Re: [PATCH v2] generic/795: add unaligned boundary test cases for WRITE_ZEROES

Re: [PATCH v2] generic/795: add unaligned boundary test cases for WRITE_ZEROES

[Zhang Yi]

+CC ext4 list

On 7/9/2026 4:30 PM, Pankaj Raghav wrote:
> During the review of WRITE_ZEROES support to XFS, Zhang Yi pointed out
> some important semantics for the boundary blocks when WRITE_ZEROES
> command is used.[1]
> 
> The test cases check the boundary blocks when they are in different
> states and WRITE_ZEROES are issued that straddle the boundary. Along with that,
> test cases have been added when a EoF straddles an allocation unit.
> 
> [1] https://lore.kernel.org/linux-xfs/0b7f1a4f-da1c-4297-8099-98d738070ab7@huaweicloud.com/
> 
> Signed-off-by: Pankaj Raghav <p.raghav@samsung.com>

Thank you for the test! The changes look good to me, and it works well
with my ext4 fixes as well.

Reviewed-by: Zhang Yi <yi.zhang@huawei.com>

Unfortunately, I've found a bug in filefrag that causes delalloc extents
to be reported with a length of 0. This means this test case can miss
delalloc extents. The issue is in misc/filefrag.c's print_extent_info(),
where ext_len is incorrectly set to 0 for the FIEMAP_EXTENT_UNKNOWN or
EXT4_FIEMAP_EXTENT_HOLE cases. Could you please send a patch to fix this
as well?

Thanks,
Yi.

