[PATCH 05/10] common: add _filter_trailing_whitespace

[PATCH 05/10] common: add _filter_trailing_whitespace

[Anand Jain]

The command 'btrfs inspect-internal dump-tree -t raid_stripe'
introduces trailing whitespace in its output.
Apply a filter to remove it. Used in btrfs/30[4-8][.out].

Signed-off-by: Anand Jain <anand.jain@oracle.com>
---
 common/filter | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/common/filter b/common/filter
index 509ee95039ac..016d213b8bee 100644
--- a/common/filter
+++ b/common/filter
@@ -651,5 +651,10 @@ _filter_bash()
 	sed -e "s/^bash: line 1: /bash: /"
 }
 
+_filter_trailing_whitespace()
+{
+	sed -e "s/ $//"
+}
+
 # make sure this script returns success
 /bin/true
-- 
2.39.3

[PATCH v7 0/10] fstests: add tests for btrfs' raid-stripe-tree feature

[Anand Jain]

Changes in v7:
- Fixed trailing whitespace in the .out files
- Fixed the following test statement in 30[4-8]:
     test _get_page_size -eq 4096
- Link to v6: https://lore.kernel.org/r/20231213-btrfs-raid-v6-0-913738861069@wdc.com

--- original cover page from Johannes ----
Add tests for btrfs' raid-stripe-tree feature. All of these test work by
writing a specific pattern to a newly created filesystem and afterwards
using `btrfs inspect-internal -t raid-stripe $SCRATCH_DEV_POOL` to verify
the placement and the layout of the metadata.

The md5sum of each file will be compared as well after a re-mount of the
filesystem.

---
Changes in v6:
- require 4k pagesize for all tests as output depends on page size
- Add Filipe's Reviewed-by
- Link to v5: https://lore.kernel.org/r/20231207-btrfs-raid-v5-0-44aa1affe856@wdc.com

Changes in v5:
- add _require_btrfs_free_space_tree helper and use in tests
- Link to v4: https://lore.kernel.org/r/20231206-btrfs-raid-v4-0-578284dd3a70@wdc.com

Changes in v4:
- add _require_btrfs_no_compress to all tests
- add _require_btrfs_no_nodatacow helper and add to btrfs/308
- add _require_btrfs_feature "free_space_tree" to all tests
- Link to v3: https://lore.kernel.org/r/20231205-btrfs-raid-v3-0-0e857a5439a2@wdc.com

Changes in v3:
- added 'raid-stripe-tree' to mkfs options, as only zoned raid gets it
  automatically
- Rename test cases as btrfs/302 and btrfs/303 already exist upstream
- Link to v2: https://lore.kernel.org/r/20231205-btrfs-raid-v2-0-25f80eea345b@wdc.com

Changes in v2:
- Re-ordered series so the newly introduced group is added before the
  tests
- Changes Filipe requested to the tests.
- Link to v1: https://lore.kernel.org/r/20231204-btrfs-raid-v1-0-b254eb1bcff8@wdc.com

Anand Jain (1):
  common: add _filter_trailing_whitespace

Johannes Thumshirn (9):
  fstests: doc: add new raid-stripe-tree group
  common: add filter for btrfs raid-stripe dump
  common: add _require_btrfs_no_nodatacow helper
  common: add _require_btrfs_free_space_tree
  btrfs: add fstest for stripe-tree metadata with 4k write
  btrfs: add fstest for 8k write spanning two stripes on
    raid-stripe-tree
  btrfs: add fstest for writing to a file at an offset with RST
  btrfs: add fstests to write 128k to a RST filesystem
  btrfs: add fstest for overwriting a file partially with RST

 common/btrfs        |  17 +++++++
 common/filter       |   5 +++
 common/filter.btrfs |  14 ++++++
 doc/group-names.txt |   1 +
 tests/btrfs/304     |  59 ++++++++++++++++++++++++
 tests/btrfs/304.out |  58 ++++++++++++++++++++++++
 tests/btrfs/305     |  64 ++++++++++++++++++++++++++
 tests/btrfs/305.out |  82 ++++++++++++++++++++++++++++++++++
 tests/btrfs/306     |  62 ++++++++++++++++++++++++++
 tests/btrfs/306.out |  75 +++++++++++++++++++++++++++++++
 tests/btrfs/307     |  59 ++++++++++++++++++++++++
 tests/btrfs/307.out |  65 +++++++++++++++++++++++++++
 tests/btrfs/308     |  63 ++++++++++++++++++++++++++
 tests/btrfs/308.out | 106 ++++++++++++++++++++++++++++++++++++++++++++
 14 files changed, 730 insertions(+)
 create mode 100755 tests/btrfs/304
 create mode 100644 tests/btrfs/304.out
 create mode 100755 tests/btrfs/305
 create mode 100644 tests/btrfs/305.out
 create mode 100755 tests/btrfs/306
 create mode 100644 tests/btrfs/306.out
 create mode 100755 tests/btrfs/307
 create mode 100644 tests/btrfs/307.out
 create mode 100755 tests/btrfs/308
 create mode 100644 tests/btrfs/308.out

-- 
2.39.3

[PATCH v7 01/10] fstests: doc: add new raid-stripe-tree group

[Anand Jain]

From: Johannes Thumshirn <johannes.thumshirn@wdc.com>

Add a new test group for testing the raid-stripe-tree feature of btrfs
with fstests.

Reviewed-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Anand Jain <anand.jain@oracle.com>
---
 doc/group-names.txt | 1 +
 1 file changed, 1 insertion(+)

diff --git a/doc/group-names.txt b/doc/group-names.txt
index fec6bf71abcb..2ac95ac83a79 100644
--- a/doc/group-names.txt
+++ b/doc/group-names.txt
@@ -94,6 +94,7 @@ punch			fallocate FALLOC_FL_PUNCH_HOLE
 qgroup			btrfs qgroup feature
 quota			filesystem usage quotas
 raid			btrfs RAID
+raid-stripe-tree	btrfs raid-stripe-tree feature
 read_repair		btrfs error correction on read failure
 realtime		XFS realtime volumes
 recoveryloop		crash recovery loops
-- 
2.39.3

[PATCH v7 02/10] common: add filter for btrfs raid-stripe dump

[Anand Jain]

From: Johannes Thumshirn <johannes.thumshirn@wdc.com>

Reviewed-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Anand Jain <anand.jain@oracle.com>
---
 common/filter.btrfs | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/common/filter.btrfs b/common/filter.btrfs
index 8c6fe5793663..e570648bb5e6 100644
--- a/common/filter.btrfs
+++ b/common/filter.btrfs
@@ -126,5 +126,19 @@ _filter_btrfs_cloner_error()
 	sed -e "s/\(clone failed:\) Operation not supported/\1 Invalid argument/g"
 }
 
