* [PATCH v2 0/3] btrfs: scrub: fix scrub_lock
@ 2018-11-29 9:25 Anand Jain
2018-11-29 9:25 ` [PATCH v2 1/3] btrfs: scrub: maintain the unlock order in scrub thread Anand Jain
` (2 more replies)
0 siblings, 3 replies; 6+ messages in thread
From: Anand Jain @ 2018-11-29 9:25 UTC (permalink / raw)
To: linux-btrfs
Idea was to fix the circular locking dependency warning as in patch 2/3,
and in the process also fixes the other identified cleanups patches 1/3,3/3
and they aren't dependent on 2ttch /3.
Anand Jain (3):
btrfs: scrub: maintain the unlock order in scrub thread
btrfs: scrub: fix circular locking dependency warning
btrfs: add lockdep check for scrub_lock in scrub_workers_get
fs/btrfs/scrub.c | 22 +++++++++++-----------
3 files changed, 13 insertions(+), 13 deletions(-)
--
1.8.3.1
^ permalink raw reply [flat|nested] 6+ messages in thread
* [PATCH v2 1/3] btrfs: scrub: maintain the unlock order in scrub thread
2018-11-29 9:25 [PATCH v2 0/3] btrfs: scrub: fix scrub_lock Anand Jain
@ 2018-11-29 9:25 ` Anand Jain
2018-11-29 10:36 ` Filipe Manana
2018-11-29 9:25 ` [PATCH v2 2/3] btrfs: scrub: fix circular locking dependency warning Anand Jain
2018-11-29 9:25 ` [PATCH v2 3/3] btrfs: scrub: add scrub_lock lockdep check in scrub_workers_get Anand Jain
2 siblings, 1 reply; 6+ messages in thread
From: Anand Jain @ 2018-11-29 9:25 UTC (permalink / raw)
To: linux-btrfs
The device_list_mutex and scrub_lock creates a nested locks in
btrfs_scrub_dev().
During lock the order is device_list_mutex and then scrub_lock, and during
unlock, the order is device_list_mutex and then scrub_lock.
Fix this to the lock order of scrub_lock and then device_list_mutex.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
---
v1->v2: change the order of lock acquire first scrub_lock and then
device_list_mutex, which matches with the order of unlock.
The extra line which are now in the scrub_lock are ok to be
under the scrub_lock.
fs/btrfs/scrub.c | 13 +++++++------
1 file changed, 7 insertions(+), 6 deletions(-)
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index 902819d3cf41..a9d6fc3b01d4 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -3813,28 +3813,29 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
return -EINVAL;
}
-
+ mutex_lock(&fs_info->scrub_lock);
mutex_lock(&fs_info->fs_devices->device_list_mutex);
dev = btrfs_find_device(fs_info, devid, NULL, NULL);
if (!dev || (test_bit(BTRFS_DEV_STATE_MISSING, &dev->dev_state) &&
!is_dev_replace)) {
mutex_unlock(&fs_info->fs_devices->device_list_mutex);
+ mutex_unlock(&fs_info->scrub_lock);
return -ENODEV;
}
if (!is_dev_replace && !readonly &&
!test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state)) {
mutex_unlock(&fs_info->fs_devices->device_list_mutex);
+ mutex_unlock(&fs_info->scrub_lock);
btrfs_err_in_rcu(fs_info, "scrub: device %s is not writable",
rcu_str_deref(dev->name));
return -EROFS;
}
- mutex_lock(&fs_info->scrub_lock);
if (!test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &dev->dev_state) ||
test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &dev->dev_state)) {
- mutex_unlock(&fs_info->scrub_lock);
mutex_unlock(&fs_info->fs_devices->device_list_mutex);
+ mutex_unlock(&fs_info->scrub_lock);
return -EIO;
}
@@ -3843,23 +3844,23 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
(!is_dev_replace &&
btrfs_dev_replace_is_ongoing(&fs_info->dev_replace))) {
