[PATCH v3 0/6] xfstests: add missing _require_scratch checks

[PATCH v3 0/6] xfstests: add missing _require_scratch checks

[Prabhakar Pujeri]

Fully consolidated resend of all remaining _require_scratch fixes, as
requested by Zorro on the v2 cover.

All of these patches add a single missing _require_scratch check; with
the exception of btrfs/333 (which places it at the top of
_require_btrfs_iouring_encoded_read(), per Qu Wenruo's earlier review)
they are plain one-line adds directly after the resource checks.  No
other changes.

v3:
- Merge ALL six single-line fixes into one series, per Zorro's request
  of 2026-09-03 on the v2 cover letter.
- Add Johannes Thumshirn's R-b for btrfs/273 and Chao Yu's R-b for
  f2fs/015,016 (both since v2).
- No code changes; patch contents are identical to the v2/v3 versions
  previously posted individually.

v2 pieces (this and the three it pulls forward) for reference:
- [PATCH v2 0/3] https://lore.kernel.org/fstests/20260903035344.752-1-prabhakar.pujeri@dell.com/
- btrfs/333 v3 https://lore.kernel.org/fstests/20260831104200.2269-1-prabhakar.pujeri@dell.com/ (v1; v3 was posted but is not archived)
- btrfs/273 v2 https://lore.kernel.org/fstests/20260902050145.2439-1-prabhakar.pujeri@dell.com/ (v1; v2 was posted but is not archived)
- overlay/081 v2 https://lore.kernel.org/fstests/20260901132438.759-1-prabhakar.pujeri@dell.com/

Reviewed-bys carried (per the individual threads):
- Darrick J. Wong: generic/590,735 and generic/795
- Zorro Lang + Chao Yu: f2fs/015,016
- Qu Wenruo: btrfs/333 (v3) and btrfs/273 (v2)
- Johannes Thumshirn: btrfs/273 (v2, second R-b)
- Amir Goldstein: overlay/081 (as Acked/Reviewed on v1; kept)

Prabhakar Pujeri (6):
  generic/590,735: add missing _require_scratch check
  generic/795: add missing _require_scratch check
  f2fs/015,016: add missing _require_scratch checks
  btrfs/333: require scratch before checking encoded reads
  btrfs/273: add missing _require_scratch check
  overlay/081: add missing _require_scratch check

 common/btrfs      | 1 +
 tests/btrfs/273   | 1 +
 tests/f2fs/015    | 1 +
 tests/f2fs/016    | 1 +
 tests/generic/590 | 1 +
 tests/generic/735 | 1 +
 tests/generic/795 | 2 ++
 tests/overlay/081 | 1 +
 8 files changed, 9 insertions(+)

Base: upstream master a370dcbed43563f0462801e889e0eceb93c7cfad
-- 
2.55.0

[PATCH v3 1/6] generic/590,735: add missing _require_scratch check

[Prabhakar Pujeri]

Both tests use the scratch device without first requiring one, so they
fail messily instead of skipping when no scratch device is configured.

Reviewed-by: "Darrick J. Wong" <djwong@kernel.org>
Signed-off-by: Prabhakar Pujeri <prabhakar.pujeri@dell.com>
---
 tests/generic/590 | 1 +
 tests/generic/735 | 1 +
 2 files changed, 2 insertions(+)

diff --git a/tests/generic/590 b/tests/generic/590
index 54c26f2a..0775d6ff 100755
--- a/tests/generic/590
+++ b/tests/generic/590
@@ -13,6 +13,7 @@ _begin_fstest auto prealloc preallocrw
 
 . ./common/filter
 
+_require_scratch
 _require_xfs_io_command "falloc"
 
 maxextlen=$((0x1fffff))
diff --git a/tests/generic/735 b/tests/generic/735
index 60fa3dce..d7e732bc 100755
--- a/tests/generic/735
+++ b/tests/generic/735
@@ -20,6 +20,7 @@ if [[ "$FSTYP" =~ ext[0-9]+ ]]; then
 fi
 
 _require_odirect
+_require_scratch
 _require_xfs_io_command "falloc"
 _require_xfs_io_command "finsert"
 _require_max_file_range_blocks $(( (1 << 32) - 1 ))
-- 
2.55.0

[PATCH v3 2/6] generic/795: add missing _require_scratch check

[Prabhakar Pujeri]

The test runs _scratch_mkfs and _scratch_mount without first requiring
a scratch device, so it fails messily instead of skipping when no
scratch device is configured.

Reviewed-by: "Darrick J. Wong" <djwong@kernel.org>
Signed-off-by: Prabhakar Pujeri <prabhakar.pujeri@dell.com>
---
 tests/generic/795 | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tests/generic/795 b/tests/generic/795
index fa5474c5..4c0e7bdb 100755
--- a/tests/generic/795
+++ b/tests/generic/795
@@ -27,6 +27,8 @@ _cleanup()
 
 . ./common/filter
 
+_require_scratch
+
 fsv_file_size=10000000
 nproc_enabler=$((4 * LOAD_FACTOR))
 nproc_reader=$((6 * LOAD_FACTOR))
-- 
2.55.0

[PATCH v3 3/6] f2fs/015,016: add missing _require_scratch checks

[Prabhakar Pujeri]

Both tests run _scratch_mkfs in a loop without first requiring a
scratch device, so they fail messily instead of skipping when no
scratch device is configured.