+# filter output of "btrfs inspect-internal dump-tree -t raid-stripe"
+_filter_stripe_tree()
+{
+	sed -E -e "s/leaf [0-9]+ items [0-9]+ free space [0-9]+ generation [0-9]+ owner RAID_STRIPE_TREE/leaf XXXXXXXXX items X free space XXXXX generation X owner RAID_STRIPE_TREE/" \
+		-e "s/leaf [0-9]+ flags 0x1\(WRITTEN\) backref revision 1/leaf XXXXXXXXX flags 0x1\(WRITTEN\) backref revision 1/" \
+		-e "s/checksum stored [0-9a-f]+/checksum stored <CHECKSUM>/"  \
+		-e "s/checksum calced [0-9a-f]+/checksum calced <CHECKSUM>/"  \
+		-e "s/[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}/<UUID>/" \
+		-e "s/item ([0-9]+) key \([0-9]+ RAID_STRIPE ([0-9]+)\) itemoff [0-9]+ itemsize ([0-9]+)/item \1 key \(XXXXXX RAID_STRIPE \2\) itemoff XXXXX itemsize \3/" \
+		-e "s/stripe ([0-9]+) devid ([0-9]+) physical [0-9]+/stripe \1 devid \2 physical XXXXXXXXX/" \
+		-e "s/total bytes [0-9]+/total bytes XXXXXXXX/" \
+		-e "s/bytes used [0-9]+/bytes used XXXXXX/"
+}
+
 # make sure this script returns success
 /bin/true
-- 
2.39.3

[PATCH v7 03/10] common: add _require_btrfs_no_nodatacow helper

[Anand Jain]

From: Johannes Thumshirn <johannes.thumshirn@wdc.com>

Reviewed-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Anand Jain <anand.jain@oracle.com>
---
 common/btrfs | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/common/btrfs b/common/btrfs
index f91f8dd869a1..9dd2a7f49e16 100644
--- a/common/btrfs
+++ b/common/btrfs
@@ -120,6 +120,13 @@ _require_btrfs_no_compress()
 	fi
 }
 
+_require_btrfs_no_nodatacow()
+{
+	if _normalize_mount_options "$MOUNT_OPTIONS" | grep -q "nodatacow"; then
+		_notrun "This test requires no nodatacow enabled"
+	fi
+}
+
 _check_btrfs_filesystem()
 {
 	device=$1
-- 
2.39.3

[PATCH v7 04/10] common: add _require_btrfs_free_space_tree

[Anand Jain]

From: Johannes Thumshirn <johannes.thumshirn@wdc.com>

Reviewed-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Anand Jain <anand.jain@oracle.com>
---
 common/btrfs | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/common/btrfs b/common/btrfs
index 9dd2a7f49e16..e1b29c613767 100644
--- a/common/btrfs
+++ b/common/btrfs
@@ -127,6 +127,16 @@ _require_btrfs_no_nodatacow()
 	fi
 }
 
+_require_btrfs_free_space_tree()
+{
+	_scratch_mkfs > /dev/null 2>&1
+	if ! $BTRFS_UTIL_PROG inspect-internal dump-super $SCRATCH_DEV | \
+		grep -q "FREE_SPACE_TREE"
+	then
+		_notrun "This test requires a free-space-tree"
+	fi
+}
+
 _check_btrfs_filesystem()
 {
 	device=$1
-- 
2.39.3

[PATCH v7 05/10] common: add _filter_trailing_whitespace

[Anand Jain]

The command 'btrfs inspect-internal dump-tree -t raid_stripe'
introduces trailing whitespace in its output.
Apply a filter to remove it. Used in btrfs/30[4-8][.out].

Signed-off-by: Anand Jain <anand.jain@oracle.com>
---
 common/filter | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/common/filter b/common/filter
index 509ee95039ac..016d213b8bee 100644
--- a/common/filter
+++ b/common/filter
@@ -651,5 +651,10 @@ _filter_bash()
 	sed -e "s/^bash: line 1: /bash: /"
 }
 
+_filter_trailing_whitespace()
+{
+	sed -e "s/ $//"
+}
+
 # make sure this script returns success
 /bin/true
-- 
2.39.3

[PATCH v7 06/10] btrfs: add fstest for stripe-tree metadata with 4k write

[Anand Jain]

From: Johannes Thumshirn <johannes.thumshirn@wdc.com>

Test a simple 4k write on all RAID profiles currently supported with the
raid-stripe-tree.

Reviewed-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Anand Jain <anand.jain@oracle.com>
[Fixed the test statement and trailing white space in the .out file.]
---
 tests/btrfs/304     | 59 +++++++++++++++++++++++++++++++++++++++++++++
 tests/btrfs/304.out | 58 ++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 117 insertions(+)
 create mode 100755 tests/btrfs/304
 create mode 100644 tests/btrfs/304.out

diff --git a/tests/btrfs/304 b/tests/btrfs/304
new file mode 100755
index 000000000000..186fa1646e48
--- /dev/null
+++ b/tests/btrfs/304
@@ -0,0 +1,59 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2023 Western Digital Cooperation.  All Rights Reserved.
+#
+# FS QA Test 304
+#
+# Test on-disk layout of RAID Stripe Tree Metadata writing 4k to a new file on
+# a pristine file system.
+#
+. ./common/preamble
+_begin_fstest auto quick raid remount volume raid-stripe-tree
+
+. ./common/filter
+. ./common/filter.btrfs
+
+_supported_fs btrfs
+_require_btrfs_command inspect-internal dump-tree
+_require_btrfs_mkfs_feature "raid-stripe-tree"
+_require_scratch_dev_pool 4
+_require_btrfs_fs_feature "raid_stripe_tree"
+_require_btrfs_fs_feature "free_space_tree"
+_require_btrfs_free_space_tree
+_require_btrfs_no_compress
+
+test $(_get_page_size) -eq 4096 || _notrun "this tests requires 4k pagesize"
+
+test_4k_write()
+{
+	local profile=$1
+	local ndevs=$2
+
+	_scratch_dev_pool_get $ndevs
+
+	echo "==== Testing $profile ===="
+	_scratch_pool_mkfs -d $profile -m $profile -O raid-stripe-tree
+	_scratch_mount
+
+	$XFS_IO_PROG -fc "pwrite 0 4k" "$SCRATCH_MNT/foo" | _filter_xfs_io
+
+	_scratch_cycle_mount
+	md5sum "$SCRATCH_MNT/foo" | _filter_scratch
+
+	_scratch_unmount
+
+	$BTRFS_UTIL_PROG inspect-internal dump-tree -t raid_stripe $SCRATCH_DEV_POOL |\
+		_filter_trailing_whitespace |\
+		_filter_btrfs_version |  _filter_stripe_tree
+
+	_scratch_dev_pool_put
+}
+
+echo "= Test basic 4k write ="
+test_4k_write raid0 2
+test_4k_write raid1 2
+test_4k_write raid10 4
+
+# success, all done
+status=0
+exit
diff --git a/tests/btrfs/304.out b/tests/btrfs/304.out
new file mode 100644
index 000000000000..39f56f32274d
--- /dev/null
+++ b/tests/btrfs/304.out
@@ -0,0 +1,58 @@
+QA output created by 304
+= Test basic 4k write =
+==== Testing raid0 ====
+wrote 4096/4096 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+5fed275e7617a806f94c173746a2a723  SCRATCH_MNT/foo
+
+raid stripe tree key (RAID_STRIPE_TREE ROOT_ITEM 0)
+leaf XXXXXXXXX items X free space XXXXX generation X owner RAID_STRIPE_TREE
+leaf XXXXXXXXX flags 0x1(WRITTEN) backref revision 1
+checksum stored <CHECKSUM>
+checksum calced <CHECKSUM>
+fs uuid <UUID>
+chunk uuid <UUID>
+	item 0 key (XXXXXX RAID_STRIPE 4096) itemoff XXXXX itemsize 24
+			encoding: RAID0
+			stripe 0 devid 1 physical XXXXXXXXX
+total bytes XXXXXXXX
+bytes used XXXXXX
+uuid <UUID>
+==== Testing raid1 ====
+wrote 4096/4096 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+5fed275e7617a806f94c173746a2a723  SCRATCH_MNT/foo
+
+raid stripe tree key (RAID_STRIPE_TREE ROOT_ITEM 0)
+leaf XXXXXXXXX items X free space XXXXX generation X owner RAID_STRIPE_TREE
+leaf XXXXXXXXX flags 0x1(WRITTEN) backref revision 1
+checksum stored <CHECKSUM>
+checksum calced <CHECKSUM>
+fs uuid <UUID>
+chunk uuid <UUID>
+	item 0 key (XXXXXX RAID_STRIPE 4096) itemoff XXXXX itemsize 40
+			encoding: RAID1
+			stripe 0 devid 1 physical XXXXXXXXX
+			stripe 1 devid 2 physical XXXXXXXXX
+total bytes XXXXXXXX
+bytes used XXXXXX
+uuid <UUID>
+==== Testing raid10 ====
+wrote 4096/4096 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+5fed275e7617a806f94c173746a2a723  SCRATCH_MNT/foo
+
+raid stripe tree key (RAID_STRIPE_TREE ROOT_ITEM 0)
+leaf XXXXXXXXX items X free space XXXXX generation X owner RAID_STRIPE_TREE
+leaf XXXXXXXXX flags 0x1(WRITTEN) backref revision 1
+checksum stored <CHECKSUM>
+checksum calced <CHECKSUM>
+fs uuid <UUID>
+chunk uuid <UUID>
+	item 0 key (XXXXXX RAID_STRIPE 4096) itemoff XXXXX itemsize 40
+			encoding: RAID10
+			stripe 0 devid 1 physical XXXXXXXXX
+			stripe 1 devid 2 physical XXXXXXXXX
+total bytes XXXXXXXX
+bytes used XXXXXX
+uuid <UUID>
-- 
2.39.3