btrfs_dev_replace_read_unlock(&fs_info->dev_replace);
- mutex_unlock(&fs_info->scrub_lock);
mutex_unlock(&fs_info->fs_devices->device_list_mutex);
+ mutex_unlock(&fs_info->scrub_lock);
return -EINPROGRESS;
}
btrfs_dev_replace_read_unlock(&fs_info->dev_replace);
ret = scrub_workers_get(fs_info, is_dev_replace);
if (ret) {
- mutex_unlock(&fs_info->scrub_lock);
mutex_unlock(&fs_info->fs_devices->device_list_mutex);
+ mutex_unlock(&fs_info->scrub_lock);
return ret;
}
sctx = scrub_setup_ctx(dev, is_dev_replace);
if (IS_ERR(sctx)) {
- mutex_unlock(&fs_info->scrub_lock);
mutex_unlock(&fs_info->fs_devices->device_list_mutex);
+ mutex_unlock(&fs_info->scrub_lock);
scrub_workers_put(fs_info);
return PTR_ERR(sctx);
}
--
1.8.3.1
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH v2 2/3] btrfs: scrub: fix circular locking dependency warning
2018-11-29 9:25 [PATCH v2 0/3] btrfs: scrub: fix scrub_lock Anand Jain
2018-11-29 9:25 ` [PATCH v2 1/3] btrfs: scrub: maintain the unlock order in scrub thread Anand Jain
@ 2018-11-29 9:25 ` Anand Jain
2018-11-29 9:25 ` [PATCH v2 3/3] btrfs: scrub: add scrub_lock lockdep check in scrub_workers_get Anand Jain
2 siblings, 0 replies; 6+ messages in thread
From: Anand Jain @ 2018-11-29 9:25 UTC (permalink / raw)
To: linux-btrfs
Circular locking dependency check reports warning[1], that's because
the btrfs_scrub_dev() calls the stack #0 below with, the
fs_info::scrub_lock held. The test case leading to this warning..
mkfs.btrfs -fq /dev/sdb && mount /dev/sdb /btrfs
btrfs scrub start -B /btrfs
In fact we have fs_info::scrub_workers_refcnt to tack if the init and
destroy of the scrub workers are needed. So once we have incremented
and decremented the fs_info::scrub_workers_refcnt value in the thread,
its ok to drop the scrub_lock, and then actually do the
btrfs_destroy_workqueue() part. So this patch drops the scrub_lock
before calling btrfs_destroy_workqueue().
[1]
[ 76.146826] ======================================================
[ 76.147086] WARNING: possible circular locking dependency detected
[ 76.147316] 4.20.0-rc3+ #41 Not tainted
[ 76.147489] ------------------------------------------------------
[ 76.147722] btrfs/4065 is trying to acquire lock:
[ 76.147984] 0000000038593bc0 ((wq_completion)"%s-%s""btrfs",
name){+.+.}, at: flush_workqueue+0x70/0x4d0
[ 76.148337]
but task is already holding lock:
[ 76.148594] 0000000062392ab7 (&fs_info->scrub_lock){+.+.}, at:
btrfs_scrub_dev+0x316/0x5d0 [btrfs]
[ 76.148909]
which lock already depends on the new lock.
[ 76.149191]
the existing dependency chain (in reverse order) is:
[ 76.149446]
-> #3 (&fs_info->scrub_lock){+.+.}:
[ 76.149707] btrfs_scrub_dev+0x11f/0x5d0 [btrfs]
[ 76.149924] btrfs_ioctl+0x1ac3/0x2d80 [btrfs]
[ 76.150216] do_vfs_ioctl+0xa9/0x6d0
[ 76.150468] ksys_ioctl+0x60/0x90
[ 76.150716] __x64_sys_ioctl+0x16/0x20
[ 76.150911] do_syscall_64+0x50/0x180
[ 76.151182] entry_SYSCALL_64_after_hwframe+0x49/0xbe
[ 76.151469]
-> #2 (&fs_devs->device_list_mutex){+.+.}:
[ 76.151851] reada_start_machine_worker+0xca/0x3f0 [btrfs]
[ 76.152195] normal_work_helper+0xf0/0x4c0 [btrfs]
[ 76.152489] process_one_work+0x1f4/0x520
[ 76.152751] worker_thread+0x46/0x3d0
[ 76.153715] kthread+0xf8/0x130
[ 76.153912] ret_from_fork+0x3a/0x50
[ 76.154178]
-> #1 ((work_completion)(&work->normal_work)){+.+.}:
[ 76.154575] worker_thread+0x46/0x3d0
