b/tests/generic/737.out
@@ -0,0 +1,2 @@ 
+QA output created by 737
+Silence is golden