[PATCH v7 07/10] btrfs: add fstest for 8k write spanning two stripes on raid-stripe-tree

[Anand Jain]

From: Johannes Thumshirn <johannes.thumshirn@wdc.com>

Add a test-case writing 8k to a raid-stripe-tree formatted filesystem with
one stripe pre-filled to 60k so the 8k are split into a 4k write finishing
stripe 1 and a 4k write starting the next stripe.

Reviewed-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Anand Jain <anand.jain@oracle.com>
[Fixed the test statement and trailing white space in the .out file.]
---
 tests/btrfs/305     | 64 +++++++++++++++++++++++++++++++++++
 tests/btrfs/305.out | 82 +++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 146 insertions(+)
 create mode 100755 tests/btrfs/305
 create mode 100644 tests/btrfs/305.out

diff --git a/tests/btrfs/305 b/tests/btrfs/305
new file mode 100755
index 000000000000..f3bf0faa414c
--- /dev/null
+++ b/tests/btrfs/305
@@ -0,0 +1,64 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2023 Western Digital Cooperation.  All Rights Reserved.
+#
+# FS QA Test 305
+#
+# Test on-disk layout of RAID Stripe Tree Metadata by writing 8k to a new file
+# with a filesystem prepropulated, so that 4k of the write are written to the
+# 1st stripe and 4k start a new stripe.
+#
+. ./common/preamble
+_begin_fstest auto quick raid remount volume raid-stripe-tree
+
+. ./common/filter
+. ./common/filter.btrfs
+
+_supported_fs btrfs
+_require_btrfs_command inspect-internal dump-tree
+_require_btrfs_mkfs_feature "raid-stripe-tree"
+_require_scratch_dev_pool 4
+_require_btrfs_fs_feature "raid_stripe_tree"
+_require_btrfs_fs_feature "free_space_tree"
+_require_btrfs_free_space_tree
+_require_btrfs_no_compress
+
+test $(_get_page_size) -eq 4096 || _notrun "this tests requires 4k pagesize"
+
+test_8k_new_stripe()
+{
+	local profile=$1
+	local ndevs=$2
+
+	_scratch_dev_pool_get $ndevs
+
+	echo "==== Testing $profile ===="
+	_scratch_pool_mkfs -d $profile -m $profile -O raid-stripe-tree
+	_scratch_mount
+
+	# Fill the first stripe up to 64k - 4k
+	$XFS_IO_PROG -fc "pwrite 0 60k" -c fsync "$SCRATCH_MNT/bar" | _filter_xfs_io
+
+	# The actual 8k write
+	$XFS_IO_PROG -fc "pwrite 0 8k" "$SCRATCH_MNT/foo" | _filter_xfs_io
+
+	_scratch_cycle_mount
+	md5sum "$SCRATCH_MNT/foo" | _filter_scratch
+
+	_scratch_unmount
+
+	$BTRFS_UTIL_PROG inspect-internal dump-tree -t raid_stripe $SCRATCH_DEV_POOL |\
+		_filter_trailing_whitespace |\
+		_filter_btrfs_version | _filter_stripe_tree
+
+	_scratch_dev_pool_put
+}
+
+echo "= Test 8k write to a new file so that 4k start a new stripe ="
+test_8k_new_stripe raid0 2
+test_8k_new_stripe raid1 2
+test_8k_new_stripe raid10 4
+
+# success, all done
+status=0
+exit
diff --git a/tests/btrfs/305.out b/tests/btrfs/305.out
new file mode 100644
index 000000000000..7090626c3036
--- /dev/null
+++ b/tests/btrfs/305.out
@@ -0,0 +1,82 @@
+QA output created by 305
+= Test 8k write to a new file so that 4k start a new stripe =
+==== Testing raid0 ====
+wrote 61440/61440 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 8192/8192 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+9d3940adb41dd525e008a847e01b15f4  SCRATCH_MNT/foo
+
+raid stripe tree key (RAID_STRIPE_TREE ROOT_ITEM 0)
+leaf XXXXXXXXX items X free space XXXXX generation X owner RAID_STRIPE_TREE
+leaf XXXXXXXXX flags 0x1(WRITTEN) backref revision 1
+checksum stored <CHECKSUM>
+checksum calced <CHECKSUM>
+fs uuid <UUID>
+chunk uuid <UUID>
+	item 0 key (XXXXXX RAID_STRIPE 61440) itemoff XXXXX itemsize 24
+			encoding: RAID0
+			stripe 0 devid 1 physical XXXXXXXXX
+	item 1 key (XXXXXX RAID_STRIPE 4096) itemoff XXXXX itemsize 24
+			encoding: RAID0
+			stripe 0 devid 1 physical XXXXXXXXX
+	item 2 key (XXXXXX RAID_STRIPE 4096) itemoff XXXXX itemsize 24
+			encoding: RAID0
+			stripe 0 devid 2 physical XXXXXXXXX
+total bytes XXXXXXXX
+bytes used XXXXXX
+uuid <UUID>
+==== Testing raid1 ====
+wrote 61440/61440 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 8192/8192 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+9d3940adb41dd525e008a847e01b15f4  SCRATCH_MNT/foo
+
+raid stripe tree key (RAID_STRIPE_TREE ROOT_ITEM 0)
+leaf XXXXXXXXX items X free space XXXXX generation X owner RAID_STRIPE_TREE
+leaf XXXXXXXXX flags 0x1(WRITTEN) backref revision 1
+checksum stored <CHECKSUM>
+checksum calced <CHECKSUM>
+fs uuid <UUID>
+chunk uuid <UUID>
+	item 0 key (XXXXXX RAID_STRIPE 61440) itemoff XXXXX itemsize 40
+			encoding: RAID1
+			stripe 0 devid 1 physical XXXXXXXXX
+			stripe 1 devid 2 physical XXXXXXXXX
+	item 1 key (XXXXXX RAID_STRIPE 8192) itemoff XXXXX itemsize 40
+			encoding: RAID1
+			stripe 0 devid 1 physical XXXXXXXXX
+			stripe 1 devid 2 physical XXXXXXXXX
+total bytes XXXXXXXX
+bytes used XXXXXX
+uuid <UUID>
+==== Testing raid10 ====
+wrote 61440/61440 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 8192/8192 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+9d3940adb41dd525e008a847e01b15f4  SCRATCH_MNT/foo
+
+raid stripe tree key (RAID_STRIPE_TREE ROOT_ITEM 0)
+leaf XXXXXXXXX items X free space XXXXX generation X owner RAID_STRIPE_TREE
+leaf XXXXXXXXX flags 0x1(WRITTEN) backref revision 1
+checksum stored <CHECKSUM>
+checksum calced <CHECKSUM>
+fs uuid <UUID>
+chunk uuid <UUID>
+	item 0 key (XXXXXX RAID_STRIPE 61440) itemoff XXXXX itemsize 40
+			encoding: RAID10
+			stripe 0 devid 1 physical XXXXXXXXX
+			stripe 1 devid 2 physical XXXXXXXXX
+	item 1 key (XXXXXX RAID_STRIPE 4096) itemoff XXXXX itemsize 40
+			encoding: RAID10
+			stripe 0 devid 1 physical XXXXXXXXX
+			stripe 1 devid 2 physical XXXXXXXXX
+	item 2 key (XXXXXX RAID_STRIPE 4096) itemoff XXXXX itemsize 40
+			encoding: RAID10
+			stripe 0 devid 3 physical XXXXXXXXX
+			stripe 1 devid 4 physical XXXXXXXXX
+total bytes XXXXXXXX
+bytes used XXXXXX
+uuid <UUID>
-- 
2.39.3