[ 76.154828] kthread+0xf8/0x130
[ 76.155108] ret_from_fork+0x3a/0x50
[ 76.155357]
-> #0 ((wq_completion)"%s-%s""btrfs", name){+.+.}:
[ 76.155751] flush_workqueue+0x9a/0x4d0
[ 76.155911] drain_workqueue+0xca/0x1a0
[ 76.156182] destroy_workqueue+0x17/0x230
[ 76.156455] btrfs_destroy_workqueue+0x5d/0x1c0 [btrfs]
[ 76.156756] scrub_workers_put+0x2e/0x60 [btrfs]
[ 76.156931] btrfs_scrub_dev+0x329/0x5d0 [btrfs]
[ 76.157219] btrfs_ioctl+0x1ac3/0x2d80 [btrfs]
[ 76.157491] do_vfs_ioctl+0xa9/0x6d0
[ 76.157742] ksys_ioctl+0x60/0x90
[ 76.157910] __x64_sys_ioctl+0x16/0x20
[ 76.158177] do_syscall_64+0x50/0x180
[ 76.158429] entry_SYSCALL_64_after_hwframe+0x49/0xbe
[ 76.158716]
other info that might help us debug this:
[ 76.158908] Chain exists of:
(wq_completion)"%s-%s""btrfs", name --> &fs_devs->device_list_mutex
--> &fs_info->scrub_lock
[ 76.159629] Possible unsafe locking scenario:
[ 76.160607] CPU0 CPU1
[ 76.160934] ---- ----
[ 76.161210] lock(&fs_info->scrub_lock);
[ 76.161458]
lock(&fs_devs->device_list_mutex);
[ 76.161805]
lock(&fs_info->scrub_lock);
[ 76.161909] lock((wq_completion)"%s-%s""btrfs", name);
[ 76.162201]
*** DEADLOCK ***
[ 76.162627] 2 locks held by btrfs/4065:
[ 76.162897] #0: 00000000bef2775b (sb_writers#12){.+.+}, at:
mnt_want_write_file+0x24/0x50
[ 76.163335] #1: 0000000062392ab7 (&fs_info->scrub_lock){+.+.}, at:
btrfs_scrub_dev+0x316/0x5d0 [btrfs]
[ 76.163796]
stack backtrace:
[ 76.163911] CPU: 1 PID: 4065 Comm: btrfs Not tainted 4.20.0-rc3+ #41
[ 76.164228] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS
VirtualBox 12/01/2006
[ 76.164646] Call Trace:
[ 76.164872] dump_stack+0x5e/0x8b
[ 76.165128] print_circular_bug.isra.37+0x1f1/0x1fe
[ 76.165398] __lock_acquire+0x14aa/0x1620
[ 76.165652] lock_acquire+0xb0/0x190
[ 76.165910] ? flush_workqueue+0x70/0x4d0
[ 76.166175] flush_workqueue+0x9a/0x4d0
[ 76.166420] ? flush_workqueue+0x70/0x4d0
[ 76.166671] ? drain_workqueue+0x52/0x1a0
[ 76.166911] drain_workqueue+0xca/0x1a0
[ 76.167167] destroy_workqueue+0x17/0x230
[ 76.167428] btrfs_destroy_workqueue+0x5d/0x1c0 [btrfs]
[ 76.167720] scrub_workers_put+0x2e/0x60 [btrfs]
[ 76.168233] btrfs_scrub_dev+0x329/0x5d0 [btrfs]
[ 76.168504] ? __sb_start_write+0x121/0x1b0
[ 76.168759] ? mnt_want_write_file+0x24/0x50
[ 76.169654] btrfs_ioctl+0x1ac3/0x2d80 [btrfs]
[ 76.169934] ? find_held_lock+0x2d/0x90
[ 76.170204] ? find_held_lock+0x2d/0x90
[ 76.170450] do_vfs_ioctl+0xa9/0x6d0
[ 76.170690] ? __fget+0x101/0x1f0
[ 76.170910] ? __fget+0x5/0x1f0
[ 76.171157] ksys_ioctl+0x60/0x90
[ 76.171391] __x64_sys_ioctl+0x16/0x20
[ 76.171634] do_syscall_64+0x50/0x180
[ 76.171892] entry_SYSCALL_64_after_hwframe+0x49/0xbe
[ 76.172186] RIP: 0033:0x7f61d422e567
[ 76.172425] Code: 44 00 00 48 8b 05 29 09 2d 00 64 c7 00 26 00 00 00
48 c7 c0 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 b8 10 00 00 00 0f
05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d f9 08 2d 00 f7 d8 64 89 01 48
[ 76.172911] RSP: 002b:00007f61d3936d68 EFLAGS: 00000246 ORIG_RAX:
0000000000000010
[ 76.173328] RAX: ffffffffffffffda RBX: 00000000019026b0 RCX:
00007f61d422e567
[ 76.173649] RDX: 00000000019026b0 RSI: 00000000c400941b RDI:
0000000000000003
[ 76.173909] RBP: 0000000000000000 R08: 00007f61d3937700 R09:
0000000000000000
[ 76.174244] R10: 00007f61d3937700 R11: 0000000000000246 R12:
0000000000000000
