Skip to contentSkip to conversation

xfs: Check if rt summary/bitmap buffers are logged with correct xfs_buf type

PATCH Superseded Chandan Babu R 4 participants 7 messages Sep 15, 2020 – Sep 21, 2020

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 xfs: Check if rt summary/bitmap buffers are logged with correct xfs_buf type

This commit adds a test to check if growing a real-time device can end up logging an xfs_buf with the "type" subfield of bip->bli_formats->blf_flags set to XFS_BLFT_UNKNOWN_BUF. When this occurs the following call trace is printed on the console,

XFS: Assertion failed: (bip->bli_flags & XFS_BLI_STALE) || (xfs_blft_from_flags(&bip->__bli_format) > XFS_BLFT_UNKNOWN_BUF && xfs_blft_from_flags(&bip->__bli_format) < XFS_BLFT_MAX_BUF), file: fs/xfs/xfs_buf_item.c, line: 331

Call Trace:
 xfs_buf_item_format+0x632/0x680
 ? kmem_alloc_large+0x29/0x90
 ? kmem_alloc+0x70/0x120
 ? xfs_log_commit_cil+0x132/0x940
 xfs_log_commit_cil+0x26f/0x940
 ? xfs_buf_item_init+0x1ad/0x240
 ? xfs_growfs_rt_alloc+0x1fc/0x280
 __xfs_trans_commit+0xac/0x370
 xfs_growfs_rt_alloc+0x1fc/0x280
 xfs_growfs_rt+0x1a0/0x5e0
 xfs_file_ioctl+0x3fd/0xc70
 ? selinux_file_ioctl+0x174/0x220
 ksys_ioctl+0x87/0xc0
 __x64_sys_ioctl+0x16/0x20
 do_syscall_64+0x3e/0x70
 entry_SYSCALL_64_after_hwframe+0x44/0xa9

The kernel patch "xfs: Set xfs_buf type flag when growing summary/bitmap files" is required to fix this issue.

---
 tests/xfs/260     | 53 +++++++++++++++++++++++++++++++++++++++++++++++
 tests/xfs/260.out |  2 ++
 tests/xfs/group   |  1 +
 3 files changed, 56 insertions(+)
 create mode 100755 tests/xfs/260
 create mode 100644 tests/xfs/260.out
tests/xfs/260 new file +53 -0
new file mode 100755
index 00000000..078d4a11
@@ -0,0 +1,53 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2020 Chandan Babu R.  All Rights Reserved.
+#
+# FS QA Test 260
+#
+# Test to check if growing a real-time device can end up logging an xfs_buf with
+# the "type" subfield of bip->bli_formats->blf_flags set to
+# XFS_BLFT_UNKNOWN_BUF.
+#
+# This is a regression test for the kernel patch "xfs: Set xfs_buf type flag
+# when growing summary/bitmap files".
+
+seq=`basename $0`
+seqres=$RESULT_DIR/$seq
+echo "QA output created by $seq"
+
+here=`pwd`
+tmp=/tmp/$$
+status=1	# failure is the default!
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+_cleanup()
+{
+	cd /
+	rm -f $tmp.*
+}
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
+# remove previous $seqres.full before test
+rm -f $seqres.full
+
+# real QA test starts here
+_supported_fs xfs
+_supported_os Linux
+_require_realtime
+
+_scratch_mkfs -r size=10M  >> $seqres.full
+
+_scratch_mount >> $seqres.full
+
+$XFS_GROWFS_PROG $SCRATCH_MNT >> $seqres.full
+
+_scratch_unmount
+
+echo "Silence is golden"
+
+# success, all done
+status=0
+exit
tests/xfs/260.out new file +2 -0
new file mode 100644
index 00000000..18ca517c
@@ -0,0 +1,2 @@
+QA output created by 260
+Silence is golden
tests/xfs/group +1 -0
index ed0d389e..68676064 100644
@@ -257,6 +257,7 @@
 257 auto quick clone
 258 auto quick clone
 259 auto quick
+260 auto quick growfs realtime
 261 auto quick quota
 262 dangerous_fuzzers dangerous_scrub dangerous_online_repair
 263 auto quick quota
Signature · 1 line
2.28.0
Reviewed-by on v2 1/1
#1 lore raw
Quoted 108 lines · Chandan Babu R