> ---
> CHanges since v1:
> - Use filefrag instead of FIEMAP as the latter uses FIEMAP_FLAG_SYNC by
>   default when used via xfs_io (Zhang yi)
> - Add an extra testcase for dirty-unwritten edges (Zhang Yi)
> 
>  common/rc             |   2 +-
>  tests/generic/795     | 187 ++++++++++++++++++++++++++++++++++++++++++
>  tests/generic/795.out |  19 +++++
>  3 files changed, 207 insertions(+), 1 deletion(-)
>  create mode 100755 tests/generic/795
>  create mode 100644 tests/generic/795.out
> 
> diff --git a/common/rc b/common/rc
> index 79189e7e..5e59968f 100644
> --- a/common/rc
> +++ b/common/rc
> @@ -3008,7 +3008,7 @@ _require_xfs_io_command()
>  		testio=`$XFS_IO_PROG -F -f -c "$command $param 0 1m" $testfile 2>&1`
>  		param_checked="$param"
>  		;;
> -	"fpunch" | "fcollapse" | "zero" | "fzero" | "finsert" | "funshare")
> +	"fpunch" | "fcollapse" | "zero" | "fzero" | "finsert" | "funshare" | "fwzero")
>  		local blocksize=$(_get_file_block_size $TEST_DIR)
>  		testio=`$XFS_IO_PROG -F -f -c "pwrite 0 $((5 * $blocksize))" \
>  			-c "fsync" -c "$command $blocksize $((2 * $blocksize))" \
> diff --git a/tests/generic/795 b/tests/generic/795
> new file mode 100755
> index 00000000..b1e05666
> --- /dev/null
> +++ b/tests/generic/795
> @@ -0,0 +1,187 @@
> +#! /bin/bash
> +# SPDX-License-Identifier: GPL-2.0
> +# Copyright (c) 2026 Samsung Electronics.  All Rights Reserved.
> +#
> +# FS QA Test 795
> +#
> +# Verify FALLOC_FL_WRITE_ZEROES (xfs_io "fwzero") on an unaligned range,
> +# exercising every state the two boundary allocation units can be in.
> +#
> +# WRITE_ZEROES must leave the whole requested range backed by *written*
> +# (zeroed) extents while preserving the out-of-range bytes of the partial
> +# boundary units. The scenarios are: written_edges, hole_edges,
> +# clean_unwritten_edges, delalloc_edges and dirty_unwritten_edges.
> +#
> +# It then covers the EOF cases: a WRITE_ZEROES that *extends* the file to a
> +# non-unit-aligned size, followed by a further extension. WRITE_ZEROES must
> +# never leave written blocks beyond EOF.
> +
> +. ./common/preamble
> +_begin_fstest auto quick prealloc fiemap
> +
> +testfile=$TEST_DIR/$seq.testfile
> +
> +_cleanup()
> +{
> +	cd /
> +	rm -r -f $tmp.*
> +	rm -f $testfile
> +}
> +
> +. ./common/filter
> +
> +_require_test
> +_require_xfs_io_command "fwzero"
> +_require_xfs_io_command "falloc"
> +_require_xfs_io_command "truncate"
> +_require_command "$FILEFRAG_PROG" filefrag
> +
> +# The allocation unit: rt extent size on an rt config, otherwise the block
> +# size. Deriving the range from this makes the head/tail units straddle rt
> +# extents when rextsize > 1.
> +u=$(_get_file_block_size $TEST_DIR)
> +
> +# The fs block size. On rt with rextsize > 1 the allocation unit u is the rt
> +# extent (u > b). A WRITE_ZEROES that extends EOF must only write out to the
> +# block-rounded EOF and leave the rest of the straddling rt extent unwritten,
> +# rather than written past EOF.
> +b=$(_get_block_size $TEST_DIR)
> +
> +# Unaligned range within EOF: start 1.5 units in, span 4 units, so both the
> +# head unit [u, 2u) and the tail unit [5u, 6u) are only partially covered,
> +# with fully covered units in between and data on both sides.
> +off=$((u + u / 2))
> +len=$((4 * u))
> +filesz=$((8 * u))
> +pattern=0xab
> +
> +# Expected images. For the written/delalloc/dirty-unwritten cases the pattern
> +# surrounds the zeroed range; for the hole/clean-unwritten cases the file reads
> +# back all zeroes.
> +$XFS_IO_PROG -f -c "pwrite -S $pattern 0 $filesz" $tmp.pattern >> $seqres.full 2>&1
> +$XFS_IO_PROG -c "pwrite -S 0 $off $len" $tmp.pattern >> $seqres.full 2>&1
> +$XFS_IO_PROG -f -c "truncate $filesz" $tmp.zero >> $seqres.full 2>&1
> +
> +# _filter_filefrag emits "phys#len#logical#flags" in bytes. filefrag has no
> +# range option, so clip the extents to the byte window [s, e) with awk before
> +# grepping - an extent [$3, $3+$2) overlaps the window iff $3 < e && $3+$2 > s.
> +_filefrag_window()
> +{
> +	local file=$1 start=$2 end=$3
> +
> +	$FILEFRAG_PROG -v "$file" | _filter_filefrag | \
> +		$AWK_PROG -F'#' -v s=$start -v e=$end '$3 < e && $3 + $2 > s'
> +}
> +
> +# Assert the whole requested range is backed by written extents immediately