[ 76.174566] R13: 0000000000801000 R14: 0000000000000000 R15:
00007f61d3937700
[ 76.175217] btrfs (4065) used greatest stack depth: 11424 bytes left
Signed-off-by: Anand Jain <anand.jain@oracle.com>
---
v1->v2: none
fs/btrfs/scrub.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index b5a19ba38ab7..9ade0659f017 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -3757,10 +3757,13 @@ static noinline_for_stack int scrub_workers_get(struct btrfs_fs_info *fs_info,
static noinline_for_stack void scrub_workers_put(struct btrfs_fs_info *fs_info)
{
+ lockdep_assert_held(&fs_info->scrub_lock);
if (--fs_info->scrub_workers_refcnt == 0) {
+ mutex_unlock(&fs_info->scrub_lock);
btrfs_destroy_workqueue(fs_info->scrub_workers);
btrfs_destroy_workqueue(fs_info->scrub_wr_completion_workers);
btrfs_destroy_workqueue(fs_info->scrub_parity_workers);
+ mutex_lock(&fs_info->scrub_lock);
}
WARN_ON(fs_info->scrub_workers_refcnt < 0);
}
--
1.8.3.1
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH v2 3/3] btrfs: scrub: add scrub_lock lockdep check in scrub_workers_get
2018-11-29 9:25 [PATCH v2 0/3] btrfs: scrub: fix scrub_lock Anand Jain
2018-11-29 9:25 ` [PATCH v2 1/3] btrfs: scrub: maintain the unlock order in scrub thread Anand Jain
2018-11-29 9:25 ` [PATCH v2 2/3] btrfs: scrub: fix circular locking dependency warning Anand Jain
@ 2018-11-29 9:25 ` Anand Jain
2 siblings, 0 replies; 6+ messages in thread
From: Anand Jain @ 2018-11-29 9:25 UTC (permalink / raw)
To: linux-btrfs
scrub_workers_refcnt is protected by scrub_lock, add lockdep_assert_held()
function in scrub_workers_get().
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Suggested-by: Nikolay Borisov <nborisov@suse.com>
---
v2: born
fs/btrfs/scrub.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index 9ade0659f017..84ef1f0d371e 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -3726,6 +3726,8 @@ static noinline_for_stack int scrub_workers_get(struct btrfs_fs_info *fs_info,
unsigned int flags = WQ_FREEZABLE | WQ_UNBOUND;
int max_active = fs_info->thread_pool_size;
+ lockdep_assert_held(&fs_info->scrub_lock);
+
if (fs_info->scrub_workers_refcnt == 0) {
fs_info->scrub_workers = btrfs_alloc_workqueue(fs_info, "scrub",
flags, is_dev_replace ? 1 : max_active, 4);
--
1.8.3.1
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH v2 1/3] btrfs: scrub: maintain the unlock order in scrub thread
2018-11-29 9:25 ` [PATCH v2 1/3] btrfs: scrub: maintain the unlock order in scrub thread Anand Jain
@ 2018-11-29 10:36 ` Filipe Manana
2018-11-30 1:00 ` Anand Jain
0 siblings, 1 reply; 6+ messages in thread
From: Filipe Manana @ 2018-11-29 10:36 UTC (permalink / raw)
To: Anand Jain; +Cc: linux-btrfs
On Thu, Nov 29, 2018 at 9:27 AM Anand Jain <anand.jain@oracle.com> wrote:
>
> The device_list_mutex and scrub_lock creates a nested locks in
> btrfs_scrub_dev().
>
> During lock the order is device_list_mutex and then scrub_lock, and during
> unlock, the order is device_list_mutex and then scrub_lock.
> Fix this to the lock order of scrub_lock and then device_list_mutex.
>
> Signed-off-by: Anand Jain <anand.jain@oracle.com>
> ---
> v1->v2: change the order of lock acquire first scrub_lock and then
> device_list_mutex, which matches with the order of unlock.
> The extra line which are now in the scrub_lock are ok to be
> under the scrub_lock.
I don't get it.
What problem does this patch fixes?