This commit adds a test to check if growing a real-time device can end up logging an xfs_buf with the "type" subfield of bip->bli_formats->blf_flags set to XFS_BLFT_UNKNOWN_BUF. When this occurs the following call trace is printed on the console,

XFS: Assertion failed: (bip->bli_flags & XFS_BLI_STALE) || (xfs_blft_from_flags(&bip->__bli_format) > XFS_BLFT_UNKNOWN_BUF && xfs_blft_from_flags(&bip->__bli_format) < XFS_BLFT_MAX_BUF), file: fs/xfs/xfs_buf_item.c, line: 331

Call Trace:
 xfs_buf_item_format+0x632/0x680
 ? kmem_alloc_large+0x29/0x90
 ? kmem_alloc+0x70/0x120
 ? xfs_log_commit_cil+0x132/0x940
 xfs_log_commit_cil+0x26f/0x940
 ? xfs_buf_item_init+0x1ad/0x240
 ? xfs_growfs_rt_alloc+0x1fc/0x280
 __xfs_trans_commit+0xac/0x370
 xfs_growfs_rt_alloc+0x1fc/0x280
 xfs_growfs_rt+0x1a0/0x5e0
 xfs_file_ioctl+0x3fd/0xc70
 ? selinux_file_ioctl+0x174/0x220
 ksys_ioctl+0x87/0xc0
 __x64_sys_ioctl+0x16/0x20
 do_syscall_64+0x3e/0x70
 entry_SYSCALL_64_after_hwframe+0x44/0xa9

The kernel patch "xfs: Set xfs_buf type flag when growing summary/bitmap files" is required to fix this issue.

---
 tests/xfs/260     | 53 +++++++++++++++++++++++++++++++++++++++++++++++
 tests/xfs/260.out |  2 ++
 tests/xfs/group   |  1 +
 3 files changed, 56 insertions(+)
 create mode 100755 tests/xfs/260
 create mode 100644 tests/xfs/260.out
tests/xfs/260 new file +53 -0
new file mode 100755
index 00000000..078d4a11
@@ -0,0 +1,53 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2020 Chandan Babu R.  All Rights Reserved.
+#
+# FS QA Test 260
+#
+# Test to check if growing a real-time device can end up logging an xfs_buf with
+# the "type" subfield of bip->bli_formats->blf_flags set to
+# XFS_BLFT_UNKNOWN_BUF.
+#
+# This is a regression test for the kernel patch "xfs: Set xfs_buf type flag
+# when growing summary/bitmap files".
+
+seq=`basename $0`
+seqres=$RESULT_DIR/$seq
+echo "QA output created by $seq"
+
+here=`pwd`
+tmp=/tmp/$$
+status=1	# failure is the default!
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+_cleanup()
+{
+	cd /
+	rm -f $tmp.*
+}
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
+# remove previous $seqres.full before test
+rm -f $seqres.full
+
+# real QA test starts here
+_supported_fs xfs
+_supported_os Linux
+_require_realtime
+
+_scratch_mkfs -r size=10M  >> $seqres.full
+
+_scratch_mount >> $seqres.full
+
+$XFS_GROWFS_PROG $SCRATCH_MNT >> $seqres.full
+
+_scratch_unmount
+
+echo "Silence is golden"
+
+# success, all done
+status=0
+exit
tests/xfs/260.out new file +2 -0
new file mode 100644
index 00000000..18ca517c
@@ -0,0 +1,2 @@
+QA output created by 260
+Silence is golden
tests/xfs/group +0 -0
index ed0d389e..68676064 100644

tests/xfs/group line 260

Quoted 4 lines
tests/xfs/group +1 -0
@@ -257,6 +257,7 @@
 257 auto quick clone
 258 auto quick clone
 259 auto quick
+260 auto quick growfs realtime

There's not 'realtime' group before. Although I don't have a objection to add this group, if you'd like to add this one, my personal opinion is using another patch to add all 'rt' related cases into this group, or don't use this group name alone.

Besides this nitpicking, others looks good to me.