[PATCH v7 08/10] btrfs: add fstest for writing to a file at an offset with RST

[Anand Jain]

From: Johannes Thumshirn <johannes.thumshirn@wdc.com>

Add a fstest writing 4k at offset 64k to a file with one RAID tripe
already pre-filled for a raid-stripe-tree formatted file system.

Reviewed-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Anand Jain <anand.jain@oracle.com>
[Fixed the test statement and trailing white space in the .out file.]
---
 tests/btrfs/306     | 62 +++++++++++++++++++++++++++++++++++++
 tests/btrfs/306.out | 75 +++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 137 insertions(+)
 create mode 100755 tests/btrfs/306
 create mode 100644 tests/btrfs/306.out

diff --git a/tests/btrfs/306 b/tests/btrfs/306
new file mode 100755
index 000000000000..e2a9f804ac8b
--- /dev/null
+++ b/tests/btrfs/306
@@ -0,0 +1,62 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2023 Western Digital Cooperation.  All Rights Reserved.
+#
+# FS QA Test 306
+#
+# Test on-disk layout of RAID Stripe Tree Metadata by writing 4k to an emppty
+# file at offset 64k with one stripe pre-filled on an otherwise pristine
+# filesystem.
+#
+. ./common/preamble
+_begin_fstest auto quick raid remount volume raid-stripe-tree
+
+. ./common/filter
+. ./common/filter.btrfs
+
+_supported_fs btrfs
+_require_btrfs_command inspect-internal dump-tree
+_require_btrfs_mkfs_feature "raid-stripe-tree"
+_require_scratch_dev_pool 4
+_require_btrfs_fs_feature "raid_stripe_tree"
+_require_btrfs_fs_feature "free_space_tree"
+_require_btrfs_free_space_tree
+_require_btrfs_no_compress
+
+test $(_get_page_size) -eq 4096 || _notrun "this tests requires 4k pagesize"
+
+test_4k_write_64koff()
+{
+	local profile=$1
+	local ndevs=$2
+
+	_scratch_dev_pool_get $ndevs
+
+	echo "==== Testing $profile ===="
+	_scratch_pool_mkfs -d $profile -m $profile -O raid-stripe-tree
+	_scratch_mount
+
+	# precondition one stripe
+	$XFS_IO_PROG -fc "pwrite 0 64k" "$SCRATCH_MNT/bar" | _filter_xfs_io
+
+	$XFS_IO_PROG -fc "pwrite 64k 4k" "$SCRATCH_MNT/foo" | _filter_xfs_io
+
+	_scratch_cycle_mount
+	md5sum "$SCRATCH_MNT/foo" | _filter_scratch
+
+	_scratch_unmount
+
+	$BTRFS_UTIL_PROG inspect-internal dump-tree -t raid_stripe $SCRATCH_DEV_POOL |\
+		_filter_trailing_whitespace |\
+		_filter_btrfs_version | _filter_stripe_tree
+
+	_scratch_dev_pool_put
+}
+echo "= Test 4k write to an empty file at offset 64k with one stripe prefilled ="
+test_4k_write_64koff raid0 2
+test_4k_write_64koff raid1 2
+test_4k_write_64koff raid10 4
+
+# success, all done
+status=0
+exit
diff --git a/tests/btrfs/306.out b/tests/btrfs/306.out
new file mode 100644
index 000000000000..25065674c77b
--- /dev/null
+++ b/tests/btrfs/306.out
@@ -0,0 +1,75 @@
+QA output created by 306
+= Test 4k write to an empty file at offset 64k with one stripe prefilled =
+==== Testing raid0 ====
+wrote 65536/65536 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 4096/4096 bytes at offset 65536
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+381b0e7d72cb4f75286fe2b445e8d92a  SCRATCH_MNT/foo
+
+raid stripe tree key (RAID_STRIPE_TREE ROOT_ITEM 0)
+leaf XXXXXXXXX items X free space XXXXX generation X owner RAID_STRIPE_TREE
+leaf XXXXXXXXX flags 0x1(WRITTEN) backref revision 1
+checksum stored <CHECKSUM>
+checksum calced <CHECKSUM>
+fs uuid <UUID>
+chunk uuid <UUID>
+	item 0 key (XXXXXX RAID_STRIPE 65536) itemoff XXXXX itemsize 24
+			encoding: RAID0
+			stripe 0 devid 1 physical XXXXXXXXX
+	item 1 key (XXXXXX RAID_STRIPE 4096) itemoff XXXXX itemsize 24
+			encoding: RAID0
+			stripe 0 devid 2 physical XXXXXXXXX
+total bytes XXXXXXXX
+bytes used XXXXXX
+uuid <UUID>
+==== Testing raid1 ====
+wrote 65536/65536 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 4096/4096 bytes at offset 65536
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+381b0e7d72cb4f75286fe2b445e8d92a  SCRATCH_MNT/foo
+
+raid stripe tree key (RAID_STRIPE_TREE ROOT_ITEM 0)
+leaf XXXXXXXXX items X free space XXXXX generation X owner RAID_STRIPE_TREE
+leaf XXXXXXXXX flags 0x1(WRITTEN) backref revision 1
+checksum stored <CHECKSUM>
+checksum calced <CHECKSUM>
+fs uuid <UUID>
+chunk uuid <UUID>
+	item 0 key (XXXXXX RAID_STRIPE 65536) itemoff XXXXX itemsize 40
+			encoding: RAID1
+			stripe 0 devid 1 physical XXXXXXXXX
+			stripe 1 devid 2 physical XXXXXXXXX
+	item 1 key (XXXXXX RAID_STRIPE 4096) itemoff XXXXX itemsize 40
+			encoding: RAID1
+			stripe 0 devid 1 physical XXXXXXXXX
+			stripe 1 devid 2 physical XXXXXXXXX
+total bytes XXXXXXXX
+bytes used XXXXXX
+uuid <UUID>
+==== Testing raid10 ====
+wrote 65536/65536 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 4096/4096 bytes at offset 65536
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+381b0e7d72cb4f75286fe2b445e8d92a  SCRATCH_MNT/foo
+
+raid stripe tree key (RAID_STRIPE_TREE ROOT_ITEM 0)
+leaf XXXXXXXXX items X free space XXXXX generation X owner RAID_STRIPE_TREE
+leaf XXXXXXXXX flags 0x1(WRITTEN) backref revision 1
+checksum stored <CHECKSUM>
+checksum calced <CHECKSUM>
+fs uuid <UUID>
+chunk uuid <UUID>
+	item 0 key (XXXXXX RAID_STRIPE 65536) itemoff XXXXX itemsize 40
+			encoding: RAID10
+			stripe 0 devid 1 physical XXXXXXXXX
+			stripe 1 devid 2 physical XXXXXXXXX
+	item 1 key (XXXXXX RAID_STRIPE 4096) itemoff XXXXX itemsize 40
+			encoding: RAID10
+			stripe 0 devid 3 physical XXXXXXXXX
+			stripe 1 devid 4 physical XXXXXXXXX
+total bytes XXXXXXXX
+bytes used XXXXXX
+uuid <UUID>
-- 
2.39.3