Doesn't seem any functional fix to me, nor performance gain (by the
contrary, the scrub_lock is now held for a longer time than needed),
nor makes anything more readable or "beautiful".
> fs/btrfs/scrub.c | 13 +++++++------
> 1 file changed, 7 insertions(+), 6 deletions(-)
>
> diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
> index 902819d3cf41..a9d6fc3b01d4 100644
> --- a/fs/btrfs/scrub.c
> +++ b/fs/btrfs/scrub.c
> @@ -3813,28 +3813,29 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
> return -EINVAL;
> }
>
> -
> + mutex_lock(&fs_info->scrub_lock);
> mutex_lock(&fs_info->fs_devices->device_list_mutex);
> dev = btrfs_find_device(fs_info, devid, NULL, NULL);
> if (!dev || (test_bit(BTRFS_DEV_STATE_MISSING, &dev->dev_state) &&
> !is_dev_replace)) {
> mutex_unlock(&fs_info->fs_devices->device_list_mutex);
> + mutex_unlock(&fs_info->scrub_lock);
> return -ENODEV;
> }
>
> if (!is_dev_replace && !readonly &&
> !test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state)) {
> mutex_unlock(&fs_info->fs_devices->device_list_mutex);
> + mutex_unlock(&fs_info->scrub_lock);
> btrfs_err_in_rcu(fs_info, "scrub: device %s is not writable",
> rcu_str_deref(dev->name));
> return -EROFS;
> }
>
> - mutex_lock(&fs_info->scrub_lock);
> if (!test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &dev->dev_state) ||
> test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &dev->dev_state)) {
> - mutex_unlock(&fs_info->scrub_lock);
> mutex_unlock(&fs_info->fs_devices->device_list_mutex);
> + mutex_unlock(&fs_info->scrub_lock);
> return -EIO;
> }
>
> @@ -3843,23 +3844,23 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
> (!is_dev_replace &&
> btrfs_dev_replace_is_ongoing(&fs_info->dev_replace))) {
> btrfs_dev_replace_read_unlock(&fs_info->dev_replace);
> - mutex_unlock(&fs_info->scrub_lock);
> mutex_unlock(&fs_info->fs_devices->device_list_mutex);
> + mutex_unlock(&fs_info->scrub_lock);
> return -EINPROGRESS;
> }
> btrfs_dev_replace_read_unlock(&fs_info->dev_replace);
>
> ret = scrub_workers_get(fs_info, is_dev_replace);
> if (ret) {
> - mutex_unlock(&fs_info->scrub_lock);
> mutex_unlock(&fs_info->fs_devices->device_list_mutex);
> + mutex_unlock(&fs_info->scrub_lock);
> return ret;
> }
>
> sctx = scrub_setup_ctx(dev, is_dev_replace);
> if (IS_ERR(sctx)) {
> - mutex_unlock(&fs_info->scrub_lock);
> mutex_unlock(&fs_info->fs_devices->device_list_mutex);
> + mutex_unlock(&fs_info->scrub_lock);
> scrub_workers_put(fs_info);
> return PTR_ERR(sctx);
> }
> --
> 1.8.3.1
--
Filipe David Manana,
“Whether you think you can, or you think you can't — you're right.”
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v2 1/3] btrfs: scrub: maintain the unlock order in scrub thread
2018-11-29 10:36 ` Filipe Manana
@ 2018-11-30 1:00 ` Anand Jain
0 siblings, 0 replies; 6+ messages in thread
From: Anand Jain @ 2018-11-30 1:00 UTC (permalink / raw)
To: fdmanana; +Cc: linux-btrfs
On 11/29/2018 06:36 PM, Filipe Manana wrote:
> On Thu, Nov 29, 2018 at 9:27 AM Anand Jain <anand.jain@oracle.com> wrote:
>>
>> The device_list_mutex and scrub_lock creates a nested locks in
>> btrfs_scrub_dev().
>>
>> During lock the order is device_list_mutex and then scrub_lock, and during
>> unlock, the order is device_list_mutex and then scrub_lock.
>> Fix this to the lock order of scrub_lock and then device_list_mutex.
>>
>> Signed-off-by: Anand Jain <anand.jain@oracle.com>
>> ---
>> v1->v2: change the order of lock acquire first scrub_lock and then
>> device_list_mutex, which matches with the order of unlock.
>> The extra line which are now in the scrub_lock are ok to be
>> under the scrub_lock.
>
> I don't get it.
> What problem does this patch fixes?