Thanks,
Zorro

Quoted 6 lines
tests/xfs/group +0 -0
 261 auto quick quota
 262 dangerous_fuzzers dangerous_scrub dangerous_online_repair
 263 auto quick quota
Signature · 1 line
2.28.0
#2 lore raw
Quoted 110 lines · IST Zorro Lang
Quoted 109 lines · Chandan Babu R

This commit adds a test to check if growing a real-time device can end up logging an xfs_buf with the "type" subfield of bip->bli_formats->blf_flags set to XFS_BLFT_UNKNOWN_BUF. When this occurs the following call trace is printed on the console,

XFS: Assertion failed: (bip->bli_flags & XFS_BLI_STALE) || (xfs_blft_from_flags(&bip->__bli_format) > XFS_BLFT_UNKNOWN_BUF && xfs_blft_from_flags(&bip->__bli_format) < XFS_BLFT_MAX_BUF), file: fs/xfs/xfs_buf_item.c, line: 331

Call Trace:
 xfs_buf_item_format+0x632/0x680
 ? kmem_alloc_large+0x29/0x90
 ? kmem_alloc+0x70/0x120
 ? xfs_log_commit_cil+0x132/0x940
 xfs_log_commit_cil+0x26f/0x940
 ? xfs_buf_item_init+0x1ad/0x240
 ? xfs_growfs_rt_alloc+0x1fc/0x280
 __xfs_trans_commit+0xac/0x370
 xfs_growfs_rt_alloc+0x1fc/0x280
 xfs_growfs_rt+0x1a0/0x5e0
 xfs_file_ioctl+0x3fd/0xc70
 ? selinux_file_ioctl+0x174/0x220
 ksys_ioctl+0x87/0xc0
 __x64_sys_ioctl+0x16/0x20
 do_syscall_64+0x3e/0x70
 entry_SYSCALL_64_after_hwframe+0x44/0xa9

The kernel patch "xfs: Set xfs_buf type flag when growing summary/bitmap files" is required to fix this issue.

---
 tests/xfs/260     | 53 +++++++++++++++++++++++++++++++++++++++++++++++
 tests/xfs/260.out |  2 ++
 tests/xfs/group   |  1 +
 3 files changed, 56 insertions(+)
 create mode 100755 tests/xfs/260
 create mode 100644 tests/xfs/260.out
tests/xfs/260 new file +53 -0
new file mode 100755
index 00000000..078d4a11
@@ -0,0 +1,53 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2020 Chandan Babu R.  All Rights Reserved.
+#
+# FS QA Test 260
+#
+# Test to check if growing a real-time device can end up logging an xfs_buf with
+# the "type" subfield of bip->bli_formats->blf_flags set to
+# XFS_BLFT_UNKNOWN_BUF.
+#
+# This is a regression test for the kernel patch "xfs: Set xfs_buf type flag
+# when growing summary/bitmap files".
+
+seq=`basename $0`
+seqres=$RESULT_DIR/$seq
+echo "QA output created by $seq"
+
+here=`pwd`
+tmp=/tmp/$$
+status=1	# failure is the default!
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+_cleanup()
+{
+	cd /
+	rm -f $tmp.*
+}
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
+# remove previous $seqres.full before test
+rm -f $seqres.full
+
+# real QA test starts here
+_supported_fs xfs
+_supported_os Linux
+_require_realtime
+
+_scratch_mkfs -r size=10M  >> $seqres.full
+
+_scratch_mount >> $seqres.full
+
+$XFS_GROWFS_PROG $SCRATCH_MNT >> $seqres.full
+
+_scratch_unmount
+
+echo "Silence is golden"
+
+# success, all done
+status=0
+exit
tests/xfs/260.out new file +2 -0
new file mode 100644
index 00000000..18ca517c
@@ -0,0 +1,2 @@
+QA output created by 260
+Silence is golden
tests/xfs/group +0 -0
index ed0d389e..68676064 100644
@@ -257,6 +257,7 @@
 257 auto quick clone
Quoted 8 lines

tests/xfs/group line 260

Quoted 3 lines
tests/xfs/group +1 -0
 258 auto quick clone
 259 auto quick
+260 auto quick growfs realtime

There's not 'realtime' group before. Although I don't have a objection to add this group, if you'd like to add this one, my personal opinion is using another patch to add all 'rt' related cases into this group, or don't use this group name alone.

Ok. I will write another patch to collect all realtime tests in one group. I will base the current patch on top of that.

Quoted 13 lines

Besides this nitpicking, others looks good to me.