Reviewed-by: Zorro Lang <zlang@kernel.org>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Prabhakar Pujeri <prabhakar.pujeri@dell.com>
---
 tests/f2fs/015 | 1 +
 tests/f2fs/016 | 1 +
 2 files changed, 2 insertions(+)

diff --git a/tests/f2fs/015 b/tests/f2fs/015
index 54cada4a..75ae91f2 100755
--- a/tests/f2fs/015
+++ b/tests/f2fs/015
@@ -12,6 +12,7 @@ _begin_fstest auto quick mount
 _require_kernel_config CONFIG_F2FS_FS_XATTR
 _require_kernel_config CONFIG_F2FS_FS_POSIX_ACL
 _require_kernel_config CONFIG_F2FS_FAULT_INJECTION
+_require_scratch
 
 options=(
 #	mount_option			mkfs_option
diff --git a/tests/f2fs/016 b/tests/f2fs/016
index a6df827d..e0857e0a 100755
--- a/tests/f2fs/016
+++ b/tests/f2fs/016
@@ -15,6 +15,7 @@ _require_kernel_config CONFIG_F2FS_FS_LZORLE
 _require_kernel_config CONFIG_F2FS_FS_LZ4
 _require_kernel_config CONFIG_F2FS_FS_LZ4HC
 _require_kernel_config CONFIG_F2FS_FS_ZSTD
+_require_scratch
 
 options=(
 #	mount_option			mkfs_option
-- 
2.55.0

[PATCH v3 4/6] btrfs/333: require scratch before checking encoded reads

[Prabhakar Pujeri]

_require_btrfs_iouring_encoded_read() formats and mounts the scratch
device while checking for encoded-read support.  Without a configured
scratch device, callers therefore fail inside the helper before any
scratch requirement is checked.

Add _require_scratch at the top of the helper so that all of its
callers skip cleanly when no scratch device is present.

Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: Prabhakar Pujeri <prabhakar.pujeri@dell.com>
---
 common/btrfs | 1 +
 1 file changed, 1 insertion(+)

diff --git a/common/btrfs b/common/btrfs
index a75fbf1c..ccac8626 100644
--- a/common/btrfs
+++ b/common/btrfs
@@ -1055,6 +1055,7 @@ _require_btrfs_iouring_encoded_read()
 	local tmpfile
 	local ret
 
+	_require_scratch
 	_require_command src/btrfs_encoded_read
 
 	_scratch_mkfs &> /dev/null
-- 
2.55.0

[PATCH v3 5/6] btrfs/273: add missing _require_scratch check

[Prabhakar Pujeri]

The test passes $SCRATCH_DEV to _require_zoned_device() and mkfs/mount
without first requiring a scratch device, so it fails messily instead
of skipping when no scratch device is configured.

Reviewed-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Johannes Thumshirn <Johannes.Thumshirn@wdc.com>
Signed-off-by: Prabhakar Pujeri <prabhakar.pujeri@dell.com>
---
 tests/btrfs/273 | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tests/btrfs/273 b/tests/btrfs/273
index 60c3967b..197cfb02 100755
--- a/tests/btrfs/273
+++ b/tests/btrfs/273
@@ -28,6 +28,7 @@ _fixed_by_kernel_commit 2ce543f47843 \
 # which is further fixed by
 _fixed_by_kernel_commit d5b81ced74af \
 	"btrfs: zoned: fix API misuse of zone finish waiting"
+_require_scratch
 _require_zoned_device "$SCRATCH_DEV"
 _require_limited_active_zones "$SCRATCH_DEV"
 
-- 
2.55.0

[PATCH v3 6/6] overlay/081: add missing _require_scratch check

[Prabhakar Pujeri]

The test runs _scratch_mkfs without first requiring a scratch device, so it fails messily instead of skipping when no scratch device is configured.  Add _require_scratch so the test skips cleanly when no scratch device is present.

Signed-off-by: Prabhakar Pujeri <prabhakar.pujeri@dell.com>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
---
 tests/overlay/081 | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tests/overlay/081 b/tests/overlay/081
index 2270a047..e0ab558b 100755
--- a/tests/overlay/081
+++ b/tests/overlay/081
@@ -14,6 +14,7 @@ _begin_fstest auto quick
 . ./common/filter
 . ./common/attr
 
+_require_scratch
 
 _scratch_mkfs >>$seqres.full 2>&1
 
-- 
2.55.0

Re: [PATCH v3 2/6] generic/795: add missing _require_scratch check

[Zorro Lang]

On Tue, Sep 15, 2026 at 02:17:41PM +0000, Prabhakar Pujeri wrote:
> The test runs _scratch_mkfs and _scratch_mount without first requiring
> a scratch device, so it fails messily instead of skipping when no
> scratch device is configured.
> 
> Reviewed-by: "Darrick J. Wong" <djwong@kernel.org>
> Signed-off-by: Prabhakar Pujeri <prabhakar.pujeri@dell.com>
> ---

Thanks for consolidating the scattered patches! As mentioned in my previous
review comment, I think patches 1/6 and 2/6 can be squashed together at least.
Anyway, let me take care of squashing them upon merging. Overall, the patchset
looks good to me.

Reviewed-by: Zorro Lang <zlang@kernel.org>

>  tests/generic/795 | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/tests/generic/795 b/tests/generic/795
> index fa5474c5..4c0e7bdb 100755
> --- a/tests/generic/795
> +++ b/tests/generic/795
> @@ -27,6 +27,8 @@ _cleanup()
>  
>  . ./common/filter
>  
> +_require_scratch
> +
>  fsv_file_size=10000000
>  nproc_enabler=$((4 * LOAD_FACTOR))
>  nproc_reader=$((6 * LOAD_FACTOR))
> -- 
> 2.55.0
>