> Doesn't seem any functional fix to me, nor performance gain (by the
> contrary, the scrub_lock is now held for a longer time than needed),
> nor makes anything more readable or "beautiful".
btrfs_scrub_dev() isn't following the lock and unlock FILO order.
Such as lock-a lock-b .. unlock-b unlock-a. So this patch is
trying to fix it.
This patch fixes the order but I think you mean to say as
__scrub_blocked_if_needed() calls unlock scrub_lock. oops my
bad this patch is wrong.
Scrub concurrency needs overhaul including the dependency on the
user land btrfs-progs, which I was trying to avoid. but looks like
its better to fix that as well. As of now I am NACK this patch.
Thanks, Anand
>> fs/btrfs/scrub.c | 13 +++++++------
>> 1 file changed, 7 insertions(+), 6 deletions(-)
>>
>> diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
>> index 902819d3cf41..a9d6fc3b01d4 100644
>> --- a/fs/btrfs/scrub.c
>> +++ b/fs/btrfs/scrub.c
>> @@ -3813,28 +3813,29 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
>> return -EINVAL;
>> }
>>
>> -
>> + mutex_lock(&fs_info->scrub_lock);
>> mutex_lock(&fs_info->fs_devices->device_list_mutex);
>> dev = btrfs_find_device(fs_info, devid, NULL, NULL);
>> if (!dev || (test_bit(BTRFS_DEV_STATE_MISSING, &dev->dev_state) &&
>> !is_dev_replace)) {
>> mutex_unlock(&fs_info->fs_devices->device_list_mutex);
>> + mutex_unlock(&fs_info->scrub_lock);
>> return -ENODEV;
>> }
>>
>> if (!is_dev_replace && !readonly &&
>> !test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state)) {
>> mutex_unlock(&fs_info->fs_devices->device_list_mutex);
>> + mutex_unlock(&fs_info->scrub_lock);
>> btrfs_err_in_rcu(fs_info, "scrub: device %s is not writable",
>> rcu_str_deref(dev->name));
>> return -EROFS;
>> }
>>
>> - mutex_lock(&fs_info->scrub_lock);
>> if (!test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &dev->dev_state) ||
>> test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &dev->dev_state)) {
>> - mutex_unlock(&fs_info->scrub_lock);
>> mutex_unlock(&fs_info->fs_devices->device_list_mutex);
>> + mutex_unlock(&fs_info->scrub_lock);
>> return -EIO;
>> }
>>
>> @@ -3843,23 +3844,23 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
>> (!is_dev_replace &&
>> btrfs_dev_replace_is_ongoing(&fs_info->dev_replace))) {
>> btrfs_dev_replace_read_unlock(&fs_info->dev_replace);
>> - mutex_unlock(&fs_info->scrub_lock);
>> mutex_unlock(&fs_info->fs_devices->device_list_mutex);
>> + mutex_unlock(&fs_info->scrub_lock);
>> return -EINPROGRESS;
>> }
>> btrfs_dev_replace_read_unlock(&fs_info->dev_replace);
>>
>> ret = scrub_workers_get(fs_info, is_dev_replace);
>> if (ret) {
>> - mutex_unlock(&fs_info->scrub_lock);
>> mutex_unlock(&fs_info->fs_devices->device_list_mutex);
>> + mutex_unlock(&fs_info->scrub_lock);
>> return ret;
>> }
>>
>> sctx = scrub_setup_ctx(dev, is_dev_replace);
>> if (IS_ERR(sctx)) {
>> - mutex_unlock(&fs_info->scrub_lock);
>> mutex_unlock(&fs_info->fs_devices->device_list_mutex);
>> + mutex_unlock(&fs_info->scrub_lock);
>> scrub_workers_put(fs_info);
>> return PTR_ERR(sctx);
>> }
>> --
>> 1.8.3.1
>
>
>
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2018-11-30 1:00 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-11-29 9:25 [PATCH v2 0/3] btrfs: scrub: fix scrub_lock Anand Jain
2018-11-29 9:25 ` [PATCH v2 1/3] btrfs: scrub: maintain the unlock order in scrub thread Anand Jain
2018-11-29 10:36 ` Filipe Manana
2018-11-30 1:00 ` Anand Jain
2018-11-29 9:25 ` [PATCH v2 2/3] btrfs: scrub: fix circular locking dependency warning Anand Jain
2018-11-29 9:25 ` [PATCH v2 3/3] btrfs: scrub: add scrub_lock lockdep check in scrub_workers_get Anand Jain
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).