[PATCH v7 09/10] btrfs: add fstests to write 128k to a RST filesystem

[Anand Jain]

From: Johannes Thumshirn <johannes.thumshirn@wdc.com>

Add a test writing 128k to a file on an empty filesystem formatted with a
raid-stripe-tree.

Reviewed-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Anand Jain <anand.jain@oracle.com>
[Fixed the test statement and trailing white space in the .out file.]
---
 tests/btrfs/307     | 59 ++++++++++++++++++++++++++++++++++++++++
 tests/btrfs/307.out | 65 +++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 124 insertions(+)
 create mode 100755 tests/btrfs/307
 create mode 100644 tests/btrfs/307.out

diff --git a/tests/btrfs/307 b/tests/btrfs/307
new file mode 100755
index 000000000000..30656bcf0d96
--- /dev/null
+++ b/tests/btrfs/307
@@ -0,0 +1,59 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2023 Western Digital Cooperation.  All Rights Reserved.
+#
+# FS QA Test 307
+#
+# Test on-disk layout of RAID Stripe Tree Metadata by writing 128k to a new
+# file on a pristine filesystem
+#
+. ./common/preamble
+_begin_fstest auto quick raid remount volume raid-stripe-tree
+
+. ./common/filter
+. ./common/filter.btrfs
+
+_supported_fs btrfs
+_require_btrfs_command inspect-internal dump-tree
+_require_btrfs_mkfs_feature "raid-stripe-tree"
+_require_scratch_dev_pool 4
+_require_btrfs_fs_feature "raid_stripe_tree"
+_require_btrfs_fs_feature "free_space_tree"
+_require_btrfs_free_space_tree
+_require_btrfs_no_compress
+
+test $(_get_page_size) -eq 4096 || _notrun "this tests requires 4k pagesize"
+
+test_128k_write()
+{
+	local profile=$1
+	local ndevs=$2
+
+	_scratch_dev_pool_get $ndevs
+
+	echo "==== Testing $profile ===="
+	_scratch_pool_mkfs -d $profile -m $profile -O raid-stripe-tree
+	_scratch_mount
+
+	$XFS_IO_PROG -fc "pwrite 0 128k" "$SCRATCH_MNT/foo" | _filter_xfs_io
+
+	_scratch_cycle_mount
+	md5sum "$SCRATCH_MNT/foo" | _filter_scratch
+
+	_scratch_unmount
+
+	$BTRFS_UTIL_PROG inspect-internal dump-tree -t raid_stripe $SCRATCH_DEV_POOL |\
+		_filter_trailing_whitespace |\
+		_filter_btrfs_version | _filter_stripe_tree
+
+	_scratch_dev_pool_put
+}
+
+echo "= Test 128k write to empty file  ="
+test_128k_write raid0 2
+test_128k_write raid1 2
+test_128k_write raid10 4
+
+# success, all done
+status=0
+exit
diff --git a/tests/btrfs/307.out b/tests/btrfs/307.out
new file mode 100644
index 000000000000..2815d17d7f03
--- /dev/null
+++ b/tests/btrfs/307.out
@@ -0,0 +1,65 @@
+QA output created by 307
+= Test 128k write to empty file  =
+==== Testing raid0 ====
+wrote 131072/131072 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+d48858312a922db7eb86377f638dbc9f  SCRATCH_MNT/foo
+
+raid stripe tree key (RAID_STRIPE_TREE ROOT_ITEM 0)
+leaf XXXXXXXXX items X free space XXXXX generation X owner RAID_STRIPE_TREE
+leaf XXXXXXXXX flags 0x1(WRITTEN) backref revision 1
+checksum stored <CHECKSUM>
+checksum calced <CHECKSUM>
+fs uuid <UUID>
+chunk uuid <UUID>
+	item 0 key (XXXXXX RAID_STRIPE 65536) itemoff XXXXX itemsize 24
+			encoding: RAID0
+			stripe 0 devid 1 physical XXXXXXXXX
+	item 1 key (XXXXXX RAID_STRIPE 65536) itemoff XXXXX itemsize 24
+			encoding: RAID0
+			stripe 0 devid 2 physical XXXXXXXXX
+total bytes XXXXXXXX
+bytes used XXXXXX
+uuid <UUID>
+==== Testing raid1 ====
+wrote 131072/131072 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+d48858312a922db7eb86377f638dbc9f  SCRATCH_MNT/foo
+
+raid stripe tree key (RAID_STRIPE_TREE ROOT_ITEM 0)
+leaf XXXXXXXXX items X free space XXXXX generation X owner RAID_STRIPE_TREE
+leaf XXXXXXXXX flags 0x1(WRITTEN) backref revision 1
+checksum stored <CHECKSUM>
+checksum calced <CHECKSUM>
+fs uuid <UUID>
+chunk uuid <UUID>
+	item 0 key (XXXXXX RAID_STRIPE 131072) itemoff XXXXX itemsize 40
+			encoding: RAID1
+			stripe 0 devid 1 physical XXXXXXXXX
+			stripe 1 devid 2 physical XXXXXXXXX
+total bytes XXXXXXXX
+bytes used XXXXXX
+uuid <UUID>
+==== Testing raid10 ====
+wrote 131072/131072 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+d48858312a922db7eb86377f638dbc9f  SCRATCH_MNT/foo
+
+raid stripe tree key (RAID_STRIPE_TREE ROOT_ITEM 0)
+leaf XXXXXXXXX items X free space XXXXX generation X owner RAID_STRIPE_TREE
+leaf XXXXXXXXX flags 0x1(WRITTEN) backref revision 1
+checksum stored <CHECKSUM>
+checksum calced <CHECKSUM>
+fs uuid <UUID>
+chunk uuid <UUID>
+	item 0 key (XXXXXX RAID_STRIPE 65536) itemoff XXXXX itemsize 40
+			encoding: RAID10
+			stripe 0 devid 1 physical XXXXXXXXX
+			stripe 1 devid 2 physical XXXXXXXXX
+	item 1 key (XXXXXX RAID_STRIPE 65536) itemoff XXXXX itemsize 40
+			encoding: RAID10
+			stripe 0 devid 3 physical XXXXXXXXX
+			stripe 1 devid 4 physical XXXXXXXXX
+total bytes XXXXXXXX
+bytes used XXXXXX
+uuid <UUID>
-- 
2.39.3

[PATCH v7 10/10] btrfs: add fstest for overwriting a file partially with RST

[Anand Jain]

From: Johannes Thumshirn <johannes.thumshirn@wdc.com>

Add a test writing 128k to an empty file with one stripe already
pre-filled on-disk. Then overwrite a portion of the file in the middle.