Thanks,
Zorro

Quoted 3 lines
tests/xfs/group +0 -0
 261 auto quick quota
 262 dangerous_fuzzers dangerous_scrub dangerous_online_repair
 263 auto quick quota
Signature · 1 line
chandan
#3 lore raw
Quoted 120 lines · Chandan Babu R
Quoted 110 lines · IST Zorro Lang
Quoted 109 lines · Chandan Babu R

This commit adds a test to check if growing a real-time device can end up logging an xfs_buf with the "type" subfield of bip->bli_formats->blf_flags set to XFS_BLFT_UNKNOWN_BUF. When this occurs the following call trace is printed on the console,

XFS: Assertion failed: (bip->bli_flags & XFS_BLI_STALE) || (xfs_blft_from_flags(&bip->__bli_format) > XFS_BLFT_UNKNOWN_BUF && xfs_blft_from_flags(&bip->__bli_format) < XFS_BLFT_MAX_BUF), file: fs/xfs/xfs_buf_item.c, line: 331

Call Trace:
 xfs_buf_item_format+0x632/0x680
 ? kmem_alloc_large+0x29/0x90
 ? kmem_alloc+0x70/0x120
 ? xfs_log_commit_cil+0x132/0x940
 xfs_log_commit_cil+0x26f/0x940
 ? xfs_buf_item_init+0x1ad/0x240
 ? xfs_growfs_rt_alloc+0x1fc/0x280
 __xfs_trans_commit+0xac/0x370
 xfs_growfs_rt_alloc+0x1fc/0x280
 xfs_growfs_rt+0x1a0/0x5e0
 xfs_file_ioctl+0x3fd/0xc70
 ? selinux_file_ioctl+0x174/0x220
 ksys_ioctl+0x87/0xc0
 __x64_sys_ioctl+0x16/0x20
 do_syscall_64+0x3e/0x70
 entry_SYSCALL_64_after_hwframe+0x44/0xa9

The kernel patch "xfs: Set xfs_buf type flag when growing summary/bitmap files" is required to fix this issue.

---
 tests/xfs/260     | 53 +++++++++++++++++++++++++++++++++++++++++++++++
 tests/xfs/260.out |  2 ++
 tests/xfs/group   |  1 +
 3 files changed, 56 insertions(+)
 create mode 100755 tests/xfs/260
 create mode 100644 tests/xfs/260.out
tests/xfs/260 new file +53 -0
new file mode 100755
index 00000000..078d4a11
@@ -0,0 +1,53 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2020 Chandan Babu R.  All Rights Reserved.
+#
+# FS QA Test 260
+#
+# Test to check if growing a real-time device can end up logging an xfs_buf with
+# the "type" subfield of bip->bli_formats->blf_flags set to
+# XFS_BLFT_UNKNOWN_BUF.
+#
+# This is a regression test for the kernel patch "xfs: Set xfs_buf type flag
+# when growing summary/bitmap files".
+
+seq=`basename $0`
+seqres=$RESULT_DIR/$seq
+echo "QA output created by $seq"
+
+here=`pwd`
+tmp=/tmp/$$
+status=1	# failure is the default!
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+_cleanup()
+{
+	cd /
+	rm -f $tmp.*
+}
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
+# remove previous $seqres.full before test
+rm -f $seqres.full
+
+# real QA test starts here
+_supported_fs xfs
+_supported_os Linux
+_require_realtime
+
+_scratch_mkfs -r size=10M  >> $seqres.full
+
+_scratch_mount >> $seqres.full
+
+$XFS_GROWFS_PROG $SCRATCH_MNT >> $seqres.full
+
+_scratch_unmount
+
+echo "Silence is golden"
+
+# success, all done
+status=0
+exit
tests/xfs/260.out new file +2 -0
new file mode 100644
index 00000000..18ca517c
@@ -0,0 +1,2 @@
+QA output created by 260
+Silence is golden
tests/xfs/group +0 -0
index ed0d389e..68676064 100644
@@ -257,6 +257,7 @@
 257 auto quick clone
Quoted 8 lines

tests/xfs/group line 260

Quoted 3 lines
tests/xfs/group +1 -0
 258 auto quick clone
 259 auto quick
+260 auto quick growfs realtime

There's not 'realtime' group before. Although I don't have a objection to add this group, if you'd like to add this one, my personal opinion is using another patch to add all 'rt' related cases into this group, or don't use this group name alone.

Quoted 2 lines

Ok. I will write another patch to collect all realtime tests in one group. I will base the current patch on top of that.

Thank you. :)

I've started using the 'realtime' tag for a bunch of rt regression tests that I've started collecting in my dev tree, which is why I suggested it here.