> +# after fwzero (before any sync/remount): no unwritten, no delalloc.
> +# The extent state is queried with filefrag: xfs_io's fiemap always sets
> +# FIEMAP_FLAG_SYNC, which would flush and convert delalloc to written before
> +# reporting.
> +_check_range_written()
> +{
> +	local file=$1
> +	local bad
> +
> +	$FILEFRAG_PROG -v "$file" >> $seqres.full 2>&1
> +	bad=$(_filefrag_window "$file" $off $((off + len)) | \
> +		grep -E -c 'unwritten|delalloc|unknown')
> +	if [ "$bad" -eq 0 ]; then
> +		echo "extents: range fully written"
> +	else
> +		echo "extents: FAIL - $bad unwritten/delalloc extent(s) in range"
> +	fi
> +}
> +
> +_check_data()
> +{
> +	local file=$1
> +	local want=$2
> +
> +	if cmp -s "$file" "$want"; then
> +		echo "data: matches expected image"
> +	else
> +		echo "data: FAIL - mismatch against expected image"
> +		cmp "$file" "$want" | head
> +	fi
> +}
> +
> +# Assert the tail past the block-rounded EOF holds no *written* extents: it must
> +# be unwritten or a hole.
> +_check_eof_tail()
> +{
> +	local file=$1 start=$2 len=$3
> +	local bad
> +
> +	if [ "$len" -le 0 ]; then
> +		echo "extents: eof tail unwritten"
> +		return
> +	fi
> +	$FILEFRAG_PROG -v "$file" >> $seqres.full 2>&1
> +	bad=$(_filefrag_window "$file" $start $((start + len)) | \
> +		grep -E -v -c 'unwritten|delalloc|unknown')
> +	if [ "$bad" -eq 0 ]; then
> +		echo "extents: eof tail unwritten"
> +	else
> +		echo "extents: FAIL - $bad written extent(s) past block-rounded EOF"
> +	fi
> +}
> +
> +_run_case()
> +{
> +	local name=$1
> +	local want=$2
> +
> +	echo "=== $name ==="
> +
> +	$XFS_IO_PROG -c "fwzero $off $len" $testfile >> $seqres.full 2>&1
> +
> +	echo "== $name filefrag after fwzero ==" >> $seqres.full
> +	_check_range_written $testfile
> +	_test_cycle_mount
> +	_check_data $testfile $want
> +
> +	rm -f $testfile
> +}
> +
> +# 1) Boundary units already written, within i_size.
> +$XFS_IO_PROG -f -c "pwrite -S $pattern 0 $filesz" -c fsync $testfile \
> +	>> $seqres.full 2>&1
> +_run_case "written_edges" $tmp.pattern
> +
> +# 2) Boundary units are holes (sparse file).
> +$XFS_IO_PROG -f -c "truncate $filesz" $testfile >> $seqres.full 2>&1
> +_run_case "hole_edges" $tmp.zero
> +
> +# 3) Boundary units are clean unwritten preallocation: read back as zeroes.
> +$XFS_IO_PROG -f -c "falloc 0 $filesz" $testfile >> $seqres.full 2>&1
> +_run_case "clean_unwritten_edges" $tmp.zero
> +
> +# 4) Boundary units are dirty/delalloc.
> +$XFS_IO_PROG -f -c "pwrite -S $pattern 0 $filesz" $testfile >> $seqres.full 2>&1
> +_run_case "delalloc_edges" $tmp.pattern
> +
> +# 5) Boundary units are dirty unwritten.
> +$XFS_IO_PROG -f -c "falloc 0 $filesz" -c "pwrite -S $pattern 0 $filesz" \
> +	$testfile >> $seqres.full 2>&1
> +_run_case "dirty_unwritten_edges" $tmp.pattern
> +
> +# 6) EOF cases. A WRITE_ZEROES that extends the file to a non-unit-aligned
> +#    size, followed by a further extend.
> +eof=$((3 * u + 1))
> +
> +echo "=== eof_then_truncate ==="
> +$XFS_IO_PROG -f -c "fwzero 0 $eof" $testfile >> $seqres.full 2>&1
> +# The write must stop at the block-rounded EOF; the rest of the rt extent that
> +# straddles EOF must be unwritten, not written past EOF.
> +eof_ru=$(( (eof + b - 1) & ~(b - 1) ))
> +_check_eof_tail $testfile $eof_ru $((4 * u - eof_ru))
> +
> +$XFS_IO_PROG -c "truncate $filesz" $testfile >> $seqres.full 2>&1
> +_test_cycle_mount
> +_check_data $testfile $tmp.zero
> +rm -f $testfile
> +
> +# success, all done
> +status=0
> +exit
> diff --git a/tests/generic/795.out b/tests/generic/795.out
> new file mode 100644
> index 00000000..dca6e571
> --- /dev/null
> +++ b/tests/generic/795.out
> @@ -0,0 +1,19 @@
> +QA output created by 795
> +=== written_edges ===
> +extents: range fully written
> +data: matches expected image
> +=== hole_edges ===
> +extents: range fully written
> +data: matches expected image
> +=== clean_unwritten_edges ===
> +extents: range fully written
> +data: matches expected image
> +=== delalloc_edges ===
> +extents: range fully written
> +data: matches expected image
> +=== dirty_unwritten_edges ===
> +extents: range fully written
> +data: matches expected image
> +=== eof_then_truncate ===
> +extents: eof tail unwritten
> +data: matches expected image
> 
> base-commit: ffc8bad17e5b2f56e48dbac43f7c5ae8ac368fe5