Reviewed-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: Anand Jain <anand.jain@oracle.com>
Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Anand Jain <anand.jain@oracle.com>
[Fixed the test statement and trailing white space in the .out file.]
---
 tests/btrfs/308     |  63 ++++++++++++++++++++++++++
 tests/btrfs/308.out | 106 ++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 169 insertions(+)
 create mode 100755 tests/btrfs/308
 create mode 100644 tests/btrfs/308.out

diff --git a/tests/btrfs/308 b/tests/btrfs/308
new file mode 100755
index 000000000000..7d651b2b4c9c
--- /dev/null
+++ b/tests/btrfs/308
@@ -0,0 +1,63 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2023 Western Digital Cooperation.  All Rights Reserved.
+#
+# FS QA Test 308
+#
+# Test on-disk layout of RAID Stripe Tree Metadata by writing 128k to an empty
+# file on a filesystem that has one stripe already pre-filled. Afterwards
+# overwrite a portion of the file.
+#
+. ./common/preamble
+_begin_fstest auto quick raid remount volume raid-stripe-tree
+
+. ./common/filter
+. ./common/filter.btrfs
+
+_supported_fs btrfs
+_require_btrfs_command inspect-internal dump-tree
+_require_btrfs_mkfs_feature "raid-stripe-tree"
+_require_scratch_dev_pool 4
+_require_btrfs_fs_feature "raid_stripe_tree"
+_require_btrfs_fs_feature "free_space_tree"
+_require_btrfs_free_space_tree
+_require_btrfs_no_compress
+_require_btrfs_no_nodatacow
+
+test $(_get_page_size) -eq 4096 || _notrun "this tests requires 4k pagesize"
+
+test_128k_write_overwrite()
+{
+	local profile=$1
+	local ndevs=$2
+
+	_scratch_dev_pool_get $ndevs
+
+	echo "==== Testing $profile ===="
+	_scratch_pool_mkfs -d $profile -m $profile -O raid-stripe-tree
+	_scratch_mount
+
+	$XFS_IO_PROG -fc "pwrite -W 0 32k" "$SCRATCH_MNT/bar" | _filter_xfs_io
+	$XFS_IO_PROG -fc "pwrite -W 0 128k" "$SCRATCH_MNT/foo" | _filter_xfs_io
+	$XFS_IO_PROG -fc "pwrite -W 64k 8k" "$SCRATCH_MNT/foo" | _filter_xfs_io
+
+	_scratch_cycle_mount
+	md5sum "$SCRATCH_MNT/foo" | _filter_scratch
+
+	_scratch_unmount
+
+	$BTRFS_UTIL_PROG inspect-internal dump-tree -t raid_stripe $SCRATCH_DEV_POOL |\
+		_filter_trailing_whitespace |\
+		_filter_btrfs_version | _filter_stripe_tree
+
+	_scratch_dev_pool_put
+}
+
+echo "= Test 128k write to empty file with 1st stripe partially prefilled then overwrite ="
+test_128k_write_overwrite raid0 2
+test_128k_write_overwrite raid1 2
+test_128k_write_overwrite raid10 4
+
+# success, all done
+status=0
+exit
diff --git a/tests/btrfs/308.out b/tests/btrfs/308.out
new file mode 100644
index 000000000000..23b31dd32959
--- /dev/null
+++ b/tests/btrfs/308.out
@@ -0,0 +1,106 @@
+QA output created by 308
+= Test 128k write to empty file with 1st stripe partially prefilled then overwrite =
+==== Testing raid0 ====
+wrote 32768/32768 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 131072/131072 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 8192/8192 bytes at offset 65536
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+d48858312a922db7eb86377f638dbc9f  SCRATCH_MNT/foo
+
+raid stripe tree key (RAID_STRIPE_TREE ROOT_ITEM 0)
+leaf XXXXXXXXX items X free space XXXXX generation X owner RAID_STRIPE_TREE
+leaf XXXXXXXXX flags 0x1(WRITTEN) backref revision 1
+checksum stored <CHECKSUM>
+checksum calced <CHECKSUM>
+fs uuid <UUID>
+chunk uuid <UUID>
+	item 0 key (XXXXXX RAID_STRIPE 32768) itemoff XXXXX itemsize 24
+			encoding: RAID0
+			stripe 0 devid 1 physical XXXXXXXXX
+	item 1 key (XXXXXX RAID_STRIPE 32768) itemoff XXXXX itemsize 24
+			encoding: RAID0
+			stripe 0 devid 1 physical XXXXXXXXX
+	item 2 key (XXXXXX RAID_STRIPE 65536) itemoff XXXXX itemsize 24
+			encoding: RAID0
+			stripe 0 devid 2 physical XXXXXXXXX
+	item 3 key (XXXXXX RAID_STRIPE 32768) itemoff XXXXX itemsize 24
+			encoding: RAID0
+			stripe 0 devid 1 physical XXXXXXXXX
+	item 4 key (XXXXXX RAID_STRIPE 8192) itemoff XXXXX itemsize 24
+			encoding: RAID0
+			stripe 0 devid 1 physical XXXXXXXXX
+total bytes XXXXXXXX
+bytes used XXXXXX
+uuid <UUID>
+==== Testing raid1 ====
+wrote 32768/32768 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 131072/131072 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 8192/8192 bytes at offset 65536
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+d48858312a922db7eb86377f638dbc9f  SCRATCH_MNT/foo
+
+raid stripe tree key (RAID_STRIPE_TREE ROOT_ITEM 0)
+leaf XXXXXXXXX items X free space XXXXX generation X owner RAID_STRIPE_TREE
+leaf XXXXXXXXX flags 0x1(WRITTEN) backref revision 1
+checksum stored <CHECKSUM>
+checksum calced <CHECKSUM>
+fs uuid <UUID>
+chunk uuid <UUID>
+	item 0 key (XXXXXX RAID_STRIPE 32768) itemoff XXXXX itemsize 40
+			encoding: RAID1
+			stripe 0 devid 1 physical XXXXXXXXX
+			stripe 1 devid 2 physical XXXXXXXXX
+	item 1 key (XXXXXX RAID_STRIPE 131072) itemoff XXXXX itemsize 40
+			encoding: RAID1
+			stripe 0 devid 1 physical XXXXXXXXX
+			stripe 1 devid 2 physical XXXXXXXXX
+	item 2 key (XXXXXX RAID_STRIPE 8192) itemoff XXXXX itemsize 40
+			encoding: RAID1
+			stripe 0 devid 1 physical XXXXXXXXX
+			stripe 1 devid 2 physical XXXXXXXXX
+total bytes XXXXXXXX
+bytes used XXXXXX
+uuid <UUID>
+==== Testing raid10 ====
+wrote 32768/32768 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 131072/131072 bytes at offset 0
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 8192/8192 bytes at offset 65536
+XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+d48858312a922db7eb86377f638dbc9f  SCRATCH_MNT/foo
+
+raid stripe tree key (RAID_STRIPE_TREE ROOT_ITEM 0)
+leaf XXXXXXXXX items X free space XXXXX generation X owner RAID_STRIPE_TREE
+leaf XXXXXXXXX flags 0x1(WRITTEN) backref revision 1
+checksum stored <CHECKSUM>
+checksum calced <CHECKSUM>
+fs uuid <UUID>
+chunk uuid <UUID>
+	item 0 key (XXXXXX RAID_STRIPE 32768) itemoff XXXXX itemsize 40
+			encoding: RAID10
+			stripe 0 devid 1 physical XXXXXXXXX
+			stripe 1 devid 2 physical XXXXXXXXX
+	item 1 key (XXXXXX RAID_STRIPE 32768) itemoff XXXXX itemsize 40
+			encoding: RAID10
+			stripe 0 devid 1 physical XXXXXXXXX
+			stripe 1 devid 2 physical XXXXXXXXX
+	item 2 key (XXXXXX RAID_STRIPE 65536) itemoff XXXXX itemsize 40
+			encoding: RAID10
+			stripe 0 devid 3 physical XXXXXXXXX
+			stripe 1 devid 4 physical XXXXXXXXX
+	item 3 key (XXXXXX RAID_STRIPE 32768) itemoff XXXXX itemsize 40
+			encoding: RAID10
+			stripe 0 devid 1 physical XXXXXXXXX
+			stripe 1 devid 2 physical XXXXXXXXX
+	item 4 key (XXXXXX RAID_STRIPE 8192) itemoff XXXXX itemsize 40
+			encoding: RAID10
+			stripe 0 devid 1 physical XXXXXXXXX
+			stripe 1 devid 2 physical XXXXXXXXX
+total bytes XXXXXXXX
+bytes used XXXXXX
+uuid <UUID>
-- 
2.39.3