--D

Quoted 19 lines
Quoted 13 lines

Besides this nitpicking, others looks good to me.

Thanks,
Zorro

Quoted 3 lines
tests/xfs/group +0 -0
 261 auto quick quota
 262 dangerous_fuzzers dangerous_scrub dangerous_online_repair
 263 auto quick quota
Signature · 1 line
chandan
Reviewed-by on v2 1/1
#4 lore raw
In reply to Chandan Babu R
Quoted 24 lines · Chandan Babu R

This commit adds a test to check if growing a real-time device can end up logging an xfs_buf with the "type" subfield of bip->bli_formats->blf_flags set to XFS_BLFT_UNKNOWN_BUF. When this occurs the following call trace is printed on the console,

XFS: Assertion failed: (bip->bli_flags & XFS_BLI_STALE) || (xfs_blft_from_flags(&bip->__bli_format) > XFS_BLFT_UNKNOWN_BUF && xfs_blft_from_flags(&bip->__bli_format) < XFS_BLFT_MAX_BUF), file: fs/xfs/xfs_buf_item.c, line: 331

Call Trace:
 xfs_buf_item_format+0x632/0x680
 ? kmem_alloc_large+0x29/0x90
 ? kmem_alloc+0x70/0x120
 ? xfs_log_commit_cil+0x132/0x940
 xfs_log_commit_cil+0x26f/0x940
 ? xfs_buf_item_init+0x1ad/0x240
 ? xfs_growfs_rt_alloc+0x1fc/0x280
 __xfs_trans_commit+0xac/0x370
 xfs_growfs_rt_alloc+0x1fc/0x280
 xfs_growfs_rt+0x1a0/0x5e0
 xfs_file_ioctl+0x3fd/0xc70
 ? selinux_file_ioctl+0x174/0x220
 ksys_ioctl+0x87/0xc0
 __x64_sys_ioctl+0x16/0x20
 do_syscall_64+0x3e/0x70
 entry_SYSCALL_64_after_hwframe+0x44/0xa9
Quoted 4 lines

The kernel patch "xfs: Set xfs_buf type flag when growing summary/bitmap files" is required to fix this issue.

Looks good to me

--D

Quoted 90 lines
---
 tests/xfs/260     | 53 +++++++++++++++++++++++++++++++++++++++++++++++
 tests/xfs/260.out |  2 ++
 tests/xfs/group   |  1 +
 3 files changed, 56 insertions(+)
 create mode 100755 tests/xfs/260
 create mode 100644 tests/xfs/260.out
tests/xfs/260 new file +53 -0
new file mode 100755
index 00000000..078d4a11
@@ -0,0 +1,53 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2020 Chandan Babu R.  All Rights Reserved.
+#
+# FS QA Test 260
+#
+# Test to check if growing a real-time device can end up logging an xfs_buf with
+# the "type" subfield of bip->bli_formats->blf_flags set to
+# XFS_BLFT_UNKNOWN_BUF.
+#
+# This is a regression test for the kernel patch "xfs: Set xfs_buf type flag
+# when growing summary/bitmap files".
+
+seq=`basename $0`
+seqres=$RESULT_DIR/$seq
+echo "QA output created by $seq"
+
+here=`pwd`
+tmp=/tmp/$$
+status=1	# failure is the default!
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+_cleanup()
+{
+	cd /
+	rm -f $tmp.*
+}
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
+# remove previous $seqres.full before test
+rm -f $seqres.full
+
+# real QA test starts here
+_supported_fs xfs
+_supported_os Linux
+_require_realtime
+
+_scratch_mkfs -r size=10M  >> $seqres.full
+
+_scratch_mount >> $seqres.full
+
+$XFS_GROWFS_PROG $SCRATCH_MNT >> $seqres.full
+
+_scratch_unmount
+
+echo "Silence is golden"
+
+# success, all done
+status=0
+exit
tests/xfs/260.out new file +2 -0
new file mode 100644
index 00000000..18ca517c
@@ -0,0 +1,2 @@
+QA output created by 260
+Silence is golden
tests/xfs/group +1 -0
index ed0d389e..68676064 100644
@@ -257,6 +257,7 @@
 257 auto quick clone
 258 auto quick clone
 259 auto quick
+260 auto quick growfs realtime
 261 auto quick quota
 262 dangerous_fuzzers dangerous_scrub dangerous_online_repair
 263 auto quick quota
Signature · 1 line
2.28.0
#5 lore raw
In reply to Chandan Babu R
Quoted 24 lines · Chandan Babu R