Re: [PATCH v2] generic/795: add unaligned boundary test cases for WRITE_ZEROES

[Pankaj Raghav]

On 7/10/26 10:23, Zhang Yi wrote:
> +CC ext4 list
> 
> On 7/9/2026 4:30 PM, Pankaj Raghav wrote:
>> During the review of WRITE_ZEROES support to XFS, Zhang Yi pointed out
>> some important semantics for the boundary blocks when WRITE_ZEROES
>> command is used.[1]
>>
>> The test cases check the boundary blocks when they are in different
>> states and WRITE_ZEROES are issued that straddle the boundary. Along with that,
>> test cases have been added when a EoF straddles an allocation unit.
>>
>> [1] https://lore.kernel.org/linux-xfs/0b7f1a4f-da1c-4297-8099-98d738070ab7@huaweicloud.com/
>>
>> Signed-off-by: Pankaj Raghav <p.raghav@samsung.com>
> 
> Thank you for the test! The changes look good to me, and it works well
> with my ext4 fixes as well.
> 
> Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
> 

Thanks.

> Unfortunately, I've found a bug in filefrag that causes delalloc extents
> to be reported with a length of 0. This means this test case can miss
> delalloc extents. The issue is in misc/filefrag.c's print_extent_info(),
> where ext_len is incorrectly set to 0 for the FIEMAP_EXTENT_UNKNOWN or
> EXT4_FIEMAP_EXTENT_HOLE cases. Could you please send a patch to fix this
> as well?
>


I think I see it here:
	if ((fm_extent->fe_flags & FIEMAP_EXTENT_UNKNOWN) ||
	    (fm_extent->fe_flags & EXT4_FIEMAP_EXTENT_HOLE)) {
		ext_len = 0;
		ext_blks_phys = 0;
	} else
		ext_blks_phys = ext_blks;

I will send a fix soon as a separate patch.

--
Pankaj

Re: [PATCH v2] generic/795: add unaligned boundary test cases for WRITE_ZEROES

[Christoph Hellwig]

On Fri, Jul 10, 2026 at 04:23:27PM +0800, Zhang Yi wrote:
> Thank you for the test! The changes look good to me, and it works well
> with my ext4 fixes as well.
> 
> Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
> 
> Unfortunately, I've found a bug in filefrag that causes delalloc extents
> to be reported with a length of 0. This means this test case can miss
> delalloc extents. The issue is in misc/filefrag.c's print_extent_info(),
> where ext_len is incorrectly set to 0 for the FIEMAP_EXTENT_UNKNOWN or
> EXT4_FIEMAP_EXTENT_HOLE cases. Could you please send a patch to fix this
> as well?

We can't really rely on having a bleeding edge filefrag for xfstests
(well, at least without a feature check for it, but even that
would not a good usability story).  I guess at least for now
we might have to add  a little tool that just calls FIEMAP for
us to xfstests?

Re: [PATCH v2] generic/795: add unaligned boundary test cases for WRITE_ZEROES

[Pankaj Raghav]

>> Unfortunately, I've found a bug in filefrag that causes delalloc extents
>> to be reported with a length of 0. This means this test case can miss
>> delalloc extents. The issue is in misc/filefrag.c's print_extent_info(),
>> where ext_len is incorrectly set to 0 for the FIEMAP_EXTENT_UNKNOWN or
>> EXT4_FIEMAP_EXTENT_HOLE cases. Could you please send a patch to fix this
>> as well?
> 
> We can't really rely on having a bleeding edge filefrag for xfstests
> (well, at least without a feature check for it, but even that
> would not a good usability story).  I guess at least for now
> we might have to add  a little tool that just calls FIEMAP for
> us to xfstests?
> 

What about adding an option to io/filemap.c to take a flag that allows calling
fiemap without FIEMAP_FLAG_SYNC? So this test could depend on xfs_io fiemap command
with the flag support enabled? I can revert this test to the previous version which uses
xfs io fiemap instead of filefrag.

--
Pankaj

Re: [PATCH v2] generic/795: add unaligned boundary test cases for WRITE_ZEROES

[Christoph Hellwig]

On Mon, Jul 13, 2026 at 03:21:36PM +0200, Pankaj Raghav wrote:
> What about adding an option to io/filemap.c to take a flag that allows calling
> fiemap without FIEMAP_FLAG_SYNC? So this test could depend on xfs_io fiemap command
> with the flag support enabled? I can revert this test to the previous version which uses
> xfs io fiemap instead of filefrag.

That could work, and adding this flag would be useful in general.

It still has the issue that we require bleeding edge tools, but at least
feature detection for the flag would be easy.