Re: [PATCH 05/10] common: add _filter_trailing_whitespace

[Anand Jain]

pls ignore this sole patch.

Re: [PATCH v7 0/10] fstests: add tests for btrfs' raid-stripe-tree feature

[Anand Jain]

On 29/12/2023 17:52, Anand Jain wrote:
> Changes in v7:
> - Fixed trailing whitespace in the .out files
> - Fixed the following test statement in 30[4-8]:
>       test _get_page_size -eq 4096
> - Link to v6: https://lore.kernel.org/r/20231213-btrfs-raid-v6-0-913738861069@wdc.com
   - Rebased on Zorro's for-next
     Staged at  https://github.com/asj/fstests/tree/ext-rst


> 
> --- original cover page from Johannes ----
> Add tests for btrfs' raid-stripe-tree feature. All of these test work by
> writing a specific pattern to a newly created filesystem and afterwards
> using `btrfs inspect-internal -t raid-stripe $SCRATCH_DEV_POOL` to verify
> the placement and the layout of the metadata.
> 
> The md5sum of each file will be compared as well after a re-mount of the
> filesystem.
> 
> ---
> Changes in v6:
> - require 4k pagesize for all tests as output depends on page size
> - Add Filipe's Reviewed-by
> - Link to v5: https://lore.kernel.org/r/20231207-btrfs-raid-v5-0-44aa1affe856@wdc.com
> 
> Changes in v5:
> - add _require_btrfs_free_space_tree helper and use in tests
> - Link to v4: https://lore.kernel.org/r/20231206-btrfs-raid-v4-0-578284dd3a70@wdc.com
> 
> Changes in v4:
> - add _require_btrfs_no_compress to all tests
> - add _require_btrfs_no_nodatacow helper and add to btrfs/308
> - add _require_btrfs_feature "free_space_tree" to all tests
> - Link to v3: https://lore.kernel.org/r/20231205-btrfs-raid-v3-0-0e857a5439a2@wdc.com
> 
> Changes in v3:
> - added 'raid-stripe-tree' to mkfs options, as only zoned raid gets it
>    automatically
> - Rename test cases as btrfs/302 and btrfs/303 already exist upstream
> - Link to v2: https://lore.kernel.org/r/20231205-btrfs-raid-v2-0-25f80eea345b@wdc.com
> 
> Changes in v2:
> - Re-ordered series so the newly introduced group is added before the
>    tests
> - Changes Filipe requested to the tests.
> - Link to v1: https://lore.kernel.org/r/20231204-btrfs-raid-v1-0-b254eb1bcff8@wdc.com
> 
> Anand Jain (1):
>    common: add _filter_trailing_whitespace
> 
> Johannes Thumshirn (9):
>    fstests: doc: add new raid-stripe-tree group
>    common: add filter for btrfs raid-stripe dump
>    common: add _require_btrfs_no_nodatacow helper
>    common: add _require_btrfs_free_space_tree
>    btrfs: add fstest for stripe-tree metadata with 4k write
>    btrfs: add fstest for 8k write spanning two stripes on
>      raid-stripe-tree
>    btrfs: add fstest for writing to a file at an offset with RST
>    btrfs: add fstests to write 128k to a RST filesystem
>    btrfs: add fstest for overwriting a file partially with RST
> 
>   common/btrfs        |  17 +++++++
>   common/filter       |   5 +++
>   common/filter.btrfs |  14 ++++++
>   doc/group-names.txt |   1 +
>   tests/btrfs/304     |  59 ++++++++++++++++++++++++
>   tests/btrfs/304.out |  58 ++++++++++++++++++++++++
>   tests/btrfs/305     |  64 ++++++++++++++++++++++++++
>   tests/btrfs/305.out |  82 ++++++++++++++++++++++++++++++++++
>   tests/btrfs/306     |  62 ++++++++++++++++++++++++++
>   tests/btrfs/306.out |  75 +++++++++++++++++++++++++++++++
>   tests/btrfs/307     |  59 ++++++++++++++++++++++++
>   tests/btrfs/307.out |  65 +++++++++++++++++++++++++++
>   tests/btrfs/308     |  63 ++++++++++++++++++++++++++
>   tests/btrfs/308.out | 106 ++++++++++++++++++++++++++++++++++++++++++++
>   14 files changed, 730 insertions(+)
>   create mode 100755 tests/btrfs/304
>   create mode 100644 tests/btrfs/304.out
>   create mode 100755 tests/btrfs/305
>   create mode 100644 tests/btrfs/305.out
>   create mode 100755 tests/btrfs/306
>   create mode 100644 tests/btrfs/306.out
>   create mode 100755 tests/btrfs/307
>   create mode 100644 tests/btrfs/307.out
>   create mode 100755 tests/btrfs/308
>   create mode 100644 tests/btrfs/308.out
> 

Re: [PATCH 05/10] common: add _filter_trailing_whitespace

[Filipe Manana]

On Fri, Dec 29, 2023 at 12:30 PM Anand Jain <anand.jain@oracle.com> wrote:
>
> pls ignore this sole patch.
>
How can it be ignored if other patches in this patchset that introduce
tests use the filter introduced by this patch?

Thanks.