This commit adds a test to check if growing a real-time device can end up logging an xfs_buf with the "type" subfield of bip->bli_formats->blf_flags set to XFS_BLFT_UNKNOWN_BUF. When this occurs the following call trace is printed on the console,

XFS: Assertion failed: (bip->bli_flags & XFS_BLI_STALE) || (xfs_blft_from_flags(&bip->__bli_format) > XFS_BLFT_UNKNOWN_BUF && xfs_blft_from_flags(&bip->__bli_format) < XFS_BLFT_MAX_BUF), file: fs/xfs/xfs_buf_item.c, line: 331

Call Trace:
 xfs_buf_item_format+0x632/0x680
 ? kmem_alloc_large+0x29/0x90
 ? kmem_alloc+0x70/0x120
 ? xfs_log_commit_cil+0x132/0x940
 xfs_log_commit_cil+0x26f/0x940
 ? xfs_buf_item_init+0x1ad/0x240
 ? xfs_growfs_rt_alloc+0x1fc/0x280
 __xfs_trans_commit+0xac/0x370
 xfs_growfs_rt_alloc+0x1fc/0x280
 xfs_growfs_rt+0x1a0/0x5e0
 xfs_file_ioctl+0x3fd/0xc70
 ? selinux_file_ioctl+0x174/0x220
 ksys_ioctl+0x87/0xc0
 __x64_sys_ioctl+0x16/0x20
 do_syscall_64+0x3e/0x70
 entry_SYSCALL_64_after_hwframe+0x44/0xa9
Quoted 2 lines

The kernel patch "xfs: Set xfs_buf type flag when growing summary/bitmap files" is required to fix this issue.

Thanks for the patch! Also thanks to Darrick and Zorro for reviewing!

The test would crash kernel without above fix, so I'd merge it after the fix landing upstream.

Would you please remind me when the fix is merged by replying this thread? And perhaps with the correct commit ID updated :)

Quoted 53 lines
---
 tests/xfs/260     | 53 +++++++++++++++++++++++++++++++++++++++++++++++
 tests/xfs/260.out |  2 ++
 tests/xfs/group   |  1 +
 3 files changed, 56 insertions(+)
 create mode 100755 tests/xfs/260
 create mode 100644 tests/xfs/260.out
tests/xfs/260 new file +37 -0
new file mode 100755
index 00000000..078d4a11
@@ -0,0 +1,53 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2020 Chandan Babu R.  All Rights Reserved.
+#
+# FS QA Test 260
+#
+# Test to check if growing a real-time device can end up logging an xfs_buf with
+# the "type" subfield of bip->bli_formats->blf_flags set to
+# XFS_BLFT_UNKNOWN_BUF.
+#
+# This is a regression test for the kernel patch "xfs: Set xfs_buf type flag
+# when growing summary/bitmap files".
+
+seq=`basename $0`
+seqres=$RESULT_DIR/$seq
+echo "QA output created by $seq"
+
+here=`pwd`
+tmp=/tmp/$$
+status=1	# failure is the default!
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+_cleanup()
+{
+	cd /
+	rm -f $tmp.*
+}
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
+# remove previous $seqres.full before test
+rm -f $seqres.full
+
+# real QA test starts here
+_supported_fs xfs

tests/xfs/260 line 47

Quoted 10 lines
tests/xfs/260 new file +10 -0
+_supported_os Linux
+_require_realtime
+
+_scratch_mkfs -r size=10M  >> $seqres.full
+
+_scratch_mount >> $seqres.full
+
+$XFS_GROWFS_PROG $SCRATCH_MNT >> $seqres.full
+
+_scratch_unmount

Any reason to do umount manually here? The test harness will umount it after test anyway.