Re: [PATCH 05/10] common: add _filter_trailing_whitespace

[Filipe Manana]

On Fri, Dec 29, 2023 at 11:02 AM Anand Jain <anand.jain@oracle.com> wrote:
>
> The command 'btrfs inspect-internal dump-tree -t raid_stripe'
> introduces trailing whitespace in its output.
> Apply a filter to remove it. Used in btrfs/30[4-8][.out].
>
> Signed-off-by: Anand Jain <anand.jain@oracle.com>
> ---
>  common/filter | 5 +++++
>  1 file changed, 5 insertions(+)
>
> diff --git a/common/filter b/common/filter
> index 509ee95039ac..016d213b8bee 100644
> --- a/common/filter
> +++ b/common/filter
> @@ -651,5 +651,10 @@ _filter_bash()
>         sed -e "s/^bash: line 1: /bash: /"
>  }
>
> +_filter_trailing_whitespace()
> +{
> +       sed -e "s/ $//"
> +}

If we're having such a generic filter in common file, than I'd rather
have it delete any number of trailing white spaces, not just a single
one, and also account for tabs and other white spaces, so:

sed -e "s/\s+$//"

Also, since this is so specific to the raid stripe tree, I'd rather
have this filter included in the raid stripe tree filter introduced in
patch 2, _filter_stripe_tree(). That would make the tests shorter and
cleaner by avoiding piping yet over another filter that is used only
for the raid stripe tree dump...

Thanks.




> +
>  # make sure this script returns success
>  /bin/true
> --
> 2.39.3
>
>

Re: [PATCH 05/10] common: add _filter_trailing_whitespace

[Anand Jain]

On 12/29/23 20:57, Filipe Manana wrote:
> On Fri, Dec 29, 2023 at 11:02 AM Anand Jain <anand.jain@oracle.com> wrote:
>>
>> The command 'btrfs inspect-internal dump-tree -t raid_stripe'
>> introduces trailing whitespace in its output.
>> Apply a filter to remove it. Used in btrfs/30[4-8][.out].
>>
>> Signed-off-by: Anand Jain <anand.jain@oracle.com>
>> ---
>>   common/filter | 5 +++++
>>   1 file changed, 5 insertions(+)
>>
>> diff --git a/common/filter b/common/filter
>> index 509ee95039ac..016d213b8bee 100644
>> --- a/common/filter
>> +++ b/common/filter
>> @@ -651,5 +651,10 @@ _filter_bash()
>>          sed -e "s/^bash: line 1: /bash: /"
>>   }
>>
>> +_filter_trailing_whitespace()
>> +{
>> +       sed -e "s/ $//"
>> +}
> 
> If we're having such a generic filter in common file, than I'd rather
> have it delete any number of trailing white spaces, not just a single
> one, and also account for tabs and other white spaces, so:
> 
> sed -e "s/\s+$//"
> 

  I'll amend.


> Also, since this is so specific to the raid stripe tree, I'd rather
> have this filter included in the raid stripe tree filter introduced in
> patch 2, _filter_stripe_tree(). That would make the tests shorter and
> cleaner by avoiding piping yet over another filter that is used only
> for the raid stripe tree dump...

  I kept this as a separate function so that it can be used elsewhere
  when needed. Doesn't that make sense?

Thanks, Anand

Re: [PATCH 05/10] common: add _filter_trailing_whitespace

[Filipe Manana]

On Tue, Jan 2, 2024 at 8:02 AM Anand Jain <anand.jain@oracle.com> wrote:
>
>
>
> On 12/29/23 20:57, Filipe Manana wrote:
> > On Fri, Dec 29, 2023 at 11:02 AM Anand Jain <anand.jain@oracle.com> wrote:
> >>
> >> The command 'btrfs inspect-internal dump-tree -t raid_stripe'
> >> introduces trailing whitespace in its output.
> >> Apply a filter to remove it. Used in btrfs/30[4-8][.out].
> >>
> >> Signed-off-by: Anand Jain <anand.jain@oracle.com>
> >> ---
> >>   common/filter | 5 +++++
> >>   1 file changed, 5 insertions(+)
> >>
> >> diff --git a/common/filter b/common/filter
> >> index 509ee95039ac..016d213b8bee 100644
> >> --- a/common/filter
> >> +++ b/common/filter
> >> @@ -651,5 +651,10 @@ _filter_bash()
> >>          sed -e "s/^bash: line 1: /bash: /"
> >>   }
> >>
> >> +_filter_trailing_whitespace()
> >> +{
> >> +       sed -e "s/ $//"
> >> +}
> >
> > If we're having such a generic filter in common file, than I'd rather
> > have it delete any number of trailing white spaces, not just a single
> > one, and also account for tabs and other white spaces, so:
> >
> > sed -e "s/\s+$//"
> >
>
>   I'll amend.
>
>
> > Also, since this is so specific to the raid stripe tree, I'd rather
> > have this filter included in the raid stripe tree filter introduced in
> > patch 2, _filter_stripe_tree(). That would make the tests shorter and
> > cleaner by avoiding piping yet over another filter that is used only
> > for the raid stripe tree dump...
>
>   I kept this as a separate function so that it can be used elsewhere
>   when needed. Doesn't that make sense?

Not so much if there's only one use case for it... specially if it's
such a trivial filter...

Even if we had multiple cases, doing this pattern in the tests:

$BTRFS_UTIL_PROG inspect-internal dump-tree (... ) |
_filter_trailing_whitespace | _filter_btrfs_version |
_filter_stripe_tree

Is ugly and verbose. The filtering could be done in
_filter_stripe_tree() by calling "_filter_triling_whitespace" there...
And mentioning that, we could also call _filter_btrfs_version there,
since it's always wanted and to make tests shorter and easier to read.

So in the end it would only be

$BTRFS_UTIL_PROG inspect-internal dump-tree (... ) | _filter_stripe_tree

With _filter_stripe_tree() as:

_filter_stripe_tree()
{
    _filter_trailing_whitespace | _filter_btrfs_version | sed -E -e (....)
}

Or:

_filter_stripe_tree()
{
    _filter_btrfs_version | sed -E -e "s/\s+$//" -e (...)
}

A lot more clean.

Thanks.


>
> Thanks, Anand

Re: [PATCH 05/10] common: add _filter_trailing_whitespace

[Anand Jain]

>>> Also, since this is so specific to the raid stripe tree, I'd rather
>>> have this filter included in the raid stripe tree filter introduced in
>>> patch 2, _filter_stripe_tree(). That would make the tests shorter and
>>> cleaner by avoiding piping yet over another filter that is used only
>>> for the raid stripe tree dump...
>>
>>    I kept this as a separate function so that it can be used elsewhere
>>    when needed. Doesn't that make sense?
> 
> Not so much if there's only one use case for it... specially if it's
> such a trivial filter...
> 

  Hmm. Yes. Also, a completely avoidable coding nitpick in btrfs-progs.

> Even if we had multiple cases, doing this pattern in the tests:
> 
> $BTRFS_UTIL_PROG inspect-internal dump-tree (... ) |
> _filter_trailing_whitespace | _filter_btrfs_version |
> _filter_stripe_tree
> 
> Is ugly and verbose. The filtering could be done in
> _filter_stripe_tree() by calling "_filter_triling_whitespace" there...
> And mentioning that, we could also call _filter_btrfs_version there,
> since it's always wanted and to make tests shorter and easier to read.
> 

  Yeah.

> So in the end it would only be
> 
> $BTRFS_UTIL_PROG inspect-internal dump-tree (... ) | _filter_stripe_tree
> 
> With _filter_stripe_tree() as:
> 
> _filter_stripe_tree()
> {
>      _filter_trailing_whitespace | _filter_btrfs_version | sed -E -e (....)
> }
> 
> Or:
> 
> _filter_stripe_tree()
> {
>      _filter_btrfs_version | sed -E -e "s/\s+$//" -e (...)
> }
> 
> A lot more clean.
> 

  I'm fine with either way. Since there is a choice here, I will keep 
the former.

  I'll send a reroll.

Thanks, Anand