Thanks,
Eryu

Quoted 28 lines
tests/xfs/260 new file +6 -0
+
+echo "Silence is golden"
+
+# success, all done
+status=0
+exit
tests/xfs/260.out new file +2 -0
new file mode 100644
index 00000000..18ca517c
@@ -0,0 +1,2 @@
+QA output created by 260
+Silence is golden
tests/xfs/group +1 -0
index ed0d389e..68676064 100644
@@ -257,6 +257,7 @@
 257 auto quick clone
 258 auto quick clone
 259 auto quick
+260 auto quick growfs realtime
 261 auto quick quota
 262 dangerous_fuzzers dangerous_scrub dangerous_online_repair
 263 auto quick quota
Signature · 1 line
2.28.0
#6 lore raw
Quoted 28 lines · IST Eryu Guan
Quoted 24 lines · Chandan Babu R

This commit adds a test to check if growing a real-time device can end up logging an xfs_buf with the "type" subfield of bip->bli_formats->blf_flags set to XFS_BLFT_UNKNOWN_BUF. When this occurs the following call trace is printed on the console,

XFS: Assertion failed: (bip->bli_flags & XFS_BLI_STALE) || (xfs_blft_from_flags(&bip->__bli_format) > XFS_BLFT_UNKNOWN_BUF && xfs_blft_from_flags(&bip->__bli_format) < XFS_BLFT_MAX_BUF), file: fs/xfs/xfs_buf_item.c, line: 331

Call Trace:
 xfs_buf_item_format+0x632/0x680
 ? kmem_alloc_large+0x29/0x90
 ? kmem_alloc+0x70/0x120
 ? xfs_log_commit_cil+0x132/0x940
 xfs_log_commit_cil+0x26f/0x940
 ? xfs_buf_item_init+0x1ad/0x240
 ? xfs_growfs_rt_alloc+0x1fc/0x280
 __xfs_trans_commit+0xac/0x370
 xfs_growfs_rt_alloc+0x1fc/0x280
 xfs_growfs_rt+0x1a0/0x5e0
 xfs_file_ioctl+0x3fd/0xc70
 ? selinux_file_ioctl+0x174/0x220
 ksys_ioctl+0x87/0xc0
 __x64_sys_ioctl+0x16/0x20
 do_syscall_64+0x3e/0x70
 entry_SYSCALL_64_after_hwframe+0x44/0xa9
Quoted 2 lines

The kernel patch "xfs: Set xfs_buf type flag when growing summary/bitmap files" is required to fix this issue.

Quoted 7 lines

Thanks for the patch! Also thanks to Darrick and Zorro for reviewing!

The test would crash kernel without above fix, so I'd merge it after the fix landing upstream.

Would you please remind me when the fix is merged by replying this thread? And perhaps with the correct commit ID updated :)

Sure, I will keep an eye on when the patch gets merged with upstream kernel and will respond to the mail thread.

Quoted 59 lines
Quoted 59 lines
---
 tests/xfs/260     | 53 +++++++++++++++++++++++++++++++++++++++++++++++
 tests/xfs/260.out |  2 ++
 tests/xfs/group   |  1 +
 3 files changed, 56 insertions(+)
 create mode 100755 tests/xfs/260
 create mode 100644 tests/xfs/260.out
tests/xfs/260 new file +42 -0
new file mode 100755
index 00000000..078d4a11
@@ -0,0 +1,53 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2020 Chandan Babu R.  All Rights Reserved.
+#
+# FS QA Test 260
+#
+# Test to check if growing a real-time device can end up logging an xfs_buf with
+# the "type" subfield of bip->bli_formats->blf_flags set to
+# XFS_BLFT_UNKNOWN_BUF.
+#
+# This is a regression test for the kernel patch "xfs: Set xfs_buf type flag
+# when growing summary/bitmap files".
+
+seq=`basename $0`
+seqres=$RESULT_DIR/$seq
+echo "QA output created by $seq"
+
+here=`pwd`
+tmp=/tmp/$$
+status=1	# failure is the default!
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+_cleanup()
+{
+	cd /
+	rm -f $tmp.*
+}
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
+# remove previous $seqres.full before test
+rm -f $seqres.full
+
+# real QA test starts here
+_supported_fs xfs
+_supported_os Linux
+_require_realtime
+
+_scratch_mkfs -r size=10M  >> $seqres.full
+
Quoted 8 lines

tests/xfs/260 line 47

Quoted 5 lines
tests/xfs/260 new file +5 -0
+_scratch_mount >> $seqres.full
+
+$XFS_GROWFS_PROG $SCRATCH_MNT >> $seqres.full
+
+_scratch_unmount

Any reason to do umount manually here? The test harness will umount it after test anyway.

This was pointed out by Darrick as well. I have removed the invocation of _scratch_unmount() and posted V3 version of the patch
(https://www.spinics.net/lists/linux-xfs/msg45162.html).

Signature · 1 line
chandan